‘3'

I - o. “- , -

X . $_ A _' ' ,, ._ .

I. Ami: ' «.73.?
.1 ._ _

33%;); . .. .‘L.’"'
,

POSTGLACIAL FOREST SUCCESSION
IN THE. LANSING AREA OF
MICHIGAN?
A STUDY OF POLLEN SPECTRA

Thesis for the Degree of M. S.

MICHIGAN STATE. COLLEGE
George Wyman Parmelee

3.947

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE return on or before date due.

we. we m

Ha:
--—n——=
-==

-==

 

..:WU‘F'"W

 

v_:v':-n:':1.-v'_-':' .

 

 

 

 

This is to certifg that the

thesis entitled

"Postglacial Forest Succession in the Lansing Area of
Michigan: A Study of Pollen Spectra".

presented b1]

George Wyman Parmelee

has been accepted towards fulfillment

of the requirements for

It“ __ _7 degree in_ B_0t3_a1_'11_

 

 

M-795

POSTGLACIAL FOREST SUCCESSION IN THE LANSING AREA
01" III GIIGAN: A STUDY OF POLIEN SPECTRA

3!
GEORGE mam: films

A ‘IHESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and‘Applied.Scienee
in partial fulfillment of the requirements

for the degree of
MASTER OF S CIENCE

Department of Botany and Plant Pathology
1947

THLSiS

 

(DNTENTS

INTRODUCTION0.0.0.000....OOOOCOCOOOOC
REVIEW OF LITERATURE”...H..........

THE IANsmG AREAOOOOOOOOOOOOOOOOOOOO.

LOCATION AND DEB CRIPTION OF B003 0000.
MEmomCCOOCOOOOOCOOOOCOOOOCOCCOCCOOO

REULE00....OOOOOOOOOOOOOOOOOCO0....

DISCIJSSIONO....OOOOOOOOOCOOOOCOOOCOOO
SUMMARY AND CDNCLIISIONsoou..u.o..u

LITERATURE CITED.....................
APPmDIXCOOOOOOOOOOOO...0.0.0.0000...

15f $3953

A C K N O'W L E D G E M E N T

The writer wishes to express sincere thanks to
‘William.B. Drew, Department of Botany, under whose super-

vision this research was conducted, for valuable sugges-

tions and.advice during the progress of the study and
untiring assistance in the preparation of the manuscript.
He wishes also to express appreciation to J. O. Veatch,
Department of’Sodls, for advice on pedologic problems,

to H. T. Darlington for data concerning Chandler Marsh,
and to R. J. Lowry, Department of Botany, for advice
regarding laboratory technique.

POSTGLACIAL FOREST SUCCEBSION IN THE MISSING AREA
OF MICHIGAN: A STUDY OF POLLEN SPECTRA

I N T R 0 D U C T I O N

Students of plant distribution in regions of Pleisto-
cene glaciation have long reOOgnised the influence of
glacial and postglacial history on.present vegetational
patterns and.have sought by every available device to
evaluate its effect. The young science of pollen analy-

sis, while a.most useful adjunct to many other lines of

inquiry, has proved particularly valuable in this quest.
Through its application, conclusions regarding postglacial

forest history may be based upon quantitative data con-
temporary with the period in question, rather than largely

on previously unavoidable inference from reliot communi-

ties and present forest distribution. Although, as Potzger
(1946) points out, pollen.data have served to correct many

preexisting misconceptions based on these phenomena, it is
not to be assumed that pollen analysis provides a master
key to the past, for it is by no means a precision tool
nor is it universally applicable. Nevertheless, where

conducted in a region favorable for large scale correla-

tions, pollen profile studies provide dependable factual
evidence concerning the magnitude and composition of post-

glacial forest formations as well as information with re-

gard to the routes and rates of migration of the component
genera.

That portion of northeastern and.north-central United

States mantled with glacial drift of the‘Wisconsin period
comprises such a favorable region. The range of latitude
and.longitude encompassed provides ample opportunity for

broad regional correlation, as well as the tracing of mi-

gration routes, while the abundant occurrence of pollen-

receptive bogs and.1akes permit ready use of the group-
study method to establish local profile sequences. It is
from the integration and correlation of such firmly estab-
lished local sequences that an evaluation of the determina-
~tive roles of climate, physiography, soil and time in.moldp
ing,the postglacial patterns of forest vegetation.will ulti-

mately emerge.
The aims of the present study are twofold: first, to

provide a valid.profile sequence for the Lansing Area, and
second, to interpret therefrom the history of the post-

glacial vegetation of the area. The study has consisted

primarily of pollen analysis of three profiles from as
many bOgs located in Clinton and Ingham Countieswsithin six

miles of Lansing. In addition, an attempt has been made to
survey such environmental factors, other than climate, as
might conceivably be reflected in the pollen spectra.
‘While it woudd appear presumptuous to assume that an inter-

pretation based on so few analyses could represent more

than a working'hypothesis, attainment of the first objec-
tive is claimed with some confidence. The profiles are

seen to be consistent among themselves to the point that

stratigraphic correlations on the basis of pollen.spectra

are clearly indicated. This congruity is believed suffi-
cient to establish the validity of the area-sequence and,

further, to Justify the belief that, with the aid of the
supplemental information provided, it may be harmonized

with sequences elsewhere .
REVIEW OF LITERATURE

According to Sears (1935). European students of pos t-

glacial history, integrating the contributions of archeol-
ogy, pollen analysis and glaciology, have come to accept

the climatic schedule of Von Post, who has divided post-

glacial time into phases of increasing, maximum, and de-

creasing warmth. The Blytt-Sernander hypothesis, previously
widely accepted, postulates five periods in which moisture,

as well as temperature, fluctuated: l, Pre-boreal (cool-

humid); 2, Boreal (warm-dry, continental); 3, Atlantic
(warm-humid, climatic Optimum); 4, Subboreal (drier, con-

tinental); 5, Sub—Atlantic (return to higher humidity).
According to Gain (194“), the suggestive terminology of
this latter hypothesis led to rejection of the concept by
Von Post .

The stimulating collaboration of investigators employ-
ing different approaches to postglacial chronology, so man-
ifest in the Old World, is largely lacking in the New. This

failure is largely due to natural causes . As pointed out
by Deevey (1943), the arid West is most fruitful from the

archeological standpoint, while pollen analytical methods

4

are best adapted to the humid, glaciated regions of the

East and Northwest. In the latter regions the contribution
of geology toward a postglacial chronology has been limited

by the paucity of varved deposits as compared with those of

northern Eumpe (Antevs, 1936; Griggs, 1942) . In consequence,

chronologies based on North American pollen sequences gener-
ally 1ack confirmation from other sources. Nevertheless,

the regional parallelism displayed by these sequences, t0-
gether with their indications of widespread climatic rever-

sion in the final period (Sears, 1938; Smith, 1940), faith-
fully bear out well-confirmed EurOpean trends, and so supply
valid grouna for acceptance.

The deepest spectra from profiles in the north-central

United States and southeastern Canada normally show an

overwhelming percentage of gigs; pollen. In the succeed-
ing levels, important regional differences appear which will
be briefly summarized.

Profiles from central Indiana (Houdek, 1933; Pretty-
man, 1937; Barnett, 1937; Smith, 1937; Otto, 1938; Howell,
1938; Richards, 1938; Swickard, 1941; Keller, 1943) and
northern Illinois (Voss, 1934, 1937) indicate a long period
of gigs; dominance succeeded usually by a brief and weak
BEEP. maximum. With decline of the conifers, deciduous
dominance is marked and persists to the present.

In Ohio (Sears, 1932, 1942), northern Indiana (Ramp,

1940; Moss, 1940; Swickard, 1941; Potsger and Wilson, 1941;
Keller, 1943), central Wisconsin (Voss, 1934, Hansen, 1937)

5

and southern.nichigan (Potzger andfiWilson, 1941; Sears,
1942; Potzger, 1946) the sequence is similar\except that

the Picea phase comprises fewer foot-levels, and the Pinus
maximum is stronger and more prolonged. In many profiles

a slight increase in Picea and.Pinus pollen is evident in

the upper levels.
Studies in Minnesota (Voss, 1934;.Artist, 1939), north-

ern‘Wisconsin.(Wilson and Galloway, 1937; Wilson, 1938;
Potzger and Richards, 1942; Potzger, 1942, 1943a), northern
Michigan (Potsger, 1942; Wilson and Potzger, 1943b) and
southeastern Canada (Auer, 1930; Bowman, 1931) show that
Pings, after succeeding Picea, remains abundant throughout.
Pollen representation of deciduous species increases in

the middle third of some profiles, and in areas of loamy

soils (Potzger, 1942) continues to increase to the surface.
In the upper layers of most bogs, Picea pollen increases
concomitant with decreasing representation of deciduous
types.

Pollen profiles from northeastern seaboard states fail
to show the distinct initial dominance of Piggg|character-
istic of the foregoing regions. Eight profiles from the
unglaciated.Pine Barrens in southern and central New Jersey
(Potzger, 1945) indicate a heterogeneous forest of conifer-
ous and.deciduous genera characterized by scarcely percep-
tible succession during the period of peat accumulation.

Pollens of Cagya, Castanea and.Tsug§ occur in the deepest
foot-levels.

In northern.New Jersey (Potzger and Otto, 1943) and
Connecticut (Deevey, 1939, 1943) the pollen record beings

with a high representation of Picea, Pinus and Abies

followed by successive maxima of Pinus, Quercus and/or
Tsuga and Gag: . Castanea pollen appears late in the

record, and, with Tsuga, rises in the upper levels as
Carla declines. A

In four profiles from.New Hampshire (Krauss and Kent,

1944) Pinus representation is maximal, and Tsuga pollen is

present, at or near the beginning of sedimentation. Pollen
of Quercus, at no point abundant, reaches highest represen-
tation in the middle foot-levels, while Tsuga attains a

definite maximum in the upper levels.
complementary to the efforts of field workers in

amassing the local and.regional sequences, briefly sum-

marized above, have been the integration.and.correlation
of profile data achieved by Sears (1932, 19353, 1935b,
1938, 1941a, 1942a, 1942b), Smith (1940) and, on a more
regional scape, by Fuller (1935, 1939), Voss (1934), and
Potsger (1946). A

In recent years, particularly, American workers have

shown increasing ingenuity in widening the frontiers of
pollen analytical research. Individual profiles have been

employed to date approximately geological disturbances
(Hansen, 1942a, 1942b; Potzger, 1943b), relics of pre-

historic man.(Deevey, 1943) and.mammal bones (Potzger,

1941), as well as to demonstrate wider former ranges for

7

certain forest genera or species (Wilson and Webster,
1942; Potzger and Thorp, 1943; Cain, 1944a) and a migra-
tion route of plants (Sears, 1941b).

_ Group studies, in the form of transects, have been
used to detect influences of altitude (Deevey, 1943),
longitude (Potzger and Keller, 1944) and latitude
(Potzger, 1946). They have been further employed in un-
glaciated regions (Hansen, 1939; Potder, 1945) for the
purpose of comparing forest sequences therein with those
on adjacent drift-mantled areas.

local sequences have been correlated.by Sears (1941a,

1942b) to determine the postglacial migration routes of

prominent forest genera and, on a much reduced scale, by
Potzger and Friesner (1939) as an aid in tracing plant

migration ianndiana.

Several workers, cognizant of the great benefits to
be obtained by refinements in pollen analytical technique,

have been active in attempting to eliminate various of the

sources of error or difficulty in technique and inter-
pretation. For example, size-frequency studies (Cain,

1940, 1943, 1944b, 1944c; Deevey, 1939) on certain conifer-

ous pollens from species native to the eastern United States
have been made in an effort to evaluate the possibility of

specific determinations. Geisler (1945), on the basis of a
similar study of 32 species of grasses, has pointed out

the overlap in size of pollens from upland and.marsh species.

In the Pacific Northwest, Hansen (1940a, 1941a, 1941b)

has discussed the characters used in distinguishing pollens
of coniferous species whereby he has been able to reconstruct

very detailed forest sequences for that region (1943, 1944).
Boring equipment designed to penetrate sandy sedhments,

and to eliminate the bottom-truncation so prevalent in pro-

files obtained by movable-cylinder borers, has resulted in

significant additions to the pollen record (Potder and
Wilson, 1941;'Wilson and.Potzger, 1943a).

Carroll (1943), employing bryophytic polsters retentive
of pollen, has devised a technique which permits direct

comparison of pollen representation with existing forest

composition. The usefulness of this particular study is

limited by a mountain locale exhibiting altitudinal dis-
tribution of farests. However, the technique offers promise

as a means of measuring over- and under-representation, as

well as portability, when employed under suitable conditions.
T H E L A.N S I N G A R E A

.An intimate relationship exists between pollen bear-
ing sediments in a given basin and its geologic history.
Scarcely less significant, in view of the mobility of wind-

borne pollen, are the gross aspects of surrounding physio-
graphy, soils, and vegetation. Sears' (1941a) statement
that supplementary study of the basin can wait, obviously

refers to complex.prob1ems awaiting solution by specialists
in other fields, and in no way repudiates a general study
of the surrounding area as an aid to profile interpretation.

9

A positive example of the value of such supplemental in-

formation is provided by comparison of pollen profiles_from
the Trout Lake Area of Wisconsin and the Gillen.Nature Re-

serve, on the‘Wisconsin-Michigan border'(Potzger and

Richards, 1942; Potzger, 1942). In the former area, char-
acterized by sandy soils and pine forests, the profiles
indicate undiminished.persistence of pine, while in the

Reserve, a region of loamy soils and hemlock-deciduous
forests, the profiles register a distinct decline of pine

in the upper one-half to one-third. Neglect to record
edaphic conditions in these studies could obviously have

lead to highly contradictory climatic interpretations.
It is held necessary, therefore, to include in this

study such available supplemental information as might
aid in interpreting profile data, which might appear other-

wise to be anomalous.
PH!SIOGRAPHY’AND,§Q;Q§

The lensing Area is a component of the Southern Up-
land Division of Michigan, a broad glaciated plain rising

from 200 to 600 feet above the bordering Great Lakes and

from 200 to 400 feet above the lowland plain to the north,
which traverses the state from Saginaw Bay to Lake Michi-

gan. The drift covering is everywhere sufficiently thick

to mask direct influence of the bedrock on tepography,
except in the vicinity of Grand Ledge, where the Grand

River has cut through a ledge of outcrOpping Eaton.Sand-
stone of the Grand River group (Martin, 1936). Confining

10

the river to a narrow channel for nearly a mile, the pre-

cipitous slepes are about 50 feet high and are incised by
a few short and narrow ravines, of which the mouth of Sand

Creek is the largest.

