J HESIS

1.!83'."-RIFS
MICHIGAN 3TH: Ljim 133:7
EAST LANSING. I‘fliC—H. 443324

This is to certify that the
thesis entitled
Late Wisconsin Ice-flow Reconstruction

for the Central Great Lakes Region

presented by

Stephen Irving Dworkin

has been accepted towards fulfillment
of the requirements for

Masters degree in Geological Sciences

 

 

 

Date My. sz/j #76“?

0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution

fllflil Ifl'ITlI liflimitflfliliflflfifllflifim'flflfl ...

3 1293 00089 4562

 

 

 

RETURNING MATERIALS:
‘PV1SSI_J Place in book drop to
LJBRARJES remove this checkout from
.—_ your record. FINES W111

 

 

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

 

.4—

 

 

 

LATE WISCONSINAN ICE-FLOW RECONSTRUCTION
FOR THE CENTRAL GREAT LAKES REGION

BY

Stephen Irving Dworkin

A Thesis

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

MASTER OF SCIENCE

Department of Geological Sciences

1984

ABSTRACT

LATE WISCONSINAN ICE-FLOW RECONSTRUCTION
FOR THE CENTRAL GREAT LAKES REGION

BY

Stephen Irving Dworkin

\

Tills associated with three Late Wisconsinan ice lobes
(Lake Michigan, Huron-Erie, and Saginaw) in Michigan were
differentiated based on their heavy mineral assemblages.
Using discriminant analysis, the heavy mineral assemblages
of tills in Michigan were also associatated with specific
source areas on the Canadian Shield. These associations
suggest that: l) The Lake Michigan Lobe flowed
southwestward across a region along the north shore of Lake
Huron and then into southwestern Michigan. 2) The
Huron-Erie Lobe flowed southwestward across .a region
northeast of Georgian Bay and then southward into southeast
Michigan. And 3) the Saginaw Lobe flowed southwestward
across both of these regions and then into southcentral

Michigan.

Comparison of heavy mineral assemblages from tills in
southeastern Michigan with those from younger tills just
south of Lake Huron indicates that a significant shift
occurred in ice-flow direction during the final retreat of

ice from southeastern Michigan.

to Karen

ii

ACKNOWLEDGMENTS

My deepest thanks and appreciation go to my thesis
advisor and friend, Dr. Grahame Larson and to my friend,
Bill Monaghan; both of whom gave selflessly of themsevles
in order that I could complete this research. I would also
like to thank my commitee members, Dr. Duncan Sibley and
Dr. James Trow for their help and advice. I will always
treasure the support and love given to me by my parents and

my inlaws, without which, I never would have gone this far.

iii

TABLE OF CONTENTS

INTRODUCTION . . . . .

PREVIOUS ICE FLOW RECONSTRUCTIONS . .

METHODS . . . . . . .

Data Collection .

HEAVY MINERAL ASSEMBLAGES IN

0 O O O O O O O

MICHIGAN TILLS .

HEAVY MINERAL ASSEMBLAGES ON THE CANADIAN SHIELD

ICE FLOW RECONSTRUCTIONS . . . . . . .

DISCUSSION . . . . . .
CONCLUSIONS . . . . .
APPENDIX A . . . . . .
Methodology . . .
APPENDIX B . . . . . .
Heavy Mineral Data

LIST OF REFERENCES . .

from Individual

iv

Samples

(DQNH

ll
l4
17
21
22
25
25
28
28
29

Table
Table

Table

Table

u N H
O. .0

Heavy mineral species identified
Heavy mineral data from Michigan tills

Heavy mineral data for Canadian

LIST OF TABLES

Shield tills

Heavy mineral data for individual

samples .

O

O

11
12

16

28

Figure

Figure

Figure

Figure

Figure

LI ST OF FIGURES

Late Wisconsinan ice-flow in the Great
Lakes region, after Leverett and Taylor
(1915) O O O O O O O O O O O O O O O O I 0

Late Wisconsinan ice-flow reconstruction
in the Great Lakes region, after Fullerton
(1980) O O O O O O O I O O O O O O O O O 0

Late Wisconsinan ice-flow reconstructions
in Ohio, after Taylor and Faure (1981), and
in Indiana, after Bleuer (1983) . . . . .

Sample locations with respect to major
moraines in southern Michigan . . . . . .

Late Wisconsian ice-flow reconstruction
for the central Great Lakes region . . . .

vi

10

19

I NTRODUCTI ON

The establishment of objective ice-flow reconstructions
for the Laurentide ice sheet has been the focus of much
recent research in Illinois (Willman, Glass and ‘Frye,
1963), Indiana (Bleuer, 1983), Ohio (Taylor and Faure,
1981), and southern Ontario (Gwyn and Dreimanis, 1979).
While most of these investigations have relied on
provenance discriminators to reconstruct ice-flow patterns,
the flow reconstructions within Michigan are still based
mainly on morphologic evidence compiled during the original
mapping of the state by Frank Leverett around the turn of

the century (Leverett and Taylor, 1915).

