PROVENANCE AND TECTONIC
INFERENCES CONCERNING THE
KEWEENAWAN INTERFLOW SEDIMENTS’
OF THE LAKE SUPERIOR REGION

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
GEORGE P. MERK
19 72

 

 

III III III I II I II IIIIIIIII

3 1293 10122

 

This is to certify that the
thesis entitled
PROVENANCE AND TECTONIC INFERENCES CONCERNING
THE KEWEENAWAN INTERFLOW SEDIMEN‘I‘S OF
THE LAKE SUPERIOR REGION

presented by

George P. Mark

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Geology

W

Major professor I

Datew

0-7639

  
   
    

   

   

I BINDING or v I
‘. IIUAB 8: WW \
« III 300K BINDERY IIIlI. Il '

I: . '

p. fl. usumr muons
L were!!!“

 

 

 

PP

 

 

 

The Kt
Significan‘
EXtent of
any immat
Marne th
of a majOr
accmulati
relatiOnSE
ho:

derlar1C

In on

ABSTRACT
PROVENANCE AND TECTONIC INFERENCES CONCERNING

THE KEWEENAWAN INTERFLOW SEDIMENTS OF
THE LAKE SUPERIOR REGION

BY
George P. Merk

The Keweenawan geology of the Lake Superior basin is
significant because of the great thickness and lateral
extent of the extrusive rocks it contains, the composition-
ally immature red beds which are intercalated with and
overlie the flows,and the fact that this basin forms a part
of a major tectonic feature of North America. The Keweenawan
accumulation also invites speculation as to the structural
relationship between the volcanic pile and the surrounding
borderland.

In order to determine the structural and physiographic
background of Keweenawan events in the Lake Superiorbasin
~and the relationship between the volcanic accumulation and
the surrounding borderland during the Keweenawan time, this
.study concentrated upon the sediments intercalated with the
flows as that portion of the sequence uniquely capable of

carrying such information.

 

Repre
collected ‘
Keweenawan
and the Eli:
of this re
indicates :
from local
volcanics.
caoics was
from prese:
Keweenawan

tectonic in

 

 

local areas
record;
Althou
diSPIOVen,
:egmn dim

F + w

George P. Merk

Representative interflow sedimentary rock samples were
collected from the Keweenawan sections exposed on the
Keweenawan Peninsula, Michigan; Mamainse Point, Ontario;
and the Minnesota shore region of Lake Superior. The results
of this petrographic study coupled with the work of others
indicates that much of the Keweenawan sediments were derived
from local tectonic highs which were mantled by Keweenawan
volcanics. Therefore, the area covered by Keweenawan vol-
canics was probably much more extensive than that indicated
from present outcrops. The general tectonic pattern for the
Keweenawan is then one of local positive and negative
tectonic instability coupled with volcanism. At least three
local areas of uplifts can be discerned from the sedimentary
record.

Although a rift valley hypothesis cannot be totally
disproven, inferred structural relationships in the outcrop
region during at least lower and middle Keweenawan time do

not support such a conclusion.

 

 

PRC

 

PROVENANCE AND TECTONIC INFERENCES CONCERNING
THE KEWEENAWAN INTERFLOW SEDIMENTS OF

THE LAKE SUPERIOR REGION

BY

George P. Merk

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Geology

1972

 

 

Pores:
are the ma:-
of Keweer.a'.-
been drawn
this study.

The at,
llooaas Vogei
critically
suggestions
siderations'
3‘~‘Pai:tir.e:at
Emanuel Hac

Suppor
National Sc
Penrose Becl
Of America;
The Calumet
Company gra
I than-k the

 

ACKNOWLEDGMENTS

Foremost among those to whom the author is indebted
are the many geologists who have preceded him in the study
of Keweenawan geology. Their published works, which have
been drawn upon freely, make up the references cited in

this study.

The author is also indebted to Drs. Robert Ehrlich,
Thomas Vogel, Bennett Sandefur and Harold Stonehouse for
critically reading this manuscript and offering constructive
suggestions. I am also glad to acknowledge the many con—
siderations rendered by my colleagues in the Natural Science
Department at Michigan State University, particularly Drs.
Emanuel Hackel, Richard Seltin, and Raymond Hollensen.

Support for field and lab expenses was provided by a
National Science Foundation Science Faculty Fellowship, a
Penrose Bequest Research Grant from the Geological Society
of America, and a Michigan State University Research Grant.
The Calumet and Hecla Corporation and the White Pine Copper
Company graciously made drill cores available for sampling.
I thank these agencies for their support.

 

ii

 

LIST
LI I

GEKHR

"on
\-£“.

. I!

. In
“A

 

GEN“.

II.

TABLE OF CONTENTS

CHAPTER

LIST OF TABLES . . . . . . . . . . . .

LIST OF FIGURES. . . . . . . . . . . .

 

I O IN'I‘RODUCT ION I 0 O O 0 O O O O O O O 0

Objectives of This Study. . . . . .
Methods Used in This Study. . . . .

II. GENERAL GEOLOGY OF KEWEENAWAN SEQUENCE
LAKE SUPERIOR REGION. . . . . . . .

Lower Keweenawan Rocks. . . . . . .
Middle Keweenawan Rocks . . . . . .
Upper Keweenawan Rocks. . . . . . .
Keweenawan Intrusive Rocks. . . . .

III. THE KEWEENAWAN GEOLOGY OF THE EAST SHORE

REGION OF LAKE SUPERIOR . . . . . .

The Alona Bay Sequence. . . . . . .
The Gargantua Cape Sequence . . . .
The.Michipicoten Island Sequence. .
The.Mamainse Point Volcanic Series.
Interflow Sediments. . . . . . .
Conglomerates. . . . . . . . . .
Sandstones and Sandy Phases. . .
Vertical Compositional Variation
Sedimentary Synapsis. . . . . . . .
Provenance . . . . . . . . ‘
Paleogeoqraphy and Sedimentation

IV. THE SOUTH SHORE REGION .'. . . . . . .

The South Range Lava Series . . . .
The Portage Lake Lava Series. . . .

iii

Page

vii

10
10

12

16
17
18
19
21
21
25
.29
37
37
39

41

43
44

 

 

TABLE OF CC

I Chem

(/7

V- THE

}

 

 

 

 

 

 

TABLE OF CONTENTS—-Continued

CHAPTER

Tne Portage Lake Interflow Sediments. .

Conglomerate. . . . . . . . . . .
Sandstone and Sandy Phases. . . .

Vertical Compositional variation in

Calumet Area Section . . . . .
Lateral variation in Composition.
The COpper Harbor Conglomerate . . .
The Conglomerates . . . . . . . .
Sandstone . . . . . . . . . . . .
Vertical compositional variation.
Lateral compositional variation .
Sedimentary SynOpsis _.° . . . . . .
Paleogeography and Sedimentation.

the

V. THE NORTHWEST SHORE REGION OF LAKE SUPERIOR .

The Isle Royale Sequence . . . . . .
The Isle Royale Volcanic Series .
The Isle Royale Conglomerate. . .

The North Shore Volcanic Series. . .
Interflow Sediments . . . . . . .
Vertical Compositional Variation.
Lateral Variation in Composition.

Sedimentary Synopsis . . . . . . . .
Provenance. . . . . . . . . . . .
»Paleogeography and‘SedimentatiOn.

VI. SUMMARY AND CONCLUSIONS . . . . . . .

REFERENCES CITED . . . . . . . . . . . . . . .

iv

Page

45
47

48

52
54
58
59
61
63
66
77
81

83

85
86
86
88
90
96
104
106
106
108

111
119

 

 

SOLE

L Corre.

ll.

12,

. Lithol

Superf

gravel

 

.Cmpa

mafic
measui

. The cc

stitut
Sedime
analyg

- Vertii

Masai:

' Kewes

Penin

. Mean

Porta

-Verti

LIST OF TABLES

TABLE

1.

2.

10.

11.

12.

Correlation chart for the Keweenawan of Lake
super ior O O O O O O O O O O O O O O O O O O O O

Lithology of the Mamainse Point interflow
gravels coarser than 19 mm. . . . . . . . . . .

Comparison between the size (long axis) of
mafic and plutonic clasts at Mamainse Point as
measured in outcrop . . . . . . . . . . . . . .

The composition of the detrital sand-size con-
stituents within the Mamainse Point interflow
sediments, as represented by the mean and modal
analyses of 64 thin sections. . . . . . . . . .

Vertical variation within the sandy phases of
Mamainse Point interflow sediments. . . . . . .

Keweenawan stratigraphic column on Keweenaw
Peninsula . . . . . . . . . . . . . . . . . . .

Mean composition of the sandy phases within the
Portage Lake interflow sediments. . . . . . . .

Vertical compositional variation within the
Portage Lake interflow sediments at Calumet,
Michigan. . . . . . . . . . . . . . . . . . . .

Lateral compositional variation within the
Allouez Conglomerate. . . . . . . . . . . . . .

Lateral compositional variation within the
St. Louis Conglomerate. . . . . . . . . . . . .

Mean composition of the sandy phases within the
COpper Harbor Conglomerate. . . . . . . . . . .

Vertical compositional variation within the
sandy phases of the Copper Harbor Conglomerate
at Calumet. . . . . . . . . . . . . . . . . . .

Page

23

24

26

30

46

49

53

55

56

62

64

 

 

LIST or 1‘}.

l4.

f—o
L]!

16.

17.

18.

19.

Verti’
sandy
at C0;

Vertic
sandy
at P0:

Latera
phases

hean
Series

 

 

Vertic
southw
interf

Vertic
Wlthi:
V01car

LIST OF TABLES--Continued

TABLE

13.

14.

15.

16.

17.

18.

19.

Vertical compositional variation within the
sandy phases of the COpper Harbor Conglomerate
at COpper Harbor . . . . . . . . . . . . . . . .

Vertical compositional variation within the
sandy phases of the Copper Harbor Conglomerate
at Porcupine Mountain, Michigan. . . . . . . . .

Lateral compositional variation within the sandy
phases of the COpper Harbor Conglomerate . . . .

Mean composition of the North Shore Volcanic
Series interflow sediments . . . . . . . . . . .

Vertical compositional Variation within the
southwest limb of the North Shore Volcanic
interflow sediments. . . . . . . . . . . . . . .

Vertical compositional variation of constituents
within the northeast limb of the North Shore
Volcanic interflow sediments . . . . . . . . . .

Lateral compositional variation within the North
Shore Volcanic Series interflow sediments. . . .

vi

Page

65

71

72

94

98

101

105

 

1. Gene;
I891.

 

. Geol:
Supe:

N

3. Geolc

Bay a
4
1.”!

. Grap}
with:

 

 

 

 

 

5. Grapl
with.

6- Geol.
Lake

7. @901
adja
Stud

Witt
Cal;

. era;
Wit]
Cali

10, Lat.

3am

ii. Geo

of

12‘ LQC
SEr

13. Gre

LIST OF F IGURES

FIGURE Page
1. General geological map of the Lake Superior
region. . . . . . . . . . . . . . . . . . . . . 7
2. Geological map of the east shore region of Lake
superior 0 O 0 O O O D O O I O O O O O O I O O O 13
3. Geological map of the Mamainse Point and Alona
Bay areas . . . . . . . . . . . . . . . . . . . 27
4. Graphs of the vertical compositional variation
within the Mamainse Point interflow sediments . 34
5. Graphs of the vertical compositional variation
within the Mamainse Point interflow sediments . 36

10.

11.

.12.

13.

Geological map of the south shore region of
Lake Superior, Upper Peninsula of Michigan. . . 41

Geological map of the Keweenaw Peninsula and
adjacent area, showing sites sampled for this
study . . . . . . . . . . . . . . . . . . . . . 50

Graphs of the vertical compositional variation
within the Portage Lake interflow sediments at
Calumet, Michigan . . . . . . . . . . . . . . . 68

Graphs of the vertical compositional variation
within the Portage Lake interflow sediments at
Calumet, Michigan . . . . . . . . . . . . . . . 7O

Lateral compositional variation within the
sandy phases of the COpper Harbor Conglomerate. 74

Geological map of the northwest shore region
of Lake Superior. . . . . . . . . . . . . . . . 84

Location map showing the North Shore Volcanic
Series and the sites sampled. . . . . . . . . . 92

Graphs of the vertical compositional variation

within the southwest limb of the North Shore
Volcanic Series interflow sediments . . . . . . 100

vii

 

 

 

I
I
1

LIST OF E:

FIGU
14. Gra;
with
V015

15. Pale
Kewe
posi

 

 

 

 

 

LIST OF FIGURES——Continued

FIGURE Page

14. Graphs of the vertical compositional variation
within the northeast limb of the North Shore
Volcanic Series interflow sediments . . . . . . 103

15. Paleogeography (If Lake Superior Syncline in

Keweenawan time showing location of postulated
positive and negative tectonic areas. . . . . . 117

viii

 

 

 

 

The Ki.
significars
extent of t
ally immat;

overlie the

 

0f a major,
Shows many
rift zones.
the Center
Plateau has
Other Such

COT-tine“

1‘.

The Kc

3f the CUr:

:elation Sh;
3f '9

articu;

the Keween.

4“

borderland

EVen 1

CHAPTER I

INTRODUCTION

The Keweenawan geology of the Lake Superior basin is
significant because of the great thickness and lateral
extent of the extrusive rocks it contains, the composition—
ally immature red beds which are intercalated with and
overlie the flows, and the fact that this basin forms a part
of a major, linear, tectonic feature of North America, which
shows many characteristics similar to those of continental
rift zones. The Keweenawan basalt accumulation is now in
the center of the continent and may be the oldest such
plateau basalt known in the geologic record. It differs from
other such accumulations which tend to be marginal to the
continent.

The Keweenawan accumulation is of much interest because
of the current attention to continental structure and the
relationship between the crust and the underlying mantle,

Of particular interest is the structurel relationship between
the Keweenawan volcanic accumulation and the surrounding
borderland.

Even though the structural relationship between the

basalt and the borderland is of great interest, most studies

I]

 

’.
r7
'.
u. 11. a."

 

bow-

of the Ke'w
petrology.
flmse stud
rocks and
ofthe par
this seque
land is th
Some
been perfo
(1929); 5a
Niel (194
Bite (1968
no attempt
°na unifie
This i
examinatior
From these
ilnamic rel

III in? I.

 

of the Keweenawan have concentrated on the stratigraphy,
petrology, and the geochemistry of the volcanic rocks.

These studies serve to shed light on a time scale for these
rocks and present evidence on the origin and differentiation
of the parent silicate melt. However, the only portion of
this sequence which carries evidence of a contempory border—
land is the sediments intercalated with the flows.

Some examination of the interflow sediments has already
been performed; i.e., Lane (1911); Butler, Burbank et a1.
(1929); Sandberg (1938); Grogan (1940) Tyler, Marsden, Grout,
Thiel (1940); Thomson (1955); White (1957, 1960, 1967);

Hite (1968); and Hubbard (1968). However, to my knowledge,
no attempt has been made to examine the interflow sediments
on a unified basis all around the Lake Superior basin.

This report describes the result of such a detailed
examination and is integrated with studies already performed.
From these results, an attempt to deduce the passive and
dynamic relationship between the volcanic rocks and the

bordering uplands and basins has been made.

Objectives of This Study

A comprehensive study was made of the Keweenawan inter—
flow sediments, especially the sandy phases, in each of the
three major Keweenawan outcrop areas around Lake Superior.
The areas studied were at Mamainse Point in Ontario, the

Keweenaw Peninsula of Michigan, and the North—west Shore

 

 

 

SC

V8

gr

Re

90‘
be

The -
ltcrops V
Ithl'eQ Re,“
{dildotr‘ly :
II the r;
SIIh Samp
IreIKeWee
Sitioflal
interfloW

55‘s.

*sled f

area of Minnesota and Ontario. The objectives of this study
were to:

1. determine the degree of contribution of local
sources to the sediment;

2. determine the degree of compositional variation
vertically and laterally within each area;

3. determine the structural and physiographic back-
ground of events in each of the three major
Keweenawan outcrOp areas, and

4. if possible, build models showing the range in
potential stratigraphic and tectonic relationships
between these three now isolated major Keweenawan

outcrop areas.

Methods Used in This Study

The purpose of this study was to compare Keweenawan
outcrops within each area and then compare and contrast the
three Keweenawan outcrOp areas. Outcrops and cores were
randomly sampled to obtain representative sand samples show-
ing the range of variability present. Approximately 800
such samples were collected to include (1) possible clasts of
pro-Keweenawan source rocks and (2) the textural and compo-
sitional range present in the interflow sediments. The
interflow sediments present on the Keweenawan Peninsula were

sampled from both core and outcrop collections, whereas the

 

 

samples C’

 

Point ser:

The c!
binocular
thin sect;
Bailey an:
recogniza:
stone were
technique.
counted if“
QEIIOUS str
ates was d
zones with

per Site w

 

 

 

or thin Se

samples collected from the North Shore Volcanic and Mamainse
Point series were taken solely from.outcr0ps.

The collected samples were studied in hand specimen by
binocular microscope and over 252 were also examined in
thin section. The thin sections were stained according to
Bailey and Stevens' (1960) technique to facilitate feldspar
recognization and discrimination. The modes of the sand-
stone were determined in thin section by point—counter
technique. At least two hundred grains per thin section were
counted in traverses parallel to the bedding within homo—
genous structural units. The composition of the conglomer—
ates was determined at the outcrOp by point counting selected
zones within conglomeratic units. At least 100 point counts
per site were made. The identities of certain problematical
pebbles were determined by subsequent binocular microscope

or thin section study.

 

 

 

truc
cow occupi

ZXPOSUIES

targins of

 

up to vert

CHAPTER II

GENERAL GEOLOGY OF KEWEENAWAN SEQUENCE
OF LAKE SUPERIOR REGION

Structurally, the Keweenawan rocks form a large syncline
now occupied by most of Lake Superior itself (Figure l).
Exposures of Keweenawan rocks are primarily confined to the
margins of the lake, where dips range from only a few degrees

up to vertical.

Lower Keweenawan Rocks

Keweenawan rocks in the Lake Superior area may be sub-
divided into three lithologic divisions (Table 1). Sedimen-
tary rocks lying between the Animikean and the earliest
Keweenawan lava flows are referred to as Lower Keweenawan
in age. Their time equivalence around Lake Superior has not
necessarily been established. The lower Keweenawan attains
its maximum develOpment on the Sibley Peninsula where it
consists of sandstones, marls, and shales. The sequence
thins southwestward to form only a thin conglomeratic sand-
stone near Duluth, called the Puckwunge formation (Table 1).
Other sediments thought to be lower Keweenawan are the

Bessemer formation east of Mellen in Wisconsin and.Michigan,

 

00‘ liiililill 'lfl‘: ”IVER. {L

talk-310 I580.- IO 2<§ilhl§il Riki: K N

I‘...!‘0N“ ‘5‘.-
!(.t:d.l Oh lulllltu tilt: . 0.0NOIdQI H

 

DUI.

.muomoccHZ mo mowuom

muwsmoao> whoam suuoz one Ame cam .cmmflnoflz mo madmcflsom

3mcoo3mM one ANV .owHMDGO mo mono unflom omcMmEmz one Adv

"m3oHHoy no ooHonEss ohm coaummflumo>cfl mesa MOM Goddamn mmonm

oza .xoou o>flmshuca 6cm .humucoaflvom .Uflcmuao> Gm3mso03oM

6 Mo mououso mo mmoum same may mcfl3onm Amoma .mHHmm Houmm
woMMfiUOEV coflmou Moduomsm oxmq mg» no mmE HMOflmoaoom Houocoo .H ousmflm

 

"flu!
and! o. o o. m._<oo

2.045 80.xmaaa wx(._ no ¢.x(

h0<h800 0.0040u0

2(.¢II<0W¢¢ cm’OJ

2(.¢fll(0fl¢1 UJOO.I

 

 

§ 0

.. z<o=._o_2 1. .m
\v a D a 0

.\\\§ 0 00...,“

u “(I
NN a. a

O x? I 2:32:24 4w”:

. / goes:
2:9. I
32.3.3. _. I
a
on; ”fun“...
9:3 1.2.4050 III .\
0 22..»
2:002:82.
m4<>oc use.
$2.830 0.4%
a. O_m<._.zo 522:8»
a
202...: 3.3
no

-50

H ousmfim

 

 

 

 

 

 

 

 

 

[III
lllllll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

OU>.C:¢h-= 8(‘(8UH’UX (03‘.
8¢’¢80u’U8 CU’OJ
8(3<8UU3W8 Hana—I

8(’¢80&’U8 80“:

8<.CII<0 IN’OJ IO 8(‘(8uu’08 Inc...-
8<.¢II<0U¢& Hui.—
Z<.¢=J.I Oh 8(.¢.I<0 38“: . 0.0NOIJ‘;

 

 

 

 

 

 

 

 

 
 

..ono~

.CIOJ... uOqu V0.u«nv0::

ucendaam Izeq no cnlezso)tx 0:. nt\ Ceca; ::_.e~csutb

.~ 67:3».