GLACIOLOGY

0

So direct is the relation of the Cary substage of
Wisconsin glaciation to the physiography of the area, and

particularly to the origin of its numerous bogs and bag

lakes, that it is necessary to treat this phase of the
tapographic development in some detail.

The physiography and glaciology of Hichigan have been

described by Leverett and Taylor (1915). Lansing lies near
the center of the exposure of Cary drift (Late Wisconsin of

Leverett) resulting from depositional activity of the Saginaw

Lobe of Huron ice.
Retreat 91 the Can Ice

The manner of retreat of this ice mass may be in-
ferred, in part, from the type, magnitude and position of
the drift features left in its wake. Thus, two distinct
modes of retreat are implicit in the sharply contrasting
drift features to the north and south of the Lansing
Moraine, a low, narrow upland passing in an east-west dir—
ection through the southern environs of Lansing. To the
south, the numerous eskers traversing the till plains of

the‘Kalamazoo and Charlotte till plains, as indicated on

11

the map of Leverett and.Taylor (1924), are features thought

to mark the courses of subglacial streams (Davis, 1892) or
ice-walled canyons (Goldthwait, 1939) associated with ex-

tensive fields of stagnant, downwasting ice. Such a fonm
of decay is believed.by Thwaites (1946) to result in dis-

continuous ice remnants characterized by differential rates

of melting, the latter in response to varying thickness of
debris on the surface., An idea of the magnitude of this

inferred ablation zone is provided by the Mason esker which,

from its point of origin within Mount HOpe Cemetery, near

Lansing, pursues a serpentine southeasterly course for over
20 miles.

The total absence of eskers north of the Lansing moraine

heightens the contrast between the area previously noted and
the morainal belts to the north. In this direction the

Lansing moraine comprises the southernmost of seven closely-

spaced, narrow moraines arranged in concentric fashion about
a center at the head of the Saginaw Lake Plain. Their
spacing and width, tagether with relief in most places of

less than 30 feet, and a universal failure to be associated
with well-developed outwash plains, forcefully suggest de-

position by a rapidly retreating active ice front.
Additional conditions contemporary with retreat of the

Cary Ice Front from this area, and.having a bearing on the

ecesis of vegetation and.preservation of pollen will be

further discussed following presentation of the profile

8 equen 683 e

12

Secondary Tapoggaphic Features

The mode of retreat appears to have influenced neither
the type nor abundance of secondary tepographic features

other than eskers. Comprising landforms common to moraine

and till plain, they consist of rounded, low hills with a
complexity of short slopes; shallow potholes or swales

alternating with gentle swells of upland; and widely dis-

tributed depressional features of diverse size and shape.
The latter occur chiefly as (l) elongated basins with a

north-south trend, probably resulting from glacio-fluvial
activity; (2) irregular, sprawling, relatively shallow

depressions of greatly varying size; and (3), of deep,

strongly concave basins derived from slumping of overlying
drift following melting of buried ice blocks. In the

Lansing Area, the first type of basin is confined to till

plain, and in.Ingham County reaches best expression in
Alaiedon and White Oak Townships. The irregular, shallow

basins are far>more numerous and more widely distributed,

but reach best deve10pment on till plain also. The deeper

basins, commonly termed kettles, and characteristic of
morainal tOpography, are of rather common occurrence near

Lansing on till plain as well.
Chandler Marsh, confined chiefly to Dewitt and Bath

Townships, Clinton County, comprises an additional type of

basin. Located between the Ionia and Grand Ledge Moraines
on the till plain of the latter, it was, in Cary time, an

ice-front lake receiving water from ice standing on the

13

Ionia Moraine. Retreat of the ice to the Portland.Moraine

and consequent cutting off of water supply ended its brief
tenure as an ice—front lake, although a considerable vol-

ume of water remained as indicated.now by marl and peat

deposits to a depth of 15 feet in the center of the basin.
c/Drainage. In an area characterized by such complexity

and disarray of land form, isolation of a considerable pro-

portion of the basins from any drainage system was inevit-
able. Only streams possessed of the erosive power inherent

in torrents of glacial meltwaters were able to negotiate

the irregularities of the drift surface, and for the most
part, the principal drainage patterns then instituted per-
sist to the present day.

POSTGLACIAL MODIFICATIONS

TOpographic changes, effected by erosion and sedi-

mentation through the centuries of postglacial time, are
considered here in the light of their capacity for inducing

forest succession in this area. or added concern are the

principles and processes involved in the develOpment of

bogs and bog lakes from youth to maturity and.senescence.
Geological Erosion and.Btream gissection

The present upland landforms of the area reflect in

major part constructional activity of glacial origin,

pointing to the relative resistance of such t0pography to

geological erosion and stream dissection. Concerning the

14

effectiveness of these forces on Late‘Wisconsin (Cary)

drift, Leverett (1909) remarks that "the small hummocks

and basins still preserve their sharpness of contour,"
and.adds the/estimate that “scarcely one-tenth of the

surface has been reduced.below the original level as a

result of drainage.” This follows quite logically from
the vast number of water-collecting depressions or basins

completely devoid of surface drainage, and hence greatly

diminishing the volume-of water available for stream dis—
section. As previously stated, the stream patterns re-

flect glacio-fluvial activity, and it does not seem in-
valid to assume that considerable of the down-cutting

achieved.by these streams may have occurred during that

time.’ Incontrovertible evidence of such meltwater ero-

sion is provided by the vast discrepancy between valley
and present stream magnitudes of Hayworth and Stony

creeks in the northern half of Clinton County. These

valleys are not comparable to those in the vicinity of
Lansing, however, since during the Late Maumee stage of

Great Lakes history they carried the entire discharge of
that great glacial lake (Leverett and Taylor, 1915).

Nevertheless, the narrow modern channels, flanked through

much of their course by extensive muck deposits, reaffinm

weak erosion and at the same time suggest more active

sedimentation.during postglacial time.

15
Sedimentation and.Bog Qevelopment

The evidence of Hayworth and.Stony Creeks as regards

relative efficiency of erosion and sedimentation is every-

where substantiated in the Lansing.Area. That this re-
lationship has long obtained, follows from the large

number of seepage lakes in the late stages of development

from maturity to senescence and extinction. Sedhmentation
in the majority, if not all such basins, has been achieved

by the bog mechanism. This, as defined by Gates (1942),

is ”preeminently a type of vegetation which controls a
habitat and changes the habitat, in the course of its

deveIOpment, from an cpen area of water to a mat and then

to a grounded.mat and finally to dry land.” That annihi-
lation of Open water may long antedate completion of the

bog phase as thus defined is demonstrated by the partly
cultivated County Line bag in Ohio, described by Rigg

(19#0) as having a mat four to six feet thick underlain

by about 40 feet of water. This same condition, to a
lesser extreme, repeatedly frustrated attempts by the

writer to obtain complete core series from the bogs of

the Lansing Area.
Determinants of Sedimentation gyp_. (Studies on the

developmental details of lakes can be considered valid

only in the specific region of their occurrence. Develop-

mental generalities, on the other hand, have a dispro-
portionately wide significance. This follows, no doubt,

from the monotonous uniformity of the water environment,

16

a condition evidently carried over even to bogs, as indi-

cated by the studies of Rigg (1940), who found more than

half the common bOg plants of eastern and western North
‘America to be identical species.

‘Whether a given water area will evolve in the direc-

tion of a bog or a marsh appears to be determined by three
interrelated factor complexes, viz., tapographic, chemical,

and vegetative. ‘Wilson (1935), in a careful study of lake

development in‘Wisconsin, attaches causal significance to
the drainage isolation of bog lakes. Gates (1942), while

concurring, stresses the contribution of shelter to bog

initiation, stating that bOgs may develop at quiet places
in stream channels, or be restricted to protected bays in

large basins subject to wave action.

The topographic complex, tOgether with mineral compo-
sition of the earth materials of the watershed also in-

fluences the course of lake development by chemical means.

Thus the institution of bog conditions as the usual after-
math of settling mineral salts and leaching of peripheral
soils (with concomitant increase in acidity) is presumably

accelerated in an undrained basin with a small, lime-
impoverished watershed. However, subsequent development

of such a bOg is likely to be impaired by a deficiency

of the leachate nutrients needed to replace those held
unavailable in the accumulating organic debris. So great

may be this deficiency in regions of sterile sandy soils
that bag conditions may fail to deve10p entirely despite

17

otherwise highly favorably conditions. Such appears to
be true of Crystal Lake, Vilas County,‘Wisconsin, described

by Twenhofel and Broughton (1939) as possessing the softest

waters of the state./

Assuming the validity of such suggested edaphic con-
trol, one might employ it further in connection with the

findings of Gates (1942) in northern Lower Michigan. Here
the association of many begs with drainage systems might

be explained by the acid drift over which the streams

flow, and from whence they derive insufficient mineral

nutrients to support marsh vegetation. Through default,
then, the slower growing bag vegetation is able to utilize

the small but continuously renewed nutrient supply.

Antithetical conditions conducive to a marsh type of
sedimentation are those in which a basin forms part of a

drainage system traversing earth materials well supplied

with mineral salts. Such a basin ordinarily supports a
profusion of calcicolous aquatic and.shore forms. However,

in cases where marl is precipitated in great quantities,

as in some southern‘Wisconsin lakes studied by‘Wilson
(1935), vegetation.may be almost completely absent. Accord-

ing to‘Welch (1935), such impoverishment is usually the
result of various filling processes which periodically

isolate the marl from contact with the overlying water,
and thus effectively prevent conversion to the soluble

bicarbonate during periods of abundant carbon dioxide

production. The precipitation is thus permanent, and

18

where it exceeds replacement by inflowing streams or

leachate from contiguous uplands, productivity declines

and with time reaches a level below the requirements of
calcicolous plants.

Such a decline in productivity might conceivably
have contributed to the deve10pment of certain of the

bogs in the Lansing Area, inasmuch as marl deposits from

12 to 17 feet deep were regularly encountered during pre-
liminary borings within the morainal plexus area north-

eastward from Chandler Marsh. In view of the continuing

alkalinity of the drainage waters in the latter area,

however, a more tenable hypothesis would attach causal
significance to the shoaling effects of such deposits,

rather than to the possible competitive advantage prob
vided calcifugal plants. Such a view would then be in

harmony with that expressed by Veatch (1933) who believes

that once filling has progressed to the point where exp

panses of very shallow water occur, calcifugal plants may
ecize regardless of water reaction or prior occupancy by

calcicolous communities. ‘Welch (1934) has attributed such
a capacity to the lack of circulation with resultant

increase in acidity, within organic accumulations formed

by such vegetation. In support, he cites the frequent
occurrence, in.Sphagnum mats, of water having a strongly
acid reaction and separated from distinctly alkaline

waters of the cpen lake by a gradient of only a few inches.

That such a capacity to ecize might lead to bog

19

formation even in a hardwater lake is also suggested by
Gates (1942), who, after dismissing water reaction as a

critical factor, states that "the type of vegetation
which deve10ps along the shore determines the fate of the

area." Elaborating, he specifies Sgigpus validus, g.

americanus, etc. as forerunners of marshes or woody

swamps, and Carex lasiccarnfi: in sufficient abundance
to form an ecological association, as necessary to ini-

tiate a bOg. The extent to which the occurrence of these

types is a function of previous soil and topographic
development, or of pure chance, is not discussed.

In the Lansing Area, Decodon verticillatus appears

frequently to assume the role of Carex as bog pioneer.
It has been observed to be actively extending the mat

framework on three bog lakes, with a similar function

in the past suggested by its occurrence as a relict on
two additional bogs now closed. Subsequent stages of the

local bog priseres are typical of northern United States

(Rigg, 1940), there being evolved communities dominated
successively by Chamaedaphne in association with.Sphagnum,

high bog shrubs, and bog conifers. The latter community

will be discussed later in connection with relict com-
munities.

Conditions Influencing Preservation of Pollen. In

bog development the cold, anaerobic, highly acid conditions
conducive to mass preservation of organic debris are

first encountered in the Chamaedaphne482haggum zone. That

20

such drastic antisepsis is not essential for the preserva-
tion of pollen, however, is demonstrated by abundant pollen

deposition in oligotrophic lakes (Twenhofel and Broughton,
l939;‘Wilson.and.Cross, 1943), eutrOphic lakes (Potzger and

‘Wilson, 1941; Deevey, 1943), or at levels long antedating

institution of bog conditions, as exemplified by Forestry
Bog Lake (Potzger and Richards, 1942). Rather, advance of

a bog mat imposes a distinct barrier to movement of pollen

to the deeper levels, where inhibited bacterial activity

and reduced oxidation favor more certain preservation. .An
apparent manifestation of this effect is provided by the

profile samples from Chandler Marsh where the transition

at five feet from detritus ooze to coarse sedge-peat is
coincident with a sharp decrease in pollen frequency.