The purpose of this paper is four fold: 1) To establish
lithologic criteria that can be used to differentiate
between tills of the Michigan, Saginaw, and Huron-Erie ice
lobes in Michigan. 2) To associate the tills of each lobe
with a distinct provenance area on the Canadian Shield. 3)
To reconstruct ice-flow patterns in the central Great Lakes
region on the basis of these associations. And 4) to

compare this reconstruction with those already proposed.

PREVIOUS ICE FLOW RECONSTRUCTIONS

One of the earliest attempts at ice-flow reconstruction
for the central Great Lakes region was by Frank Leverett
(Leverett and Taylor, 1915; plate V) aproximately 70 years
ago. He based his reconstruction (Figure l) essentially on
the morphology of glacial deposits and suggested that the
relative positions of recessional moraines in Indiana and
Michigan indicate that ice flowed across the region in
three main lobes; the Michigan Lobe, the Saginaw Lobe, and

the Huron-Erie Lobe.

In ensuing years, various aspects of till lithology have
been used to determine drift provenance and this has
resulted in some modifications of Leverett's original
reconstruction. For instance, Fullerton (1980) has
suggested that the surface drift in southeastern Michigan,
central and western Ohio, and central Indiana was deposited
by ice that flowed through the Lake Huron basin (Figure 2).
He based this reconstruction on the presence of tillite and
banded slate indicators, the source of which is most likely
northeast of Lake Huron (White, 1939). On the other hand,
Bleuer (1983), has suggested that the Erie Lobe flowed
.through both southeastern Michigan and northern Ohio and
extended much farther west into Indiana than was indicated
by Leverett (Figure 3). This reconstruction is based on
high (> 3) garnet-epidote ratios from the sand sized

fraction of till in western Indiana. The source of these

 

 

 

 

 

 

 

 

 

 

Imam dnfl.‘ I mun-a.
n oflc. cut-rum snown onuy in
r‘lwm w wlsconaln an“.

a' no' ’r
Figure 1. Late Wisconsinan ice-flow in the Great Lakes region,
after Leverett and Taylor (1915).

Di

 

 

 

 

 

 

Figure 2. Generalized ice-flow paths, after
Fullerton (1980).

1. Green Bay Lobe 2. Michigan Lobe

3. Saginaw Lobe 4. Huron Lobe

5. Georgian Bay Lobe 6. Lake Simcoe sublobe
7. Ontario Lobe 8. Erie Lobe

9. Grand River sublobe 10. Kilbuck sublobe

ll. Scioto sublobe 12. Miami sublobe
13. East White sublobe l4 Decatur sublobe

 

 

 

. ». _.

 

 

 

 

aoo KlOMEIERS

zoo
+—~—J —-r—L—-—-—'~T—-—J
N!)

200 MlES

3m

mo

0
0

Figure 2 .

 

  

-.(Rlo.n

    

I
l

100 zoo 0. .. ,,
Illwilfl

3:! !:o“.V 9"'('."“

“01'“ do”

'alnm': bus

CZ]
E23
CZ]
Gnu-u n I. 0mm.
C23
525

seen" 'wv'nu. lulu!”
but" Menu
Soc-cum Mr,

 

 

 

Figure 3. Late Wisconsinan ice-flow reconstructions in
Ohio, after Taylor and Faure (1981), and in
Indiana, after Bleuer (1983).

two minerals is most likely north and northeast of Lake
Ontario (Gwyn and Driemanis, 1979). A similar
reconstruction by Taylor and Faure (1981) (Figure 3) was
based on the age of detrital feldspar from tills in Ohio.
The feldspar dates indicate a Grenville province source and

thus transport by the Erie Lobe.

METHODS

Although indicators, mineral ratios, and radiogenic
dates on feldspars provide valuable information regarding
ice-flow directions, limitations exist with respect to
these methodologies. For example, indicators are generally
rare and they can have more than one source area (White,
1939). Likewise, the application of both mineral ratios
and feldspar dates can only discriminate between broad
regional source areas and thus are of limited use for local

reconstructions.

A more precise approach to ice-flow reconstructions has
been developed by Gwyn and Driemanis (1979) and was applied
to ice-flow in southern Ontario. This method first defines
provenance areas by multivariant analysis of heavy mineral
assemblages in tills on the Canadian Shield and then
classifies assemblages from tills down ice with the
assemblages of the source areas. The classifications are
then used to reconstruct ice flow patterns. This method is

also used for the reconstruction of ice-flow patterns in

this investigation.