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

”a l w J a
. _ “ .... .e
_ 1 33m 3on Hoar". a..— 26: 26m 3365. 0.3332 Saunrflmumfl
q<xxoz w . ell-aanuu: . a .
1%,
a 1 033.390 L um oozing M
_ . Om ul'IIOQ BONN“ :«HIW M I
{ rflflm ."‘H m u
_ . m ounuuom M... g
1 I m 93.5 “W
M _ . 1. J .2,- 1
335 u u . no no. a». :93 :«uom U .m. .— m
1 . a: u A i I 9.320.. r.
. . oust..— Ausou . hangar—3:0: aid-o w... w
1 o No
1 u a.
_ a»: A m
_ .9.“ 03325: t a. n m
n . acousuou sous-sou m. «w
. I} meson-05. monounoqao a n
t.-- r .. i u :33: 552.832: I m m Wanna:
. swam
.23.— . _ w d sun-5:025:
qaxxoz aqueouauauu: , nous-n osna . .osuo. as.” m gusset-nous
iiidumduaau
. Nona ”ECU .068 vs /(1
cacao-:3. announce.
. 502.3!
nm um noon» .no ouaouo on one so each
an sagas-nouns .m odaasonooan .no ouoauxan .m sudxunuu
massages cans-H oauauao canines: caucoo-u: use avenues gunman saga» woo-uncut . . ensues
emphases couooumaAqu .ucmnm .4 chug-nu some: anonuuoz Iceman“: .
manoeuQa pi

 

 

 

 

..oho~ .auouo “some coauavoa. uosuansm ages «0 auxin-onus oau uou unusu couuaaouuoo .~ «Anna

 

the Barrc

 

formation

Thes
lava flow
netic pol
thought t
sequences
Series or.
Puclmmge

Series at

 

North Sho
limb of ti
between t:
IiChigan

in wiSCOn
hOWeverl

pre‘Kewee

Bay: and

the Barron quartzite in northern Wisconsin and the Sioux
formation in southwestern Minnesota.

These sediments are conformably overlain locally by
lava flow sequences whose age, based upon a reversed mag—
netic polarization for both the sediments and volcanics, is
thought to be lower Keweenawan. Examples of such lava
sequences overlying sedimentary rocks are the Sibley—Osler
Series on the north limb of the Lake Superior syncline, the
Puckwunge-Grand Portage Series and the Puckwunge-Nopeming
Series at the north and south end, respectively, of the
North Shore Volcanic Group in.Minnesota. And on the south
limb of the Lake Superior basin, a similar situation occurs
between the Bessemer Sandstone-South Range Lava Series in!
Michigan and the Bessemer Sandstone4flellen and Hurley Lavas
in Wisconsin (Table 1). On the east limb of the syncline,
however, lower Keweenawan lava flows rest unconformably on
pre—Keweenawan crystalline rocks at Mamainse Point, Alona

Bay, and Cape Gargantua.

The Middle Keweenawan Rocks

The middle Keweenawan is made up of a thick succession
of mafic lavas, with lesser amounts of felsic rocks and
interflow sediments. Rhyolite occurs as both flows and small
intrusive bodies. Prominent exposures of middle Keweenawan

rocks are found along the Minnesota shore of Lake Superior,

 

il -_-___....-._..__ _.%

 

in east c
Northern
In additi

Island ar

The
0f sandst
shales.
the Kewee

1518 ROYa

Superior .

 

 

10

in east central Minnesota, on the Keweenaw Peninsula, in
Northern Wisconsin, and at Mamainse Point in Ontario.
In addition, middle Keweenawan rocks form Michipicoten

Island and most of Isle Royale.

The Upper Keweenawan Rocks

The upper Keweenawan sequence consists predominantly
of sandstones and conglomerates overlain by sandstones and
shales. Upper Keweenawan rocks are extensively exposed on
the Keweenawan Peninsula, in Northern Wisconsin, and on
Isle Royale, and possibly along the east shore of Lake

Superior.

Keweenawan Intrusive Rocks

 

Keweenawan intrusives are widespread and abundant in
the areas marginal to Lake Superior. The location of these
intrusives bodies marginal to the Lake Superior syncline, the
exposed Keweenawan lava flows, and the interflow sediments
is significant. These intrusives represent not only potential
feeder systems for the flows, but could also indicate the site
of borderlands which shed detritus in times of volcanic
quiescence.

The largest intrusive in the Lake Superior area is the
Duluth gabbro complex which intrudes the lower and middle
Keweenawan rocks in Minnesota. Other mafic sills, dikes, and

irregular bodies known collectively as the Logan intrusives

 

also int:
the nort'r.
Another C1
the Kewee

Rhyolete

 

portion 0
is Mount II
shyre an:
on the nol
the basisl
mostly th
dant in a
especiall
pOint. th

the South

 

11

also intrude middle Precambrian and Keweenawan rocks along
the northwest shore of the lake from Duluth to Schreiber.
Another complex of gabbro and granOphyre rocks intrudes

the Keweenawan volcanics in the.Mellen area of Wisconsin.
Rhyolete intrusive bodies are present throughout northern
portion of the Keweenawan Peninsula. The largest of these

is Mount Bohemia, an intrusive complex consisting of grano-
phyre and syenodiorite. A large syenite complex at Coldwell
on the north shore is also thought to be Keweenawan in age on
the basis of radiometric studies. Diabase dike swarms,
mostly thought to be lower Keweenawan in age, are also abun-
dant in areas marginal to the lake. These swarms are
especially abundant along the northwest shore near Pigeon
Point, the east shore from Schreiber to Mamainse Point; along
the south shore in the Huron Mountains, near Lake Gogebic,

near Marquette; and also along the north shore of Lake Huron.

 

 

Seve
age, occu
Recent ra
isolated
metrical]
rocks e1:
also ind
KWEenaw
COIrelat

Alt
O"‘télrio
Lower Pr
'a’.0:1(;0riG

VOICani,

CHAPTER III

THE KEWEENAWAN GEOLOGY OF THE EAST SHORE
REGION OF LAKE SUPERIOR

Several isolated volcanic sequences, Keweenawan in
age, occur on the east shore of Lake Superior (Figure 2).
Recent radiometric data indicates that the ages of these
isolated outcrops fall within the same range as other radio-
metrically derived dates for lower and middle Keweenawan
rocks elsewhere around Lake Superior. Paleomagnetic studies
also indicate that magnetic divisions within these isolated
Keweenawan sequences permit the various sequences to be
correlated on the basis of magnetic polarity.

Although the Lake Superior shore between Schreiber,
Ontario and Sault Ste Marie, Ontario is formed largely of
Lower Precambrian rocks, several prominent, isolated, pro-
montories along the east coast are made up of Keweenawan
volcanic flows and sedimentary rocks. The thickest and most
prominent exposure is at Mamainse Point where the succession
is at least 14,000 feet thick (Figure 2). Michipicoten
Island also contains a middle Keweenawan rock series that
is thought to be about 11,000 feet thick. Lesser occurrences
of Keweenawan rocks are found at Cape Gargantua and Alona
Bay; however, the thickness of the series~at these locations

does not exceed 3,000 feet.
12

 

 

7—"-

—/_1./

scuggist

mi

UPPEI

IflDDL
LOWE

 

Figure 2 .

13

 

 

 

 

 

e 32' a?
-/5./ a, , svemrr
- I“ co r rx
sqggm Iowwapz M ‘
65“" ‘3
ISIAND
O’N'IA.RIC)
G G
d
a?
\ §'
woe surmoc o:
suous \ h
45' \I g II NI 8'
§\\ \\\\ 4
4T Micmrgar'sg
L A K E .% a. \\\N\\
I ‘9» \\ ‘\\ G
s u p E RI 0 R mama. \\‘°
CAPE
CAslllgg a,
clui‘ .w
up 0 "‘
II \ I \\\\
L7. 816. ALONA IAY / db §§
\ \
open Pnecuunuu “‘22,,st =1... \\\\\\
UPPER Keweenaw“ on CAIIRMN , '" _§—= '
JAcoaswLLs Moonstone W §"_“"I 47
recon sAuosroue 7 GENE? t If.

UNDIFFERENTIATED

MIDDLE KEWEENAWAN

LOWER KEWEENAWAN
DIAIACE DIKE SWAN“

HIDDLE PRE c AMBRIA N

LOWER PRE CAMBRIAN
«moon: couracr
uoomo DIP AND smote

SCAM

 

l.

 

  

MICHIGAN

  

 

 

3.5

 

Figure 2.

Geological map of the east shore region of Lake
Superior (modified after Halls,

1966) .

14

Generally speaking, the strike of the various series
is parallel to the broad curve-of the shore line and the
axis of the Lake Superior Syncline, with dips inclined
lakeward at angles from 150 to 50°. Burwash (1905) suggested
that these isolated Keweenawan exposures along the lake may
be associated with anticlinal arches marginal to the lake
but plunging lakeward. Structural data on Michipicoten
Island (Burwash, 1905) and Mamainse Point (Thomson, 1953)
appear to lend credence to this idea.

The first detailed geologic map of these Keweenawan
rocks on the eastern shore area of Lake Superior were made
of the Michipicoten Island sequence by Burwash (1905) and
revised more recently by Annells (1970). The first detailed
map of the Mamainse Point succession was made by Moore (1925),
followed by Thomson (1953) and Giblin (1965, 1969). Nuffield
(1955) has described the geology at Alona Bay. The Cape »
Gargantua Keweenawan rocks were described by Ayres (1969).
Recently a detailed stratigraphic-petrographic study relating
the Keweenawan volcanic rocks at Mamainse Point, Alona Bay,
Cape Gargantua, and Michipicoten Island was carried out by
Annells (1970, 1971). PaImer (1970) made paleomagnetic
studies which are useful in correlating these Keweenawan
volcanic exposures.

Van Schmus (1971) states that the best-estimate for the
age of middle Keweenawan igneous activity along the east shore

is 1,070 1 50 m.y. By conducting Rb-Sr analyses on felsites

15

interlayered with volcanic sequences, he assigned the
following over-all ages to these sequences: Mamainse Point
(1,070 i 50 miy.); Cape Gargantua (1.091 r 40 m.y.) and
Michipicoten Island (953 i 50 m.y.).

The over-all age range for these igneous rocks around
the east shore is within the same range of other radiometric
data for middle Keweenawan rocks elsewhere in the Lake
Superior region. Paleomagnetic studies by Du Bois (1962)
and Palmer (1970) suggest a lower Keweenawan age for the
Alona Bay sequence, the lower 1,000 feet of the Cape Gargantup
series, and the lower 3,000 feet of the Mamainse Point succes—
sion on the basis of their reversed polarity. Palmer (1970)
indicates that normal directions of magnetization prevail in
the upper parts of the Mamainse Point, Cape Gargantua, and
Michipicoten Island sections and magnetically these units
should be considered middle Keweenawan in age.

Any reconstruction of Keweenawan volcanic and tectonic
events must account for at least two factors. One factor
concerns the geology of those areas marginal to Lake Superior
and adjacent to the presently exposed Keweenawan outcrops;
it is assumed that these adjacent areas were also blanketed
by Keweenawan lavas at one time, but the flows have subse-
quently been removed by erosion. The second factor concerns
the observation that large scale intrusive complexes and dike
swarms of Keweenawan age occur closer to the margins rather

than to the axis of the basin.

 

 

occur in
of Lake ’
Wawa, an
Keweenaw;
at all 16
Annells,
the Mich:

Diat

the Mont:

 

PhOSEd, s
in region
been remc

The
0f laVa C

16

A number of Keweenawan alkalic intrusives complexes
occur in areas marginal to the northeastern and eastern shore
of Lake Superior. The largest such bodies occur at Coldwell,
Wawa, and Chapleau, Ontario (Kerr, 1910; Parsons, 1961).
Kewhenawan rhyolitic plugs, dikes and intrusive sheets occur
at all levels in the Mamainse Point section (Thomson, 1953;
Annells, 1971). Felsic intrusives have also been reported in
the Michipicoten Island sequence (Annells, 1971).

Diabase dike swarms occur from the Schreiber area to
the Montreal River regions. The dikes are fresh, unmetamar—
phosed, strongly magnetic (HalI, 1966), and presently exposed
in regions where the arched cover of Keweenawan rocks have
been removed by erosion.

The dikes may represent a widespread perepheral system
10f lava conduits which served as feeders for an extensive.

Keweenawan volcanic terrain that has been eroded.

The Alona Bay Sequence

.A series of lava flows of lower Keweenawan age (Du Bois,
1962; Palmer, 1970) is eXposed as a narrow rim around the
southeastern shore of Alona Bay (Figure 2). These mafic
flows unconformably overlie pre-Keweenawan rocks.

Nuffied (1955) notes that no intervlow sediments were

observed intercalated with the volcanic succession.

 

 

I
I
l

conglome:

The

 

pprox in;

Keweenawe

The
lay be cq
separated
section c

studies 1;

 

of sectio
lower Kew
Should be

AYIe
least 5 p
Gargantua
tictic in
Percent b

Pebbles o

17

The Gargantua Cape Sequence

The Keweenawan section in the Gargantua Cape area is
approximately 2,700 feet thick and consists of basalts,
conglomerates and rhyolite unconformally overlying pre-
Keweenawan grantic rocks.

The Keweenawan succession in the Gargantua Cape area
may be conveniently divided into three basalt members
separated by 2 conglomerate members, with the t0p of the
section capped by Rhyolite (Ayres, 1969). Paleomagnetic
studies by Palmer (1970), indicate that the lower 1,000 feet
of section (within the lower basalt member) is prdbably
lower Keweenawan, whereas the upper half of the section
should be considered middle Keweenawan.

Ayres (1969) indicates that the conglomerates form at
least 5 percent of the total Keweenawan series at Cape
Gargantua. He also notes that the conglomerates are poly—
mictic in nature, with a composition comprised of 80 to 90
percent basaltic clasts and minerals derived from the basalts;
pebbles of diabase and.plutonic rocks are also observed.

The clasts have an average diameter of 3 inches; however,
boulders as much as 2 feet in diameter occur. This composi—
tion is similar to that of the lower interflow sedimentary

horizons at Mamainse Point.

 

 

The
as a row
1911).
of mafic
zentary 1
Island a:
localitic
”P at 1e:
0f the M:
parts of

The
occur mai

TheSe Sed

 

island,
northeast
9€Sted th
3 10ca1 d
eroSiOn a
The

“rived f
but also
gneisuses
in 1‘ESSer
“mm

b .
«01
at and

18

The Michipicoten Island Sequence

The Keweenawan rocks occupy the entire island as well
as a row of smaller islands of its south shore (Van Hise,
1911). The section on the island consists of 11,230 feet
of mafic and felsic extrusive, felsic intrusive, and sedi-
mentary rocks (Burwash, 1905). The rocks of Michipicoten
Island are unique in comparison with other middle Keweenawan
localities around Lake Superior in that felsitic lavas make
up at least half of the succession. The more mafic members
of the Michipicoten sequence are most abundant in the lower
parts of the succession.

The interflow sedimentary rocks on Michipicoten Island
occur mainly in the lower horizons of the sequence (Figure 2).
These sediments are confined to the northwestern part of the
island, being thickest in the west and thinning out of the
northeast. Because of their location, Burwash (1905) sug-
gested that the interflow conglomerates are associated with
a local anticlinal uplift which resulted in more rapid
erosion and deposition of elastic sediments in that area.

The conglomerates are composed primarily of material
derived from felsic rocks. Mafic clasts are subordinate
but also abundant; granites, greenstones, and biotite
gneisses derived from pre—Keweenawan rocks are also present
in lesser amounts (Burwash, 1905; Van Hise, 1911). These
polymictic conglomerates are similar to those at Mamainse

Point and suggest the nearby presence of a tectonically

19

active borderland which was shedding both volcanic-surficial

and nonvolcanic-basement type rocks simultaneously.

The Mamainse Point Volcanic Series

The thickest and most prominent Keweenawan section on
the east shore of Lake Superior occurs at Mamainse Point,
Ontario (Figure 1). It contains appreciable amounts of pre-
Keweenawan plutonic rock fragments in addition to volcanic
clasts. This combination of observations suggests the
presence of a tectonically active borderland which was able
to supply coarse detritus from both volcanic flows and an
underlying nonvolcanic basement simultaneously throughout
the middle Keweenawan succession preserved at Mamainse Point.

Thomson (1953) noted that the Mamainse Point Volcanic
section forms a broadly plunging anticline, which is accentu-
ated by faulting. The dips range from 200 to 48°, with the
steepest dips appearing at the top of the section. The
fanning of dips characteristic of most Keweenawan sections
exposed elsewhere around Lake superior appears to be re-
versed at Mamainse Point. This observation, coupled with the
presence of abundant dike swarms at nearby Alona Bay, indi-
cates that some kind of tectonic folding and fissuring was
in progress during Keweenawan time. The coupling of folding
and dike-intrusion tectonism does not appear to be unique
to the east shore of Lake Superior. Similar occurrences

along with sedimentological and lithologic criteria will be

 

 

 

 

presente‘
diastrop
Superior
flows ir.
as on th
shore, i.
shore.
The
Point Keg.
Slightly
t0 14.30:
ness esti
good outc
exposed,

prOduCe S

 

Mafi
series at
eXtrusivE

1955; Ann.
PlUg;
at all le‘
are rare
Nuffj
Keweenawar
ofMamains

K 9
Semen aWa ‘

20

presented in the course of this study to suggest that such
diastroPhism and volcanism may have occurred around Lake
Superior whenever Keweenawan sediments are intercalated with
flows in areas where diabase dikes are present nearby, such
as on the south shore, the northwest shore, and the east
shore, if not perhaps completely around the Lake Superior
shore.

The estimates of the apparent thickness of the Mamainse
Point Keweenawan succession made in the past vary from
slightly over 16,000 feet (MacFarland, 1866; Moore, 1925)
to 14,300 feet (Annells, 1971). This variation among thick-
ness estimates is understandable in view of the fact that
good outcrOps are scarce, a continuous succession is nowhere
exposed, and the effects of possible strike faulting may
produce spurious thickness values.

Mafic lavas make up the greater part of the Keweenan
series at Mamainse Point. Some rhyolite sheets, possibly
extrusive, are also present throughout the sequence (Thomson,
1955; Annells, 1971).

Plugs, dikes and sheets of fine grained rhyolite occur
at 311 levels in the Mamainse section. However, mafic dikes
are rare (Thomson, 1925).

Nuffield (1955) found numerous dikes cutting the pre-
Keweenawan terrain at the Point aux Mines, just to the north
of Mamainse Point, but only one dike appears to cut the

Keweenawan lavas. Therefore, most diabase dikes are either

 

 

younger ‘

 

flows wh:

Interflo:

Inte
are at 16
make up a
Xamainse

averages

 

section e
horizoss
feet with
30“ clos
Change ir
SequenCe
0f the i:
sed'ments
KeweenawE
“hare an

Mamainse

Such a 3

graphic

coalesq;

The

Point We

21

younger than the exposed Keweenawan flows, or they fed

flows which have since been eroded away.

Interflow Sediments

Interbedded with the lava, in the Mamainse Point area,
are at least 9 sedimentary horizons which in total thickness
make up about 13 percent of the aggregate thickness of the
Mamainse section. The thickness of the conglomerate beds
averages over 100 feet, with one near the middle of the
section exceeding 1,700 feet in thickness. The conglomerate
horizons occur on the average at intervals of about 1,900
feet within the Mamainse Point Series. However, they appear
more closely spaced in the upper half of the section. This
change in spacing is also characteristic of the Portage Lake
sequence on the Keweenaw Peninsula. The greater thickness
of the interflow sedimentary beds, and the fact that these
sediments comprise a greater percentage of the total middle
Keweenawan volcanic succession at Mamainse Point than else—
where around the lake, suggests that the terrain east of the
Mamainse depositionary site was a major physiographic feature
such as a fault scarp or a mountain range. This physio-
graphic feature was flanked on its westward margins by

coalescing alluvial fans.

Conglomerates.
The lithology and size (long axis) of gravel coarser
than 19 mm (-4.25 phi) in the interflow sediments of Mamainse

Point were determined in the field to obtain data pertaining

 

 

to: (1‘2

(2) rela

 

and (3)
award It
‘ I
lected 2
were cho‘
availabi.
The ider‘.‘
determine
section a
A g:
COluilomer
0f plutor

pre‘KEWee

 

cor‘iglomer
Vein qua:
generally
however,

thIOughOu
fangIOmer
BecauSe t
bEEQ far

ibSenCe 0

front 0 f .

1

,.
.11
es W851

 

22

to: (1) the lithology present in the source terrain,

(2) relative location of the western margin.of the highland,
and (3) whether the highland underwent a single or several
upward movements. The analyses were made by counting se-
lected zones within the conglomeratic units. The sites

were chosen on the bases of stratigraphic distribution and
availability. At least 100 point counts per site were made.
The identities of certain prdeematical pebbles were
determined in the laboratory by binocular microsc0pe or thin
section study.

A great variety of rock types are represented among the
conglomerate clasts (Table 2). The largest boulders consist
of plutonic rocks and mafic lava fragments (Table 3). Other
pre—Keweenawan, but smaller, rock fragments present in the
conglomerates are Keewatin-type greenstones, and schists,
vein quartz and sedimentary clasts. Mafic lava fragments
generally form the majority of the conglomerate clasts;
however, plutonic rock fragments are generally abundant
throughout the Mamainse Point section (Table 2). No angular
fanglomerates were observed in the Mamainse Point succession.
Because the source of many of the rock fragments may not have
been far east of the present Mamainse outcrop belt, the
absence of fanglomerates suggests that either the mountain
front of the highland did not involve an abrupt break in
slope, or that sections along the outcrop belt are several

miles west of the possible break in slope.

23

.88 ma can» Homumoo mucosuflumcoo omega mace wmuuuom
ou Hopuo cw Xboa ou woumasoamoou mmB .ucoemo can XHHum8 msflpsaoxm .ucsoo ucflom

 

 

 

 

 

 

 

hoaonvoum
cone u z o.ooa o.- o.mo m.eo n.mm mssaa>
mmmuo><
N 0.008 n u I 0.008 «.2 souuom
m 0.008 o.m~ s.mo m.wo o.mm mu:
m o.ooa «.mo o.ea_ . H.~m an:
a o.ooH m.om «use e.ao m.om was
m o.ooH s.~m H.mo m.oa o.me 5-2
m o.ooa ~.m¢ o.mo n.m ~.s¢ «.2
a o.ooa ¢.H~ m.ao m.m m.so on: doe
mmnouno Mom acmuuom mummau mummao mummao mummao couwuom
Umucsou Hmuoa vacoudam mumu808wvommuoz_ ouwmaom owmmz. 3OamnmucH
mmuem mocmspeumcoo venues

 

 

 

 

 

 

mm3 ucoou ucwom

.ouam Hod husuoo ooa coma comma

.mououso any meaucsoo unwom kn posflBHouov mm Assn m~.vlv

.88 ma can» Homnmoo mam>mum 3oamuoucfl ucwom mmcwm8mz_onu mo aboaoguwn

N OHQMB

{Clairenzu C.— uuflenfinzlnzescz mmmw UCflAUe~ 3.312% «3:72

Jfl. 096.:de UfiCOuJHQ 3:1 UdHEE NO «flux-r. WCOHV QNwmn «av—«u n.2nv53LUQ nuomwznntgftnurv .m... meanwrN.