Low pollen frequency is occasionally associated.with

marl, particularly in the upper portion of thick deposits.
This has been attributed by Voss (1937) to cumulating pOp-

ulations of calciphilous molluscs which feed on organic

‘ matter including, presumably, pollen grains.’ Whether the
paucity of pollen at the 29th foot-level of Mud Lake Bog

may have resulted from such scavenging is cpen to question,

since the marl deposit here is only two feet thick.
SOILS
Textural and Drainage Relations

Veatch (1930) has included the Lansing Area in a
natural land division designated by him the Clinton Rolling

21

Plains, and described as level to rolling clay plains with
soils chiefly of loams over compact clay. Some indication
of the predominance of imperfectly, or well drained, fine
textured soils in this division is provided by reference to
Table 1, comprising data taken from soil surveys for the
three counties of which the Lansing Area is a part (Moon,
1933; Veatch, 1941a; Johnsgard, 1942).
ance of heavy soils is only partially expressed, is sub-

That this preponder-

stantiated by the occurrence of considerable areas of a
contiguous land division, the Hillsdale-Lapeer'Sandy High-
land, in the southern parts of Ingham and Eaton Counties.
Despite this inclusion of lighter-textured.soils in the
data, it will be observed that, with but two exceptions,
fine textured soils overwhelmingly dominate all three drain-

age categories of mineral soils in the three counties.

TABLE 1. Percentage representation and drainage relations
of principal soil types of Eaton, Ingham, and
Clinton Counties. Finer textured series are on

left in each drainage category.

 

 

 

 

 

 

MINERAL SOILS ORGANIC
SOILS
Well Imperfectly Poorly
Drained Drained Drained
m
n
a) .4
E: m -3 C .
33 '3! 8 e c: .‘3 '0 2:
09 I3 m; m o- w H w
-H w o b >. w .M o .4 m
.4 E H r—l O '0’ £2 0 ‘H H H
»4 m r1 r4 N g g g g C: E :1
8 a :E‘ 38 r3. 0 m m “1 <5 0 ‘1‘
Eaton 31 e2 1.0 e8 3 e0 2 e6 20 02 406 3 e7 2 e7 3 09 6 03 2 0}
Ingham 12.6 16.6 4.3 5~O 21. 7.0 -- 7.0 -- 9.7 4.1
Clinton 27.5 3.1 2.5 3.5 23.5 3.8 2.4 11.7 1.1 3.2 3.9

 

 

 

 

 

 

 

 

 

 

22
Postglacial Soil DevelOpment

The youthful stage of the geologic erosion cycle in

this area suggests soil develOpment under relatively con-

stant surface drainage conditions. That changes have
occurred, nevertheless, is indicated by Veatch (1938) who

cites as evidence the common occurrence of dry valleys and

basins in many parts of the state. The former commonly
display faint dendritic patterns attesting to postglacial

origin, while the basin floors were submerged or water-

logged.sufficiently long to develOp mineral soil profiles
characteristic of inundation. The many centuries neces-

sary to develop a soil profile leads Veatch to correlate

the disappearance of these water bodies with the well
documented (Raup, 1937; Sears, 1938; Transeau, 1935. 1941)

xerothermic period of relatively late postglacial time.
Regarding the degree of weathering evidenced by Late

Wisconsin (Cary) drift, Leverett (1909) stresses the
remarkably slight depth of leaching, estimating that the

average distance to unleached till is somewhat less than
a meter. The foregoing evidence would.indicate that soil

develOpment, other than that dependent upon climatic change

or plant reaction, may be largely discounted.as a factor
of forest succession.

Organic deposits supply direct evidence of postglacial
soil deve10pment. Besides materially contributing to the

space available for ecesis (see Table I), they provide

sanctuary for characteristic communities of boreal relicts,

cognizance of which must be taken in interpretation of

23
pollen spectra.
FOREST VEGETATION //

To the pollen analyst, present forest cover is of
interest primarily as a means of checking the fidelity
with which pollen spectra record the presence and rela—
tive abundance of its components. It is, however, a check

subject to limited application and.difficu1ties of execu-
tion. Obviously, despite the evident stability of late

postglacial edaphic and topographic conditions, the present

generic composition of the surrounding forest is only of

value for comparison with the tap-most spectra of the pro-
file. floreover, its dependability, insofar as an indicator

of pollen over- or under-representation is concerned, is

further limited by the reliability of quantitative distri-
butional data of forest species for comparison with pollen

spectra percentages.

As a result of these shortcomings, the existing forest
vegetation is often summarily dismissed by pollen workers.

Certain phases of the present profiles, however, appear

sufficiently eccentric to warrant marshalling of all avail-

able forest-cover data as an aid to interpretation. ‘
Climatically Favored Communities

Climax forest communities display local groupings of

dominants in response to edaphic variations. Veatch (1932)
has pointed out that this distributional equivalency may be

24

practicably exploited in units as small as the soil type.
Later (1941a), in summarizing the probable virgin forest

groupings for the larger soil bodies of Ingham County, he
attained as close a degree of such correlation as appears
possible from the fragmentary records and remnants of
original cover available for study. It is evident from

this tabulation, in which frequency is expressed in a gen-
eral way by order of listing, that relative abundance of a

species, rather than its presence or absence, must serve as
a criterion for forest type differentiation. Thus, while
elm is listed as a component on every soil series, it is
only in the imperfectly and poorly drained categories that
it attains controlling influence. There appears no ground
for supposing that groupings so devised, while referring
specifically to Ingham County soils, should not prove
equally valid for similar soil types in Clinton and‘Eaton
Counties. Accordingly, the groupings typical of the soils
included in Table I, and therefore considered co-extensive
with them, are here presented with the view of expressing
their approximate areal extent as a percentage of the
original total cover of the three counties:

Sugar maple-beech, with white oak, elm, white ash

“m1 COO-IIO-m-I-OCII-c-c-D ------ O--&--.‘--------- 24%

Oak-hickory, with sugar maple, beech, elm
(Hillsdale, Bellefontaine, Fox) -------------- - 17%

Elm-ash-basswood, with oak, hickory, beech, sugar
ma le, walnut, butternut
Conover)----- ------------------------------- - 22%

Elmpsilver.maple-ash, with shagbark hickory,
swamp white oak, basswood

(Brady, Brookston, Carlisle muck)------- ------ 19%

25

Tamarack-aspen, with red maple, elm, occasional
black spruce, paper birch
(Rifle peat)-- --------------------------- ----- 3%

32;;33 Communities

Postglacial advance and consolidation of deciduous
forests in Clinton, Ingham, and Eaton Counties has reduced

the pre-existing boreal formations to less than 4% of the

total area. However, to the student of postglacial vegeta—
tion, these communities merit considerably more attention

than their areal extent would indicate. Their normal beg

habitat assures incidence on a receptive surface of a high
prOportion of the pollen produced. Moreover, the extensive.

and rapid expansion of mat surface coincident with the

closing phases of bog deveIOpment might well be reflected
in contemporary pollen spectra through an increase in the

area available for ecesis of boreal plants.

ggg Relicts. In the few bogs of the Lansing Area not
profoundly altered by drainage and repeated fires, the seral
stages for the most part parallel those described by Gates
(1942) for northern Lower Michigan. Or the forest stages,
the chief divergence from those of the more northern bogs

lies in the restricted distribution of Thuja and Picea mari-

ana. ‘No Larixfig..mariana-Thuja sequences, and but two in-
stances of Larix-Thuja succession are known to the*writer.

Although Larixe§.mariana sequences are frequent, only two

cases have been observed where the latter species achieves

more than the most meager representation.

26

Larix is common and.widespread throughout the area, an

occurrence perhaps correlated with its normal habitat on
floating mats or otherwise waterlogged substrates beyond the
zone of bog fires. Picea mariana is also typically found

on such sites, though usually occurring out of successional

order, as much stunted individuals within the high bog-shrub
zone. Far better deve10ped, however, is the nearly pure
stand at Mud Lake Bog in Delta Township, Ingham County, where
20-30-foot trees occupy a quaking mat six feet thick and sep-
arated from detritus ooze by about nine feet of water. The
other well develOped stand.of g. mariana, while not confined
to a floating mat, occurs on an undrained area north of Park
Lake in Bath Township, Clinton County. Such a present dis-
tribution pattern suggests the possibility of a much more
common occurrence of this species prior to widespread bog
drainage and the repeated fires that followed on dried-out,

stranded mats. By their much poor growth on such mats, in

comparison.with peripheral stands, the aspen invaders pro-
vide further’substantiation of an abbreviated bog forest

tenure.

The Thuja stands, while located in basins whose water
relations have been altered by drainage, show no indication

of fire. This is most obvious at the Cedar Lake station
where Thuja grades into a stand of mature Bgtula lutea,

developed by sucker growth from large stumps showing no

scars of fire. ‘While the priseral succession was inter-

rupted by removal of the primeval birch stand, its rapid

27

regeneration apparently achieved stabilization quickly, as

evidenced by restoration of nearly pure dominance. The

shoreward Larix-ThujagQ. lutea sequence may therefore be
considered normal, with only the succession from bag forest

to lowland forest subject to influence by man. The pre-

ponderance of Ulmus americana, Fraxinus nigga, and.Acer
rubrum in the fragments of lowland forest remaining at the

bog periphery probably reflects the original composition,

however, and suggests ultimate preemption of the bog site
by these species. Such succession is likely to be long

delayed since it is ”beyond the bOg“ and hence contingent
in large degree upon the elimination of edaphic factors

favoring bog forests. The effect of these compensatory

factors at Cedar Lake is well expressed by the vigorous
young colony of Thuja which in recent years has invaded a

small peaty depression about one-quarter mile to the east.

The second.Thuja stand located on the broad marshy
flats north of Potter's Lake in Bath Township, Clinton

County, is in an advanced.state of decadence owing to heavy

windthrow and perennial trampling by grazing animals.
Upland Relicts. On only two upland sites have com.

pensating habitat factors been sufficiently operative to
prevent the dominance of climatically favored genera of
trees. The influence of physiography is evident at Grand
Ledge in Oneida Township, Eaton County, where M persists

on precipitious north and.northeast facing ledges centering
about the deeply incised mouth of Sand Creek.

28

Less obvious are the factors responsible for persis-

tence of the second relict, a Pinus Strobus stand.esst of

Pine Lake in Meridian Township, Ingham county. Its occur-
rence in a locality characterized.by dry “islands“ of

Coloma loamy sand.interspersed.with Rifle peat, however,
suggests the influence of both isolation and edaphic com-

pensation. Thus, of the deciduous genera, only the more

xeric species of oak might be expected to compete favor-

ably with pine established on such soil. ‘With falling
water tables and removal of pine in recent years, oaks have

been invading the site with increasing ease, especially

since 1940 when the pines were decimated by lumbering Oper-

ations.
LOCATION AND DESCRIPTION OF BOGS

In this study the selection of bogs for investigation

was conditioned by a desire to obtain profiles sufficiently
localized to yield a valid area-profile, while at the same

time of adequate depth to avoid the danger of surface trun-

cation. Of the numerous bogs visited during reconnaissance

work, the three whose descriptions follow most nearly satisfy
these conditions. All have develOped in kettle basins more

than 20 feet deep and are located less than 12 miles apart
on a nearly perfect straight line oriented approximately

30° E of North.

29 ‘
BEAR LAKE BOG

Bear Lake and the major portion of its surrounding bog
is located in Section 35, T 4N, R 2W. In area, the irregular

basin tOgether with the lake approximates a quarter section.

Of this area, the lake occupies somewhat in excess of 12
acres, being located near the western.margin at the broadest

portion of the bOg. The boring, through 23% feet of sedi-

ments to sand, was made on the east side of the lake about
40 feet back from the bog margin. .A test boring at15he edge

of the mat revealed 26 feet of deposition, but eight feet
of this immediately below the mat lacked sufficient consis-

tency to Operate the borer mechanism.

Surrounding Tapography and.Soils

Bear Lake is located near the northern margin of the

Lansing recessional moraine in a zone transitional from

moraine to till plain. The prevailing tapography is gently
undulating and everywhere marked by small undrained depres-

sions, some well above the water table, others containing

small marshes or bags. The greatest relief, about 20 feet,
is provided by a series of rather abrupt slopes which form

the western and southern rum of the basin. Hillsdale sandy

loam on the north, west and south, and.Conover loam to the

east, comprise the upland soils of the vicinity.

Recent Histogy

County records indicate that a bog forest existed

30 ’

intact until sometime after 1870, for at that date the
County Surveyor, in commenting on his failure to run the

line between.Sections 34rand.35, described the area as

“totally inaccessible swampland.“ By 1909, however, the
area was at least partially cleared as it was then pos-

sible to run the line and establish the half-mile corner.

This was confirmed by a conversation with Mr. Otto Andrews,
for nearly 60 years a resident of the vicinity, who stated

that within his memory a tamarack stand covered most of the

bog, but that fence post demands at the turn of the century
resulted in almost complete clear-cutting of the area.

Destruction of the bog forest led to conditions highly
favorable for the advent of fires which, according to Mr.

Andrews, have raged periodically for the last 40 years.

Especially clear in his memory are the fires of 1907 and
1908, ashes of the former conflagration having covered the

lake to considerable depth and.killed fish "by the wagon

load.”
Present Vegetation

The least disturbed plant communities comprise those

protected from fire by virtue of their habitat on the float-
ing mat. A marginal zone of Decodon verticillatus completely

encircles the lake and.by its buoyant rootstalks provides

framework for the advance of Chamaedaphne calyculata and a
deep, resilient carpet of Sphagnum. Shoreward, this leather-
leaf-sphagnum zone grades into a high bog-shrub community,

31

forming an almost impassable tangle 25-50 feet wide, con-
sisting of successively taller representatives of Gaylus-

sacia baccata, Aronia melanocarpa, Vaccinium corymbosum and

NemOpanthus mucronata. Less common are Rhus vernix, Ilex
verticillata and Picea mariana, while in the eastern and

southwestern segments mature Larix laricina and Acer rubrum

occur as remnants of the former bog forest.
Beyond this marginal fringe, leatherleaf, characterized

by its quick recovery following fire, occurs in pure stands,

or more often, in.mixture with scattered, low Vaccinium
corygbosum, Aronia melanocarna. and.Nem9panthus mucronata.