Data Collection: A total of 31 till samples, 11 associated
with the Michigan Lobe, 8 associated with the Saginaw Lobe,
and 12 associated with the Huron-Erie Lobe, were collected
from exposures in moraines and till plains within the
southern half of the lower penninsula of Michigan.
Generally these samples were taken at least five feet below
the ground surface. The locations of the samples and the
moraines from which they were collected are shown in figure
4. The ages of these moraines are believed to be between
15,000 and 13,000 yr B.P. (Farrand and Eschman, 1974;

Dreimanis and Goldthwait, 1973; and Dreimanis, 1977).

Heavy minerals from each sample were gravity separated
using tetrabromoethane (5.9. 2.95) following the procedure
described by Gwyn (1971). Only the fine sand fraction (2
to 3 phi) was studied because it has been demonstrated to
be the most diagnostic for heavy mineral studies of till
(Gwyn and Dreimanis, 1979). The weight of the heavy
minerals was measured and reported as a weight percent of
the fine sand fraction. The magnetic minerals were removed
with a hand magnet, weighed, and reported as a weight
percent of the heavy mineral fraction. The separated heavy
minerals were mounted on glass slides with a high

refractive index mounting medium

Figure 4. Sample locations with respect to major moraines

in southern Michigan.
Abreviations: LBM- Lake Border Moraine, VM- Valparaiso

Moraine, KMM- Kalamazoo Moraine Michigan Lobe,
KMS- Kalamazoo Moraine Saginaw Lobe, CM- Charlotte Moraine,
LM- Lansing Moraine, GLM- Grand Ledge Moraine,
MM- Mississinawa Moraine, FWM- Fort Wayne Moraine,

DM- Defiance Moraine.

 

\

 

 

‘
.-
é

/’
/ Ice Lobe Boundary

0 Sample Location

oonia
CZIZIZJ

NV‘JIHDIW Div]

 

 

lO

 

Figure A.

(Ladd High Mount, R.I. 1.686). On each slide 450 to 500
grains were identified and counted using an Olympus (model
BH) petrographic microscope. Heavy minerals were
identified optically by their color, refractive index,
pleochroism, cleavage, habit, extinction angle, and optic
sign. The mineral species identified are listed in table
1. All other mineral species and grains which could not be

identified were combined into a category termed "others".

Table 1. Heavy mineral species identified.

Hornblende

Tremolite- Actinolite
Orthopyroxenes- Enstatite, Hypersthene
Clinopyroxenes- Augite, Diopside
Garnet

Epidote

Rutile

Sphene

zircon

Tourmaline

Opaque minerals

HEAVY MINERAL ASSEMBLAGES IN MICHIGAN TILLS

The frequency percent of heavy mineral species occurring
in samples from the Michigan, Saginaw, and Huron-Erie lobes
in Michigan is shown in table 2. From this data, it is
apparent that tills associated with the three lobes can be
differentiated on the basis of their heavy mineral

assemblages. The Michigan Lobe is characterized by high

11

percentages of epidote and opaque minerals (primarily
ilmenite, hematite, and pyrite) and low percentages of
clinopyroxene and low weight percent of heavy minerals.
High weight percent of heavy minerals, high percentages of
hornblende and clinopyroxene, and low percentages of
epidote and opaque minerals are characteristic of the
Huron-Erie Lobe . The Saginaw Lobe generally has heavy
mineral percentages intermediate between the Michigan and

Erie Lobes except for garnet which tends to be low.

Table 2. Heavy mineral data from Michigan tills .

Lobe

. l Michigan Saginaw Huron-Erie
E:23£E£_-------_-_----g--_-z-_-_-----:----z;-_-_---z--_-z;-
Opaque Minerals 20.3 6.6 12.7 2.9 6.0 1.5
Weight Heavies2 0.9 0.3 1.5 0.2 2.5 0.8
Rutile 0.2 0.2 0.1 0.1 0.2 0.2
Epidote 7.9 1.8 5.8 1.6 5.4 1.6
Zircon 0.1 0.1 0.2 0.2 0.1 0.1
Sphene 0.4 0.3 0.7 0.4 0.9 0.7
Clinopyroxene 3.8 1.0 4.5 1.7 7.4 2.2
Orthopyroxene 1.2 0.4 2.5 0.9 2.3 1.2
Tremolite 0.6 0.4 1.0 0.5 0.9 1.2
Magnetics Fraction3 5.8 3.3 5.2 2.5 5.2 2.6
Hornblende 34.2 5.8 33.9 2.9 42.5 5.3
Tourmaline 0.8 0.6 0.6 0.3 0.7 0.6
Garnet 8.8 2.9 7.1 1.7 9.5 2.7
Others 21.2 4.9 30.7 4.0 24.2 4.6
Sample Size 12 8 ll

1- In descending order of importance for differentiation of
lobes.

2- Units: weight percent of fine sand fraction.

3- Units: weight percent of heavy minerals.