 

.. a. L; f. g ... s .3: u-

 

 

24

.80 H.o “mouse: on mucosmusmmoz

 

 

 

m.m¢ N.Hm mEBEaxmz_mo .m>¢
H.0H h.mo mcmoz.mo .m>¢ Mnm88sm
lllllllllllllllll llll'lllll'l.l.l.l.l.ll.l..l.ll IIII'IIIIIIIIIII..I.I|L lllnllll-llll
I m.m¢ oNflm .xmz cm3mcoo3mx
I m.ma ouam cows. «:2 Ho3oq
IIIIIIIIIIIIIIIII llllll.|l..l.|lllnlll|ll|lllul lul'ullllullnll'lulllllllluL 'IuIllIlullI-IIII
m.om m.om Guam .xmz
>.NH N.NH oNHm Gmmz ..mwz.
m.mn m.om mafia .xmz
N.mH 0.50 ouam and: ml:
«.00 m.om Guam .xmz
H.mo 0.00 mnam cmoz mlz
IIIIIIIIIIIIIIIIII IllllllllllllllllllllIIIIIIIIIIIIIIIIIIIJIIIIIIIIIII
o.mm N.mH ouam .xmz cmBmcwoBmx
b.0H ¢.mo mNam cmmz hi2 mawvwz
IIIIIIIIIIIIIIIIII I'll-allIIIIIIIIIIIIIIIIIllllllllllllllllli.Il.|.|.ll..|l|.|.|
m.mm h.m¢ mNHm .xmz
a.mo m.mo mNHm cmmz NIE
¢.m¢ m.hm mnam .xmz
m.hH o.oa mean and: 0:2
A80 Gav A80 cwv
Amwxm mcoHV onam Amflxm mcoav muwm whomoumo coufluom mod
ummao oflcousam ummao uemmz Amexm mcoav muflm ucmeflsmm

 

 

 

 

 

 

um mummao vascusam

.mouuuso cw penanm08,nm ucaom omcwm8mz
can oammE mo Amflxm mcoav mafia onu coo3umn somwummeoo .m magma

25

It appears that a nearby source terrain consisting
of metasedimentary and plutonic rocks was being eroded
during Keweenawan time. The fact that the lower-most flows
rest on weathered granites possessing considerable relief
(Nuffield, 1955; Ayres, 1969) rather than sedimentary or
low ranked metamorphic rocks probably indicates that much
unroofing had occurred prior to Keweenawan time.

It is worth noting that the lower Keweenawan interflow
sediments at both Cape Gargantua and Mamainse Point are
comprised almost solely of coarse mafic volcanic clasts
whereas the stratigraphically higher interflow sediments are
polymictic. This observation about the lower Keweenawan
sediments lends credence to the idea that extensive basalts
flows probably blanketed much of the source terrain in
lower Keweenawan time. Uplift of the areas marginal to the
present Lake Superior syncline during periods of volcanic
quiescence allowed these flows to be eroded, thus forming
a source of detritus for the interflow sediments. Continued
erosion with time laid bare the underlying pre-Keweenawan
rocks which then also contributed detritus to the deposi—

tionary site throughout middle Keweenawan time.

Sandstones and Sandy Phases

Sandstones occur in the.Mamainse Point Keweenawan series
in thin beds, as sandy lenses within the conglomerates, or
as conglomeratic matrix. Sixty-four thin sections were

made from samples collected from within these sandy phases.

26

All samples were collected solely from outcrOps (Figure 3).

Table 4. The composition of the detrital sand-size con-
stituents within the sandy phases of the Mamainse

Point interflow sediments,

as represented by the

mean of the modal analyses of 64 thin sections.

 

 

Volume Frequency

 

 

 

 

Constituents Yr S
Mafic rock fragments 29.9 30.3
Plutonic rock fragments 28.6 25.5
Metasedimentary rock fragments 21.8 23.7
Felsic rock fragments 13.6 15.7
Simple quartz 01.8 1.4
Plagioclase 02.1 2.8
Undalatory Quartz 01.1 1.5
Potassium feldspar 00.8 1.7
Polycrystalline quartz 00.2 0.5

99.9

mean of 64 thin sections.

0')
ll

standard deviation of 64 thin sections

 

The mafic rock fragments appearing in the sandy phases

of the Mamainse Point interflow sediments comprise approxi-

mately 30 percent of the total volume of these sediments.

These sand sized mafic fragments display a wider range of

textures than other mafic clasts in exposures elsewhere in

Lake Superior. The mafic clasts at Mamainse Point range in

texture from the fine basaltic to basaltic amygdaloidal to

27

 

3.91: 5

”WW. "3" , 51/
num:umanma Ffitfit ‘égg?

m... //

 

me an m 3'41" n4 .
canmununa

a
-
q
T“
/-."
‘(“'
‘(C

0'
mm p, I. ..

vi¢
(Ԥ
'
¢<<<¢c

C

8

.b
.\\
A. \
n—

Loeetion of interflow .. ~ v
sou-outer! horizon: E .332 s‘
0.91.11 for this study .1 £53!”

a , W vvvv v vvuv
H gr_ 7 *\vvv‘v vv-v‘ vvvvv vavvu /
‘

K
5
I

34

‘
\
‘
«(Na .

(C(((((‘I

i
k
C.
(<
(I¢¢(¢(((ll(
<¢<

l v
*VVvvh vvv,o
.2 ‘ vswvvv vvv Va
vvvvvvvvv‘vC‘VVV
'VV'VVVV'V“‘U- “.‘
vvvvvvvvvvvvflvv v \:
V'V'VVVUVVVVV ‘V
VVVvvvvvvvvvvv vv
V" V

 

-
v33 . ~rp vavvvvvvv
v'_ V? vvvvvvvvv
v \ ‘ vvvsvvvvv
v
V

I

‘ vvvvvv
\. \ vvvvvv u- ,
‘. vvvv vvva
\ E v vvvv

"VV'V V

    

3* _-_=

c
c
c
<
1
(((
¢¢¢¢¢c
¢¢¢¢¢¢

(<‘(1
(‘(((
‘((((
(((‘K
C((Q‘
((utd
qc<et
“‘C\
‘(t‘(
<((‘
((‘
(Q '
o

, 4

I

new? . vvvvvvvvvvvv
vvv ~

Q

30 1°
4.4/0
nut! "

w

 

 

 

 

Figure 3. Geological map of the Mamainse Point and Alona
Bay areas (after Palmer, 1970).

28

diabasic to coarse diabasic° This variety is often present
in a single thin section.

The plutonic rock fragments comprise approximately 29
percent of the volume of the sandy phases within the interflow
sediments. These clasts appear to be either gneiss or
granites.

Metasedimentary fragments constitute approximately 22
percent of the volume of the interflow sediments and
consist primarily of micaceous schists and immature sand-
stones in an approximate ratio of 4 to 1.

The felsite rock fragments comprise 13.6 percent of
the sandy phases of the interflow sediments at Mamainse.
The felsite clasts commonly consist of uniformly fine-
grained quartz and feldspar, which commonly show gradations
in their relative prOportions, although the feldspars are
generally dominant. No granOphyre fragments were observed
in the various sedimentary horizons at Mamainse Point.

The various kinds of detrital quartz grains comprise
3.1 percent of the average volume of the sandy phases.
Unstrained quartz is the dominant quartz variety and makes
up 58 percent of the total quartz volume within the sedi-
ment. Quartz grains showing moderate to strong undulose
extinction comprise 35.5 percent, while polycrystalline
quartz makes up the remainder (6.5%) of the total quartz
volume.

Detrital feldspars constitute 2.9 percent of the total

volume of the sandy phases within the interflow sediments.

Plagic
of the
sist a
ally k
0.8 pe
most a
howeve
ingly

Point

 

E
within
Sedime

Occur:

*3

re»

e r
Chaige
sition;
SiZe’ (
in gra:

dEtritE

29

Plagioclase is the dominant feldspar comprising 72 percent
of the total feldspar volume. The plagioclase grains con-
sist most commonly of labrodorite and andesine and gener-
ally have a fresh appearance. Potassium feldspar makes up
0.8 percent of the total volume of the sandy phases. The
most abundant potassium feldspar is an untwinned variety,
however, grains showing gridiron twinning become increas-
ingly abundant in the upper-most horizons in the Mamainse

Point succession.

Vertical Compositional Variation

Evaluation of the vertical compositional variation
within the sandy phases of the Mamainse Point Interflow
sediments was made to detect any changes in provenance
occurring during Keweenanaw time.

The sedimentary samples were collected from outcrops
in such a way as to portray a composite vertical section
representative of the interflow succession at Mamainse Point
during lower and middle Keweenawan time. Figure 3 shows the
location of the outcrops sampled for this study.

If it is assumed that the location of the source ter—
rene remains constant throughout the vertical section, then
changes in the volume frequencies among the various compo-
sitional constituents can only arise from variation in grain
size, depositional environment, or provenance. As variation
in grain size effects the volume frequencies of the various

detrital constituents within the sediment even in the absence

3O

 

 

 

 

 

couwuon humucoaprm some mo sawumfl>ou phoneme» u m
couauon immunovaou nose mo hocosvouw ossao> some n m
I 0.0 0.m n.~ a.~ «.0 ~.H m.0 v.~ 0.H m.0~ H.00 n.0a n.a~ ¢.v m.m 0.H 0.0 m: 209900
I 0.0 I 0.0 I 0.0 I 0.0 I 0.0 I 0.0 I 0.0 m.m 0.m m.m 0.50 v:
I 0.0 m.a 0.H ~.H h.0 m.H 0.a 0.H 0.H v.0 m.¢a m.mH ~.m ~.¢ m.~ 0.0H n.00 gun:
I 0.0 5.0 0.0 I 0.0 I 0.0 ~.H 0.0 m.ma m.~m e.m m.m 0.8 m.a~ m.~ H.Hm m:
I 0.0 I 0.H I 0.0 m.~ H.~ m.v ~.v m.0a m.m~ ~.m a.¢ I 0.0 m.a« 0.0m m:
I 0.0 I 0.0 I 0.0 I 0.0 m.a 0.0 H.H~ ~.ma H.Hm m.~m m.mH n.0H m.m~ m.m~ m:
I 0.0 I 0.0 I 0.0 I 0.0 m.~ H.A I 0.0 I 0.0 o.¢a ~.m¢ H.m~ h.~m a!
I 0.0 m.n m.a I 0.0 H.m n.0 H.~ v.0 o.¢a ~.nm m.mm 0.0m 5.0 n.v m.~a c.5H n:
v.~ ~.H m.m 0.n I 0.0 o.m n.~ m.m o.~ m.¢~ m.0a m.mH m.m~ n.8H v.- ¢.h~ n.m~ NS
H.m m.m m.v m.0 I 0.0 0.m a.m m.m 0.m 0.mH 0.0m «.ma 0.na H.m~ ¢.h~ v.0 m.m~ 0: mos
m m m m m m m m m m m m m m m m m m
HmomUAmm ommaooflumam uuumna Nuummq nuuMSOI mummHo wommHu mummau mummHU couwuom
EndanmuOd .uu»uu>~om huousasvca mamsfim escapadm xumucosapommuoz camaom Dawn:

 

 

.muCUEanm Bonumucw ucflom mmc«m8mz may wo mommnm husmm mnu secua3 :oHumfium> HMCOauflmoafioo amofiuum> .m manna

31

mEi other effects, any analysis of compositional variation
must first take this effect into account. Therefore, the
meathod chosen to evaluate the vertical variation among
v<>1ume frequencies of the various detrital constituents was
tlaat procedure utilizing the least square regression analysis.

Regression analysis was employed to allow the deviation
:in the volume frequency (dependent variable) to be minimized
as the grain size (independent variable) for that composi-
tional component was assumed fixed. Thus the volume frequency
‘was regressed about the grain size. This method allows the
variation in the volume frequency to be explained in terms
of the variation in grain size. Any portion of the total
variation not accounted for by the differences in grain size
remains as a residual and must be due to changes in proven-
ance or the depositional environment.

The following method was used to evaluate vertical
compositional variation. The existence of a linear relation-
ship between grain size and volume frequency was verified by
graphing, thus meeting the assumption of linearity necessary
for use of regression analyses. The least square line of best
fit and the correlation coefficient were calculated. The
(observed-expected) values of the volume frequency versus
depth were plotted. All plots were then evaluated by 2 x 2
Chi square contingency analysis to judge if the scatter of
points about the zero line showed a trend possibly indicating

a signigicant (szI: 0.05) residual.

32

Figures 4 and 5 graphically portray the relationships
between the (observed-expected) volume frequency and depth
for the important sand-size compositional constituents
present within the interflow sedimentary horizons at Mamainse
Point. When the grain size effect is statistically removed
from.the compositional elements present in the Mamainse Point
section and results are plotted the only compositional ele-
ments which show a significant (PX2 < 0.05) residual trend
are the mafic, plutonic, and felsite rock fragments. The
mafic and plutonic rock fragments show a decrease in volume
frequency upward in the section, while the felsite fragments
show an increase. The increase in felsite clasts appears to
correlate with the Michipicoten Island sequence which
Annels (1971) believes is stratigraphically higher than the
Mamainse Point sequence. In the Michipicoten Island succes-
sion the felsite clasts dominate the composition of the inter-
flow sediments.

Analysis of the vertical variation in composition and
grain size in the Mamainse Point interflow succession with
time suggests that the grain—size effect exerts the greatest
degree of control on composition vertically. Gradual changes
in the source terrain are also suspected to have occurred,
based on the significant changes in the volume frequency of
the mafic, plutonic, and felsite constituents. The analysis
indicates that a definite lessening of importance of the

volcanic and plutonic terrains as a source of rock fragments

 

Figure 4.

33

Graphs of the vertical compositional variation
within the.Mamainse Point interflow sediments.
Graphs show the residuals when the (observed-
expected) volume frequencies of the various
detrital constituents are plotted versus their

depth of occurrence in the.Mamainse Point section.

Graph A is the plot for mafic clasts, Graph B is
for plutonic clasts, and Graph C is for felsite
clasts.

N.-- M. W_ _. q.

am:
an

(a.

:mx
an:

R} n .:

J“ U "
I

use

3H!
1‘

i

3

 

 

 

 

mm~¢mcufl mm mm: or Ian:

an
I"
“I
0".»
1‘.

I9.

IO-

0.9

 

0 ‘0
a.» 78.

34

 

3

00500

 

 

 

 

 

«n -u 0 on «o no
Qumran-ulcer") M W or man “an

Figure 4

M I o o
‘ -
”I o o 6 o
C -m o o o
t ' no
a- n A o o o o o
- 1 O ‘20 0.0 ‘0
(commune) a“ my 0 “.070“ ““7.

 

 

Figure 5.

35

Graphs of the vertical compositional variation
within the Mamainse Point interflow sediments.
Graphs show the residuals when the (observed-
expected) volume frequencies of the various
detrital constituents are plotted versus their
depth of occurrence in the Mamainse Point section.
Graph D is the plot for metasedimentary clasts,
Graph E is for unstrained quartz clasts.

I, s

’0’

3

 

 

 

 

“‘0

~80

‘10

0”

060

36

W~KMfl mm ma ”MMNNNWV M3

'0-

Figure 5

all

3

 

 

‘1':

 

M'M) mm WV W mar: um

 

51
1e
Ke
re

 

r1
Ru.

37

occurred during Keweenawan time. The analysis also suggests
that felsitic rocks as a source of detritus are becoming
increasingly important upward in the sequence with time.
This trend of decreasing mafic and plutonic clasts could
signify decreasingxtectonism with time.’ Erosion then would
less vigorously be attacking the basalt mantled pre-
Keweenawan non volcanic rocks, thereby permitting a
relatively greater contribution of the more weathered mono-

mineralic fragments to be present in the sediment.

Sedimentary Synopsis

Provenance

The compositional immaturity of the interflow sediments
within the Mamainse Point series allows deductions to be
readily made concerning the nature of the Keweenawan source
terrain. The composition of the Keweenawan interflow sedi—
ments suggests a nearby source terrain consisting of mafic,
felsic, metasedimentary and plutonic rocks.

Annels (1971) points out the striking similarity of the
flows in the lower levels of the Mamainse Point, Alona Bay,
and the Cape Gargantua volcanic sequences, which are petro-
graphically similar to those forming the basal part of the
Michipicoten Island sequence. He suggests that these latter
flows may represent the upper part of an extensive and
largely uniform flood basalt pile whose earliest members

rest directly on the Lower Precambrian rocks of the east shore

IMMMMM

38

of Lake Superior. These observations, plus the presence of
coarse interflow conglomerates near the base of both the
Mamainse Point and Cape Gargantua which consist entirely of
detritus from mafic flows, in addition to the presence of
widespread diabase dike swarms of lower Keweenawan age located
eastward of the presently exposted volcanic series, indicates
that the basal flows were once more extensive. It is con-
ceivable that they could have covered much of the source
terrain during Keweenawan time.

The source area, dominated by an extensive and thick
sequence of volcanic rocks, was tectonically unstable, which
encouraged the production of considerable amounts of coarse
detritus to be shed from local sources intermittently during
periods of little or no volcanic activity.

The coarse texture of the Mamainse Point Keweenawan
interflow sediments, in Which the volcanic fraction locally
may be as coarse as the pre—Keweenawan plutonic fraction,
suggests that the Keweenawan lavas and the pre—Keweenawan
rocks, acting dually as a source, existed as a continually
active borderland from which the sediments were derived.
Concomitant with this active borderland, the deposition and
preservation of such a thick series of flows and sedments
suggests a nearby rapidly subsiding basin in which continued
downwarping produced a topographic depression into which not
only lavas but streams flowed and deposited detritus shed
from the nearby tectonically active borderland during periods

of volcanic quiescence.

39

The fact that the sediment intercalated between the
flows appears periodically in the section need not necessar-
ily imply that there were discrete periods of volcanism,
uplift, sedimentation, and subsidence, but that continued
erosion and sedimentation were taking place elsewhere as
their locii where shifting laterally as given drainage areas

were locally blocked by flows.

Paleogeography and Sedimentation

The coarseness and compositional immaturity of the con-
glomerates and sandstones constituting the Mamainse Point
series indicates a nearby source of some relief. These
characteristics also suggest that the sediment underwent a
short transportational history and rapid deposition in a
sedimentary basin which was rapidly sinking.

The uplift of the marginal source terrain coupled with
subsidence of the central portions of the basis initiated
erosion, transportation and deposition of immature, poly—
mictic conglomerates and arenites by a flurial system on a
piedmont fan.

The lack of shale and fine sand, the coarse nature of
the sand and conglomerate, the crude bedding, and the
typically poor sorting and lack of rounding existing in the
sandy phases suggests that the interflow sediments comprising
the Mamainse Point series were transported by a fluvial system
characterized by relatively high energy conditions and high

gradients.

4O

Steep gradients and rugged relief in the source area
provided the tOpographic control necessary to allow lava
flows to be dissected by streams. Renewed vertical erosion
by incised streams cut completely through the flows, locally
exposing the underlying pre-Keweenawan rocks. This enabled
fresh volcanic rock fragments showing a complete range of
textures, pre—Keweenawan rocks, and weathered detritus from
the interfluve areas to be mixed together.

The evidence presented in this study suggests that the
most likely location for the source area borderland of both
the east shore Keweenawan volcanic rocks and their inter-
calated sediments was just eastward of the presently exposed
Keweenawan outcrops. This area is represented by the
occurrence of abundant lower Keweenawan diabase dikes exposed

in lower Precambrian metasedimentary and plutonic rocks.

CHAPTER IV

THE SOUTH SHORE REGION

The Keweenawan rocks of the south shore region of Lake
Superior are found on the Upper Peninsula of Michigan and
in northern Wisconsin (Figure 6).

The rock units exposed on the Keweenaw Peninsula are
the Bessimer sandstone and the South Range Lava Series of
lower Keweenawan age, the Portage Lake Lava Series of Middle
Keweenawan age, the overlying COpper Harbor conglomerate,
the Nonesuch Shale, and the Freda sandstone of Lake Keweenawan
age, and the Jacobsville sandstone of late Keweenawan or
Cambrian age. Modal analyses were made only on the Portage
Lake and Copper Harbor interflow sediments for this phase of
the investigation.

Hubbard (1967, 1968) has described the South Range
Series. He suggests that these rocks, which are located
adjacent to both the Huron Mountain dike swarms and the.Mellen
Complex, may be the source of the interflow sediments present
in the overlying Portage Lake Series and the COpper Harbor
conglomerate.

The proposed source area for the Keweenawan interflow

sediments is thought to have consisted of pre-Keweenawan

41

42

.363 6.3.8 “when 633695 338...: mo 33:33 “and:
"Howuomsm mxmq mo coflmou ouonm Anson onu mo mm8 Hmowmoaoow .0 musmfim

 

{ J...

'
O-
'-
-‘--""-'-

ck

240.10.!

 

 

B B BBB BB BBBB

Inguuc. H48”!

gun a: a3~a

MAS—km «>3 #9.: 5m

53%.—

gu 3

8.23m (>4.— 3 "i=0.—

13334 38“!