.Although Pogulus tremuloides readily invades the periphery
of the deposit after each fire, intensity of burning in

this zone virtually assures its destruction with each suc-

ceeding conflagration.
CHANDLER MARSH

Although so designated forwwant of a better name, the
basin confined to the NE } of Sec. 32, T 5N, R IW, differs

markedly in origin and subsequent development from the

contiguous Chandler Marsh proper. Only imperfectly isolated
from the Chandler Marsh-Park Lake basin by low sandy divides,

its status as an independent kettle basin is nevertheless
confirmed by abrupt peripheral bottom slapes to depths con-
siderably below those of the former. This is particularly
manifest at the site of the boring, on the west side, just

south of the drainage ditch, where 24 feet of organic sedi-
ments occur within 150 feet of the sandy rhm. ‘A further

32

unique feature in this locality is the complete absence of

marl deposition, an indication of water so deep, or perhaps

so turbid, as to prevent the growth of Chara and other lime-
precipitating organisms.

A long period of cpen water deposition, preceding a

recent and rapid advance over shoaling waters, is implicit
in the extent and nearly uniform thickness of the floating

mat. Occupying fully three-quarters of the basin of about

80 acres, it assumes a definite quaking aspect at an aver-
age of 150-200 feet from the periphery, and though every-

where sufficiently tenacious to support a man's weight, it

grows progressively more tenuous toward the center.
Surrounding Topography and Soils

This young bog forms part of the eastern margin of a
peat-filled, multiple basin, attaining a depth of 15 feet

and exceeding four miles in width and.aeven square miles

in area. Park Lake, at the extreme eastern end, represents
a deeper portion in which bog development has been strongly

retarded by wave action except in the southwest part. Two

test borings along the margin in this area revealed a nearly
level bottom overlain by about 12 feet of marl and 5 feet
of fibrous peat.

From north and west of the lake, an irregular salient ,
of sandy soils, chiefly Oshtemo loamy sand and.Berrien loamy

sand, trends southwest for more than a mile into the marsh.

Elsewhere the upland soils are of heavier texture. Moderate-

sized bodies of Brady loam and Conover loam form the southern

33

margin of the bog, with Hillsdale sandy loam of a mostly
rolling phase prevailing to the east.

mm:

The magnitude of change induced by the Remy-Chandler
drain is perhaps best portrayed by the soil map for Clin-

ton County where essentially the whole of Chandler Marsh,

once famed.as a mecca for migratory fowl, is now mapped as
a burned phase of Rifle peat. Burning in the kettle basin

was of negligible extent, however, owing to the great pre-

ponderance of floating mats which remained water-logged by
subsidence to the new level. Nevertheless, important

drainage effects are indicated by the rapid closure of the

met during recent years, a process accelerated by the six
foot drop inywater level and the consequent inhibition of

‘wave action. Subsequently, except for such effects inci-

dental to reclamation of the adjacent organic soils, the
bog has been little disturbed, its mat and peripheral fringe

of bog forest being so immature as to preclude any possible
utilization by man.

Present Vegetation

Further confirmation of recent and rapid bog development
is provided by the curious interpenetrations, and in places

even mixtures, of calcifugal and normally calcicolous species.

This is evident on the sedge mat where colonies of Sarracenia

pugpurea, Betula pumila, and occasionally Sphagnum, regularly

34

alternate with socies of Typha latifolia and Scigpus validus.

In more shoreward sites, particularly on the west and south,
Chamaedaphne calyculata and Typha latifolia frequently occur

in mixture with about equal expression of dominance.

Betula pumila is everywhere the pioneer of the shrub
stage, followed at a considerable interval by Cornus stolon-

ifera, and infrequently, by vaccinium corymbosum. ‘Where

present, Chamaedaphne calyculata shows remarkably little
aggressiveness, having been overrun usually by the advanc-

ing tamarack bog-forest. Sphagnum shows a similar tendency,

even where leatherleaf is best developed, and over much of
the area is supplanted by Aulacomnium palustre, and to a

lesser extent, Polytrichum strictum. The peripheral belt
of Larix laricina, averaging about 150-200 feet in width

is not everywhere continuous and at such points Populus

tremuloides and Acer rubrum invade the bog a short distance.

Elsewhere these species, together with Ulmus americana and
Prunus serotina, are established on the mineral soils border-

ing the bog and are invading the tamarack zone.
MUD LAKE BOG

The basin so named is centered about 120 rods south of
Holt Road in the NE% of Section 21, T 3N, R 2W. Its surface

is completely closed by a mat which at the approximate

center is five feet thick and underlain by nine feet of

water and.peat too fluid to permit sampling. Insufficient
borer extensions prevented depth determination at this site,

but increasing compactness and.marl content of the sediments

35

at 31 feet indicated the bottom was close at hand. The

samples for analysis were collected from a 30-foot boring
on the grounded mat Just south of a fence 80 rods south

of, and paralleling Holt Road.
Surrounding ToEoggaphy and.Soils

This basin, tOgether with that of Mud.Lake proper

one-quarter of a mile to the north, lies in one of the
shallow but very extensive swamp depressions ramifying

throughout the Charlotte till plain. Originally serving

as drainage ways for glacial meltwaters, such depressions
became swamps with cessation of that flow and are today

filled.with deposits sufficiently decomposed to be mapped
as Carlisle muck. The uplands of the vicinity are gently

undulating and in general characterized by imperfectly to

well-drained clayey soils of the Conover and Miami series.

Recent History

Although its marginal fosse has been connected by a
shallow ditch with Mud Lake and an outlet to the north,

the bog remains today essentially priseral, the evidence of

fire following drying out being confined to a narrow peri-

pheral portion at the northeast corner. In this sector
Chamaedaphne calyculata and‘Vaccinium corymbosum daminate

about an acre. To the north, on the senescent bog, between

the two kettle basins, frequent peat fissures, fire scarred

aspen stubs, and a nearly pure stand of even aged Aronia

36

melanocarpa covering about 40 acres attest to more exten-
sive and severe fires in the locality. Within the bog

forest numerous decapitated black spruoes, apparently re-
sulting from cutting to fulfill local Christmas tree de-

mands, comprise the only visible disturbance.
Present Vegetation

Presenting a remarkably faithful boreal aspect, the
bog forest of 20-odd acres is largely dominated by Picea

mariana. Individual trees commonly reach 20 to 30 feet
in height, which, according to Otis (1931), is the maximum

size for this species in Michigan. Scattered throughout
are much taller specimens of Larix laricina, which though

reproducing to a minor extent, appear relegated to relict

status by the more aggressive spruce. This relationship

is best exemplified in the burned are to the northeast
where black spruce is vigorously invading without competi-
tion from tamarack.

Vaccinium corymbosum dominates the Shrub layer within
the forest, with Chamaedaphne calyculata and Leda! gggg_-

landicum frequent throughout. 0f more restricted distri-
bution are Vaccinium macrocarpon and Andromeda glaucophylla.

Sphagnum is generally distributed and especially well
developed, often forming mounds as high as a foot above the

deep carpet. Aulacomnium palustre and golytrichum strictum,
where present, are usually associated with such mounds.

Completely surrounding the bog forest and serving.as

a line of demarcation between bog and swamp forest on east,

37

south and west, is a zone of NemOpanthus mucronata, varying
in width from about 25 to'75 feet; on the north, it is wider

and grades into the extensive chokeberry consocies. Co-

dominants of the swamp forest encroaching on three sides
are chiefly POEulus tremuloides, g. grandidentata, Acer

rubrum, Ulmus americana and Prunus serotina. Herbaceous

members of this community include Lycgpodium obscurum, L.

lucidulum and Coptis groenlandica.
M E T H O [)5

Field

 

In this study the boring technique outlined by Erdt-

mann (1931), in which the need for pollen samples from the

deepest part of the basin.wadstressed, was followed so far
as bottom slopes and consolidation of the overlying sedi-

ments indicated its applicability. Bottom slopes in the
vicinity of all three borings were so gradual as largely

to preclude sediment drifting. In consequence, slightly

shallower borings from inner margins of the grounded mats
appeared Justified in order to avoid the six to nine-foot

hiatus beneath the sub-aerial peat of the floating mats.
Sediment samples were collected at one foot intervals

by means of a rotating-cylinder borer of the Hiller type,

described and illustrated by Erdtmann (1943). Thirty feet

of extension rod, in detachable five-foot sections, were
adequate to reach sand, the functional limit of the borer.

To guard against contamination of pollen from different

' layers, the samples were taken from the center of the core

38

and the head of the whole borer was thoroughly washed after

each sampling. The samples were immediately placed in glass
containers and, on arrival at the laboratory, were sealed

with paraffin to maintain their original moisture content

until ready for processing.

Laboratory

During the preliminary laboratory work, attention was

directed to the more widely known methods of preparing
fossil pollen for counting (Sears, 1930; Geisler, 1935;

Hansen, 1940b; Erdtmann, 1943). The modified alkali tech-

nique, finally selected as most efficient with.respect to
time and results, approximated that of Artist (1939), with

the exception that a 10 percent solution of NaOH, rather

than a one percent concentration, was used to free the

pollen grains from their colloidal matrix.
Samples of well-decomposed peat dispersed in dilute

‘NaOH were heated to near the boiling point for a minimum of
30 minutes, during which time sufficient safranin had been

added to impart a deep color to the solution. ‘While still

hot, the latter was centrifuged. After decanting, small
amounts of the surface film from the sediments in each tube
were transferred to a drOp of warm glycerin Jelly on each

of two 22 square mm. cover slips. These were then inverted

on a slide, which was checked for approximate pollen fre-
quency before discarding the sediments. In all cases the
frequency was sufficiently high to indicate that no addi-

tional slides were needed.

39

Samples of sub-aerial peat and the contiguous coarser

material, after initial heating for 30 minutes, were filtered
through a 1 mm. screen prior to centrifuging. The filtering

was necessary to render the centrifuged sediments sufficiently

coherent to permit decantation. Subsequent procedure paral-
leled that for well-decomposed.peat, except that declining

pollen frequency toward the surface necessitated additional

slides, of which as many as four per sample were prepared.
Samples composed chiefly of marl were treated with an

excess of concentrated.hydrochloric acid, which on cessa-

tion of carbon dioxide liberation, was washed out by cen-
trifuging. With the calcium carbonate fraction thus re-

moved, subsequent treatment of such samples was identical

with that of well-deve10ped peat .

Pollen Identification and Counting

Identification of fossil pollen was facilitated by
comparison with a reference set of herbarium pollens which

had'been treated with NaOH and.stained with safranin to
simulate the fossil condition. Particularly useful in

directing attention to critical diagnostic features of pol-
len grains were the illustrated key of'Sears (1930) and
the works of Wodehouse (1935) and.Erdtmann.(l945). In the
key of Sears, the omission of giggg glauca,.pointed out by
Potzger (19#4), was remedied by basing the distinction be-
tween giggg and.Abigg on differences in the insertion,

shape and relative size of the bladders. In distinguishing

40

the pollen of Quercus and Fagus the usual size distinction

was complemented by noting the presence or absence of dis-

tinct equatorial germ pores in the furrows; in Pa us, such
pores are plainly evident even in polar view, whereas in

Quercus they are absent or only rudtmentary.

The slides were examined under a Bausch and Lomb
binocular’microscope equipped with condenser, mechanical

stage and.acular micrometer. Counts and.identifications

were made at a magnification of 430 diameters, and strips
across both cover glasses the width of the field were

examined completely. Ordinarily, only a few traverses

were necessary to obtain a minimum count of 200 pollens
of trees; but in five instances, low pollen frequencies

prevented attainment of this objective. Counts at two
of these levels were terminated between 150 and 200 grains,

but the remaining three, averaging less than 30 grains,

fell well below the number required for a satisfactory

coefficient of reliability (Barkley, 1934), and hence were
excluded from the graphs. ‘Non-tree pollen and spores were

also recorded but were not used in the computation of per-

centages.
R.E:S U L TES

Results of the pollen analyses are presented in the
form.of bar graphs (Figs. 1 to 3) and Tables III to V. The

former express the percentage of total tree pollen for 14

of the 17 arborescent genera represented in the profiles,

SOOIUBd OliVWl'lO

 

 

 

 

ILI

 

O

 

0

J4

 

 

LIJ

 

 

 

 

 

E
glll-I-Isl--I- .-.slell
§
3
S

P'."......'.-I.....Io

010c>m ouaolbc)m OKJOIO

J

gun's-Illuunnel...

.2 _
SIIII'II"||.JI“IIIII.
JIIM lllllul- .-

US

 

3......0- ..-.--

LARIX TS GA

PINUS

0 IO 0 l02030405060708090 O I02030405O 0 IO 0 IO 01020304050 0 |02030 0 l02030

 

PERCENTAGE

Illlll'Illll.llu. .

vw¢°92:52253gg

HSBVW SUB-W40

(\

2030 0102030 010 010 0:0 0:0 0:0 0:0 010

E

-9
o

0
ID
_0
Q’
-8
_O
N

-9
O
-9

0
<1
9—
z
m
o
0:

LL!

0.
O

Q
o

8

0:0 0|02030405000i06090 010203040

0

'sII'll'l'lsllelllssllllIl'. - O

s
28-

24
26-

 

30-

0 IO 0 l02030405060708090 O l020304050 O (O 0 IO 0 l020304050 O |02.030 O IOZOBO 0 IO 0 lO 0 IO 0 IO 0 IO 0 IO

PERCENTAGE

Area of Michigan.

he Lansing

r +
- U

. F1 0;

files tvpic

Pollen prof

P and 3.

’

1

Figs.