12

In an attempt to further differentiate between the three
lobes the heavy mineral data in table 2 was analyzed using
a discriminant analysis computer program (Klecky, 1975).
The program mathematically establishes the best combination
of mineral variables for differentiating heavy mineral
group populations and through a linear discriminant
function transforms the heavy mineral data for each sample
into a single discriminant score. That score, or
transformed variable, represents the samples position along
a line defined by the linear discriminant function. The
program minimizes the difference between discriminant
scores among samples within groups and maximizes the mean
discriminant scores between groups. The program also has
the capability of classifying ungrouped samples into their
most likely group association based on the their

discriminant scores.

Discriminant analysis of the heavy mineral assemblage
data in table 2 correctly assigned 93 percent of the till
samples to their respective lobes. Two samples, l-l and
25-1, that were collected from moraines assciated with the
Huron-Erie Lobe were classified with the Saginaw Lobe.
These samples, which were obtained very near the
interlobate area between the Huron-Erie and Saginaw lobes,
could have been derived from Huron-Erie Lobe ice that

overrode and incorporated older Saginaw Lobe deposits.

13

 

HEAVY MINERAL ASSEMBLAGES IN TILLS ON THE CANADIAN SHIELD

Heavy mineral assemblages of tills on the periphery of
the Canadian shield north of the Great Lakes region have
been studied by Dreimanis et a1. (1957) and by Gwyn and
Dreimanis (1979). They found that the most marked change
in heavy minerals in tills occurs between the Grenville and
the Superior-Southern structural provinces and suggested
'that these changes are due to "a fundemental difference" in
the rocks underlying these two areas. For example, fills
in the Superior-Southern province are characterized by a
high epidote and occasionally high magnetic and opaque
mineral content, and low amounts of garnet, tremolite,
sphene, and total heavy mineral content. On the other
hand, tills in the~ Grenville province are generally
identified by their high weight percent of heavy minerals
and their high content of garnet, tremolite, and to a
lesser extent sphene and orthopyroxene . Epidote, opaque
minerals, and the magnetic fraction also tend to be lower
in the Grenville province than the Superior-Southern
province.

Within the Superior-Southern and Grenville provinces
several local areas with distinct heavy mineral assemblages
have also been suggested by Gwyn and Dreimanis (1979). Two
of these occur in the Grenville province; a western area
named here "Grenville west" can be identified by abundant

garnet and a low purple-red garnet ratio, and an eastern

l4

area named here "Grenville east" can be identified by high
garnet and tremolite, and a garnet ratio generally greater
than one. Two local areas are also located in the
Superior-Southern province, one along the north shore of
Lake Huron and the other near the Thunder Bay area of Lake
Superior. The area on the north shore of Lake Huron is
named here "Superior east" and typically has very high
percentages of clinopyroxene and relatively low percentages
of hornblende and garnet. The "Thunder Bay" assemblage
area has high percentages of clinopyroxene and magnetic
minerals, however, Gwyn and Dreimanis (1979) have pointed
out that this provenance area may be too limited in extent

to be a useful provenance area.

From Gwyns' data (1971), two additional provenance areas
are proposed. One is the "Province Boundary" area which
occurs along the boundary line that separates the Grenville
and Superior-Southern provinces. This areas assemblage is
identified by its high pecentage of hornblende,
intermediate percentages of clinopyroxene, garnet, epidote,
opaque minerals and total heavy minerals, and low
percentages of magnetic minerals. The other area is
located between Thunder Bay and the Superior east area
along the northeast shore of Lake Superior. This area,
named "Superior west", can be characterized by high epidote
and low garnet. Table 3 shows the mean percentages of the

heavy minerals in all six areas described above. The

15

.opmaumom Hoaocfie >>don on» ca manhoofia caponwca mo pcoocom pswaoz "mafia: um
.oOHpodnm coda onflm map :a madaocfla h>don mo pcoonom pcmaoz Hanan: um

.moocdcm>0hm mo coapofipcouommfio now monophogaa mo nocno mawccoomoc :H Ia

pmam aaaa>cmac -.m.o .pmaz maan>amao -.3.o .aaaeasom
mocfi>onm u.m.m .pmwm uOHLomsm u.m.m .pmoz hoaaomsm 1.3.m .hcm poucsna u.m.a

"mnoapofi>onn¢

r-i
MPGNKOQMMMNMLAH
OHO\OHOU\U\NHt—IO\O

NL‘OOr-ILRONA’NQDNH
LnOCDNc-iOMMMr-iooo

JONPHNJWHOQQH
FOMCDHONOONONO

O

:HNMHNJOHQCDMH
(IDNNCDHONOOHOQO

J—dzmmmmr—(NHNONH
MOJJMONOOOHMO
r-Ir-i

O O O O 0

ONOQONQMHMQMH
:HOHJOOOOOONO

cum

0

ozox0L~moxm0.::-M\oa
JOCIDLnr-(OQOOOOJO

I-IH

H

.10.]