V. 4.323080 35— go

33.»; gnu-nu
Bahama g
33$ g

g wag—>285
.432... fig 8 3.3. .5

I33! 25

gun-n “has EC.— 9 nag g

 

 

monumsm

HV~<A

 

 

43

metasedimentary and plutonic rocks mantled by a thick cover

of Keweenawan volcanic rocks. The source area was uplifted
and intruded by a large gabbro-granophyre complex in the
Mellen area of Wisconsin (Aldrich, 1929; Tyler, 1940; Leighton,
1954; White, 1966). Goldich et a1. (1961) and Silver and
Green (1963) have reported that the granitic portion of body
intruding the Keweenawan lavas at Mellen, Wisconsin has been
dated by the K/A method as being approximately 1.0 b.y. old.
This age determination corresponds well with ages for
Keweenawan time.

Furthermore, paleomagnetic studies by Du Bois (1962),
and Books (1968) indicate that the South Range Lava flows are
of reverse polarity and are thus magnetically defined as
lower Keweenawan in age. The overlying Portage Lake Lava
series and the Copper Harbor Conglomerate are considered to
be of middle Keweenawan age on the basis of having normal .
polarity and a remanent magnetization significantly different

from that of the South Range Series.

The South Range Lava Series

The oldest Keweenawan rocks on the Keweenaw Peninsula
are the quartzose sandstones of the Bessemer formation.
These are overlain by the South Range Lavas which locally also
overlie.Middle Precambrian rocks. The South Range Lava Series
crops out in an east-west trending band near Ironwood, Michigan
which extends approximately to the southern end of Lake

Gogebic (Figure 6). Hubbard (1967, 1968) indicates that the

44

lava series is at least 10,000 feet thick. The rocks of

the South Range Series are separated from the overlying
Portage Lake Lava Series by a covered interval approximately
2 miles wide near Ironwood. The rocks underlying the covered
interval have smaller magnetic anomalies than either of the
two mafic sequences above or below it. Hubbard (1968) indi-
cates that the magnetic data are compatible with the covered
interval being underlain by sedimentary rocks or by sili-
ceous volcanic rocks. Outcrops of felsite are reported on
strike with the covered interval in adjacent Wisconsin
(Aldrich, 1929; Hubbard, 1968).

The dips of the South Range Series increase toward Lake
Superior. This observation is diametrically opposed to the
normal fanning of dips characteristic of the Portage Lake
and Copper Harbor successions over most of the Keweenaw
Peninsula. Hubbard ascribes this variation in dips between
the South Range Series and other Keweenawan rocks to anti—
clinal flexure on the limb of the syncline.

The middle member of the South Range Series contains a
few interflow sandstones, while the upper member contains
some intercalated conglomerates, but exposures are rare

(Hubbard, 1968) .

The Portage Lake Lava Series

 

The Portage Lake Lava Series extends as a belt one to

three miles wide from the tip of Keweenaw Point southward

45

into Wisconsin. The southern margin of the lava series over
most of the peninsula is marked by the Keweenaw fault
(Figure 6). The rocks of the South Range Series, the Portage
Lake Series, and Copper Harbor Conglomerate all thin westward
west of Lake Gogebic, with the higher Portage Lake flows
extending the furtherest west of Lake Gogebic (Hubbard, 1968).
White (in Hubbard, 1968) attributed this thinning to progres-
sive lapping against a tOpographic high near Mellen, Wisconsin.
The Portage Lake flows are coarser in texture than the
South Range series (Hubbard, 1967) and consist primarily of
basalt and andesite flows. Rhyolite flows are rare and on
the average constitute no more than 0.5 percent of the
sequence on the Keweenaw Peninsula (Cornwall, 1951); however,
the upper portion of the section at Porcupine Mountain con-
sists of over 2000 feet of rhyolite extrusives (Rohrbacker,

1969).

The Portage Lake Interflow Sediments
The Portage Lake Lava Series are interbedded with con-

glomerate and sandstone beds; these rocks constitute l to 3
percent, and locally as much as 11 percent of the aggregate
rock volume of the lava series on the Keweenaw Peninsula
(Johnson and White, 1969). Marvin (1873) recognized and
enumerated 22 such sedimentary horizons within the lava series
(Table 6). The conglomerate and sandstone beds range in
”thickness from a fraction of an inch to as much as 370 feet

locally, and occur at stratigraphic intervals of approximately

46

Table 6. Keweenawan stratigraphic column on Keweenaw
Peninsula (after Sprioff and Slaughter, 1961).

 

 

Freda

 

Nonesuch

 

Outer Con lomerate
COpper Harbor Lake Shore Traps
Great (Eagle River) Conglomerate

Upper Keweenawan

 

Hancock Conglomerate (No. 17)
Ashbed Lode

Pewabic West Conglomerate (No. 16)
Pewabic Lode

Portage The Greenstone Flow

Allouez Conglomerate (No. 15)
Houghton Conglomerate (No. 14)
Iroquois Amygdaloid

Calumet & Hecla Conglomerate (No. 13)
Lake Calumet Amygdaloid

Osceola Amygdaloid

Kingston Conglomerate (No. 12)
Kearsarge Flow

National Sandstone

Lava Wolverine Sandstone (No. 9)
Old Colony Sandstone

Scales Creek Flow

Gratiot Flow

Isle Royal Lode

Series Behemia Conglomerate (No. 8)
Copper City Flow

St. Louis Conglomerate (No. 6)
Lac La Belle Conglomerate
Baltic Lode

Baltic Conglomerate (No. 3)

Upper Precambrian

Middle Keweenaw

 

 

 

----------- Keweenaw Fault - - - - - - - - - - - —

 

 

 

I Inf-.Inul

47

1,000 feet, however they are more closely spaced in the
upper part of the column above the Greenstone flow and in
the lower portion of the series beneath the Bohemia con-
glomerate (Table 6). The sedimentary beds within the
Portage Lake series are excellent marker horizons because
of their persistence. Two of the beds, the Allouez and the
Bohemia conglomerates have been recognized from the east end
of the Keweenaw Peninsula to Victoria, Michigan, a distance
of approximately 100 miles along the strike (Cornwall, 1955).
The conglomerate horizons are sub—parallel to one
another and the aggregate thickness of the lava flows sep—
arating them remains relatively uniform along the strike.
The conglomerate beds prObably represent relative planar sur-
faces suggesting that the depositionary site was probably a

piedmont alluvial fan.

Conglomerate

The conglomerate beds within the Portage Lake Lava
Series are lenticular, poorly sorted, crudely stratified,
and commonly a dark red to reddish brown in color. The bulk
of the clastic particles range from a fraction of an inch to
6 inches in diameter. The coarsest clasts occur in the
thicker beds, but rarely exceed a foot in diameter (Butler,
and Burbank, 1929; White, Cornwall, Swanson, 1953). The
pebbles and cobbles are generally well rounded. The beds
are compact and tightly cemented. Lenses of sandstone are

commonly present even in the thicker conglomerates and often

48

contain cross-bedding, ripple marks, and mud crack features.
Shales are rare indicating that the associated finer size
grains must have bypassed this region and were deposited
elsewhere.

Based upon examination of underground workings and
numerous drill cores in the mining districts throughout the
Keweenaw Peninsula, Lane (1911), and Butler, Burbank, et a1.
(1929) noted the existence of two lithologically distinct
conglomeratic types within the Portage Lake lava series:

(1) "a felsite conglomerate" consisting largely of felsitic
rock fragments with minor mafic fragments, and (2) an
"amygdaloidal conglomerate" composed of mafic amygdaloidal
rock fragments in a mafic sand. According to these ob-
servers, where both types are present in the same exposure,
a bed of the amygdaloid type almost invariably underlies

the felsite conglomerate.

Sandstone and Sandy Phases

Sandstones occur in the Portage Lake Lava Series as
thin beds, as sandy lenses within conglomerates, as sandy
phases marginal to thicker conglomerate lenses, and asimatrix
within conglomerates.

Sixty thin sections were made from samples taken within
the sandy horizons and from the sandy matrix of the con-
glomerate horizons within the Portage Lake interflow sedi-
mentary zones. Forty-seven of these thin sections were taken

from a composite vertical section composed of several cores

49

drilled through the Portage Lake Lava series in the vicinity
of Calumet, Michigan. Thirteen thin sections were made

from samples collected from cores and outcrops elsewhere on

the peninsula to indicate any lateral variation in composi-

tion When compared with the Calumet section as a standard.

Figure 7 shows the sites sampled.

Table 7. Mean composition of the sandy phases within the
Portage Lake interflow sediments. The composition
of the detrital sand size constituents within
the sandy phases of the Portage Lake Interflow
sediments as represented by the mean of the modal
analyses of 60 thin Sections.

 

 

Volume Frequency

 

 

'(%)
Constituents 3? s
Felsite rock fragments 35.4 23.8
Rhyolite rock fragments 28.6 23.6
Mafic rock fragments 23.5 25.6
Unstrained quartz 4.6 8.5
Plagioclase 4.1 5.8
Granophyre rock fragments 2.9 4.6
Potassium feldspar 1.0 3.1

 

Total 100.1

 

= mean of 60 thin sections ,
= standard deviation for 60 thin sections

if
S
The felsite fragments most commonly possess a xeno-

morphic granular texture and consist primarily of plagioclase

or potassium feldspar with only minor amounts of mafic

50

.Mpsum menu now poamEmm mmuwm mcH3onm
.moum unoomnpm can masmcficom 3mcoo3m& mo mm8 osmoaooo .5 ousmam

 

 

swdusso: unannouom Ioa poasado
and: I0 gonna: oawdm

season In oxen sonsao

nosoad< I» sonudm uuanoo

oudsaaon Iw gonad: oonuouuox

.hvspu

use: sundae» xoou hhdpsoafivou scan: some msofipouoa on» on nuouou «ecu honeys one

 

1 u 4 (1 J1
988.8. an a. 8 .u 2

mowuom o>oq oxdq omspuom
ouduoEOHNsoo songs: humqoo
accumccem spasm use oadnm nosuosoz

vacuuvsdm oaaa>upoouh

In
I:
In
IN
IH
many so» evades»

 

 

 

w<m 2<nmm3mu

Baa<m 3<zmm3mx

         

monmmDm mu<4

.cadpsso:
unannouom

 

 

51

minerals. The mafic minerals commonly occur as thin needles
or as small disseminated clots.

The mafic rock fragments in thin section most commonly
possess either a basaltic or Ophitic texture; however, those
possessing an amygdaloidal texture, although rare, may be
locally abundant. The mafic fragments are commonly hematized
to some degree. Often, both fresh and altered mafic frag-
ments will occur in the same thin section, but for the most
part fresher mafic fragments predominate.

The rhyolite fragments most commonly consist of uniformly
fine grained potassium feldspar and quartz. In some horizons
large phenocrysts of quartz or feldspar were observed in the
rhyolite rock fragments.

The granophyre fragments consist of radially fibrous or
vermicular quartz grains inclosed within potassium feldspar
host grains. Rock fragments showing textural characteristics
transitional between rhyolite and granOphyre were observed in
certain horizons.

Feldspars constitute approximately 5 percent of the volume
of the interflow sediment. Plagioclase is dominant, compris-
ing over 80 percent of the total feldspar volume. The
plagioclase consists most commonly of labradorite and ande-
sine, of which the majority present a fresh appearance.
Potassium feldspar is usually present in subordinate amounts..
The most abundant potassium feldspar is an untwinned variety.
Grains with gridiron twinning also occur but are rare.

Perthitic feldspar was not observed.

52

Unstrained quartz constitutes 3 percent of the total
volume of the interflow sediments. The quartz grains show
simple extinction or only slightly undulose extinction.

The larger grains are often embayed.
Vertical Compositional Variation in
the Calumet Area Section

An analysis of the vertical compositional variation in
the sedimentary succession at Calumet was made to detect
any changes in provenance during middle Keweenawan time.

Table 8 shows the vertical variation in the volume fre-
quency of the constituents in the sand phases of the Portage
Lake interflow sediments in the composite section at
Calumet, Michigan.

Because of the similarity in composition between the
Portage Lake and Copper Harbor sediments, and because of
their similar remanent magnetism, magnetic polarity, and
local interfingering; the model analysesof both units within
the Calumet section were combined to make a composite
Keweenawan sequence. Therefore, the data, detailed analyses,
and conclusions concerning the Portage Lake interflow sedi-
ments will be deferred and discussed along with the discus-
sion of the vertical variation within the CoPper Harbor
Conglomerate. CursOry examination of Table 8, however, sug-
gests that the detrital constituents of the Portage Lake inter-
flow sediments were derived from a volcanic source terrain

consisting of felsic, mafic, rhyolitic and locally granophyric

53

 

 

 

 

 

 

 

 

 

 

macayoom cflnu 5% mo coaumw>op pumpcmum u m
mcoapoom zany 5¢ mo cmoE u.m
mm.a 05.H 05.H ¢0.0I mm.0 00.0 No.0 oNHm and
ommuo>¢
0.¢ N.m 0.N 0.H v.0 0.m 0.0a ¢.Hm 0.m v.m H.0H ¢.5m H.mm H.mm o80H0>
, ommuo><
m.m m.0H I I I I 5.5 0.mm 5.N 0.0 m.0H N.0H 0.5 v.5m mm sceaom
m.¢ 5.0 I I 5.0 m.0 m.m 5.H I I m.~ ¢.m o.¢ m.mm mm
m.ma «.ma 0.5 H.v m.m m.m 0.0a 0.H¢ m.m 0.m m.¢a 5.5a m.m 5.0a 00m
N.5 0.m I I H.m v.m H.0H 0.mo I I m.ma 0.mH m.HH 0.0a Rd
0.0 m.m 0.m 0.0 ¢.0H ¢.am m.ma m.m I I m.mm m.m0 0.0 0.m 08m
0.H m.m m.¢ m.m m.¢ 0.m 5.0m m.a¢ m.m m.m 0.5m H.mm ¢.0H m.0H mm
5.m m.m 5.m 0.0 m.a m.v m.HH ¢.mm m.m m.m ¢.m 5.mv m.m m.m «m
0.N m.a I I m.m N.H m.ma 0.0m I I 0.5a 0.0m H.m ¢.0H mm
I I I I I I m.mm m.ma 0.0 m.¢ m.m 0.0 5.vm N.M5 0mm
5.0 m.0 5.0HI,0.5 0.0 0.H Wham m.mm H40 a.m b.0H D.m¢ 0.0 0.wa ma doe
m m w .W m m m W m m m M m MW.
ommao Hmmmpaom uuumso ouflmaom oumnm ouaaomnm oammz coufluom
Ionmam Esflmmmuom uocflmuumco Ioamuw pcoEHpom
.cmmwnoflz .uoEUHmU
um codpoom on» Ca mpcmm Scannouca oxmq ommuuom CH cofiymflum> Hmowuno> .m wanna

54

rocks. No contribution from a non volcanic borderland
could.be detected. The local provenance did not change

significantly during Portage Lake time.

Lateral Variation in Composition

In order to determine the character of lateral varia—
tion in different outcrops of the same horizon in the Portage
Lake sediments, samples were taken from various locations on
the Keweenawan Peninsula. The horizons chosen for sampling
and comparison were those known for their lateral persistence.
The Allouez Conglomerate and the St. Louis Conglomerate were
selected for analysis and comparison on this basis (Table 6,
Figure 7). Table 9 shows the average volume frequency of
constituents within the sandy phases of the Allouez Conglom-
erate in the Calumet, Allouez, and Delaware area, Keweenaw
Peninsula.

Comparison of the modal analyses of the Allouez Conglom-
erate shown in Table 9, indicates that the same lithic clasts
are present in each area, but that the average volume fre—
quency of each varies considerably from one locality to
another in most cases. Of the rock fragments, the mafic grains
show the least variation, whereas the felsite grains possess
the most variation between outcrOps. There also appears to
be a proportional relationship between high rhyolite and high
quartz volume frequency between outcrops of the Allouez Con-

glomerate.

 

55

Table 9. Lateral composigional variation within the Allouez
Conglomerate. Y represents the mean volume fre-
quency within the sandy phases of the Allouez Con-
glomerate in the Calumet, Allouez, and Delaware,
Michigan areas, respectively.

 

 

 

 

 

Constituent Calfimet Albfiuez Delfiware
Area _Area _Area
Y S Y S Y S
Mafic rock fragment 5.9 6.2 5.1 2.5 1.7 1.6
Rhyolite rock fragment 43.7 9.4 32.1 4.4 50.1 14.1
GranOphyre rock fragment 2.9 3.5 3.8 2.5 13.6 8.6
Felsite rock fragment 38.4 11.3 55.6 1.0 17.3 6.2
Quartz, simple 4.8 1.8 3.4 3.1 17.9 2.2
Potassium feldspar 0.4 5.7 - - - -
Plagioclase 3.9 3.7 — — - —
Total Number (N) Counts N = 1200 N = 400 N = 400

 

The second persistent sedimentary horizon chosen for
sampling and comparison of lateral variation within the
Portage Lake Lava Series was the St. Louis Conglomerate.
Samples of the St. Louis Conglomerate at Mass, and Donken
were collected and compared with those collected in the Calumet
area. Analysis of Table 10 shows that considerable variation
exists laterally between the volume frequency of the various
lithic constituents contained within the St. Louis Conglom-
erate. The table suggests that mafic grains decreased south-
westward toward Mass. In addition, Table 10 indicates that

the local felsic bodies were probably not rich in porphyritic

56

Table 10. Lateral compositiona1_variation within the St.
Louis Conglomerate. Y represents the mean volume
frequency within the sandy phases of the St. Louis
Conglomerate in the Mass, Donken, and Calumet,
Michigan areas respectively.

 

 

 

 

 

 

 

Mass Donken Calumet
Constituent Area Area Area
‘ Y S Y' S E S
Mafic rock fragment 3.8 2.3 33.3 27.3 85.9 4.6
Rhyolite rock fragment 30.0 18.0 8.7 3.3 2.4 2.8
Granophyre fragments — - 2.5 5.0 — -
Felsite rock fragments 64.3 15.0 55.2 29.4 1.7 2.5
Quartz - simple 1.1 1.6 - - 0.3 0.7
Potassium Feldspar - — — — — —
Plagioclase 0.8 1.0 0.3 0.6 9.7 4.9
Total number (N) of counts N = 400 N = 800 N = 800

 

quartz at this time as there does not appear to be any corre-
lation between high rhyolite and high quartz volume frequen-
cies.

Analysis of the lateral variation between outcrops of
the same sedimentary horizon within the Portage Lake Lava
Series shows that the same compositional constituents may be
present in each outcrOp throughout the extent of a given
.sedimentary horizon. .However, the analysis also indicates
that the volume frequency of any constituent may be expected
to vary appreciably relative to the volume frequencies of
the other constituents for any given exposure within the

sedimentary horizon.

57

The lateral variation in the volume frequencies of the
various compositional constituents within the sediment may
be due to the interaction between composition and tOpographic
relief within the source area. For instance, whenever the
source terrain composition is heterogenous in its extent,
and the tOpographic relief does not vary in general character
from one sector of the source area to another, then any
lateral variation in the volume frequencies of the various
constituents within the sediment reflect a real difference
in provenance.

However, if the composition of the source terrain is
essentially homogenous in its heterogeneity throughout its
extent, but the tOpographic relief shows real changes in
character from one sector to another, then any apparent
lateral variation in volume frequency of the various constit—
uents is probably weighted in favor of the specific composi—
tion of the higher elevations. The higher elevations would
be expected to shed a disporportionately greater volume of
sediment than the lower elevations. Thus lateral variation
in the volume frequencies reflects only an apparent differ—
ence in provenance.

,Regardless of the reason for variation in volume fre-
quencies laterally, it may be said that the source for the
Portage Lake interflow sediments was a volcanic source ter-

rain consisting of mafic, felsic, rhyolitic and locally

granophyric rocks.

58

The Copper Harbor Conglomerate

The Copper Harbor Conglomerate lies with apparent con—
formity upon the lavas of the Portage Lake Series and
locally interfingers with them (Cornwall, 1955). The direc-
tion of magnetization of Middle Keweenawan Portage Lake
Lavas is not significantly different from that of the suppos-
edly Upper Keweenawan Copper Harbor Conglomerate (Du Bois,
1962, Book, 1968), indicating that a uniform magnetic field
direction existed during the emplacement of the Portage Lake
Lava Series and extended through deposition of the Copper
Harbor Conglomerate.

The COpper Harbor may be divided into an upper unit,
the Outer Conglomerate, and a lower unit, the Great Conglom—
erate, by an intervening lava sequence known as the Lake
Shore Traps. These three units are present as such in the
Copper Harbor conglomerate exposures from Houghton eastward
to the tip of the peninsula. However, south of Houghton, the
Great Conglomerate thins appreciably over a high in the
Portage Lake Lava series.

The mafic Lake Shore Traps within the COpper Harbor
Conglomerate succession attain their greatest development at
the northeastern end of the Keweenaw Peninsula. Their thick-
ness decreases to the southwest from a maximum of 2000 feet
near the tip of the peninsula (Cornwall, 1955) to zero at
Houghton (Cornwall and Wright, 1956) but reappear farther

to the southwest. Only the Outer Conglomerate member and

59

the tOp of the underlying Lake Shore Trap are present farther
south in the Porcupine Mountain area. In northern Wisconsin,
the Outer Conglomerate is the only member of the COpper
Harbor Conglomerate present.

The COpper Harbor Conglomerate varies considerably in
thickness. Its thickness ranges from 6,000 feet at the tip
of the Keweenaw Peninsula to about 2,000 feet in the
vicinity of Houghton, to about 5000 feet in Wisconsin (White
and Wright, 1960). However, in the Porcupine Mountain area
the unit thins to about 300 feet over a thick pile of rhyolite
flows at the t0p of the underlying Portage Lake series.

White and Wright (1960) suggest that these rhyolitic flows
formed a t0pographic high around a volcanic center on the
surface. This rhyolitic high in the Porcupine Mountain area
apparently persisted as a center of dispersal for siliceous

sediments until it was buried in Outer Conglomerate time.