41

while the tables record counts of all pollen grains and

spores, as well as the actual percentage representation

of all tree pollens.
A cursory inspection of Figures 1 to 3 reveals several

gross features common to all three profiles. Abies, never

significant quantitatively, but most abundant at the lowest
level, declines steadily and is not represented in the pro-

files at or near the close of the Pinus maximum. Picea,

similarly declining from a maximum at the deepest levels,
differs in magnitude of representation, in persistence and

in its limited re-occurrence in the upper levels. Pinus

attains a marked and remarkably uniform maximum concomitant
with the decrease of Picea, and then declines to a level

suggestive of the origin of pollen from local relict stands

or perhaps even from more northern stations. guercus super-
sedes Pinus to reach an early maxnmms prior to establishment

in force of other deciduous genera. Following a decline

contemporary with the increase of Ulmus and Faggs pollen,
Quercus attains a second more impressive maximum, the more

significant because of the competition of other deciduous
genera by now all well established. Ulmus reaches h15hgat

representation in the lower half of the profile between the

two levels of guercus maxima. Thereafter, it shows a steady
decline, interrupted only by the brief and small-scale re-

covery following the second Quercus maximum. The increase

of Faggs, last of the common deciduous genera represented

in the profiles, is interrupted at an early stage by the

42

second.maximum of Quercus, as indicated by its abrupt in-
crease concomitant with the decline of Quercus in the upper

levels. Tsuga pollen enters the profile with, or slightly
after, Faggs, and like the latter attains greatest represen-

tation in the upper one-third of the profiles. The whole

level of representation of Tsuga, however, is so low as to

render questionable its status as an indicator genus in the

Lansing Area.
Generalized Profile Seguence

By more thorough examination, in.which special signifi-

cance is attached to shifts in dominance of indicator gen-

era, as well as to ecologically consistent trends in their
percentage representation, it is possible to divide-logically

the sequences observed into a number of periods. In de-

limiting these periods, where a definite shift in dominance
is lacking, especial emphasis is laid on grouping t0gether

spectra wherein a reciprocal relation exists between two

genera (or groups of genera) characterized'by divergent
moisture requirements.

The periods of a generalized.sequence of pollen spectra

will first be presented, to be followed by consideration of
the individual profiles.

A. Picea and Abies decline from a pre-existing maximum.

B. Pinus attains a clearly defined maximum and.wanes.
Co guercus becomes clearly dominant, then declines

sharply as Ulmus rises to a conspicuous maximum, terminated

43

by the succeeding rise of Ragga. Quercus pollen becomes
more abundant near the end of the period, as Pinus declines

further to a minimum.
D. Quercus reaches a. marked culmination enduring

through 5 foot-levels as the pollen representation of Eggs
and 1122s; declines .

E. Eggs; and 23955 rise to maxima as guercus wanes.
gigs and filing rise, the latter again declining in the

uppermost levels .
Individual Profile Sequences

The profiles are here examined individually to note

their degree of conformity with the generalized profile
and to emphasize those trends lacking unanimity and there-

fore excluded from that sequence.

Bear Lake Bog (Fig. l.) The absence of a clearly de-
- fined A-period in this profilemay be attributed to a bot-

tom mixture of sand and sedimentary peat which repeatedly

Jammed the borer at the 24 foot-level. Not encountered to
a similar extent in the other borings, the sand fraction

segregated by centrifuging comprised at least 25% of the
23t-foot sample. Although the remainder of the profile

corresponds fairly closely to the generalized sequence,

relationships in the predominantly deciduous portion are
obscured by the anomalous fluctuations of the Quercus,

Page and Ulmus curves . These phenomena, apparently cyclic,

are most evident in the Ulmus curve but on closer examination

44

are seen to involve reciprocal fluctuation between Ulmus

and Quercus in the first three periods, and a similar
relationship between Ulmus and Faggs at the fourth and fifth

foot-levels in period E. The curves of Cagya, Tsuga and

Tilia indicate weak maxima during period E but there are no
significant variations in pollen percentages of these gen-

era within other periods.

Chandler Marsh (Fig. 2.) This profile reflects faith-
fully the generalized sequence up to period E. It is unique

in indicating a wellemarked Carya maximum and a Tsuga mini-

mum during period D.

The sudden shift in percentage representation of Fagus
and Pinus in.period E is coincident with an abrupt dr0p in

pollen frequency occasioned by the transition, at five feet,

from limnic past to coarse sedge peat. The low magnitude
of the pollen frequency in the latter material is indicated

by the area of eight to ten cover slips needed to complete
the counts for the third to fifth foot-levels. 'While count-

ing at these levels, numerous grains of Fagus, Ulmus and, to

a lesser extent, guercus were encountered which displayed

varying degrees of collapse and.arosion of the exine. £5533
grains were frequently so distorted and fragmentary as to
render identification uncertain unless the characteristic
germinal pores were in evidence.

Egg Lgkg B25 (Fig. 3.) This profile, reflecting the
succession of postglacial vegetation with singular clarity,
might well serve as the type for delimiting the periods of

45

the generalized sequence. Despite the defect of a marl
level too poor in pollen for a reliable count, the A-period

is best demonstrated here. Two feet of marl deposition

interposed between bottom sands and.limnic peat permit max-
imum penetration of the borer, so that there is demonstrable

a duration and degree of zicea control not found in the
other two borings where limnic peat grades into sandy sedi-

ments. In the deciduous phase, the successive rises of

Quercus, Ulmus and.Fagus within period C, and the recipro-
cal adjustments of period D are convincingly authenticated.

The rise of Garza in the latter period, while well-defined,
is of a lower’magnitude than that occurring in period E.

Stratigraphic Correlations

The possibility that the four foot-levels of Picea
dominance exhibited in Fig. 3 might merely indicate an

exceptionally rapid.rate of sedtmentation, demonstrated a

need to correlate at least some of the foot-levels in all

three bogs, and so arrive at a measure of their compara-
tive rates of sedimentation. An obvious starting point is

the peak of the Pinus maximum comprising a single foot-
level with percentage representation between 55 and 60

percent. The next sharply defined.point of reference is

the first Quercus maximum, likewise of a single foot-
level occurring at 20 feet in Figs. 1 and 2, and.at 23 feet

in Fig. 3. No further reference levels occur before the

D-period, the delimitation of which is shown by a definite

46

trough in the Fagus curve and.an equally definite crest in

the Quercus curve, suggesting a mass correlation of five
fOOto-l 07816 e

The number of foot-levels between these three zones

of reference is identical in Figs. 1 and 3, two levels
occurring between the Pinus maximum and.the first Quercus

maximum, and seven between the latter and the D-period

group. In Fig. 2, the number is one less in each case.
It is seen, then, that sedimentation occurred at a re—

markably uniform rate in all three basins at least as late

as the end of period D.

D I S C U’S S I O N

‘While the pollen profiles of lakes and bogs doubtless
contain the most complete and accurate record of postglacial

vegetation now available, their translation by pollen ana-

lytical methods are subject to many limitations and.sources
of error. These, termed by Cain (1944a) ”the pitfalls of

pollen analysis“, have been discussed in detail by that
author and.many others, particularly Godwin (1934), Voss

(1934), Eiseley (1939) and Erdtmann (1943). and need be

elaborated here only as they concern specific problems of

interpretation.
‘lhile many of the sources of error in pollen analysis

are inherent in the nature of pollen deposition, and.ao are

beyond control of the investigator, others involve faulty
techniques or subjective interpretation. Foremost of the
"latter group are those arising out of widely divergent

47

opinions concerning interpretation of pollen data. In

_ American literature, the extremes of such cpinion are

represented.by the views of Artist (1939) and.Smith (1940).
Artist, on‘the one hand, follows Aario in attributing cli-

matic significance only to the appearance or disappearance

of genera. Smith, however, freely postulates climatic
changes on the basis of major fluctuations in percentage

representation and finds even minor variations helpful

for purposes of correlation. The degree to which such
latitude of opinion may influence interpretation is well

illustrated by C00per (1942b) who, on the basis of pro-
files reported by Voss (1934), arrives at an interpreta-

tion the complete reverse of the‘latter'sy
Evaluation g; Indicator'Sigpificance 2; Forest Genera

It is evident that logical interpretation of profile

sequences from the standpoints of paleoecology and climatol-

-ogy require careful evaluation of the indicator significance
of pollen fluctuations recorded therein. Such an evaluation

requires at the outset the assumption that habitat relations

of forest genera remain sufficiently constant that their
indicator status throughout the radically different climatic

periods of postglacial time may be inferred from present
forest patterns. While Cain (1944b) has cited the need for

caution in interpreting the past in terms of the present,

such an assumption appears tenable insofar as the genera

involved are of restricted ecological amplitude, or where

48

the indicated shifts in dominance or successional trends
are consistent with present patterns of forest distribu-

tion and succession. 'Where a genus embraces species

widely divergent in moisture and temperature requirements,
as exemplified by Quercus, the impossibility of species

determinations leads to the further, more hazardous qual-

ification that for such a genus a mean indicator signifi-
cance may be assigned, which is applicable throughout
postglacial time. Granting the possibility of establish-

ment of such a generic mean, based upon present relative
abundance of the component species, it is nevertheless
extremely unlikely that the current species ratio, and
therefore the mean, could have descended essentially un-
changed through postglacial time. In consequence, the

pollen curve of such a genus might not only fail to reflect

postglacial climates with fidelity, but might conceivably
evmn obscure a major climatic change through shifting in

species representation from one moisture extreme to the

other. Therefore, it seems illogical to accept unqualified
the indicator significance of such a curve, but instead to

permit sufficient latitude of interpretation to bring it
into accordnwith those of more sharply delimited indicator

genera. Accordingly, in this study Quercus and Carla

pollen representation is considered a product of an as-

semblage of species varying in time as well as in space,
the mean indicator significance of which changes in the
different foot-levels and which may be inferred from

49
comparison with the curves of Fagus and Ulmus.

Ph sio ra h , Soils and.Normal Succession Ag Factors 2; Change

‘25 Stability in Pollen Profiles

According to Gain (1944s), the various factors other
than climate which might effect changes in pollen sequences

have been inadequately acknowledged by most American pollen

analysts. Among the more important of these factors he
lists (1) tapographic changes effected by erosion or deposi-
tion; (2) soil develOpment and.(3) nonmal plant succession.
In a survey of postglacial tapographic changes and soil
develOpment in the Lansing Area in a previous section of

this paper, the relative unimportance of erosion was cited

(p. 14) and contrasted with the significant increase in
land surface resulting from organic deposition. Soil

deve10pment was held to be largely dependent upon clhmate,
either directly as suggested by soils at present well

drained but displaying waterlogged profile characteristics

(p. 22), or indirectly as a result of reactions of plants.

Organic soils originating by the latter means, while of
limited extent (see Table I, p. 21), were considered to

have disproportionate significance from the pollen analyt-
ical standpoint because of their proximity to borings of

the seral forests developed thereon.

Pollen profiles themselves provide an added approach
for evaluating the capacity of non-climatic factors to

produce changes in pollen representation. As Hansen (1938)

and.Sears (1941a) have pointed out, successive maxima of

50

pollen from progressively more mesophytic genera may be
adequately explained by normal plant succession.

In the Lansing Area, however, the indicated.succes-

sions not involving retrogression bear such evident rela-
tion to regional changes in temperature and moisture that

normal plant succession seems not involved as a primary

factor of change. Nevertheless, the reactions of pre—
existing forests in facilitating ecesis of Faggs must be
recognized, since, as Sears (1942b) has pointed out, Faggs

is a specialized genus appearing late in succession.
The non-climatic factors which might obscure or

partially mask changes in pollen representation of species

responding to climatic change are in some areas a major
source of error in interpretation unless recOgnized. Thus,

all pollen profiles from within the sandy outwash plains co-
extensive with the pine forests of the Lake States manifest

a Pinus dominance, following initial decline of Picea pol-

len, which.endures unbroken to the present (Vase, 1934;

Artist, 1939). These remarkably uniform sequences were
interpreted by the authors as indicative of equally uniform

postglacial climate. However, their comparison with cli-

matic sequences now confirmed by the results of other
ecologists (Deevey, 1939. 1943; Smith, 19AO; Sears, 1942a),

serves instead to substantiate the degree to which very

sandy soils compensate climatic changes as well as to
indicate their slow rate of develOpment toward a mesophytic

condition.

51

Sandy oak uplands in southern Michigan, occurring in

an area climatically favorable to beech-maple forest, are

analogous edaphically to the pine outwash plains further
north. It is evident, then, that in the spectra predom-

inantly of deciduous genera, from the Lansing Area some

masking of climatically favored changes in forest compo-
sition is likely to have occurred as a result of edaphic

compensation. While no accurate means of determining the

magnitude of such distortion in representation is available,
the great predominance of heavy soils in the area (see

Table I, p. 21) would indicate that the distortion is rela-

tively minor. Furthermore, from the standpoint of pollen
analysis, the influence of heavy soils might be considered

here even more important. Thus, as perhaps best illustrated

by the map of agricultural land classification, compiled by
Veatch (l9Alb), the lighter soils (third and fourth-class)

are shown to occur to the northeast, south and southeast of

Lansing. These areas,on this map are located at compass
points unlikely to contribute much pollen to the bog deposits

studied, owing to the infrequency of winds from those direc-

tions.
Post-Cary Cenditions in the Lansing Area

Before undertaking an interpretation of the Lansing
sequence, it is essential to attain further perspective

concerning local and regional ice-influence subsequent to

withdrawal of active Cary ice from the Lansing Area. Of

52

' primary concern in this regard are (l) the relation be-

tween ice recession and initiation of pollen deposition

in the basins studied, and (2) regional climatic influences
associated with the succeeding Mankato substage, the ter-

(minal moraine of which lies within 75 miles of Lansing.
like most other'profiles reported from glaciated

North.America, those from the Lansing.Area were secured

from.depressions formerly occupied by buried ice masses.

In such depressiom , deposition obviously could not have
begun until melting of the overlying drift had been com-
pleted. ‘While it is thus clear that kettle basins fail to

record a complete postglacial history, no reliable means
of estimating the magnitude of the missing segments are

available. No doubt the amount of time elapsing between
glaciation and initial deposition varied through wide

limits in response to differing ice bulk or thickness of

overlying drift. However, in no recorded case was it so

short that spruce-fir forests had not been established
during the interim, as evidenced by the abundant pollen

representation of these genera in the deepest foot-levels

of all North.American profiles.
There is a paucity of information concerning the type

and distribution of vegetation at the periphery of contin-

ental glaciers. Pollen analysis fails to provide a record
until the ice margin has retreated an indefinite distance,

indicating only that coniferous forests are able to become

established before melting of buried ice masses has been

completed. In Alaska, coniferous forests rapidly invade

53

areas uncovered by receding mountain glaciers (Cocper, 1935.