(“\ONMCDHQNOONHH
\OOCDGDMOOOOOOOO

(DmCthJRMNLRLACDI-‘l
NOJMNOHOOOONO

Hr-‘l
\OQI‘LfiH-fi'JNNFCDHn-i
JONQMOJOOOOOO

m:

oawm oaqawm

oocdno>osm

.maaap afloaam anaconda eon «pee Haaoaaa a>aom

0.0 0.0 coouam
0.5 0.:H mosdomo
0.0 0.0 mamnmm
:.0 3.0 ocoxonzmonpao
0.0 0.0 moaadahsoe
m.o H.o mpflaoamna
N.“ o.e nmoapoamaz
H.0 H.0 oaflpsm
m.o o.a upoeamm
:.m a.» weamananom
:.0 m.>> oqoxouamooaau
m.a e.a ammfl>amm pawns:
>.0 v.0 vacuum

.0 x
I manage:

.m.a

.m vanes

16

location of these provenance areas is shown in figure 5.

To further substantiate the existence of these six
provenance areas, discriminant analysis was performed on
the heavy mineral data listed in table 3. Correct
assignment of 89 percent of the samples to their respective
provenance areas confirms the existance of five of these
areas. Correct assignment of only 60 percent of the
Thunder Bay area samples suggests that this suite of till
samples does not represent a distinct provenace area.
Consequently, this area is not considered any further in
this study.

ICE FLOW RECONSTRUCTIONS

On the basis of their discriminant scores, the heavy
mineral assembalges of Michigan tills have been classified
into the five established provenance assemblages of the
Canadian shield. The results of these classifications are
summarized in figure 5. The classifications show that the
Huron-Erie Lobe assemblage is best asscociated with the
assemblage of the Province boundary area. They also show
that the assemblage of the Saginaw Lobe is associated with
both the assemblages of the Province boundary and the
Superior east areas. The assemblage of the Michigan Lobe
appears to be primarily associated with the heavy mineral

assemblage of the Superior east area.

17

Figure 5. Late Wisconsinan ice-flow reconstruction for the
central Great Lakes region. Samples near the southern
shorline of Lake Huron and on the Canadian Shield were
collected and analyzed by Gwyn (1971).

18

.m ansmaa

 

33 3. 8 =5:B>o§\

Canaan 80.. oo_\\
.mmw 3:690 I

303 2.3.5.6 .u

raccoon ooc.>o.a 4

Bow 5.5ng

o 7.25 5:83 o
3:3055

 

355.1 3.2925 2.2.20“ (

32:16.“. 35.025 Sofaom oca 5:330

 

 

 

l9

  

9m. .
1030.: 022 1

 

 

 

 

 

From the above classifications late Wisconsinan ice-flow
patterns can be reconstructed for the central Great Lakes
region (Figure 5). Ice of the Huron-Erie Lobe in Michigan
appears to have been derived primarily from the Province
boundary area. From here it flowed southwestward through
Georgian Bay, then southward into the Huron Basin, and then
southwestward into Michigan. Likewise, ice associated with
the Saginaw Lobe appears to have been derived primarily
from two areas, the Province boundary and Superior east
areas. From these areas the ice flowed towards the south
and southwest across the Huron Basin and into the Saginaw
Basin. Ice of the Michigan Lobe, in Michigan, appears to
have been derived mainly from the Superior east area and
flowed southwestward into the Lake Michigan Basin, and then

into southern Michigan.

The above reconstruction is also supported by the
distribution of individual variables. For example, weight
percent of heavy minerals, which has been shown by Gwyn
and Dreimais (1979) to be one of the most diagnosic
variables for discriminating between tills from the
Grenville (high weight percent) and Superior-Southern (low
weight perecnt) structural provinces, indicates that
Huron-Erie Lobe tills in Michigan were derived from the
Grenville province because both have high weight percent of

heavy minerals. On the other hand, Michigan Lobe tills

20

have a low weight percent of heavy minerals which indicates
that they were derived from the Superior-Southern province.
Saginaw Lobe tills have intermediate percentages of weight
percent heavy minerals which indicates that they were

derived from both provinces.
DI SCUSSION

The ice-flow reconstructions presented in Figure 5 agree
well with those postulated by Leverett and Taylor (1915):
the only exception involves the source of ice in
southeastern Michigan. Leverett and Taylor suggest that
during the Late Wisconsinan ice from the Erie Basin
extended west through southeastern Michigan as far as the
Ohio-Indiana state line. This study, however, shows that
only iCe from the Huron basin extended into southeastern
Michigan. This interpretation is supported by Fullertons'
reconstruction of ice-flow patterns for the region and by
Dreimanis and Goldthwaits' (1973) contention that "the
southwestern portion of the so-called Erie Lobe" was
actually composed of ice derived from several basins. The
reconstruction shown in figure 5 also agrees well with the
regional ice-flow reconstruction suggested by Shilts (1980)
for the Laurentide ice sheet north of the Great Lakes area.
This regional flow pattern appears, however, to have been

localy diverted by the basins of the Great Lakes.