The Conglomerates

The sedimentary portion of the Copper Harbor consists
primarily of thick conglomerates interbedded with sandstone
lenses or locally of sandstone beds with scattered pebbles
and conglomerate lenses. Individual sandstone units as
thick as 400 to 700 feet have been reported (Cornwall, 1954,C;
White and Wright, 1960).

The composition of the conglomeratic phase of the Copper
Harbor varies significantly, both vertically and laterally

in the Keweenawan Peninsular region. On Manibou Island, the

6O

conglomerate phase of the COpper Harbor consists primarily
of coarse basalt, andesite and-rhyolite rock fragments,

but notable amounts of pre—Keweenawan quartzite, granite,
syenite and schist clasts appear near the tOp of the Great
Conglomerate (Cornwall and White, 1955). Throughout much
of the Keweenaw Peninsula, however, the conglomerate compo-
sition consists primarily of felsitic and mafic detritus
derived from the Keweenawan terrain. Lateral comparison of
clast composition suggests that the percentage of mafic
fragments decrease to the southwest as the felsitic debris
increases in that direction. At the southwestern end of
the Keweenaw Peninsula in Wisconsin, these volcanic frag-
ments appear to decrease upward, while preuKeweenawan
metasedimentary, grantic, and vien quartz clasts increase
in abundance upward in the section (Hite, 1968).

The appearance of the pre-Keweenawan metasedimentary
and plutonic clasts in the upper Copper Harbor Conglomerate
on both Manitou Island and the southwestern portion of the
Keweenawan Peninsula suggest that the source terrain vol-
canic cover was denuded in these areas exposing the under-
lying pre-Keweenawan rocks to the effects of erosion at this
time.

Studies by Hamilton (1965) indicate that the non volcanic
detritus in the southern areas of the Peninsula were derived
from a source terrain located south and southwest of the

present outcrOp area. According to Hamblin and Horner (1961)

61

lithologic studies and cross stratification measurements
show that the Huron Mountain area was an important center

of sediment dispersal during Upper Keweenawan time.

Sandstone

The sandstone within the Copper Harbor occurs as small
discontinuous lenses a few inches thick to more continuous
layers 400 or more feet in thickness.

Seventy-three thin sections were made from samples
collected within the sandy horizons and from the sandy
matrix of the conglomerates, which locally is abundant.
Forty-two of these thin sections were taken from core and
outcrop samplings along the section of the Copper Harbor
Conglomerate in the Calumet area, four from the section at
Horseshoe Harbor, eight from the section at Copper Harbor,
five from exposures in the Glazon Lake area, four from the
section exposes near Eagle Harbor, and ten from a core in
the Porcupine Mountain area near White Pine, Michigan
(Figure 7).

The mafic rock fragments comprise approximately 24
percent of the total volume of the sandy phases within the
Copper Harbor. They are predominantly basaltic in texture,
hematized, and have a more weathered and opaque appearance
than similar types in the underlying Portage Lake Lava
Series.

The felsite, rhyolite and granOphyre fragments consti-

tute approximately 24 percent, 30 percent, and 4 percent,

62

Table 11. Mean composition of the sandy phases within the
C0pper Harbor Conglomerate. The mean Composi—
tion of the sandy phases of the Copper Harbor
Conglomerate as determined by the modal analyses
of 73 thin sections.

 

 

Volume Frequency

 

 

 

Constituents pg%)
3? s

Rhyolite rock fragments 30.5 21.3
Mafic rock fragments 24.2 22.0
Felsite rock fragments 24.2 18.5
Plagioclase 5.9 6.5
Unstrained quartz 5.5 8.0
Potassium feldspar 3.8 5.2
Granophyre rock fragments 3.7 7.1
Undulatory quartz 1.8 4.4
Polycrystalline quartz 0.3 0.7
Total % 99.9

 

mean volume frequency of 73 thin sections

f
S standard deviation of 73 thin sections

respectively, of the total volume of the sandy phases within
the Copper Harbor Conglomerate, and have the same attributes
as these fragments found in the underlying interflow sedi-
ments of the Portage Lake Lava Series.

Detrital, quartz grains comprise approximately 5.5
percent of the total volume of the sandy phases within the
Copper Harbor succession. In most sections the quartz grains

show only slight or no undulose extinction, however, there

63

are two exceptions: strongly undulose quartz grains occur
near the top of Great Conglomerate in the Calumet section
and throughout the Outer ConglOmerate in the Porcupine
Mountain section, constituting 0.8 and 10.3, respectively,
of the total volume frequency of detrital constituents
within those sections.

Feldspars constitute approximately 10 percent of the
total volume of the sandy phases within the Copper Harbor
Conglomerate. PlagioclaSe is the dominant feldspar com-
prising approximately 61 percent of the total feldspar
volume. The plagioclase grains consist most commonly of
labradorete and andesine; they commonly show a more altered
aspect than those in the underlying Portage Lake Series.

Potassium feldspars make up 3.8 percent of the total
volume of the sandy phases. The most abundant potassium
feldspar is an untwinned variety. Perthitic feldspars occur
in the Porcupine Mountain section and in the Calumet section
near the top of the Great Conglomerate just beneath the

Lake Shore Trap.

Vertical Compositional Variation

Evaluation of the vertical compositional variation
within the sandy phases of the composite section containing
the Portage Lake interflow sediments and the COpper Harbor
Conglomerate at Calumet (Tables 7 and 11), and within the
Copper Harbor Conglomerate sections at CoPper Harbor (Table

12), and at Porcupine Mountain (Table 13) was made to detect

611

T“‘

..$I 1“ “MN"..QH .

sonauoa humusoawuon some :0 use-mum newuuw>oo caduceus husonuouu ossao> u m

sonuuon hauusOvaon none awnuwt hosesbouu Olaao> sees I w

 

 

 

 

 

 

 

 

e0.0 00.0 mn.~ 05.0 05.0 00.0 00.0 00.0 00.0 «:0 oounuww

0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 m.0 0.v 0.00 0.00 0.0 0.0 5.00 0.00 0.1 5.00000uomwcuwwmmn

«.0 «.0 0.0 n4 I I I I 0.0 0.0 0.00 0.00 0.0 0.0 0.50 0J0 0.0 0.5 000 06.3.00

¢.A 0.0 5.4 0.0 I I I I 0.0 «.0 0.0 v.00 0.0 0.00 0.40 0.00 0.0 v.00 100

0.v 0.0 n.0 0.0 I I 0.0 0.0 0.0 0.0 0.00 5.0m v.5 0.0 0.00 (.00 A.v 0.0 000

0.v 0.0 5.0 0.4 I I I I 0.00 0.0 «.50 0.00 I I 0.00 0.00 0.0 0.00 000

0.0 5.00 5.0 0.5 I I I I 0.0 0.0 0.00 0.00 0.0 v.0 0.00 5.00 0.0 0.0 000

0.v 0.0 0.0 0.0 I I I I m.~ n.0 5.0a 0.1v 0.v 0.0 0.V0 5.00 0.0 v.0H 000

5.0 5.0 0.0 0.0 I I I I 0.5 0.0 I I I I «.0 0.00 0.0 5.50 A00

0.0 0.0 0.¢ «.0 I I ¢.H 0.0 v.0 0.0 0.00 5.00 0.5 0.0 n.0n 0.00 5.0 0.00 000

n.00 H.0d 0.v 0.0 I I 0.0 5.0 0.0 0.00 I I 0.0 0.0 0.00 0.00 0.5 o.vn MUG

0.0 0.0 0.5 0.50 «.0 0.0 v.0 0.v H.0 0.0a v.0 0.0 I I 5.1 0.vn 0.04 5.00 000 009
m m m m m m m m m m m m m m m m. m m

oumu udonuaom uuudsa Rosana uuudsa ouAmAMMI ouano ouwaoxnm Ilmwwmmll couwuom
lawmeam nBAnwouom Haunmuuxaom encasvco vocwnucha Iosqu xuoucoaavom

 

 

.uoasauo um

noduuon on» :0 ounuoaoamcou sonudm Monaco can no nonmnm accu-

sanuwi sawuawun> Assamuwnonloo Hauwuuo>

.00 OdAflh

65

Table 13. Vertical compositional variation within the sandy
phases of the Copper Harbor Conglomerate in the
section at Copper Harbor.

 

 

 

 

 

 

 

Rhyo- 'Plagio—

Sedimentary Mafic lite Felsite clase

Horizon (%) (76) (%) (‘36)

TOP CH-lA 67.1 1.1 31.8 - ’7
CH—lB 62.8 2.6 34.6 — F
CH—6l 68.5 12.3 19.2 - g
CH—62 76.6 1.3 22.1 - ;
cn-71 79.2 3.9 11.7 5.2 ;
CH-72 72.8 2.5 19.8 4.9
CH- 8 72.1 27.9 — —

BOTTOM CH—lO 69.4 8.1 17.7 4.8

Avg. Vol. Freq.(%) 73.1 9.3 15.1 2.5

Standard Deviation 4.1 10.0 8.2 2.7

Average Phi Size 0.36 0.60 0.57 1.14

 

any changes in provenance occurring during Keweenawan time

as represented by these units.

Least square regression analysis and the 2 x 2 chi

square contingency method were used to evaluate the vertical

variation among the volume frequencies of the various

detrital constituents. The analysis procedure is identical

to that employed in evaluating the vertical variation within

the Mamainse Point section.

Analyses of the plots of the (observed-expected) volume

frequency with depth reveal no significant residual effects

66

for the mafic, felsite, and granOphyre rock fragments or

the quartz or plagioclase clasts upward within the combined
section at Calumet. However, significant residuals (szJ>0.05)
were found for both the rhyolite fragments and the potassium
feldspar clasts (Figures 8 and 9).

Thus it may be concluded that only the rhyolite and fl
potassium feldspar show significant changes in their volume
frequencies upward in the section; both increasing in abund-
ance upward. Since there is no real change in the volume

frequencies of the other compositional constituents, then

 

any apparent changes in the volume frequencies of these
constituents vertically in the section must be due largely
to the grain-size effect and the capriceous nature of the

depositional environment.

“Lateral Compositional Variation

The six vertical sections through the Copper Harbor
Conglomerate which were studied in this report, from south
to north, respectively, are in the vicinity of the Porcupine
Mountain, Calumet, Eagle Harbor, Glazon Lake, Copper Harbor
and Horseshoe Harbor (Figure 7).

The lateral variability of the Copper Harbor Conglom-
erate is shown in Table 15 and suggests the existence of
local compositional variations.

In the Horseshoe Harbor, Copper Harbor, and Glazon Lake
areas, mafic clasts are the most abundant detrital sand

constituent ranging from 44 to 73 percent. In the Calumet

 

Figure 8.

67

Graphs of the vertical compositional variation
within the Portage Lake interflow sediments.
Graphs show the residuals resulting when the
(observed—expected) volume frequencies of the
various detrital constituents are plotted
versus their depth in the composite Keweenawan
interflow section at Calumet. Graph A is the
plot for mafic clasts, Graph B is rhyolite
clasts, Graph C is for felsite clasts, and
Graph D is for granophyre clasts.

68

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“O- m m.
m a” m.
A mr B
II"
“I ' ' ' ‘6' o o (c
m g . g m . “
u 0 on o o o- I” n o n o D. .
' o o I o 0
I o u o o a o o O
o o
‘0 O. D II- D O .0 . C
O . ”I O O .0. .
I o oo- ' q 0- o
I ‘ ' Q 0
I“ - o n o a
~ 0 O C as 0 .- ' o u I o I
~ 0 o o 0 ~ 0 o 0- o 0
~ o 00. a ’ ~ 00- o 0
RI x
“C o o o 5 m 0 on
m o O 0 ~ 0 a
~ - . ’- ‘4 o O O
.4 o p 0.. I. ~ 0 on a
o o N o. ll 0.
'10 ‘30 O 0 IO ‘40 - .0 '80 O ‘20 0”
mam-ammo) man: my or m was (maven-m) mm MW 0' um”: am
an. mun: ‘0.
m H”: m
C M? D
I6"
“I 0 CC,
‘C‘ o
a o. o. o ’J- n 0.. o o
I I
. 0 o o I o 1
II o o I" o
’ o O O .
.:. . ’1.’ " n- ‘2 “'.. '... .
I O O O I I. O
I ' ‘ I. "
K ' F“ 0
fl 0 o o o. o o 9- "
~ 0 do I O. m .0 I
~ no 000- o 1 "1 000-
K K
N o o ‘_ I:
~ 0 o 0 ~
n e on ’- M 0
n no II o o I- P. ”(on
o o 0’ P! _
- .0 -zo o . «to "'0 4° 0 no no
mum mm) mm mommy U Fa’IYE MS}! Mum-ammo) malt mum W m m

Figure 8

 

Figure 9.

69

Graphs of the vertical compositional variation
within the Portage Lake interflow sediments.
Graphs show the residuals resulting when the
(observed—expected) volume frequencies of the
various detrital constituents are plotted

versus their depth in the composite Keweenawan
interflow section at Calumet. Graph E is the plot
for unstrained quartz clasts, Graph F is for
potassium feldspar clasts and Graph G is for
plagioclase clasts.

mm- do
much In
an
J
u
“1
[3" u
I
a
II
l
n- ’2.
I.
I.
m
’- p
as
,_ In
E
III:
‘-
N
I“
,-
I- n
o. n

 

 

 

I
“mum” mm lacuna or W cm

 

 

7O

 

 

 

 

 

 

 

ram an
arm's Ion.
E an F
I -
o on ‘ ‘Cf‘ - o o
o o o “t . . .
1" ’1 cod 0 o
I a
I o
2’ o.
I .
. u- 19 7::
52 -
m
g- N
N o u
o H 00.
1-
RI
‘° Ml:
‘-
~4
’. “I .
t- a.
o. a i
'10 “'0 '00 'IO 0 0‘0 "O “I.
MKD°M mun: ”demm mm
"mm 1!.
awn-0‘s m
an
I... ‘0 o I o
‘c‘ 0 o
1" n a. o. o
I o
I o O o o
1‘ . .0 a.
II' 7 o. be e o
I 00 o o
I. '
In
9 ~ '
u 0 can
'. ~ -
g
u
‘.
N o o
’- M o .
I- ~ . .-
0. ~ 0 o o.
'10 ~I° 0 ‘IO '10
W’ ) M It"!!! 'W m

Figure 9

 

afi‘ ;
L
I;
a
In

71

confines some sway“: :00udw>oo oudpsduo husonvouu 085009 a 0

souwuon some swguws hucuaoouu oss~o> 8008 u 0

 

 

 

 

50.0 05.0 «0.0 «0.0 0v.~ «0.0 00.0 05.0 00.0 0N00 050 omauo><
0.0 0.v 0.0 0.5 5.0 0.0 0.v 0.0 0.0 noduuwhon vudvnnum
0.0 0.0 v.~ 0.08 m.o~ m.- «.0 n.ma 5.0a aucoavouu

.~o> .m>¢
0.50 0.00 0.0 0.0 «.50 0.00 I 0.0 0.50 oaImz :08800
0.00 0.0 0.0 0.0 0.00 0.vu I 0.0 v.00 mlmx
I 5.0 I 0.0 0.00 v.00 0.1 v.00 0.0a 0Im3
5.0 0.0 0.0 0.0 0.00 5.00 I 0.00 0.00 5I03
0.¢ 0.0 0.0 5.00 0.00 5.50 I 0.va 5.0 0I03
5.v 5.v 0.0 v.0 0.00 0.00 I 0.00 H.¢H mum:
0.0 0.0 I I 0.0 0.00 0.0 0.00 0.50 VIms
0.0 0.0 «.0 0.00 0.00 0.00 v.0 v.00 0.0a fiIva
0.0 0.0 I 0.00 0.50 0.00 I 0.00 0.0 0I03
v.5 0.00 I 0.v 0.00 0.00 0.¢H 0.00 0.1 0I03 009
$3 35 38 0x.
Axe nadmcsmm uuumao «guano "guano 1x0 1x0 Axe Axe couauom
onu00000MA0 ssfinmnuom .umhuOAHom 00005000 uocamuunca ouaaaom ouicmocnuo ouwuo>£0 000m: inaucoswvom

 

 

us cofiuooo any :0

.swoucsox.ocwmsouom

oumuoeo~0cou nonuom Hoodoo 0:0 00 «omega apnea on» saga“: soflunwud> HeGOwuwuomsOo meowuuo> .VA canoe

'72

IEJJ&

3...? a.

 

 

 

 

 

 

 

 

 

 

 

 

00.0 00.0 00.0 00.0 00.0 «0.0 unwauu
Hmuosaz Huuwuuon
3.0 3.6 $6 on... 86 86 nucasmnum zoom
cowuoom sowuoow sawuoom :Owuoom cowuoom swmuqsoz Guam Mam ommuo><
nonumm oonmomuom uonumm uommou oxmq souuao Manama oamom uosfldmu ocwmauuom
Amado an uouosmfla o muo>¢
000 0000 0000 000 0000 0000 .z. deuce
0.0 5.0 5.0 0.0 0.0 0.0 5.04 5.00 0.0 0.5 0.0 0.0 oomaoowmmHm
I I I I 0.0 0.0 0.0 0.0 0.0 0.0 0.v 0.0 addendum sad-ouuom
I I I I I I I I «.0 0.0 0.0 v.0 sundae
oswnaaunhuuudom
I I I I I I I I 0.0 0.0 0.5 0.00 nausea ouoduonn
I I I I 0.0 0.0 0.0 0.0 5.v 0.0 5.0 0.00 sundae nonwmuuuab
0.00 0.v0 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0 0.00 ouonOh
I I I I ~4 0.0 a... 0.0 0.... 0.0 o... 4.0 3.3.0055
0.0 0.0 0.00 0.0 0.0 0.00 0.00 v.00 5.00 0.0¢ 0.0 0.00 ouwdoanm
0.50 v.v¢ a.v 0.05 0.00 0.00 5.00 0.00 0.v 5.0a 0.0 5.00 came:
0 .m 0 m m m 0 m 0 m 0 m
scauuowI sawuoom IldmwmwomIl sawuoom Imewwww samusmmn ucoauwuusoo
Hanson vogue-Mom Monumm Hoodoo oxmq consao nonumm 000nm noesamo oswmsuuom

 

 

.oumuosoamsoo nonunm nommou ogu mo mousse meson one saga“: sowuuwum> sowuwaomaoo Hauouma .00 canes

73

and Eagle Harbor areas rhyolite clasts rank first in abund-
ance, averaging between 44 to 40 percent. Figure 10
indicates that an inverse relationship exists between the
mafic and rhyolite volume frequencies.

Felsite rock fragments are present in all sections and
range in average volume frequency from a low of 19 percent
in the Calumet area to a high of almost 55 percent in the
vicinity of Horseshoe Harbor.

Granophyre grains are most abundant in the Calumet and
Eagle Harbor sections where the average volume is approxi-
mately 3.7 percent, but elsewhere range in abundance from O
(COpper Harbor) to 2.4 percent in the Porcupine Mountains.
This distribution suggests there were several nearby centers
of dispersal: the mafic constituents appear to increase to
a maximum in volume frequency northeastward, while the rhy-
olite and granophyre clasts reach a maximum frequency in
the Calumet-Eagle Harbor area (Figure 10).

Unstrained quartz grains are most abundant in the
Porcupine Mountain and Calumet sections with an average
volume frequency of 20 and 4.9 percent, respectively.
Elsewhere unstrained quartz grains occur in amounts of 1
percent or less. Undalose quartz grains were observed only
in the Porcupine Mountain (10.3 percent) and Calumet sections
(0.8 percent).

The unstrained quartz, undulose quartz, polycrystalline

quartz, and potassium feldspar are most abundant in the

 

74

A LAT‘ML ”In?!“ H {IN/C ”'11: CW armour/av ”mil
0 rut CDPFCR m mmurt

 

 

 

 

1D
-— H mm: “A”!
-— R wart CM"!
‘0
D
x;
E to
2
a n
H
t
3
so
lo
0
on c tn .4. an an
Jtcrrow Loan/ow
B urtm mum uv ummu mm ecu-Maura mnvm M
° coom um Noam
O —O warm em
to ”.....u. wuumv m
---n mvuvm WI
_ ...K ”moan mm
2 u. ...-a manual
g
z to
h
g I
3‘
>
O

 

 

 

 

toe-m mm
M 0 mm m ‘L l “A!” m
C “UN" Cl 0 W ”M
(N I.“ woos av ' mu“ m

Figure 10. Lateral compositional variation within the sandy
phases of the COpper Harbor Conglomerate.

 

75

Porcupine Mountain area where the volume of monomineralic
grains within the sediment collectively makes up 44 percent
of the detrital composition. The Porcupine Mountain section
is stratigraphically higher than the more northern sections
on the Keweenaw peninsula. The greater percentage of
detrital monomineralic grains within the Porcupine Mountain
section, therefore, may indicate a more advanced stage of
weathering and breakdown of the rocks in the source area,
resulting in a greater percentage of monomineralic grains
within the sedimentary detritus in late COpper Harbor time.

Throughout the sections studied on the Keweenaw peninsula
the amount of feldspar and quartz appears to be inversely
related to the amount of mafic debris. Where the volume of
mafic detrital grains are high, low values of feldspar occur
due to the failure of the lava flows to decompose and release
the individual grains. Low quartz volumes occur in any case
because of the deficiency of quartz in a mafic source rock.
Grain size undoubtedly exerts an affect on lateral composi-
tional variation, but it is thought that the major factor
causing the lateral variation is a difference in provenance
laterally.

Thus the majority of the Upper Keweenawan interflow
sediments were derived from a Keweenawan age volcanic source
terrain consisting of mafic, felsitic, rhyolitic, and grano-
phyric rocks. However, the relative scarcity of unstrained,

strained, and polycrystalline quartz in the northernmost

 

76

sections and their increasing abundance southwestward in

the Porcupine Mountain and Calumet sections (especially just
below the Lake Shore Trap horizon) suggests incursion of
detritus locally from a pre—Keweenawan plutonic and/or meta—
morphic source terrain.