1942a; Griggs, 1934, 1942) and.have even been found to occur
on ablation moraine resting on stagnant ice (COOper, 1942b).

However, many qualifications are necessary in drawing

parallels between climatic conditions at the peripheries of
mountain glaciers and continental ice sheets. Existing con-

tinental glaciers are characterized by anticyclones which

give rise to radiating winds often attaining hurricane
velocity (Thwaites, 1946). Such fierce anticyclonic winds

are believed by Hobbs (1942) to constitute the dominant
transportation agent within the extramarginal zones of

continental glaciers. According to this view, the long

winters in such zones are characterized by driving sand

and dust storms, the sand coming to rest in dunes normal
to the glacier front, and the dust spreading widely to be

deposited in the form of loess.

It must be noted, however, that these conclusions are
based upon observation of conditions in high latitude

glaciers where the ablation zones are relatively close to

the centers of accumulation. Thwaites (1946) contends that
the ice borders during Pleistocene glaciation thrust far

down into the zone of wastage, where air masses from the

Gulf of Mexico might well have brought warm climate. In
substantiation Of this conclusion, he cites the absence of

local glaciers in both the Southern Appalachians and the

Driftless.Area of”Wisconsin, as well as the occurrence of
molluscs typical of a mild climate in Iowa glacial outwash.
The absence of loess on Cary drift (Leverett, 1909), as well

54

as evidence indicating that the driftless areas of Wis-

consin.and.New Jersey served as dispersal points for

postglacial forest migration (Sears, 1942b; Potzger, 1945)
would.further'indicate that ice front conditions in this

latitude were less rigorous than those postulated by Hobbs.
a}
Accordingly, the 75emile zone between Lansing and the

point of nearest approach of the Mankato ice sheet would

appearbmore than adequate to contain the extramarginal for-

est destruction and other evidences of disturbance result-
ing from advance of the latter. Further, the wide range of

spruce-fir forests during Pleistocene time as indicated by
a pollen record.from Texas (Potzger and Tharp, 1943) would

suggest that these genera could have been expected to still

occupy the Lansing.Area at the termination of Cary retreat
to the Straits of Mackinac (Thwaites, 1946). Consequently,

secondary spruce maxima evident in profiles elsewhere (Han-

sen, 1937; Voss, 1937; Prettyman, 1937; Deevey, 1939, 1943)

would not be expected to have occurred in basins of the
Lansing Area receiving pollen deposition prior to the Man--

kato advance. Accordingly, the absence of evidences of

disturbance, such as loess deposition, as well as the
absence of a secondary spruce maximum, in no way controverts

the postulate that initiation of pollen deposition in the
Lansing Area occurred during the interglacial period be-

tween the Cary and Mankato substage. Therefore, in the ab-

sence of evidence to the contrary, and in view of the esti-

mated 3500-year duration of the interglacial period (Antevs,

55

1945), as well as the probable truncation of the Picea
period in all bogs, the Lansing sequence is here consid-

ered to include a part of pre-Mankato forest history.
Interpretationlf

Pollen profiles presented here indicate the post-

glacial forest sequences in the Lansing.Area comprise a
pattern whose major trends are recognizable elsewhere, and

are therefore attributive to postglacial climatic changes.

For convenience of interpretation, these climatically sig-
nificant trends are treated individually and in chronologic

order.

Period A. The spruce-fir period.is indicative of a
humid, microthermal climate, only a transitory phase of

which is recorded by the Lansing profiles. This is no doubt

due in part to profile truncation, as indicated especially
in Figs. 1 and.25'but a more important factor is the late
date at which pollen deposition was initiated in the Lansing

'basins. The long, unbroken dominance of spruce-fir forests
indicated.by profiles located on Tazewell drift in Indiana

and Illinois, implies the prolonged influence of a powerful
climatic stabilizing factor, Operative even during the short

,interglacial periods between subsequent substages of‘Wiscon-

sin glaciation. Interglacials do not necessarily imply

warmer climates in the zones of wastage of ice, but only the
cessation of ”nourishment" in the centers of ice accumula-

tion. Since during.the‘w1sconsin interglacials, great masses

56

of ice remained at such centers (Thwaites, 1946), continu-

ing anticyclonic winds could have provided.such stabiliza-

tion. Thus, the Wisconsin interglacials may well have been
characterized by great expansion of the spruce-fir forest

effected by northward migration to occupy the newly-exposed

land while maintaining a stable southern boundary, as sug-
gested by Gleason (1923). It is seen, then, that the spruce-

fir forests once occupying the Lansing Area were part of a

forest characterized by widespread distribution and a pre-
vious long tenure at the ice fronts of the Wisconsin glacia-

tion.

Period g. In sharp contradistinction to the pollen
record of the postglacial spruce-fir forests, the pine max-

imum depicted'by pollen profiles shows a striking increase

in importance northward. This is well shown by a transect
of nine profiles extending from southern Indiana to north-

ern Lower Michigan (Potzger, 1946). In view of the well-

known overbrepresentation of Pinus pollen, this genus must
have been nearly obliterated in the vicinity of Bacon's

Swamp, Indiana, the southernmost station, because the per-

centage representation there was very low. At the next sta-
tion northward, Kokomo Bog, Indiana, a very weak but well-

defined maxmmum of Pinus indicates prothity to the mar-

shalling area from whence the northward migration of this
genus originated. Successive stations northward show a

progressive increase in importance of Pinus pollen, both in

percentage representation and.number of spectra dominated,

until from Farwell, Michigan, northward, it dominates to

the surface of the profiles.

Despite the apparent absence of Pinus segregation in
the glacial forests, there can be no doubt of the climatic

significance of the Pine maximum developed later, owing to
its widespread occurrence (Sears, 1935b, 1942a; Deevey, 1939,

1943). However, it is clearly evident from the present dis-

tribution that the greatly increased importance of Pinus
pollen in profiles to-the north is the result of edaphic

compensation. Despite the distinct pollen-maximum of be-

tween 55 and 60 percent in the Lansing profiles, Pinus was
probably of scattered distribution at the height of its

abundance in the Lansing Area. In a report indicating

equivalent percentages (55~60%) in surface samples from A1-

Berta, Canada, Erdtmann (1943) states that Pinus "has such
a local distribution that a botanist roaming about in this
vast district might not notice it for weeks." It follows

from the present grouping of species, of which even 2.
Strobus has been shown to be of minor importance in the

climax forest (Nichols, 1935). that Pinus must be assigned
xerophytic indicator significance during postglacial time

in the lensing Area. [Accordingly, it is believed.to have

occurred at its maximum in isolated stands on the better

drained, sandy sites rendered unfavorable for Picea by an
increasingly warm climate.

Period 9. ‘While considerable prior-invasion of de-

ciduous species may likely be masked by over-representation
of Pinus pollen in Period B, in Period C the invasion of

deciduous genera is clearly evident. Part of the initial

58

Quercus maximum is interpreted to represent pollen from

xerOphytic species which were able, as a result of further
climatic amelioration, to preempt the sites occupied by

Pinus. The remaining Quercus representation is difficult
to interpret, but the dryness implied by the low percentage

of Ulmus pollen would suggest that the mean for the remain-

ing fraction of oak pollen also lies on the xerOphytic side

of mesophytism. The abrupt increase in representation of
Ulmus to a postglacial maximum is of climatic significance

as indicated by a widespread, albeit less pronounced, par-

allel occurrence in profiles from the north-central states/
(Richards, 1938; Howell, 1938; Potzger, 1943a; Keller, 1943).

It is interpreted here as indicative of an interval of in-

creasing moisture, correlating with the beginning of Period
III of Sears (1942b), C-1 of Deevey (1939. 1943), and with
the wet period postulated by Veatch (1938) on pedologic

evidence. ’The unusually strong develOpment of this maximum
in the Lansing profiles may be attributed to the large pro-

portion of poorly drained soils supporting elm stands even

at the present time, supposedly a drier periodt/ (See Table
/

I, p. 21 and p. 23) In view of such hygric indications,

the Quercus pollen occurring at the levels of highest Ulmus

representation is assumed to be the product of predominantly
mesophytic or hydrOphytic species. Drier, but still humid
conditions, are indicated by the increase of Fagus pollen

contemporaneous with a decline of Ulmus representation late
in the period. Though it is impossible to draw definite

inferences regarding the occurrence of Acer from the pollen

59

profiles, it is assumed on the basis of similar habitat re-
quirements, that maple increased in numbers corresponding
to £2535 during this latter phase of Period C.

Period 9. The xeric tendency evident at the close of
Period C leads to a culminating retrOgression in Period D,

as shown by a definite trough in the Fagug curve, associated
with a postglacial maximum of Quercus pollen. Such a xeric

period, characterized by a decline of Fagus pollen, is amply

corroborated.by profiles elsewhere: correlating with Period

IV of Sears (1942b), applicable to the north-central states
and southeastern Canada, and with Period 0-2 of Deevey (1943)

for Connecticut. ’The reciprocal relation shown by the curves

of Quercus-Carya and Fagug pollen may be taken to indicate
that the species of Quercus and Cagya were predominantly

xeric.
Period.§. The initial sharp rise in this period of

Fagus pollen indicates a return to the mesOphytism of late

Period C. The higher percentage representation reached by
Fagus, as compared with the latter period, suggests an in-

crease in the amount of mesic sites available for this

genus, and by inference, Acer. The lower amount of Ulmus
representation as compared.with Period C indicates that

such a decline in elm pollen may be indicative of a drying
up of the hygric sites of that time, and the invasion there-

on of more mesic genera. ,As a group, the upper foot levels
show a disappointing lack of agreement from profile to pro-

file. This is particularly manifest in the curves of Eggug,
Pinus, Carya and Tilia and appears confined to the levels of

6O

sub-aerial peat. Attention was previously directed (p. 44)
to the very low pollen frequency and the evidences of distinc-

tion of the exine in grains of Fggus, Quercus and.Ulmus ob-

served in the upper levels of the profile from Chandler Marsh.
Such evidence is felt to indicate that, under particularly

adverse conditions for preservation, some differential preser-

vation may occur in pollens generally considered uniformly

resistant. The abrupt increase in Pinus representation could
thereby be explained as resulting from greater resistance to
decay of pollen from that genus as compared particularly with
22525-2"

Because a high pollen frequency was maintained to the

upper foot-level, the profile from Mud Lake Bog is considered

to present the most reliable recent record of pollen deposi-
tion. The postglacial maximum of Tsuga pollen at the upper-

most level of this profile is difficult to reconcile with
the present relict status of the genus in southern.Michigan,

and.may mean that pollen has not been preserved in the bog

for a considerable period of time. This, however, is con-

sidered unlikely owing to the abundance of'Sphagnum which
is considered to provide conditions highly favorable for

pollen preservation. The increase in Picea pollen is co-

incident with the layers of sub-aerial peat. It is there-
fore interpreted to indicate invasion by g. mariana, in

accord with the belief of others (Wilson and Galloway, 1937;
Potzger and.Richards, 1942), and.bance is considered devoid

of climatic significance.

Since the Pinus representation at Chandler Marsh is

61

held to be unreliable, and since the increase in pollen of
_Picea in all bOgs is probably a consequence of invasion of

solidified mats by bOg spruce, it cannot be said that the
Lansing profile provides evidence in support of the climatic

reversion so well substantiated further north (Raup, 1941;

Griggs, 1942; CoOper, 1942b). Nor, as well demonstrated by

the Tsuga curve, can these pollen profiles be considered an
aid in reconstructing the pro-settlement forests in the

Lans ing Area .
S U M M A R Y A N D C O‘N C L U S I O N S

A pollen analytical investigation of three kettle basins
in the vicinity of lensing, Michigan is presented. In addi-

tion, an attempt has been made to include such data concern-

ing environmental factors as might aid in interpretation of

the profiles.
Postglacial erosion is shown to have been a minor factor

in altering the tepography of the Lansing Area, as Opposed to

significant increases in land area resulting from organic
sedimentation. Dominance of fine-textured soils is demon-
strated by data taken from soil surveys for Clinton, Ingham
and Eaton Counties. Forest groupings typical of the major
soil types, and therefore considered co-extensive with them,
are presented with the view of expressing their approximate
areal extent in terms of the area percentages available for
the soil types.

The profiles are sufficiently consistent among them-

selves that stratigraphic correlations on the basis of pollen

62

spectra are clearly indicated in the lower foot-levels. The

forest sequences revealed by the profiles follow a pattern

whose major trends are repeated elsewhere and are therefore
attributive to postglacial changes. Such major trends are

indicated on the overlay accompanying the pollen diagrams.

The unusually strong deve10pment of the Ulmus maximum
midway in Period C of the Lansing sequence is attributed to

the large preportion of poorly drained soils supporting elm

stanm even at the present time, supposedly a drier period.
The high percentage representation reached by Fagus early

in Period.E, as compared with late Period C, suggests an

increase in the amount of mesic sites available for this
genus. Such increase is interpreted as having resulted

from the drying up of some of the hygric sites of Period
C, and the invasion thereon of more mesic genera. This view
is supported by the lower amount of Ulmus representation in

Period E as compared with Period C.

Certain inconsistencies in pollen representation, cor-
related with sub-asrial samples of sedge peat, are held to

result from differential preservation of pollen grains gen-

erally considered uniformly resistant. Thisevidence, to-
gether with a postglacial maximum of Tsuga pollen in the
uppermost spectra, indicates that the Lansing profiles can-

not be considered a reliable aid in reconstructing the very

recent forests in the area.

63
L I T E R A T U R E C I T E D

Antevs, E. 1936. Correlations of late Quaternary chronol-
ogies. 16th Int. Geol. Cong. Rept. 1: 213-216.

1945. Correlation,of‘Wisconsin.maxima. Am. Jour.
Sci. 2432A: 1-39.

 

Artist, R. C. 1939. Pollen spectrum studies on the Anoka
Sand Plain in Minnesota. Ecol. Mon. 9: #94—535.