A comparison of the classification of heavy mineral

21

assemblages of tills in southeastern Michigan with that of
Gwyns' (1971) for tills along the southern shore of Lake
Huron suggests a shift in provenance from the Province
boundary area to the Superior east area for tills of the
Huron ice lobe (Figure 5). This would imply that a major
shift in ice flow direction occurred during the retreat of
the Huron Lobe from southeastern Michigan into the Huron
basin. This shift from a southwestward flow to a southward
flow has also been suggested for the Huron Lobe by

Dreimanis et a1. (1957).

CONCLUSIONS

The conclusions of this study are as follows:

1) The Michigan, Saginaw, and Huron-Erie lobes in Michigan
can be differentiated on the basis of their heavy mineral
assemblages. Michigan Lobe tills are characterized by high
percentages of epidote and opaque minerals and low
percentages of clinopyroxene, and low weight percent of
heavy minerals. Huron-Erie Lobe tills are characterized by
high weight percent heavy minerals as well as high
percentages of hornblende and clinopyroxene and low
percentages of epidote and opaque minerals. Tills of the
Saginaw Lobe generally have heavy mineral percentages
intermediate between those of the Michigan and Erie lobes

except for garnet which tends to be low.

2) The heavy mineral assemblage of tills from each of the

22

three lobes can be associated with distinct provenance
areas on the Canadian Shield. The assemblage - from
Huron-Erie Lobe tills is associated with a provenance area
just north of Georgian Bay. The heavy mineral assemblage
from tills of the Saginaw Lobe is associated with two
provenance areas, one area just north of Georgian Bay and
another area on the north shore of Lake Huron. The
assemblage from tills of the Michigan Lobe is associated
with a provenance area along the northeast shore of Lake

Huron .

3) The association of heavy minerals assemblages within
Michigan tills with provenance areas on the Canadian shield
indicates a general northeast to southwest ice-flow pattern
into the Great Lakes region during the late Wisccosinan

that was locally modified by the lake basins.

4) Association of heavy mineral assemblages in
southeastern Michigan with a provenance area just north of
Georgian Bay indicates that during the late Wisconsian the
Huron-Erie Lobe in southeastern Michigan was derived from
the Huron basin and not from the Erie basin as has been

previously suggested.

5) Comparison of heavy mineral assemblages from tills in
southeastern Michigan with those from younger tills just
south of Lake Huron indicates that these deposits are

associated with two differnt provenance areas. This would

23

suggest that a significant shift occurred in ice flow
direction during the final retreat of ice from southeastern

Michigan.

I suggest that multivariant analysis of heavy mineral
assemblages could also be applied to solving ice-flow
reconstructions in other areas. For example, the origin of
the Decatur sublobe in east-central Illinois (Figure 2) has
been debated. Johnson et a1. (1971) and Fullerton (1980)
agree with Leverett and Taylor that this sublobe was
associated with the Michigan Lobe. In constrast, Dreimanis
and Goldthwait (1973) have suggested that both the Saginaw
and Huron Lobes contributed ice to the Decatur sublobe
while Willman and Frye (1970; 1973) have proposed an Erie

lobe origin.

24

APPENDICIES

APPENDIX A

APPENDIX A

METHODOLOGY

Collection of Samples: Samples of till were collected from
gravel pits, road cuts, and natural exposures. After the
first several inches of the exposure were scraped off,
dilute hydrocloric acid was poured on the till. If it
effervesced, the till was assumed to be unleached and a
sample was taken. As the till was shoveled from the oucrop
into a zip-10c storage bag, care was taken to avoid highly
weathered cobbles and pebbles which might crumble and

contaminate the heavy mineral assemblage of the till.

Laboratory Procedure: Approximately 350 to 400 grams of
till from each sample was gently crushed using a mortar and
pestle. The sample was then sieved to remove cobbles and
pebbles and then soaked for 12 hours in a solution of
distilled water and Calgon (5 grams of Calgon per litre of
water) to defloculate the clays. At this point the sample
was wet sieved though a .250 mm sieve which was nested on
top of a .125 mm sieve. The fine sand which was caught on
the lower sieve was soaked in 20 percent HCl for 12 hours

to remove cabonates. The sample was then washed, resieved

25

and dried in an oven.

Heavy minerals were separated from the fine A sand
fraction of the till by gravity settling in glass
separartory funnels. The procedure was as follows:

1) A 500 ml separtory funnel, placed inside a fume
cabinet, was filled with 100 ml of tetrabromoethane.
2) The fine sand fraction was poured into the separatory
funnel and then agitated thoroughly with a glass rod.
3) After 10 minutes the mixture was stirred again without
disturbing the accumulation of heavy minerals that had
settled above the valve.