The source area borderland therefore probably consisted
of plutonic and/or metamorphic rocks mantled by volcanic
rocks. The volcanic rock cover may have been breached near
the Mellen to Porcupine Mountain area and again near the
Manitou Island area, thus exposing the underlying plutonic
and/or metamorphic rocks to erosion.

The source area borderland was probably located just
southward of the presently exposed Keweenawan outcrops and
is represented by the occurrence of abundant lower Keweenawan
diabase dikes exposed in Animikean metasedimentary and
plutonic rocks and a gabbro—granOphyre complex which intrudes
Keweenawan and possibly pre—Keweenawan rocks (Aldrich, 1929;
Tyler, 1940; Leighton, 1954; White, 1966). The bordering
source area front would at least follow a line connecting
the.Mellen—Huron.Mountain area. According to Hamblin and
Horner (1961) lithologic studies and cross stratification
measurements show that the Huron Mountain area was an
important center of sediment dispersal during upper Keweenawan

time.

Sedimentary Synopsis

 

The compositional immaturity of the interflow sedi-
ments within the Portage Lake Lava Series and the Copper
Harbor Conglomerate easily facilitates the determination of
the parent material. It appears evident that the middle
Keweenawan and the majority of the upper Keweenawan inter—
flow sediments were derived from a Keweenawan age volcanic
source terrain consisting of mafic, felsitic, rhyolitic and
granOphyric rocks. However, the presence of appreciable
undulose quartz in the Porcupine Mountain and Calumet sec-
tions of the upper C0pper Harbor Conglomerate just below the
Lake Shore Trap horizon suggests incursion of detritus
locally from a pre—Keweenawan plutonic and/or metamorphic
source terrain.

The source area, dominated by an extensive and thick
sequence of volcanic rocks was tectonically unstable which
encouraged the production of large amounts of detritus from
local sources during periods of little or no volcanic.
activity. The coarse texture of the Keweenawan interflow
sediments and their predominantly volcanic composition bear
witness that the Keweenawan lavas, acting as a source,
existed as local uplands.

The deposition and preservation of such a thick succes—
sion of flows and sediments indicates a rapidly subsiding
basin in which continued downwarping produced a topographic

depression into which not only lava flows moved but streams

78

flowed and deposited detritus shed from borderlands during
periods of volcanic quiescence.

An apparent difference in volume frequency exists be—
tween the composition of the middle and upper Keweenawan lava
flows and the composition of the interflow sediments con—
tained within them. According to Butler, Burbank et al.
(1929), Cornwall (1951) the middle Keweenawan lavas consist
of approximately 92 percent or more basalt and andesite rocks

and locally no more than 8 percent rhyolitic (0.5%p and

..1 ‘2 . L: (I'L. :‘."'.-1.‘u-I'A .0 u" , .
I

«I .

mu . 'L '.". ' "Io

 

felsitic rocks (7.5%). Yet the average composition of the
interflow sediments consists of 30 percent felsite, 29 per-
cent rhyolite and only 24 percent mafic fragments. All
fragments show little effects of extensive weathering. This
disparity suggests two possible alternative explanations:
either (1) the more silicic middle Keweenawan rock bodies
locally formed the higher elevations, consequently they were
subjected to more severe erosion than the mafic flows and
thus furnished a dispr0portionate amount of sediment, or
(2) the interflow sediments are not of immediate derivation
from middle Keweenaw rocks but came from long-lived uplands
composed of a higher percentage of felsitic-rhyolitic rocks.
Hubbard (1967, 1968) suggests that the rocks of the
South Range Series; existing locally as upland highs, fur-
nished the sediment intercalated with the middle Keweenawan
lava flows during Portage Lake and Copper Harbor time.

His deductions are based upon his observation that (l) the

79

mafic pebbles within the Portage Lake sediments are typically
fine grained--no Ophitic pebbles having been found, whereas
the Portage Lake flows tend to be much coarser with Ophitic
flows being common throughout its succession. The presence
of abundant pebbles typical of the South Range Series sug-'

gests to him that the South Range was therefore the source

1%
for the middle Keweenawan interflow sediments. ,  
2. The interpretation of the magnetic anomaly present ;
in the covered interval separating the South Range Series E
from the overlying Portage Lake Series suggests that this or J

 

interval is underlain by felsites and sedimentary rock
(Hubbard, 1968). This is corroborated by the observation
that outcrOps of felsite are reported in Wisconsin along the
strike of the covered interval (Hubbard, 1968). Here also,
Aldrich (1929) describes rhyolite flows with rhyolite
conglomerate resting upon them, in which the conglomerate
appears to be derived from the flows. Thus, the thick
succession of rhyolite postulated to occur in the covered
interval would be conveniently located to supply the abundant
felsic detritus present in the middle Keweenawan sediments.

(3) The unconformable relationship between the South
Range Series (of lower Keweenawan age) and the Portage Lake
Series (of middle Keweenawan age) agrees with paleomagnetic
data which indicate reversed polarity for the former and
normal polarity for the latter, and

(4) The dips of the South Range Series indicate that

they were part of an anticlinal flexure (Hubbard, 1968)

80

whose formation may be related to the intrusion of a large
gabbroic-granOphyre complex in the Mellen area of Wisconsin.
But regardless of the identity of the source rock, the
existence of a nearby source terrain of some height and
relief located to the south of the present Keweenawan expo-
sures is suggested by the coarseness, thickness, and direc-
tions of cross bedding and imbrication contained within the
Portage Lake and Copper Harbor sedimentary sequences (White i

and Wright, 1960; Hamilton, 1965; Hite, 1968).

 

The source terrain was probably related to the intru- . J
sions of the gabbro-granOphyric complex at Mellen, Wisconsin.
The scale of these intrusions suggests they were associated
with tectonic uplift in areas which were mantled with
Keweenawan volcanic rocks. Furthermore unroofing and much
erosion has occurred in this area as indicated by the exposure
of an extensive and widespread diabase dike swarm of lower
Keweenawan age within premKeweenawan metasedimentary and
plutonic rocks in the Huron Mountain area to the south of
present Keweenawan exposures.

In summary, it is suggested that a volcanic sequence
south of the present Keweenawan outcrops was uplifted and
was being eroded during Portage Lake and Copper Harbor time.
The thick succession of rhyolite postulated to occur in the
covered interval between the Portage Lake Series and South
Range Series would be conveniently located to supply the

abundant felsic detritus present in Portage Lake and Copper

81

Harbor sedimentary zones. The thick sequence of rhyolitic
flows in the Porcupine Mountain area, which existed as a
tOpographic high throughout much of Copper Harbor time
(White in Hubbard, 1968) and possibly earlier, may have also
furnished felsic debris to be incorporated in the interflow

sediments.

Paleogeography and Sedimentation i
The coarseness and compositional immaturity of the con-
glomerates and sands comprising the Portage Lake and Copper

Harbor sequences indicate a nearby source of some relief. .&

 

These attributes also suggest that the sediment underwent a
short transportational history and rapid deposition in a
sedimentary basin which was rapidly sinking.

The marginal uplift of the volcanic terrain coupled
with subsidence of the central portions initiated erosion,
transportation, and deposition of volcanic conglomerates
and arenites by a fluvial system on a piedmont fan. Shales
are rare, indicating that the associated finer sized grains
must have bypassed this region and were deposited elsewhere,
probably as a flood plain sequence in the more central por—
tion of the basin.

The coarse nature of the sands and conglomerates, the
presence locally of channels with disconformable bases, high
angle cross strata, and the typically poor sorting and
rounding generally present in the sandy phases suggests that

the interflow sediments comprising both the Portage Lake and

82

Copper Harbor series were transported by a fluvial system
characterized by relatively high energy conditions and high
gradients. Uplift of the highlands plus subsidence of the
basin maintained a well—drained, dissected surface consisting
of channels and flood plains. The presence of standing
bodies of water is affirmed by the presence of algal—ball-
like bodies in the Porcupine.Mountain area (Hite, 1960) and
stromatolites (Cornwall, 1951) in the section exposed at
Horseshoe Harbor. The upland was probably rugged but pos—
sessed only moderately high relief.

The evidence presented in this study suggests that the
most likely location for the source area borderland of both
the south shore Keweenawan voICanic rocks and their inter-
calated sediments was just southward of the presently exposed
Keweenawan outcrops. This area is represented by the occur-
rence of abundant lower Keweenawan diabase dikes exposed in
Animikean metasedimentary and plutonic rocks and a gabbro-
granophyre complex which intrudes Keweenawan and possibly

also pre-Keweenawan rocks (Tyler, 1940; White, 1966).

 

CHAPTER V

THE NORTHWEST SHORE REGION OF LAKE SUPERIOR

The northwest shore region of Lake Superior, extending
from Duluth to Schreiber forms a crescent shaped area occu—
pied by the Sibley-Osler Series and the North Shore Volcanics
which comprises the northwest flank of the Lake Superior
Syncline (Figure 9). It contains Keweenawan sequences which
when taken together represent all three divisions of
Keweenawan rocks. Sedimentary sequences of lower Keweenawan
age are exposed in the Thunder Bay area, where they are known
as the Sibley Series, and in northeastern Minnesota, where
they are represented by the Puckwunge formation. The Osler
Lava Series of lower and possibly middle Keweenawan age
overlaps the Sibley Series. The North Shore Volcanic Group
exposed along the Lake Superior shore of Minnesota is of
lower and middle Keweenawan age. The section on Isle Royale
consists of middle Keweenawan lavas overlain by an upper
Keweenawan sedimentary sequence.

The Keweenawan rocks of the northwest shore are in-
truded by the Duluth Gabbro complex of lower to upper
Keweenawan age, the Logan diabase dikes largely of middle

Keweenawan age (Figure 11).

83

 

84

 

. . .00500 .coouw
.0500 0090mm u0000 .m00mm HouMMV H00Homzm oxmq
mo c00mmu muogm umo3£uuoc may mo mas 0mo0000omw .00 ousm0m

 

30 «a E 0069....

o. o O—
: ngom m; H.033

 

EHKSUmm—m Mann 0:

mAAHm 24600

ZOHB<ImOL moz:3¥ozm

mmHmmm >m0m0m

mmHKmm mmqmo

IIEJBBEII

0c mqu<ogo> mmoxm mfimoz

.2 34 am wing 5030.0

 

 

      

mq<>om mamH

><m mmazbzfi

 

2m<3n mIHQ mm<m<ua
xMJAXOQ xHDJDQ
mon<qu> mmoxm thoz
m002<040> mq<>om meH

z<3<zmMBmz muoodx
m9<mmzoqwzoo m4<wom mumH

’
.
.. 52253.»... .595
. iiaxfiia muss.

 

 

 

 

85

Detailed mapping of the North Shore Volcanic Group began
with Sandberg's (1938) description of the succession between
Duluth and Two Harbors. Additional stratigraphic studies
elsewhere in the sequence have been made by Schwartz (1959,
1949), Grogan (1940), Gehman (1957), Grout et a1. (1949), and
Green (1966, 1968, a. b., 1970). The section on Isle Royale
has been mapped and described by Lane (1898, 1911), Huber
(1971), and Wolff (1968). The Sibley and Osler Series were

mapped and described by Tanton (1931).

 

The Isle Royale Sequence

 

Isle ROyale possesses the only Keweenawan section on the
northwest limit of the Lake Superior syncline which exhibits
the top of the volcanic sequence and also an overlying
Keweenawan sedimentary succession. ”The shape of the island
closely follows the strike of the Keweenawan rocks of which
it is composed. The bulk of Isle Royale is occupied by the
volcanic sequence (which is similar to the Portage Lake Lava
Series) and the remainder by the overlying sedimentary unit
(much like the Copper Harbor Conglomerate) which is limited
to the southwest part of the island (Lane 1898, 1911; Huber,
1971).

According to Huber (1971), the only major structural
distortionon the island involves the upper portion of the
'Portage Lake Lava Series, which is warped around a node of

apparent uplift and fracturing. Evidence of the presence of

86

nearby uplift seems to be a typical attribute to most

Keweenawan sections exPosed around Lake Superior.

The Isle Royale Volcanic Series

The volcanic sequence on Isle Royale consists of a series
of basaltic, andesitic, and rare rhyolitic flows, with inter-
bedded sandstones, conglomerates and tuffaceous rocks. The {—5
fragmental rocks make up less than 10 percent of the lava

series on the island (Huber, 1971). Huber (1971) also recog—

nized seven interflow conglomerates and sandstone units in

 

the uppermost part of the volcanic sequence. He notes that 54
these units are generally less than 25 feet thick and normally
persist over the length of the island.
According to Lane (1898) the interbedded conglomerates
are composed of volcanic rock fragments consisting primarily
of abundant felsite and quartz porphyry pebbles in addition
to mafic clasts. As with the Portage Lake interflow sedi-
ments, the presence of felsite and quartz porphyry pebbles is
somewhat anomalous in view of the rarity of felsite volcanic

rocks within the lava succession.

The Isle Royale Conglomerate

The Isle Royale Conglomerate overlies the Isle Royale
Volcanic series and forms a clastic wedge which thickens in
an easterly direction in a distance of 20 miles from a mini-
mum of 1,500 feet to over 6,000 feet between stratigraphib

marker horizons (Huber, 1971). Lane (1898, 1911), and Huber

87

(1971) report that with the exception of a few quartzite
pebbles, essentially all rock fragments in the Cepper
Harbor Conglomerate are of volcanic origin, with felsic
fragments predominating slightly over mafic ones.

Wolff (1969) reports that the coarse-grained sandstones

within the Isle Royale Conglomerate normally contain greater

 

than 50 percent mafic and felsic volcanic rock fragments.
However, the fine to very grained sandstones generally con- g

tain less than 2 percent volcanic rock fragments, only

 

traces of sedimentary and metamorphic rock fragments, but j

about 60 percent quartz, and between 10 to 25 percent feld-

spars, which suggests that a not-too-distant, non-volcanic

source terrain was also contributing important quantities of

fine—sized detritus to the basin of deposition of this time.
Huber (1971) noted that the felsic and mafic volcanic

pebble types found in the Isle Royale sedimentary sequence

are unlike the exposed underlying volcanic sequence on the

Island, but are similar to the flows exposed in the North

Shore Volcanic Group in Minnesota. These observations led

him to suggest the possibility of an unconformity between

the North Shore Volcanic Group and the Isle Royale volcanic

sequence, similar to that prOposed by Hubbard (1968) in

western Michigan between the South Range Series and the Portage

Lake Lava Series. Such an unconformity would permit the

erosion of the lower voICanic sequences at the uplifted mar—

gins of the basin during periods of volcanic quiesence,

88

while the sedimentary rocks were being deposited within the
basin.

The source terrain for the Isle Royale sediments, based
upon studies by Wolff (1969) and Huber (1971) utilizing data
derived from their studies of sedimentary directional struc-
tures, from textural changes within the sediment; and from
the composition of the boulders and cobbles within the con—
glomerate beds, is deduced to be just west of Isle Royale.
The direction of sedimentary transport as inferred from cur-

rent indicators was from the southwest, west, and northwest

 

and suggests that the source terrain may have been part of

the North Shore Volcanic Group.

The North Shore Volcanic Series

 

The base of the Keweenawan sequence in northeastern

Minnesota is made up of the Puckwunge formation, an ortho-
quartzite and quartz conglomerate no more than 200 feet thick.
It is immediately overlain, wherever found, by the North
Shore Volcanic Group of lower and middle Keweenawan age.
The North Shore Volcanic Series consists of a thick sequence
of Keweenawan lavas intercalated with small amounts of sand-
stone. The rocks of this series occupy a broad arcurate belt
running along almost the entire coast of Lake Superior from
Duluth to Grand Portage (Figure 12).

Shoreline exposures of the North Shore Volcanic outline

a broad synclinal structure of which the youngest horizons

89

occur in the vicinity of Tofte. The strike of the volcanics

remain parallel to the coast for much of their outcrop length,

but intercept the coast at acute angles toward either end at
Duluth and Grand Portage. The rocks generally dip lakeward
at angles ranging from about 250 to 10°.

The thickness of the North Shore volcanic succession on
the south limb of the synclinal structure in Minnesota, has
been estimated by Green (1971) to have a minimum total
thickness of 28,000 feet. The sequence of flows making up
the northeast limb of the North Shore Volcanic Group between
Tofte and Grand Portage totals about 21,500 feet (Green,
1971).

Olivine basalt is the most common lava flow-type present
in the North Shore Volcanic Group, but andesites and latities
are also present (Green, 1971). Rhyolite is thought to com-
prise about 10 to 15 percent of the total thickness of the
southwest limb of the sequence (Sandberg, 1938), Schwartz
(1959), while making up approximately 10 percent of the
volcanic series on the northeast limb (Schwartz, 1959).

The intrusive rocks, associated with the North Shore
Volcanics consist of the Duluth Gabbro Complex and the Logan
diabase dikes.

The paleo—magnetic pole positions of the gabbro and the
lavas it intrudes are very similar, which suggests that only
a short time interval separated the extrusive and intrusive

episodes (Du Bois, 1962). A portion of the Complex may even

 

tit—3a u———- <
I
l

“ins—Ii-

9C

predate the North Shore Volcanic Series (Nathan, 1969;
Davidson, 1971). Thus the intrusion of gabbro must have
occurred at intervals throughout lower and beginning upper
Keweenawan time. Therefore, partial unroofing of gabbroic
bodies may have taken place well before the end of middle
Keweenawan time, which would explain the abundant plagioclase

and pyroxene within the interflow sediments.

Interflow Sediments

The interflow sediments make up approximately 500 feet
or 3 percent of total thickness on the northeast limb of the
North Shore Volcanic group in the Grand Portage of Tofte
Sequence, but comprise only slightly more than 300 feet or
1 percent of the total succession thickness exposed on the
southwest limb.

The interflow sandstones contained within the North
Shore Volcanic Series may overlie either amygdaloidal con—
glomerates or rest directly on the amygdaloidal top of the
underlying flow units. Interflow sandstones containing con-
glomeratic units are very rare within the North Shore Volcanic
sequence.

The interflow sandstones within the North Shore Volcanic
Series are buff to reddish brown, and consist predominately
of fine grained, well—sorted, sub-rounded to rounded grains.
A fine lamination is characteristic, but beds from 1/2 inch
to several inches in thickness occur in the thicker deposits.

The thinly bedded units are often structureless, but in thin

 

 

91

section frequently show graded bedding alternating with
laminae consisting of heavy mineral concentrations. Cross
bedding occurs commonly in the thicker units with the tabu-
lar variety being most frequent, followed by the planar,
and trough types, respectively. Asymetrical ripple marks
and shale chips are not uncommon. In one instance ripple F.
marks with the coarsest grains on the crest, suggesting an ~ 1
eolian origin, were observed. 2

The lower surface of the interflow sands is conformable I

 

with the surface structure of the underlying flow, while the '~j
upper surface of the sandstone is flat or sometimes slightly 1
undulating. Grogan (1940) and others note that no soil zone
was found above any of the sandstone layers.
Interflow sandstones occur within the North Shore Volcanic
Group as thin beds, as lenses and as crevice fillings.
Fifty-five thin sections were made from samples collected
from within these sandstone units. All samples were collected
solely from outcrOps (Figure 12).
Detrital feldspars constitute almost 48 percent of the
total average volume of interflow sandstones within the North
Shore Volcanic Group. Plagioclase is the dominant feldspar
comprising 97 percent of the total average feldspar volume.
The most common varieties of Plagioclase appearing in the
interflow sands are labradorite and andesine. The majority
of the grains are fresh, expecially those appearing within
the sediments on the northeast limb of the North Shore

Volcanics.

92

.mmu50 .mwan .mUHmH .mmn00 .mOHMH .mMHNH .man00

.0mmuo0 .mmmum .mmmnm .mmms5 .ommum .nmmum .200n¢ .mqum
.m0u0 .0DDH0 "00000000 N00000E0000 000300000 000 00 000000
I000oo 008 000 00 0000 000800 008 .m0000 0000 000 0000800
0003 000050000 300000000 000000 000000o> 00000 00002 000
00003 00 00000 000 0003000 00500 .00000 000000 005 00000000

 

’1
t.

.00 000000

93

 

'09 j!“ _
-'— 76600001?

MINNESOTA
EARLY
PRECAMBRIAN

 

   

 

    
 

m 0
2 C'
Q I 2
(D
a 3:" 2
‘~ 2.3. 0 *- 3
\ Z
I § ..
“J z '4 o 3
g 0 5 4 ... ..
”I“ 4 I3 2 q 8 2
Z ./ .4 . .1 § 3 5
s ( is. g ... 0 9
2 :< we . ‘ “ " e
’2' E . “’5. '-* 5 t
J :‘. 1:1": m
‘ Z": :‘Vv‘r; 2
>7 '- '7‘;‘ ‘
3 J

   
   
 

IENER DAY

I" ma masons

MICHIGAN

I

WISCONSIN

 
     

MINN.

 

 

Figure 12

 

 

94

Table 16. Mean composition of the North Shore Volcanic
Series interflow sediments. The composition
of the detrital sand size constituents within
the interflow sediments of the North Shore
Volcanic Series as represented by the mean of
the modal analyses of 55 thin sections.

 

 

Volume Frequency

 

 

 

Constituents (%) r01
Y S ;
Plagioclase 46.4 24.5
Mafic rock fragments 25.6 23.5 I
Rhyolite rock fragments 5.6 20.8 'i”
Felsite rock fragments 5.1 10.7 ‘Ij
Pyroxene 4.8 8.2
Unstrained quartz 4.2 6.6
Opaque minerals 4.0 4.9
Potassium feldspar 1.5 4.1
Undulatory quartz 1.1 2.4
Epidote 1.0 4.9
Polycrystalline quartz 0 7 2.6

 

Total percent 100.0

 

mean of 55 thin sections

S standard deviation of 55 thin sections
Potassium feldspar comprises only 1.5 percent of the
total average volume of the interflow sandstones. The most
common variety is untwinned. Grains showing gridiron twin-
ning were observed within the 4 uppermost sedimentary zones
of the southwest limb, but were lacking elsewhere within the

North Shore Volcanic sequences.