Auer, V. 1930. Peat bogs in southeastern Canada. Geol.
Surv- Can. meme 1623 1'32.

Barkley, F. A. 1934. The statistical theory of pollen
analysis. Ecol. 15: 283-289.

Barnett, J. 1937. Pollen study of cranberry Pond, near
Emporia, Madison County, Indiana. Butler Uhiv. Bot.

Stud . 4: 55-64 .

Bowman, P. w. 1931. Study of a peat bog near the Matamek

River, Quebec, Canada, by the method.of pollen analysis.
Ecol. 12: 694~708.

Cain, S. A. 1940. The identification of species in fossil
pollen of Pinus by size-freguency determinations.
Amer. Jour. Bot. 27: 301-30 .

and L. G. Cain. 1943. Size-frequency studies of
Pinus palustris pollen. Ecol. 25: 229—232.

1944a. Foundations of plant geography. Harper and
Brothers. New York and London.

19A4b. Pollen analysis of some buried soils, Spartan-
burg County, South Carolina. Bull. Torrey Bot. Club
71: 11‘220

1944c. Size-frequency characteristics of Abies fraseri

pollen as influenced by different methods of preparation.
Amer. Midi. Nat. 31(1): 232-236.

 

Carroll, G. 1943. The use of bryOphytic polsters and mats
in the study of recent pollen deposition. Amer. Jour.
Bot. 30: 36 -366.

COOper, w. S. 1935. ‘A third expedition to Glacier Bay,
Alaska. Ecol. 12: 61-95-

1942a. Vegetation of the Prince‘lilliam Sound region,
Alaska; with a brief excursion into post-Pleistocene

climatic history. Ecol. Mon. 12: 1-22.

64

COOper, W. S. 1942b. Contributions of botanical science to
the knowledge of postglacial climates. Jour. Geol.
w: 981-9940

Cunnin , R. w., and H. G. White. 1941. Forest resources
the Upper Peninsula of Michigan. U. S . D. A. Misc.
PUbe 4290

Dansereau, P. 1944. Interpenetrating climaxes in Quebec.
Science 99(2578): 426-427.

Davis, W. M. 1892. The sub-glacial origin of certain
eskers . Boston Soc. Nat. Hist. Proc. 25: 477-499.

Deevey, E. 8., Jr. 1939. Studies on Connecticut lake sed-

iments. I. A postglacial chronolo for southern New
England. Amer. Jour. Sci. 237: 691- 24.

/ 1943 . Additional pollen analyses from southern New

England. Amer. Jour. 80.1 241: 717-725.

Eiseley, L. C. 1939. Pollen analysis and its bearing upon
Asmeiigalizagrehistory: a critique. Amer. Antiquity

Erdtmann, G. 1943. An introduction to pollen analysis.
Chronica Botanica Co. Waltham, Mass.

Fuller, G. D. 1935. Post lacial vegetation of the lake
Michigan region. Eco .16: 473-487-

1939. Interglacial and post-glacial ve station of
[Illinois . Trans. Ill. State Acad. Sci. 2: 5-15.

Gates, F. C. 1942. The bags of Northern Lower Michigan.
Ecol Mon. 12: 213- 254.

Geisler, F. 1935. A new method for separation of fossil
pollen from peat. Butler Univ. Bot. Stud. 3: 141-146.

1945. A study of pollen grains of thirty-two species
of grasses. Ibid. 7.

Godwin, H. 1934. Pollen analysis: an outline of the
problems and potentialities of the method. 1. Tech-

nique and interpretation. New Phytol. 33: 278-305.

Goldthwait, Lawrence. 1939. Enter chains of the Attle-

boro, Massachusetts, district. Amer. Jour. Sci. 237:
110-115.

Crises, R. F. 1934. The e e of the forest in Alaska and
the reasons for its pos tion. Ecol. 15: 80-96.

55

./'
Griggs, R. F. 1942. Indications as to climatic changes
£11“? 5;? timberline of Mount Washington. Science 95:

Hamp, F. A. 1940. A fossil pollen study of two northern
Indiana bogs. Butler Univ. Bot. Stud. 4: 217-225.

Hansen, H. P. 1937. Pollen analysis of two Wisconsin bogs
of different age. Ecol. 18: 136-172.

1939. Postglacial vegetation of the riftless area
of Wisconsin. Amer. Midl. Net. 21: 752-762.

1940a. Paleoecology of a montane peat deposit at
Bonaparte Lake, Washington. Northwest Sci. 14: 60-68.

1940b. Paleoecology of two peat bogs in southwestern
British Columbia. Amer. Jour. Bot. 27: 144-149.

1941a. A pollen study of post-Pleistocene lake sedi-
ments in the Upper Sonoran Life Zone of Washington.

Amer. Jour. of Sci. 239: 503-522.

1941b. Paleoecology of a bOg in the spruce-hemlock

climax of the Olympic Peninsula. Amer. Midl. Natur.
25: 290-297 -

1942a. The influence of volcanic eruptions upon post-
Pleistocene forest succession in eastern Oregon.
Amer. Jour. Bot. 29: 214-219.-

l942b. Post-Mount Mazama forest succession on the
east slope of the central Cascades in Oregon. Amer.

“idle “at. 27: 523-534.

1943. Post-Pleistocene forest succession in northern
Idaho. Amer. Midl. Nat. 3O (3): 796-802.

1944. Postglacial vegetation of eastern Washington.
Northwest Sci. 18(4): 79-87.

Hobbs, W. H. 1942. Wind, the dominant transporting agent

within extra-marginal zones to continental glaciers.
Jour. Geol. 50: 556-559.

Houdek, P. K. 1933. Pollen statistics for two Indiana bogs.
Indiana Acad. Sci. Proc. 42: 73-77.

Howell, J. W. 1938. A fossil pollen study of Kokomo bog,
1‘178'137 County, Indiana. Butler Univ. Bot. Stud. 4:

 

66

Johnsgard, G. A. 1942. Soil survey of Clinton County,
Michigan. Us Se Do A. Series 1936, N0. 12.

Keller, C. O. 1943. A comparative study of three Indiana
begs . Butler Univ. Bot. Stud. 6: 65-80.

Krauss, R. W., and G. N. Kent. 1944. Analyses and correla-

11011170: four New Hampshire begs. Ohio Jour. Sci. 44(1):

Leverett, F. 1909. Weathering and erosion as time measures.
Amer. Jour. Sci. 177: 349-368.

Leverett, F. and F. B. Taylor. 1915. The Pleistbcene of
Indiana and Michigan. U. S . Geol. Survey Mon. 53.

1924. Map of the surface formations of the southern
peninsula of Michigan. Mich. Dept. of Conservation.

Martin, H. 1936. The centennial geological map of Michigan.
Mich. G901e surv. PUD. N00 39, G90. 10 Series 330

Moon, J. W. 1930. Soil survey of Eaton County, Michigan.
Us 30 DO A. Series 1930, N00 100 ,.

Moss, B. W. 1940. A comparative pollen analysis of two bogs
within boundaries of the late Wisconsin glaciation in
Indiana. Butler Univ. Bot. Stud. 4: 207-216.

Nichols, G. E. 1935. The hemlock-white pine-northern
hardwoods region of eastern North America. Ecol. 16:

403 '422 e

Otto, J. H. 1938. Forest succession in the southern limits

of early-Wis consin glaciation as indicated by a pollen
spectrum from Bacon s Swamp Marion County, ndiana.
Butler Univ. Stud. 4: 93-1 6.

l

Potzger, J. E. 1941. Pollen spectra as time markers.
Amere “idle Nate 25: 224.227.

Potzger, J. E. 1942. Pollen spectra from four bogs on the
Gillen Nature Reserve along the Michigan-Wisconsin

state line. Amer. Midl. Nat. 28: 91-511.

1943a. Pollen study of five bogs in Price and Sawyer
Counties, Wisconsin. Butler Univ. Bot. Stud. 6: 54-64.

1943b. Pollen profile from sediments of an extinct

lake in Hendricks County, Indiana, marks time of
drainage. Proce Indiana» Acad.. $010 52: 83-86e

67

Potzger, J. E. 1944. Pollen frequency of Abies and Picea
in peat: a correction on some published records from
Indiana bOgs and lakes . Butler Univ. Bot. Stud.

6: 123-1300

1945. The Pine Barrens of New Jersey, a refugium
gurixsig Piggistocene times. Butler Univ. Bot. Stud.
: 2- .

 

,/
// 1946. Phytosociology of the primeval forest in
central-northern Wisconsin and Upper Michigan, and a

brief postglacial history of the Lake Forest formation.
Ecol. Mon. 16: 211-250.

and R. C. F iesner. 1939. Plant migration in the
southern limits of Wisconsin glaciation in Indiana.
Amer. Midl. Nat. 22: 351-369.

and I. T. Wilson. 1941. Post-Pleistocene forest
migration as indicated by sediments from three deep

inland lakes . Amer. Midl. Nat. 25:‘ 270-289.

and R. R. Richards. 1942. Forest succession in the

Trout Lake, Vilas County, Wisconsin Area: A pollen
study. Butler Univ. Bot. Stud. 5: 179-189.

and J. H. Otto. 1943. Postglacial forest succession
in northern New Jersey as shown b pollen records from
five begs. Amer. Jour. Bot. 30: 3-89o

and B. C. Tharp. 1943. Pollen record of Canadian
,spruce and fir from Texas bag. Science 98(2557): 584.

Prettyman, R. L. .1937. Fossil pollen analysis of Fox
Prairie bog, Hamilton County, Indiana. Butler Univ.

Bot. Stud. 4: 33-42.

Raup, H. M. 1937- Recent changes of climate and vegeta-

tion in southern New En and and adjacent New York.
Jour. Arnold Arb. 18: 7 -117.

1941. Botanical problems of Boreal America. Bot.
Rev. 7: 147-248.

Richards, R. R. 1938. A pollen profile of Otterbein Bog,
Zarlrgrei gohgnty, Indiana. Butler Univ. Bot. Stud.
3 - e

Riggx, G. B. 1940. Comparisors of the development of sane
Sphagnum bOgs of the Atlantic coast, the interior and

the Pacific coast. Amer. Jour. Bot. 27: 1-14.

Sears, P. B. 1930. Common fossil pollen of the Erie Basin.
Bot. Gas. 89: 95-112.

x

68
Sears, P. B. 1932. Postglacial climate in eastern North
America. Ecol. 13: 1-6.

and E. Jansen. 1933. The rate of peat growth in the
Erie Basin. Ecol. 14: 348-355.

/
L//r 1935a. Glacial and postglacial vegetation. Bot.
Rev. 1: 37'51'

 

1935b. es of‘North American pollen profiles.
Ecol . 16: 1188-499

1938. Climatic interpretation of postglacial pollen
deposits in North America. Bul. Amer. Meteor. Soc.

 

' 1941a. Postglacial vegetation in the Erie-Ohio area.
Ohio Jour. Sci. 41: 225-234.

1941b. A submerged migration route. Science 94: 301.

5” 1942a. Forest sequences in the north central states.
Bot. Gas. 103: 751-761.

1942b. Postglacial migration of five forest genera.
Amer. Joure BOte 29: 68 691. _

,/
Smith, P. 1940. Discussion: Correlation of pollen profiles
from glaciated eastern North America. Amer. Jour. Sci.

Smith, W. M. 1937. Pollen spectrum of Lake Cicott bo , Cass
County, Indiana. Butler Univ. Bot. Stud. 4: 43- .

Swickard, D. A. 1941. Comparison of pollen spectra from
bogs of Early and.Late Wisconsin glaciation in Indiana.

Butler uni-Vs BOt. Studs 5: 67-84'0

Thwaites, E. T. 1946. Outlines of glacial geology. Edwards
Brothers, Inc. Ann.Arbor, Michigan.

Transeau, E. N. 1935. The prairie peninsula. Ecol. 16:
423-437.

1941. Prehistoric factors in the development of the
vegetation of Ohio. Ohio Jouru Sci. 41: 207-211.

Twenhofel, W. H. and'W. A. Broughton. 1939. The sediments

of Crystal Lake, an oligotrOphic lake in Vilas County,
Wisconsin. Amer. Jouru Sci. 237: 231-252.

69

‘Vsatch, J. O. 1927- The dry prairies of Michigan. Mich.
Acad. Sci.,Arts and Letters. 8: 269-278.

1930. Natural geographic divisions of land. Mich.
Acad. Sci.,Arts and Letters. 14: 417-432.

1932. Soil maps as a basis for mapping original
forest cover. Mich. Acad. Sci.,Arts and Letters.
15: 267'273s

1933. Some relationships between water plants and
water soils in Michigan. Mich. Acad. Sci.,Arts and
Letters. 17: 409-413.

1938. Pedologic evidence of changes of climate in
Michigan. Mich. Acad. Sci., Arts and Letters .

23: 38 5-390 .

1941a. Soil survey of Ingham County, Michigan.
Us 5. Do As Series 1933, NOs 36s

1941b. Agricultural land classification and land

types of Michigan. Michigan State College Agr. Exp.
Stas SPOCs Buls 2310

Voss, J. 1934. Postglacial migration of forests in Illi-
nois, Wisconsin and Minnesota. Bot. Gas. 96: 3-43-

1937. Com arative study of bOgs in Ca and Taze-
well drift n Illinois. Ecol. 8: 119-1 5.

Welch, P. S. 1935- Limnology. MoGraw-HillBook 00-, Inc.
I, New Yo rk and London . -

.i/
Wilson, I. T. and J. E. Potzger. 1943a. Pollen records
from lakes in Anoka Count , Minnesota: a study on
methods of sampling. Eco . 24: 382-392.

and J . E. Potzger. 1943b. Pollen study of sediments
from Douglas Lake, Cheboygan County, and Middle Fish

Lake, Montmorency County, Michigan. Indiana Acad.
501s 52: 87'92s

Wilson, L. R. 1935. Lake develOpment and plant succession
in Vilas County, Wisconsin. Ecol. Mon. 5: 207-247.