After another 10 minutes, the following procedure was
used to drain the separatory funnel:
4) A vacuum funnel apparatus with a #1 Whatman filter
paper was placed under the separatory funnel. With the
vacuum pump operating, the valve of the separatory funnel
was turned to the open position and then closed after all
the heavy minerals that had accumulated over the valve had
been extracted.
5) Once the tetrabromoethane had drained through the
filter paper the funnel was removed and the filtered
tetrabromoethane was retuned to its original container.
6) The heavy minerals were washed thououghly with acetone.
7) Steps four through six were repeated for the light

mineral fraction.

26

Magnetic grains were separated from the heavy minerals
using a hand magnet covered with a piece of waxed paper.
The heavy minerals were then mounted on glass petrographic

microscope slide with Ladd High Mount mounting medium.

27

APPENDI X B

mm.mm
94m
mmzmm
mm.ma
3.5
30.0H
095
mmim
3.0m
Hm.em
34a
0:.mm
mmim
00.0m
mm.mm
mm.mm
8.3
mm.mm
mm.mm
8am
«lama
mm.em
mH.Om
mm.mm
me.0m
mm.mm
omém
mmém
mm.ma
3.0m
mm.am

$50

NMJMN:R‘O

Charr4;roo\or4
uxg3gazr4rb~;r

MOQOQNCDNNOO

O\\O—:i'

.ONQOJNOQOO

c>0101rifific>0Jr4b~r4oioicwuxc>Eg«w«1018;0\uxonc>;rr4b—GDriuxo\
undDgrar:t
e+c>C>c>c>C>c>c>c>r4C>c>c3<3(Dc:<3r4<3<3<3<>r4c>r4c>c>c>r4c>r4r1

O O O O O O O O O O O O O I

:zg3r4r4<><Dcuu\c>c>c><3r4r4<D<D<3r4<3<3<3<3<3cu<3r4<3<3<Dr4E3<3
g; 010101C>0Jq>c>c>cucocucu<3<D<>cu<3<3<D<D<31r<>cu<3<3<0.a <3
b3C>C>CAC>C>C>CDC>C>CD<D<D<D(D<3<3<D<D<D<D<D<D<D<D<D<D<D<D<D<D(D

gmHNOOQNHHH—i
MNOJJJ’NNN—i—d’

OOOOOOOOOHOHOOHHOOOHOOOOHNOOOOO

NONONQNNNQOQO:

\O\O

mdmawmmOJNbdr-INOMJON
mm C0

3

moaqsam Hd50H>H0GH Bonn

(fir-(Hm
:{NN
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

sssssssssss

NMu-i
N\O.d’

aaasae

F'F1F1C3CDQJ
Cycu;r<2<>cu

00.0
993m

0 O 0' O

O O

HJMNONL‘wb-Hmm

>3thme MCI)\O.:1' O\r-{ O

ommmmbhdmmwmwbw

\ON
on:
Ln

m>.m
RS

00.0
mm.®
mm.>
mm.©

5:.HH

mm.m
No.13
Hm.m
mm.m
ma.m
me.w

wo.aa

mm.0

em.oa

mm.m

we.ma
mm.oa

me.o
mo.0
me.m
as.:
am.»
ma.m
50.0
aw.“
m0.oa
mm.:
mm.0

:m.ma
m:.oa
Hm.oa

92mm

mm.m
0>.m
mm.:
$3
0:.m
mm.m

:0.0H

m0.»
om.m
a:.:
am.m
ma.m
mm.m
am.m
we.“
as.m

mm.oa

R3
ma.m
m>.m
H0.>
em.m
0s.:
mo.m
:m.a
mm.a
Ga
mm.m
m>.m
mm.:
0>.m
ammo

\OCD Fatwa

c>c>r4c>r1r1c>r+oirioir1ciojoiwwaioiwwdwawrirteiriric>c>~4r+r4r+

i

LRNCDLQHHNN

. .QHmOb-wwwmawmgc—(JJF

0‘1.\OOCON<D @0155-

OMJMQMCDMOJLRCDLRO

\DNG):

xmm

<3<o<3<3<3<5<3<3r4<o<o

sssasss

Chi-(HO
LRNNO

Hm.mm
mm.mm
mo.mm

NN.N:

no.0:

m0.mm
m9?

mfifi

0H.H:

0T§
>0.0:

0:.H :m.wm

Hm.o
em.a
m:.o
:m.a

gl‘u-HB—ITNHM
’CD—d' O\O OJMN
OOOOHOONt-IOr-lr-IOu-(zr

Chr4ATCUP‘b-Od
O\OJQDGD<3

>5.m2

mm.a:

ma.0m

mm.>m

sa.o:
om.mm
me.mm

mm.mm

Hm.mm
em.mm
so.mm
mm.mm
ma.:m

mm.wm

H:.:m
mm.m:
ma.mm
0H.:m
ma.mm

94mm

upon Hopoqdz h>coz

m XHszmmd

ECHO—22’

'3
Si;*fi2%§:t€h<>ffir4C0<N

\OU\OO\OO\NO\ (\IN
L‘MCIDNL‘Q exogoor~
:r—d'LnLACDb-

O
O
O

H:.>

mo.o

:rb~onq)oib-c>
ririririC3rlFlF1F1F1FiFiFiriQJWTFIQJQJQJQIFiQJ

Eir1c>gwo\b-r1U\c>cnu3cu:r
gtrqcur+t~cocua3;r\0Lnoo

E

28

LIST OF REFERENCES

LIST OF REFERENCES

Bleuer,N.K., 1983. Rock and Event Stratighraphy. In
Bleuer, N.K., W.N. Melhorn, R.P. Pavey, eds.,
Interlobate Stratigraphy of the Wabash valley,
Indiana: 30th Field Conference Midwest Friends of the
Pleistocene, pp. 21-28.

Dreimanis, A., Reavely, G.H., Cook, R.J.B., Knox, K.S., and
Moretti, F.J., 1957. Heavy mineral studies in tills of
Ontario and adjacent areas: Jour. Sed. Pet., v.23, pp.
148-161.

Dreimanis, A., 1977. Late Wisconsiana glacial retreat in
the Great Lakes region, North America: Annals of the
New York Acad. Sci., V. 288, pp. 70-89.

Dreimanis, A. and Goldthwait,R.P., 1973. Wisconsin
glaciation in the Huron, Erie, and Ontario lobes:
Geol. Soc. America, Memior 136, pp. 71-106.

Farrand, W.R. and Eschaman, D.F., 1974. Glaciation of the
southern peninsula of Michigan- A review: Michigan
Acadamician, v. 7, pp. 31-56.

Frye, J.C. and Willman, H.B., 1973. Wisconsinan climatic
history interpreted from Lake Michigan Lobe deposits
and soils: Geol. Soc. America, Memior 136, pp.
135-152.

Flint, R.F., 1970. Glacial and Quaternary Geology: John
Wiley and sons, New York, 829 p.

Fullerton, D.L., 1980. Preliminary correlation of post-Erie
interstadial events (16,000-10,000 radiocarbon years
before present), central and eastern Great Lakes
region, and Hudson, Champlian, and St. Lawrence
lowlands, United States and Canada: Geological Survey
Professional Paper 1089, 52p.

Gwyn, H., 1971. Heavy mineral assemblages in tills and
their use in distinguishing glacial lobes in the Great
Lakes region: Ph.D. thesis, The University of Western
Ontario, London, Ont. 192p.

Gwyn, Q.H.J., and Dreimanis, A., 1979. Heavy mineral
assemblages in tills and their use in distinguishing
glacial lobes in the Great Lakes region: Canadian
Journal of Earth Sciences, v. 16, pp. 2219-2235.

29

Johnson, W.H., 1976. Quaternary stratigraphy in Illinois-
Status and current problems, In Mahaney, W.C. eds.,

Quaternary Stratigraphy in North America:
Stroudsburg, Pa., Dowden, Hutchinson and Ross, pp.
161-196.

Leverett, F. and Taylor, F.B., 1915. Pleistocene of Indiana
and Michigan and the History of the Great Lakes: U.S.
Geol. Survey, Monograph 53, 816 p.

Klecka, W.R., 1975. Discriminant analysis, In Norman, N.H.,
C.H. Hull, J.G. Jenkins, K. Steinbrenner, D.H. Bent,
eds., SPSS: New York, McGraw-Hill, pp. 434-407.

Shilts, W.W., 1980. Flow patterns in the central North
American ice sheet: Nature, 286, pp. 213-218.

Taylor, K.S. and Faure, G., 1981. Rb-Sr dating of detrital
feldspar: A new method to study till: Journal of
Geology, 89, pp. 97-107.

Willman, H.B., Glass, H.D., and Frye, J.C., 1963.
Mineralogy of glacial tills and their weathering
profiles: Illinois Geological Survey, Circular 347, 55

p.
White, G.W., 1939. Tillite and varved slate erratics in

northern Richmond county, Ohio: Ohio Journal of
Science, 39, pp. 317-323.

30

 

A] W .a aw \ Mk): 33,. .P A t .L .. ‘I v . .IIII...I I'QGVIuki c. i3! I. : QM...- II4 I,...\. .FrANr'ildfilq I

F - - .
in II. .
Initial}! f... ,t . . .. . . .. n
. . . . . . . . . . .v ..

"Iiiiliiiliii’illiiii