95

The mafic rock fragments appearing within the interflow
sandstones of the North Shore Volcanic Group.comprise 25.6
percent of the total average volume of these sediments.

The mafic clasts range in texture from basaltic to basaltic-
amygdaloidal to diabasic, but the coarser textures predomi—
nate. These mafic clasts show less variation within a
sedimentary horizon than those in the Mamainse Point section
and appear to have a coarser texture than those in the Portage
Lake and Copper Harbor Conglomerate.

Rhyolite rock fragments make up 5.6 percent of the total
average volume of the interflow sandstones within the North
Shore Volcanic Series. The rhyolite clasts commonly consist
of uniformly fine-grained quartz and feldspar. No granophyric
fragments were observed in the various sedimentary horizons
within either limb of the North Shore Volcanic Series.

The felsite rock fragments comprise 5.1 percent of the
total average volume of the interflow sandstones. The fel—
site clasts commonly consist of feldspar with minor or no
mafic or quartz grains.

The total detrital quartz grains constitute 6.0 percent
of the average volume of the interflow sandstones within the
North Shore Volcanic Group. Unstrained quartz is the dominant
quartz variety and makes up 70 percent of the total quartz
volume within the sediment. Approximately 18 percent of the
total detrital quartz grains observed within the interflow
sediments possess moderate to strong undulose extinction,

whereas only 12 percent consist of polycrystalline quartz.

 

0......»

96

The Opaque minerals comprise 4.0 percent of the-totar
average volume of the interflow sediments and consist of
hematite, ilmenite, leucoxene and magnetite.

Detrital pyroxene makes up 4.8 percent of the total
average volume of the interflow sandstones within the North

Shore Volcanic Series. The grains are commonly pale brown-

 

r5.
ish gray and consist most commonly of augite and possibly E I
some di0pside. Detrital pyroxene was observed in eight of i
the twelve sedimentary horizons sampled on the southwest |
limb of the North Shore Volcanic sequence, but was noted in .5 J

only one of five sampled on the northeast limb (Tables 17
and 18) which may indicate that the source rock of the pyro-
xene was located somewhat to the south and west of the cur-
rent exposures of the North Shore Volcanic Series. It is
also worth noting that detrital pyroxene grains were not
observed in either the Keweenaw Peninsula or Mamainse Point
samples. Its presence in the North Shore Volcanic Series,
coupled with the abundance of detrital plagioclase grains,
may indicate that mafic intrusive rocks were being unroofed

in the bordering source area during Keweenawan time.

Vertical Compositional Variation

Evaluation of the vertical compositional variation with-
in interflow sandstones of the North Shore Volcanics was
made to detect any changes in provenance occurring during

the Keweenawan time represented by these sediments.

97

The samples used to study the North Shore Interflow
sediments were collected over an interval spanning 130
miles (Figure 13). This interval represents a line of sec-
tion across both limbs of the syncline, although not neces-
sarily at right angles to the axis of the structure. Thus,
even though each sample has a certain vertical position in
the sequence any vertical component may also have a different
lateral connotation.

The samples from both limbs of the North Shore Volcanic

Series were tested for vertical differences in composition

 

(Table 17, Figure 13; Table 18, Figure 14). The testing pro-
cedure was identical to that used on the Keweenawan Peninsula
andeamainse Point samples and consisted of regression
analysis and chi square evaluation of the residuals. Chi
square analysis of the plots of the (observed-expected)
volume frequency with depth reveals no significant residual
2‘< 0.05) for any of the compositional constituents

)4

within the interflow sediments of either limb of the North

trends (P

Shore Volcanic sequence. For brevity Figures 13 and 14 in-
clude only the plots for those constituents whose volume
frequences (Tables 17 and 18) appear to show a trend upward
within the succession.

The lack of any significant residual trends for any of
the detrital constituents after the grain size effect has
been statistically removed from the composition suggests that
there was no appreciable change in provenance for the inter—

flow sediments within the North Shore Volcanic Series.

98

 

 

 

 

 

 

 

WWW; Lilla-Jr
Eu .. I ....w... «E‘M‘hdaifi I-IMIW—
0000000 0000 00000) 0000a0>00 00000000 000000000 osu0o> n 0
0000000 .0000 00.000) abs-250.3 can—00> 000! I 0
I no.0 vn.oI 05.0 05.0 00.0 50.0 50.0I 00.0 00.0 0000 0am
omn00><
0.0 0.0 0.0 m.0 0.0 5.0 0.0 0.0 0.0 0.0 0.5 0.0 0.00 0.00 5.0 0.0 0.00 0.5 0.00 0.5 0.00 0.00 000090000
40> .92
I I I I I I I I I I I I I I I I I I I I o.o 80 «£50
0.5 0.00 0.0 5.0 I I I I I I I I 0.0 0.00 I I I I I I 0.0 0.0 00
I I 0.0 0.0 0.0 5.0 I I 0.0 5.0 0.0 0.0 0.00 0.5V 0.0 «.0 I I 0.50 0.0m 5.0 0.5 0.0
I I I I 0.0 5.0 I I 0.0 9.0 0.0 0.0 0.00 0.00 I I I I 0.0 0.0 0.00 0.00 200
I I 0.0 1.5 I I I I 0.0 0.0 v.0 0.0 0.0 0.5m 0.0 1.0 I I «.00 0.00 0.0 v.00 000
I I 0.0 0.0 0.0 0.0 I I 5.0 0.0 v.0 0.0 0.0 0.00 n.0 5.0 I I 0.00 5.00 5.0 0.00 000
I I I I 0.0 0.0 I I I I 0.0 5.0 v.00 0.00 0.0 0.0 I I I I 5.00 0.00 000
I I 0.0 n." 0.0 0.0 I I m.0 0.0 0.0 0.0 0.0 0.05 0.0 «.0 I I I I 0.0 0.00 000
I I 0.0 0.0 I I 0.0 0.0 0.0 0.0 0.00 «.00 0.00 5.m0 0.00 5.0 0.00 0.00 I I 0.0 0.0 000
I I 0.0 0.0 0.v 0.0 I I 0.0 0.0 0.0 0.0 0.0 0.00 0.0 m.n I I I I 0.0 0.00 000
I I 0.0 0.00 I I I I 0.0 0.0 I I I I I I I I 0.00 0.05 08
I I 0.0 0.00 0.0 0.5 I I I I I I m.0 0.00 I I I I I I 5.m 0.00 ab
0 I:0 h 0 W 0 ma 0 w 0 w m h m h 0 h m h 0 b. 0000000
00 . u .00 0: 0x0 N000 0004 0000 00000 me 00000 00 000.000 0000: 00000000000
fl 000000 I .u-huux0om 00000000 0000000000 I000¢00 8 I i 8
I m x x

 

 

.uusolauon 300000000 000000o> 00000 00002 000 Mo 0500 000320000 000 00000) 0000000d> 0000000IOQ300 0000000> .50 00008

 

l
1 L n‘ F “"F‘ “Ff-"k I;

Figure 13.

99

Graphs of the vertical compositional variation
within the southwest limb of the North Shore
Volcanic Series interflow sediments. Graphs
show the residuals resulting when the (Observed-
expected) volume frequencies for some of the
various detrital constituents are plotted versus
their depth of occurrence in the southern limb
of the North Shore Volcanic Series. Graph A is
the plot for plagioclase clasts, Graph B is

for mafic clasts, and Graph C is for pyroxene
clasts°

my “a l1! “0.
m ”0‘. m m.
(w max) A farm)
no. “‘ ° ' no- "' ° -
fl m
10.5" [0.5-
m .00. n, 0 on.
m ' . . a, 0‘0 ' .
9.0- "' u.
m o m
”C . I O 0 o o. ”c‘ O o.
7.,- ’0‘-
”, I ”.1 o o
m 0 o I 00 “'4 0
u- (R - o u. ‘R I o o
I.“ Q‘-
30- 3.0-
1.6- 1.:-
‘t I o ‘t O 0
aa- M 0.0- N'-
-30 -IO 0 ‘ '20 '80 °IO 0’0 0
(“Vin-(“67(0) mum: (“OWNCV 0’ mm : -— —..._- _5} mm mummy 0’ ”K um
um!
1‘7 3“.
FROM m
an C
@v 1000's)
F '4
Ito- ‘
7R 0 an
’0‘
“I C O . C.
”I
m C 0.
‘0
m C .
m
7.5
”0
'8‘ o o 0
‘.0 M o 0
a:
JD
[.5
u
a0 0“.
'10 'D ID 0’0 ‘19

 

 

100

 

 

 

 

1:31 _ .0! J .1’. _, I

 

 

 

 

 

 

 

 

 

(“firmwmmwnmm

Figure 13

Viz—Isa;

 

000000900 00000000 00aauc n m
0008 000:8. I 0

 

 

 

101

 

 

I 5o.0 00.0 ¢@.0 00.0 00.0 I 50.0 I 00.0 000m «80
oma0o><
Sc

v.0. «.0 0.0 0.0 o.n 0.6m o.0 I 0.3 00506000
OH°> 09‘
0.v 0.00 I I I I I I 5.0 0.0 0.o0 5.5V I 0.0 0.0 I 0.00 0.50 an
5.0 m.m 0.0 m.m I I I I 0.0 m.0 m.00 w.vo I 0.0 0.0 I 5.0 v.00 nu
0.0 «.00 I I o.m 0.0 I I 0.0 5.0 m.m0 0.5. I 0.0 0.0 I o.o 0.00 «U
0.0 5.0 I I I I 5.0 m.0 0.0 o.o 0.0 5.00 I v.0 m.o I 0.m n.mn an
0.0 0.0 I I I I 5.0 0.m 0.00 5.00 0.5 0.00 I I I I m.0 0.0 no
m h N m _h m .h m eh m h m N 000000!
I|||.|u0flua£ IIIJEI Illmflal j J33 lulufio flmwfll :0: fill '00. 0_.|0oul Jdull 0038570

050000 .00000000& 0000000: 000000000: I00m00m X 8 x

i .i * fl x

 

 

.0000l0000 300M00000 0000000> 000£m 00002 000 00 0500 00000000: 0:0 00:00) 000000000000 00 00000000> 000000wnoml00 000000o> .00 00009

 

Figure 14.

102

Graphs of the vertical compositional variation
within the northeast limb of the North Shore
Volcanic Series interflow sediments. Graphs
show the residuals resulting when the (observed-
expected) volume frequencies of the various
detrital constituents are plotted versus their
depth of occurrence in the northern limb of the
North Shore Volcanic succession. Graph A is the
plot for plagioclase clasts, Graph B is the

plot for mafic clasts, and Graph C is for un-
strained quartz grains.

 

 

Rn

two-0

I‘.

l.-

’8'

V

103

 

 

 

 

 

 

 

 

 

 

 

 

(wanna-ammo) mum moan" or can:
cm

Figure 14

“A an up.
A an:
(A! 1000:)
1.-
0 0 m 00
u .0 O p . . .
a 0 0 d 0 "' a 0 0 0
fl ’ ' ° ‘° ’° I.. u 0 0 a. 0
(2'
Io :-
H
a ' ' -
19 ~10 o no 920 8 '80 do 0 0° ‘”
(WIRWO- alarm) m mummy 0! Wm (clump-(mp) mm: Inn-«tau a move
cum can
an ua
M m C
(a roads)
I.‘
o. C C
v C . C
’9 c‘ 00 40
I“- u 0. “.0
ll-
ID-
.a n l
'80 -IO 0 Ola on

 

A In - I'm-z,-

 

104

Therefore any apparent changes in the values of the volume
frequencies of the various constituents upward in the sec-
tion, as derived by modal analysis, must be due to the

grain-size effect and the capriceous nature of the deposi-

tional environment.

Lateral7Variation in Composition
A comparison of the average grain size shown in Table
19, reveals that the sediments on the southwest limb tend

to be slightly coarser (1.96 $) on the average than those on

 

ITI. J'.I.2..“l- .I. -.

the northeast side (2.27 ¢) of the North Shore Volcanic Ser-
ies. The grain-size effect suggests that the coarser
sediments should have the higher rock fragment volume fre-
quency. This observation may explain why the southern limb
of the volcanic series contains a greater volume of rock
fragments (40%) on the average than those on the northeast
limb (26%). Even though the mafic rock fragments have the
same volume frequency on both limbs, Table 19 indicates

that the more siliceous rock fragments are more abundant on
the southern limb, while detrital monomineralic grains are
more important volume—wise in the sediments on the northeast
side. The relative concentration of the felsite and rhyo—
lite clasts within the southern limb may suggest that the
more siliceous—rock centers of sediment dispersal were
localized closer to the southern limb than elsewhere, whereas
those areas shedding mafic detritus were centrally located

or evenly distributed through the area.

105

 

Aouum “an I muxm muons

 

 

 

 

 

b~.m mm.d Hmumemfla 0amu0 ammum>¢

H.ooa m.mm ucmuumm amuoe

m.¢ m.H m.o a.o nuumsv mcwaamumhuohaom
m.m ~.H m.a o.H nuumsv huoumaspcb
I wanna o.m o.m 0000000
I 000MB 5.0 H.N Hmmmvawm sawmmmuom
m.v v.n ~.¢ m.~ mamumcus mswmmo
m.m o.m 0.5 m.m nuumsg cocamuumcb
m.m ¢.N ¢.m n.m wcmxoumm
I 00009 N.NH m.h mucmsmmnm MUCH Ufluflmamm
B.H 0.0 o.em m.n mucmsmmum goon mafiaomam
n.~0 H-mm m.m~ m.mm mpcmemmum goon uummz
m.mH n.0m m.mm H.N¢ mmmauonmam

m m m . w
AXV Axe Gowuamomeou

Mu0osvmum 080H0> mocmswmum 060H0>
pang ummmnuuoz 0500 ummsnusom

 

 

.mucmeanmm 30dmnmucfi mmwumm

oa000a0> 0H0£m nunoz wnu GflSuH3 coaumwum> Hm0owuwmomsoo Hmumumq .mH GHQMB

106

Analysis of Table 19 indicates that the sediment inter-
calated within both limbs of the North Shore Volcanic Series
represents detritus shed from a source terrain consisting of
mafic, felsitic, and rhyolitic volcanic rocks, and mafic

intrusive rock.

Sedimentary Synopsis

 

Provenance

The Keweenawan interflow sediments within the North

 

"r ”...".f can “and.
‘5; . I _

Shore and Isle Royale Volcanic Series, in addition to the
Isle Royale Conglomerate sequence were derived from a
Keweenawan volcanic source terrain consisting of mafic,
felsitic, and rhyolitic rocks. However, the presence of
abundant detrital plagioclase and appreciable pyroxene with-
in the sediments of the North Shore Volcanic Series suggests
that mafic intrusive bodies in the source terrain were also
being unroofed and broken down by erosion to yield mono—
mineralic detritus.

Although Wolff's analysis (1969) of the coarse sands
on Isle Royale indicates they contain abundant felsite and
mafic rock fragments, the finer sands are dominated by both
.simple and undulatory quartz and potassium feldspar detrital
grains, which suggests that a not too distant, non-volcanic
source terrain was also contributing important quantities

of fine-sized detritus to the basin of deposition at the

107

same time. This inference, based upon data from Isle Royale
sediments, does not appear to correlate with the predomi-
nantly volcanic detritus encountered in the sedimentary
horizons within the North Shore Volcanic series.

,It is possible that the differences in composition
existing between the Isle Royale and North Shore interflow
sands could be due to the difference in age of the two
deposits. It has been noted that whenever the upper parts

of the middle Keweenawan are observed, as on Isle Royale and

 

the Keweenaw Peninsula, the volcanic sequence contain a much
higher pr0portion of interflow sediments than do the North
Shore Volcanic Series, and also that they grade into an
almost entirely clastic sequence of the upper Keweenawan.
The small percentage of interflow sedimentary rock contained
within the North Shore Volcanic sequence suggested to Grout
(1959), that the North Shore Volcanics represented only the
lower part of the middle Keweenawan, the upper portion lies
submerged beneath Lake Superior off the coast of Minnesota.
Thus, the Isle Royale sequence, raised by later faulting
could represent the upper portion of a more northerly portion
of Keweenawan deposits located along the northwest limb of
the Lake Superior syncline.

The Keweenawan source area, dominated by an extensive
and thick sequence of volcanic rocks, was tectonically un-
stable as evidenced by the instrusion of the Duluth Gabbro

Complex and the Logan intrusives. However, during the lower

108

Keweenawan and lower-middle‘Keweenawan time the source ter-
rain was primarily producing magmatic products, as shown by
the small amounts of interflow sediments and their fine
grained nature within both the North Shore Volcanics and the
Osler Series. These characteristics also suggest that the
source area was not very high or rugged. However, waning of r
volcanic activity, and increasing uplift during Keweenawan {a

time apparently occurred as evidenced by the coarser nature

 

of the Isle Royale Interflow sediments and the thick over- H

lying clastic sequence. . [:

Paleogeography and Sedimentation

The fine, subangular to subrounded, well—sorted aspect
of the grains, plus the compositional immaturity and relative
scarcity of the interflow sediments within the North Shore
Volcanic and Osler Series pr0bab1y indicates a somewhat dis—
tant source terrain possessing low relief. These character—
istics also suggest that the sediment did not necessarily
undergo a short transportational history, but, in conjunc-
tion with the thick volcanic sequence, were deposited in a
sedimentary basin which was rapidly sinking. However, in the
case of the Isle Royale Cbnglomerate, the coarse nature,
generally poor sorting, and compositional immaturity of the
sediment implies the existence, at that time, of a nearby
source of higher relief, a shorter transportational history,

and more rapid deposition in a rapidly sinking basin.

109

In early Keweenawan time, the uplife of the marginally
located source terrain coupled with the subsidence 0f the
central portion of the basin initiated erosion, transporta-
tion, and deposition of immature, find sands by temporary
streams meandering across the volcanic terrain surface.
More active uplift with time probably moved the northwest

r
located source-terrain-front eastward and subsequently in- f]

I“.-‘ ;

creased the rate of erosion and transportation of immature

conglomerates and coarse sands by a fluvial system on a

 

piedmont fan. J

'7. r1.
..

The most likely source area borderland for the sedi—
ments intercalated in the North Shore Volcanic Series, the
Isle Royale succession, and the Sibley-Osler Series is an
arcuate area running from the west to the north of the
presently crescent-shaped Keweenawan belt of exposures.

This deduction is based upon the following observations.
Crossbedding within the North Shore Volcanic interflow sedi-
ments suggest that the transporting currents flowed toward
Lake Superior (Sandberg, 1938; Halls, 1965). Sedimentological
studies conducted on the Isle Royale Conglomerate sequence

by Wolff (1969) and Huber (1971) utilizing transport direc-
tion indicators, and direction of decreasing thickness and
decreasing textural maturity suggest that these sediments

came from a source area located west or northwest of the
island. Tanton (1931) notes that sandstones between the

Sibley-Osler Series show crossbedding indicating derivation

110

of material from a northerly source. This pr0posed source

area is represented variously by the occurrence of extensive
exposures of large gabbroic bodies and abundant unmeta-
morphosed diabase sills and dikes exposed in Archaen meta-

sedimentary and plutonic rocks in.Minnesota and Ontario.

a. ... dun.“

 

CHAPTER VI

SUMMARY AND CONCLUSIONS

"A-

Sedimentary rock samples were collected from Keweenawan
sections exposed at Mamainse Point, Ontario; the Keweenawan
Peninsula, Michigan; and the Minnesota shore region of Lake

Superior. The results of this petrographic study coupled.

h“..- :

 

It»

with work of others indicates that much of the Keweenawan
sediments were derived from tectonic highs which were mantled
by Keweenawan volcanics.

The coarse, immature, polymictic—conglomerates and
arenites incorporated in the Mamainse Point Volcanic Series
were deposited by a fluvial system on a piedmont fan. The
fluvial system was characterized by high energy conditions
and high gradients marginal to a source area of rugged
relief. The source area was tectonically active and shed
coarse detritus of both volcanic—surficial and non volcanic-
basement aspects simultaneously throughout middle Keweenawan
time. Analysis of the vertical compositional variation with-
in the Mamainse Point succession suggests that while there
was a definite lessening of importance of the mafic volcanic

and plutonic terrain as a source of rock fragments during

111

112

middle Keweenawan time, the felsitic rocks as a source of
detritus become of increasing importance upward in the
sequence with time. The most likely location of the source
area borderland for the interflow sediments intercalated
within the.Mamainse Point Volcanic Series was the area just
eastward of the presently exposed Keweenawan outcrops. This

area is represented by the occurrence of abundant lower

"— {Ate-11‘

diabase dike swarms exposed in lower and middle Precambrian
metasedimentary and plutonic rocks.

The immature, polymictic conglomerates and arenites _1

 

incorporated within the Portage Lake Lava Series and the
Copper Harbor Conglomerate are thought to have been deposited
on a piedmont fan by a fluvial system. The fluvial system
was characterized by relatively high energy conditions
marginal to an upland. The upland was probably rugged, but
possessed only moderately high relief. The sediments within
the Portage Lake and Copper Harbor series were derived
primarily from a nearby Keweenawan volcanic source terrain
consisting of mafic, felsitic, rhyolitic, and granophyric
rocks. However, thezpresence of appreciable undulose quartz
clasts and non-volcanic lithic clasts locally in the upper
portions of the C0pper Harbor Conglomerate suggests incursion
.of sediments from a nearby pre-Keweenawan plutonic and/0r
metamorphic source terrain. The Keweenawan volcanics
mantling the pre—Keweenawan crystalline rocks in the source
area were‘probably breached, at this time, by streams allow-

ing the underlying rocks to be exposed to erosion.

113

The most likely location for the source area borderland
which supplied the various Keweenawan sediments intercalated
with the volcanics series on the southern shore of Lake
Superior was just southward of the presently exposed
Keweenawan outcr0ps. This area is represented by the occur-
rence of abundant lower Keweenawan diabase dikes in Animikean
metasedimentary and plutonic rocks, and a gabbro—granophyric ' Fm]
complex intruding Keweenawan volcanic rocks.

The Keweenawan sediments within the North Shore Vol-

canic and Isle Royale Series were derived primarily from a

 

Keweenawan volcanic source terrain consisting of mafic,
felsitic, and rhyolitic volcanics in addition to mafic intru-
sive rocks. However, abundant undulatory quartz within the
sandy phases of the Isle Royale Conglomerate suggests that

a non—volcanic source terrain was also contributing fine-
sized detritus to the basin of deposition.

It is possible that the North Shore Volcanic Series
represents only the lower part of middle Keweenawan time,
the upper portion lying submerged beneath Lake Superior off
the coast of Minnesota. The Isle Royale sequence, raised
by later faulting, could represent the upper portion of a
more northerly portion of the Keweenawan deposits along the
northwest limb of the Lake Superior basin.

The fine, well-sorted, but immature nature of the North
Shore Volcanic sediments indicate deposition by streams

meandering across a rapidly subsiding volcanic terrain.

41m

 

114

However, the coarser nature of the Isle Royale Conglomerate,
plus its immaturity, implies derivation from a nearby source
area of higher relief, a shorter transportation history,
and more rapid deposition in a subsiding basin.

The most likely source area for the sediments inter-

calated in the North Shore Volcanic and Isle Royale Shries

s. r 10.? '14 r3

is an arcuate area running from the west to the north of the
presently crescent-shaped Keweenawan belt of exposures.
This proposed source area is represented variously by the

occurrence of extensive exposures of large gabbroic bodies

 

1w? 'ff‘r'fi'. K‘ fl ...1
...
MI.

and abundant, unmetamorphosed diabase dikes and sills in
lower Precambrian metasedimentary and plutonic rocks in
Minnesota and Ontario.

The data assembled herein indicates that much of the
Keweenawan sediments of the Lake Superior region were
derived from local tectonic highs. These source terrains
were probably in all instances initially mantled by
Keweenawan volcanics. Therefore, the area covered by the
Keweenawan volcanics was probably much more extensive than
the accumulation deduced from present outcrops would indi-
cats. The ultimate extension of Keweenawan volcanism cannot
be determined at this time, but if the basaltic dikes of
Keweenawan age are considered to be conduits for the subse—
quently eroded flows, then the regional extent would be

considerably increased.

115

The general tectonic pattern for the Keweenawan is
then one of considerable positive local tectonic instability
as well as a time of volcanism. From the evidence discussed
above at least three such local uplifts can be discerned
from the volcanic sedimentary record. These positive

tectonic areas occur on all sides of the Lake Superior basin

rm
and were responsible for contributing perhaps thousands of g 3
feet of Keweenawan sediment observed on Isle Royale, the
North Shore of Minnesota, the Keweenaw Peninsula, Mamainse 3
Point, Michipicoten Island, and Cape Gargantua. j
I!

 

It is possible that many more such uplifts occurred at
this time, whose presence cannot be determined because of a
deficiency in the sedimentary record.

The concept that the location of the present exposure
of Keweenawan rocks was as much influenced by the surround-
ing basement tectonics as the areal extent of volcanism is
not a new one. Burwash (1905) concluded that the triangular
shape of Lake Superior was much controlled by bordering up-
lifts and the possibility that Keweenawan dikes fed subse—
quently eroded basaltic terrain. The results cited herein
should be viewed as corroboration of the insight of a master
geologist.

In recent years, many inferences concerning the tectonic
style of the mid continent have been based upon the occur-
rence and outcr0p pattern of Keweenawan basalts. The present

outcr0p trend coupled with ge0physical evidence of sub-surface

116

.uuommcmuu wumucosflpom mo cospoouflp powwowcfi

onu mumUHUCH mBOHHm one .mmoum Uflcoucou o>aumm00

000 o>fluflmom poumasumom mo coflumooH @0H3ogm 08w»
003mcoo3ox 0a ocwaocmm Heaummom oxmq mo unamumoomooamm

 

III!I I

.ma whomwm

117

 

ma onsmflm

 

 

 

QZ<._ 10:... Z<0=._.u_<
m. .

 

 

 

. 025.10....

 

... 50322.2

 

 

118

trends has lead to the supposition that the Lake Superior
basin is the result of Keweenawan volcanism within a major
rift valley. However, the outcrOpping portion of the
Keweenawan rockslin light of the results presented herein,
is thought to be a remnant of a broader deposit. Although

a rift valley hypothesis cannot be totally disproven, it

can be stated that inferred structural relationships in I?

Keweenawan time in the region of outcrOps do not support I

such a conclusion. White (1966) also disfavors the rift

hypothesis, based upon patterns of detailed faulting and PJ
I

 

stratigraphy within the Keweenawan pile, and suggest that
major fault features effecting the flows post date

Keweenawan volcanism°

REFERENCES CITED

Aldrich, H. R., 1929, The Geology of the Gogebic Iron Range
of Wisconsin, Wisconsin Geol. Nat. Hist. Surv. Bull.
71.

Annells, R. N., 1971, Middle Keweenawan volcanism of eastern
Lake Superior, 17th Ann. Institute on LakeSuperior
Geol., Duluth, Minn.

Annells, R. N., 1970. Keweenawan volcanic geology of
Michipicoten Island, Lake Superior, 16th Ann. Institute
on Lake Superior Geol., Thunder Bay, Ont.

 

F‘I" xv;

Atwater, G. I., and G. M. Clement, 1935, Precambrian and
Cambrian relations in the upper Mississippi valley,
Bull. Geol. Soc. Am., 46, 1659-1686.

Ayres, L. D., 1969, Geology of Townships 31 and 30, Ranges
20 and 10; Ontario Dept. Mines Geol. Rept., 69, 38.

Bacon, L. 0., 1957, Relation of gravity to geological struc-
ture in Michigan's Upper Peninsula, in Geological
Exploration, edited by A. K. Snelgrove, pp. 54-58,
Michigan College of Mining and Technology, Houghton.

Bacon, L. 0., 1964. Investigation of the thickness of the
Jacobsville sandstone by seismic reflection methods--
.a proqress report (abstract), 10th Ann. Inst. Lake
Superior Geol., Ishpeming, Mich., p. 57.

Bailey, E. H. and Steven, R. E., 1960, "Selective Staining
of K-feldspar and plagioclase on Rock Slabs and Thin
Sections," Amer. Min., Vol. 45, pp. 1020—1025.

Balsley, J. R., H. L. James, and K. L. Wier, 1949, Aero-
magnetic survey of parts of Baraga, Iron, and Houghton
Counties, Michigan, with preliminary geologic inter—
pretation, U. S. Geol. Surv. GeOphys. Invest.

Bean, E. F., 1959, Geologic Map of Wisconsin, Wisconsin
Geol. Nat. Hist. Surv.

Books, K. G., 1968, "Magnetisation of the lowermost
Keweenawan lave flows in the Lake Superior Area, U. S.
Prof. Paper 600—D, 248-252.

119

120

Broderick, T. M., 1935, Differentiation in lavas of the
Michigan Keweenawan, Bull. Geol. Soc. Am., 46, 503-558.

Burwash, E. M., 1905, Geology of Michipicoten Island,
University of Toronto Studies, Geology Series, ho. 3.

Butler, B. S., and W. S. Burbank, 1929, The copper deposits
of Michigan. U. S. Geol. Surv. Profess. Paper 144.

Case, J. E., and J. E. Gair, 1965, Aeromagnetic map of parts
of Marquette, Dickinson, Baraga, Alger, and Schoolcraft
Counties, Michigan, and its geologic interpretation, 3.
U. S. Geol. Surv. GeOphys. Invest. Map GP-467. g]

Coleman, A. P., 1899, Copper regions of the Upper Lakes,
Ontario Bur. Mines 8th Ann. Rept., part 2, pp. 121-174.

Coleman, A. P., 1902, Iron ranges of northwestern Ontario, ‘.
Ontario Bur. Mines 11th Ann. Rept., pp. 128-151. J
L.

 

Cornwall, H. R., 1951, Differentiation in lavas of the
Keweenawan series and the origin of the copper deposits
of Michigan, Bull. Geol. Soc. Am., 62, 159-202.

Cornwall, H. R., 1954a, Bedrock geology of the Phoenix
quadrangle, Michigan, U. S. Geol. Surv. Quadrangle Map
GO 34.

Cornwall, H. R., 1954b, Bedrock geology of the Delaware quad-
rangle, Michigan, U. S. Geol. Surv. Quadrangle.Map CO 51.

Cornwall, H. R., 1954c, Bedrock geology of the Lake.Medora
quadrangle, Michigan, U. S. Geol. Surv. Quadrangle Map
G0 52.

Cornwall, H. R., 1955, Bedrock geology of the Fort Wilkins
quadrangle, Michigan, U. S. Geol. Surv. Quadrangle.Map
G0 74.

Cornwall, H. R., and W. S. White, 1955, Bedrock geology of
the Manitou Island quadrangle, Michigan, U. S. Geol.
Surv. Quadrangle Map G0 73.

Cornwall, H. R., and J. C. Wright, 1954, Bedrock geology of
the Eagle Harbor quadrangle, Michigan, U. S. Geol. Surv.
Quadrangle Map GD 36.

Cornwall, H. R., and J. C. Wright, 1956a, Geologic map of
the Hancock quadrangle, Michigan, U. S. Geol. Surv.
Mineral Invest. Field Studies Map MF 46.

 

121

Cornwall, H. R., and J. C. Wright, 1956b, Geologic map of
the Laurium quadrangle, Michigan, U. S. Geol. Surv.
Mineral Invest. Field Studies Map MF 47.

Craddock, D., E. C. Thiel, and B. Gross, 1963, a Gravity
investigation of the Precambrian of southeastern
Minnesota and western Wisconsin, J. Geophys. Res.,
68(21), 6015-6032.

Daly, R. A., 1917, The geology of Pigeon Point, Minnesota:
American Jour. Science, ser. 4, v 43, p. 423-448.

1'

Davidson, D. M., 1971, A new view of the Duluth Complex,
Minnesota, 17th Ann. Institute on Lake Superior
Geol., Duluth, Minn. 3

Davidson, B. S., G. H. Espenshade, W. S. White, and J. C.
Wright, 1955, Bedrock geology of the Mohawk quad-
rangle, Michigan, U. S. Geol. Surv. Quadrangle.Map
GQ 54.

 

“F‘s—r rrav‘. .41; -

Du Bois, P. M., 1952, Palaeomagnetism and correlation of
Keweenawan rocks, Geol. Surv. Can, Bull. 71.

Fairbairn, H. W.,H. J. Bullwinkel, W. H. Pinson, and P. M.
Hurley, 1959, Age investigation of syenites from
Coldwell, Ontario, Proc. Geol. Assoc. Can., 11, 141-
144.

Fritts, C. E., 1965, Stratigraphy, structure, and granitic
rocks in the Marenisco-Watersmeet area, Michigan
(abstract), 11th Ann. Inst. Lake Superior Geol., Univ.
Minn., St. Paul, p. 15.

Giblen, P. E., 1969, Ontario Dept. Mines, Prelim. Geol.
Maps 553 and 555.

Goldich, S. S., A. O. Nier, H. Baadsgaard, J. H. Hoffman
and H. W. Krueger, 1961, The Precambrian geology and
geochronology of Minnesota, Minn. Geol. Surv. Bull. 41.

Gordon, W. C., and A. C. Lane, 1906, A geological section
from Bessemer down-Black River, Mich. Geol. Surv. Rept.

Graham, J. W., 1953, Changes of ferromagnetic minerals and
their bearing on mageetic properties or rocks. J.
Geophys. Res., 58(2), 243-260.

Grant, U. S., 1900, Preliminary report on the copper-bearing
rocks of Douglas County, Wisconsin, Wisconsin Geol.
Nat. Hist. Surv. Bull. 6.

122

Green, J. C., 1971, The North Shore Volcanic Group, 17th
Ann. Institute on Lake Superior Geol., Duluth, Minn.

Grout, F. F., J. W. Gruner, G. M. Schwartz, and G. A.
Thiel, 1951, Pre-cambrian stratigraphy of Minnesota,
Bull. Geol. Soc. Am., 62, 10l7~1078.

Grout, F. F., R. P. Sharp, and G. M. Schwartz, 1959, The
geology of Cook County, Minnesota, Minn. Geol. Surv.
Bull. 39.

Halls, H. C., 1966, "A Review of the Keweenawan Geology
of the Lake Superior Region,” Geophys. Union,
Monograph 10, pp. 3—27.

Hamblin, W. K., 1958, The Cambrian sandstones of northern
Michigan, Mich. Geol. Surv. Publ. 51.

Hamblin, W. K., 1961, Palaeogeographic evolution of the
Lake Superior region from late Keweenawan to late
Cambrian time, Bull. Geol. Soc. Am., 72, 1—18.

Hamblin, W. K., and W. J. Horner, 1961, Sources of the
Keweenawan conglomerates of northern Michigan, J. Geol.,
69, 204—211.

Hinze, W. J., J. W. Trow, N. W. O'Hara, G. B. Secor, and
R. J. Wold, 1965, Aeromagnetic survey of Lakes Superior,
Huron, and Michigan (abstract), U. S. Progr. Rept.
Intern. Upper Mantle Project.

Hite, D. N., 1968, Sedimentology of the Upper Keweenawan
sequence of northern Wisconsin and adjacent Michigan,
unpub. Ph. D. thesis, The Univ. of Wisconsin, 217 p.

Hopkins, P. E., 1921, Schreiber-Duck Lake area, Ontario
Dept. Mines 30th Ann. Rept., part 4, pp. lu26.

Hotchkiss, W. 0., 1923, The Lake Superior geosyncline, Bull.
Geol. Soc. Am., 34, 669m678.

Hotchkiss, W. 0., E..F. Bean, and O. W. Wheelwright, 1915,
Mineral land classifications, showing indications of
iron formation, in parts of northern Wisconsin, Wiscon-
sin Geol. Nat. Hist. Surv. Bull. 44, 35-45.

Huber, N. K., 1971, The Keweenawan geology of Isle Royale,
Michigan, 17th Ann. Institute on Lake Superior Geol.,
Duluth, Minn.

 

 

£I<>11nson, C. H., and R. L. Foster, 1965, Contaminated Pre—

cambrian ash flow tuff, Cascade River, Minnesota
(abstract), Geol. Soc. Am. Special Paper 82, 102.

Joubin, F. R., and associates, 1964, Mining Geologists Ltd.,

Geological maps of the Algoma District, prepared for
the Algoma Central and Hudson Bay Railway Company.

Ibaine, A. C., 1898, Geological report on Isle Royale,

Michigan, Mich. Geol. Surv., 6.

ILEine, A. C., 1911, The Keweenawan series of Michigan,
Mich. Geol. Biol. Surv. Publ. 6, Geological Series,

no. 4.

Imeighton, M. M., 1954, Petrogenesis of a gabbro—granophyre
complex in northern Wisconsin, Bull. Geol. Soc. Am., r

65, 401-442.

 

ILeith, C. K., R. J. Lund, and A. Leith, 1935, Precambrian ;
rocks of the Lake Superior region, U. S. Geol. Surv.
Profess. Paper 181.

ILowdon, J. A., C H. Stockwell, H. W. Tipper, and R. K.
Wanless, 1963, Age determinations and geological
studies (including IsotOpic Ages, Rept. 3), Geol. Surv.
Can. Paper 62—17.

Lyons, P. L., 1959, The Greenleaf anomaly, a significant
gravity feature, in Symposium on GeOphysica in Kansas,
Kansas Geol. Surv. Bull. 137, 105—120.

Mattis, A. F., 1971, Lower Keweenawan sediments of the Lake
Superior region, 17th Ann. Institute on Lake Superior
Geol., Duluth, Minn.

McConnell, R. G., 1926, Sault Ste. Marie, District of Algoma,
Ontario Dept. Mines 35th Ann. Rept., part 2, pp. 1-52.

Moore, E. S., 1926, Batchewana area, District of Algoma,
Ontario Dept. Mines 35th Ann. Rept., part 2, pp. 53w85.

Moorhouse, W. W., 1957, The Proterozoic of the Port Arthur
and Lake Nipigon region, Ontario, in The Proterozoic
in Canada, Royal Soc. Can. Special Publ. 2, 67-76.

Nathan, H. D., 1969, The Geology of a portion of the Duluth
Complex, Cook County, unpub. Ph. D. thesis, Univl of

Minn., 198 p.

124

Nuffield, E. W., 1955, Geology of the Montreal River,
Ontario Dept. Mines 64th Ann. Rept., part 3.

Palmer, H. C-, 1970, Palaeomagnetism and correlation
of some middle Keweenawan rocks, Lake Superior,
Can. Jour. Earth Sci., 7, 1410-1436.

IPEirker, J., 1961, Sublacustrine tOpography of eastern Lake
Superior, unpublished M. S. thesis, Michigan College
of Mining and Technology, Houghton.

IEMarsons, A. L., 1918, Slate Islands, Lake Superior, Ontario
Bur. Mines 27th Ann. Rept., part 1, pp. 155-167.

ESandberg, A. E., 1938, Section across Keweenawan lavas at
Duluth, Minnesota, Bull. Geol. Soc. Am., 49, 795-830.

ESilver, L. T., and J. C. Green, 1963, Zircon ages for middle
Keweenawan rocks of the Lake Superior region (abstract),
Trans. Am. Geophys. Union, 44(1), 107.

'Tanton, T. L., 1931, Fort William and Port Arthur and Thunder
Cape map areas, Geol. Surv. Can. Mom. 167.

Tanton, T. L., 1936, Pigeon River area, Thunder Bay District,
Canada Dept. Mines Maps 354A to 356A, inclusive.

Taylor, R. B., 1964, Geology of the Duluth gabbro complex,
Minn. Geol. Surv. Bull. 44.

Thiel, E. C., 1956, Correlation of gravity anomalies with
the Keweenawan geology of Wisconsin and Minnesota,
Bull. Geol. Soc. Am., 67, lO79~llOO.

Thomson, J. E., 1953, Geology of the Mamainse Point c0pper
area, Ontario Dept. Mines 62nd Ann. Rept. part 4, pp.
1—25.

Thwaites, F. T., 1912, Sandstones of the Wisconsin coast of
Lake Superior, Wisconsin Geol. Nat. Hist. Surv. Bull.
25.

Tyler, S. A., R. W. Marsden, F. F. Grout, and G. A. Thiel,
1940, Studies of the Lake Superior pre-Cambrian by
accessory mineral methods, Bull. Geol. Soc. Am., 51,
1429—1538.

Van Hise, C. R., and C. K. Leith, 1911, The geology of the
Lake Superior region, U. S. Geol. Surv. Monograph 52,
p. 385.

 

41h] "‘

\IErn.Schmus, W. R., 1971, Rb—Sr age of Middle Keweenawan
rocks, Mamainse Point and vicinity, Ontario, Canada,
Bull. Geol. Soc. Am., V82, pp. 3221-3226.

VVlIite, W. S., 1956, Geologic map of the Chassel quadrangle,
Michigan, U. S. Geol. Surv. Mineral Invest. Field
Studies Map.MF 45.

Vqlnite, W. S., 1957, Regional structural setting of the
Michigan native copper district, in Geological Explora-
tion, edited by A. K. Snelgrove, pp. 3-16, Michigan
College of Mining and Technology, Houghton.

VVhite, W. S., 1960, The Keweenawan lavas of Lake Superior,
an example of flood basalts, Am. J. Sci., Bradley
Volume, 258—A, 367-374.

VVhite, W. S., 1966, Tectonics of the Keweenawan basin,
western Lake Superior region, U. S. Geol. Surv. Profes.
Paper 524-E.

‘White, W. S., H. R. Cornwall, and R. W. Swanson, 1953,
Bedrock geology of the Ahmeek quadrangle, Michigan,
U. S. Geol. Surv. Quadrangle Map GQ 27.

White, W. S., and J. C. Wright, 1954, The White Pine Copper
deposit, Ontonagon County, Michigan, Econ. Geol. 49,
675-716.

White, W. S., and J. C. Wright, 1956, Bedrock geology of
South Range quadrangle, Michigan, U. S. Geol. Surv.
Mineral Invest. Field Studies Map MF 48.

White, W. S., and J. C. Wright, 1960, Lithofacies of the
C0pper Harbor conglomerate, northern.Michigan, U. S.
Geol. Surv. Profess. Paper 400wB.

Wilson, A. W. G., 1910, Geology of the Nipigon Basin,
Ontario, Geol. Surv. Can. Mem. 1.

Wold, R. J., 1965, An aeromagnetic survey of western Lake
Superior (abstract), 11th Ann. Inst. Lake Superior
Geol., Univ. Minn., St. Paul, p. 36.

Wolff, R. G., 1969, Upper Keweenawan sedimentary rocks on
Isle Royale, Lake Superior, Unpub. M. S. thesis,
Univ. of Wisconsin. ‘

Wright, J. C., and H. R. Cornwall, 1954, Bedrock geology
of the Bruneau Creek quadrangle, Michigan, U. S. Geol.
Surv. Quadrangle Map GQ 35.

 

‘Young, G. M., 1966, Huronian stratigraphy of the McGregOr
Bay area, Ontario: Relevance to the palaeogeography

of the Lake Superior region, Can. J. Earth Sci., 3(2),
203-210.

EZietz, I., 1965, Aeromagnetic study of the midcontinent

gravity anomaly (abstract), U. S. Progr. Rept. Intern.
Upper Mentle Project.

:Zumberge, J. H., and P. Gast, 1961, Geological investigations
in Lake Superior, Geotimes, 6(4), 10—13.

 

 

 

 

 

 

"I7'11111’11111’11111