1938. The postglacial history of vs station in north-
western Wisconsin. Rhodora 40: 137-1 5.

and E. F. Galloway. 1937. Microfossil succession in
a beg in northern Wisconsin. Ecol. 18: 113-118.

70

Wilson, L. R. and R. M. Webster. 1942. Fossil evidence of
wider post-Pleistocene range for butternut and hickory
in Wisconsin. Rhodora 44: 409-414.

and A. T. Cross . 1943. A study of the plant micro-
fossil succession in the bottom deposits of Crystal
Lake, Vilas County, Wisconsin, and the peat of an
adjacent bog. Amer. Jour. of Sci. 241: 307-315.

Wodehouse, R. P. 1935. Pollen grains. McGraw-Hill Book
Co., Inc. New York.

71

APPENDIX

TABLEE

(/

oweacsossms

 

HocoHHeeV
H « H «HH HH. HH .H« .983
ssH
H m H H -sHsosns<
couocono
H H H H « H fiance”:
as. a.
H. H a .— d H 135
H H H m H ososHsdso
H m H m H H n n n H m n H. n H « 333350
N N b w hm «H mm m n .N H 0 H N n n ssHhLoo
H.Hceoeev
H H H. HH. H.« «H HH HH 332m
Hmv
08 08 H«« 08 HH« 8« oo« 2« HH« 8« Ho« oH« oo« 8« H«« 8« «H« ..o« «8 HH« 8« oo« H«« H.o« H38
«H 2 .2 H H H o H «H. H HH HH H HH H.H H H H o H.H «H 2 HH HH £33332
cH 9H 1H 1« 9« 9e 9: 9H 9H 9: A» SH 9H 9H as Q« 9H QH 5. 9e 9H 9H 9e 0H Hiass
H H _
H. o n. o s. o segues.»
w .H « H H H H « H H H H H H H
Ho H H o.HH H.« «H HHo H.HH H « H6 H.H 1H H.H H.« HmH H.o H.« 3588
0H 2 H H H.Ho « H « H6 H6 o.H H H6 35..
. . . . . H .H. H « c H H « H H. H H
H.H H.H H« H.H oH o.H H.o o.H H6 H.H H.H To o.« .H.H H.« o6 H.« H.H H.H H.« H.« 33%
m « H H H « H. c .H H. H « H H H H e H H .H H
eH OH {H H.« H.H o.H m.n o.« o.« n.n 9n o.H «.«.HH m.o H.H H.H H6 H6 o.H m.o uHHem
H H H H H « _H e H H « H H H H « .H «
H Hwo H...« H owe H.HH Hwo o.HH H.o o.« H..« H.H H.H ow« H.o H..H H.H o.H H.H o.H .83.
« H H H H H. c H 2 2 HH
H.o «.« o.« o.H H.H H.« H.« H.H 9m H.« o.HH H.H H... .H.H H.« H.H H.H H.H 0H 4... H..H H.HHH...
o.«H HHo o.«H HHH oH H s H o HH H H. « H H H H
. . «.« o.H H.H .H H.o H .« 1H «.H o.H o.H «.« 4.« 33H 2.
H«coHHHH.HH.H.H.HHHHHmmoHHeHHoHfi
1H 9H sH 9H 9H 3% 8H 9 tH HH e& 9« 0H 9: h t; 0« cH 9» 9H 1. 9H are
H HH 8 H« H.H HH. m. «H .H H «.H «H H« H« 8 HH H.H .H H. HH «H
H.H 0H 0.2 H.oH H.«« H.H.H H.H« H..«« 93 H.oH «.3 H. H 0.8 13 H.«H H.«H H.H. H.HH H.HH o.H H... H.HH H.H ones
« « H H o« «H H.H 8 HH HH H.« H« R 3 o« 2 HH H.H H..
To o.H H.H H6 92 o.HH ¢.H« o.HH H.HH H...HH H.HH «.3 H...« H.. H«.o o.HH o.HH H.HH H.HH Bee...
H H HH 42 «H «H He 42 8 HH 4H 8 Hm HH 8H 2: HH H «H. «H. mm HH. «H
H.« H.HH H.H.H H.H. o.H.H o.HH «.3 «.3 H.HH «.«H 99 99 o. c «.HH H.oH H.H»...HH H.HH o.HH o.HHH.o« H.HH 99.. 39.28
H H H H « H H H H «H H H H o
H.o H.o H.H H.o o.H «.« H6 H6 1H H.H o... H... «.« a... owe:
HHHHHHHHHHoHHcHH««HH H
H.« H..« H.H H.H o.« H.« o.« H.H H.H a... H.H o.« H.« o... o.H H..H H.H H.o 33..
Ho SH SH HH HH 3 H« c« H« HH H .H o Hm H Hm 2 H« o« H« o« 8 HH H«
1H: 0.8 H.HH H64 H.HH H.oH 921: 92 o.« H.H H.H H... H. H.H H. H.... H.oH H.H H.HH H.-H H.oH H.H o.HH one:
HoH HH H.H HH HH 2 H H H H. .H H H H. .H « a HH HH «H HH 0H 0H
o.HH: omo« H.«HH H.HHH H.HH. o.H H.« H.« H.« H.o o.H o.H H.H H.H H.« To o.« H.H H.H o.H H.HH e... H.H. coca
H
H.H. H... H.H H.H H6 82...
.. 2
4H« H« «.« H« 8 HH HH H.H HH HH 1 HH «H HH 3 H H H. H H c H « H .Ho: 58..

. 3.22. 8.5 8« Ho
.Ho>sH mass as .323 can us 253 H33 IHIHHH

IaldaHI s oaHassoo sHHns.osasHsnsa ssaoms use essHHoa espouse: uo sassoo
HI I no usesn sasHHoa ssh» no coHosassssuasu emsaceomsa use Hanson csansoo
Hon 3.3 soon too 34H. 5:8

H5
3:
.HHH mama

smmacsosowe

 

H.HoHHee.
H H H H sshoam
ceH
N n H . H N H . N N s -eHuosns<
. Hoaocecn
H n H H H H H H H nnacesms<
. Heceeceew
H H N H an H»
H H H H H e H oeocH-eee
H H H H H H e H 4 HH «H HH H H oeooohohao
H « H H e c c H . H « H H « H csHHeoo
Hearse:
« H H H 88......
.8
uom H«« oo« Ho« H«« oH« H«« oo« oo« oo« ««« co« eo« HH« H«« Ho« oo« oo« oo« oo« HH« HHH o« Hoses
uceoeosonm<
H H e H H H H a H H H HH HH «H oH .H H HH H .H e« HH 5
...m H... o.« H.H H.N o.H H.H H.H H... H.H Ham 0.». H.H H.H 0.: 0.0 0... H.H 0.0 o.H. o.HH 92 stones:
H a H
m.0 0.0 m.0 sscHxssh
H « H « H « H H H H « H H
H H H HMH Hwo H.o o.H H.o o.H H.o H.H H.o H.H o.H H.o oscHHeeo
m.0 H.o 4.0 ¢.o H.o cmH sscH<
H e «H -H H .. H H .H H H .. .. 2 H H. H. H H H. H
H.o «.H 0.0 u.« H.H H.H H.H H.« u.H ¢.H H.N H.H H.H H.H H.H H.H H.H o.H H.« «.H H.H HHsaom
H H H H H HH oH HH oH H H H H c H « c H a « «H H
H.o H.o H.o H.« H.H H.H H.« H.H o.H H.H H.H H.« H.H H.H H.H o.H o.« H.« o.« o.H H.H H.H xHHeH
H H « H « « « H H « H « H « H « ..
H.o «.« H.o H.H o.H o.H H.o H.H H.o H.o H.o o.H H.H o.H H.H H.o H.« .52
-H H H H H .. H. « H. H H « .. H « .. H H «
u.o H.H H.« H.« H.H o.« H.H o.H «.H. H.H H.« H.o H.H H.H o.H o.« H.o H.o H.o .2?
H e « « H H « « H H H a H « « « H H
H.H HnH o.H o.H H.« 4.H o.H o.H H.H H.o H.o H.H H.H o.H o.H H.o H.o neeHHoe
.H H « H .. H H H H. H ..H HH H HH H. S H H. H H
H.o H.H H.o «.« H.H H... H.« o... «.H H.H H.H ...H o... H.H H.H o.H H... H.H H.« H.H are
H «H H« «« HH OH H: «e -HH ea 0H HH H« H« HH «H 4H «H HH HH H
H.« o.H H.HH H.H H.H« H.«« H.H« o.«« u.eH «.H« H.HH H.HH H.«H «.oH H.H o.H o.H o.H H.H H.H H.H osaHH
« H oH H« u« H« «e H« oH H« .« H«. as HH «H HH HH H
o.H «.« H.H H.HH H.«H H.«H H.HH H.oH H.HH H.H H.oH H.«H o.o« H.H« o.H« H.H H.H «.H eswea
H« HH «« Hm ««m HoH HH HH He HH 4H HoH HH HoH «HH HoH HH HH He HH He HH H
«.HH a.» o.HH o.«H H.H« H.HH H.H« o.HH H.HH H.H. H.HH H.oH H.H. «.HH H.HH «.HH H.HH H.HH H.HH H.oH H.HH H.HH .ooeooe
H H « e H c H H « H H H H H H « H
H.o «.0 H.o o.« H.H H.H H.o H.o H.o H.H H.o H.« H.o H.H «.H H.H swans
«H H « H « H H H H H H H H « H « « H « 4 H
H.H H.« o.H H.H H.o ...o «.« H.H H... H.H H..« .H.« H.o H.o «.« o.H o.H H.o o.H H.H o.« 3.3
o« HH HHH HH H4 H HH eH .H «H H oH HH H H« H« c« H« HH H4 He oH H
H.H H.HH H.HH H.HH H.HH H.H «.H. o.H. o.H. o.H H.H H... H..H. «... «.oH H..«H o.«H H.«H H.H H.o« H.H« H.HH 3::
HNH mm mm OH H H s e H H H N e N H OH HH 0N HN c
o.MH awe: HvHH Hwe HwH H.H H.H o.« H.H H.o H.H o.H H.H o.H o.H o.H H.H H.H o.HH. eoon
H
«.2 H... o.« H.o H.o 83..

Ha.
a« H« «« H« 0« HH HH 5H HH HH eH HH «H HH OH H H a H H c H « H .HoroH-Hoom
.eseHHom some 00m Bo IschHs a wsHacsoe sHHns osasHsns» assem- ece ecoHHoa soau-coc uo Hansen Amy

. am 0 secs H o as c a H.:o cones HcsHHoa sea» no soHHsscoeesaos oweacoesoa use eucsoo HocHnsoo H<v
Ho>oH some as HcHe 00m u H H H H aH H gene: ssHvseno sou seed coHHom .>H m4m<p

«03:00.30...
H303}:
H H 0 H .230

:3omoa3om

H H . SHcaupaam
GPH

.3 09.5.:
couocog

N N H IEQCQL¢B<
. .82..on

2. H.H.

c . H N H H 23:33...

(I
HHNM
F!
n
'4
H
q
n

030E020

H
n
n
,4
H

H H N N N N H N N H « 3:88
H322:
H N a 332m
HHH

 

C
0.H 0.¢ 0.0 0.4 0.0 0.c 0.0 {.0 0.N 0.N 0.0 0mm 0.0 0.NH.0.Q

#
o.H. 0.0 0.: 0.0 o.« o..~ n.¢ o.HH 0.0 0.0 o.H 0.H 0.N
0.00 0.00 0.N0 0.3 0.NN N

ONIID

0 H. o 0 o 0 0 HHo H. 0

n N N c. n

0.H o.H 0.0 0.H 0.H 0.0 0.0
H H .
9H 0.0 0.0

N h 0 0 c o n n N 0 t

0.H 0.H 0.H 0.H 0.H 0.H

ow. H.HH 9.. m... .H. A... .. H . 3 H .
0.0 0.0 0.0 0.0 0.0 «om 0.0 0.0 0.0 0.0 0.N 0.¢
. H N 0 HH

0.0 o.H c.H 0.N

N N 0 0 H N o 0

0.H 0.0 1H 0.0 0.0 o.H 5.0 0.H

H c H c N

0.H 0.H 0.H 0.H o.H

« H H H H H H « H H
o.H H.o H.H H.H H.H o.H H.« o.H H.« H.«
HH H« 3 HH HH HH HH S H.. HH «. H«

0.N. 0.HH 0.0H 0.0 0. 0H m.¢H 0.0N N.00 0.HN o.HH 0.0H 0.HH

« H H H « « H« H« «H H«
o.H H.o «.« H.H o.H H. H.HH H.«H o.«H H.«H
oH H.H HH HHH 02 HH «H H.H 02 HoH HH H.oH

0.0 0.NN 033 0.00 0.H0 0mm: 0.04 H.0c 0.00 0.N0 0.: 0.00

0.0 0.0 0.0
0H m 0 0 HH 0. N NH 0 N

QHH 40H Hm 0c 00 00 NH {H 0 0H m 0H
.0H b.0H 0.0 0.0 0.0 0.0 Hun 0.N
0N 0 H n H

0.00 0.0.... 0.0H 0.0 0.0 0.H 0.0

0N .HN HN NN HN 0N 0H 0H H.H 0H 0H 3
.IHoHHon 3.3 OON Ho

OONOONOONOONBNOONOONHnNOHNHONNROONNONcHNOONbnNNNNNONOONOnNOONOONNHNOONHHNOONOONoHN H.500.

. Humouvuonnq
0 0 0 NH. H. 0H 0 0 NH H.H HH 0 HH

0 o o 0 mow . unioClCD
0mm h N 0.¢ 0.0 .000 0.0;. 000 0.:N 0.NoH .HNo ¢N0 c H.Ho
0.Ho HHH 0.H 0.H 0.H o.H o.H o.H 0.N H.H .3538
H
#30 10 . _ 2E:
H H « « . H H .H H H HH
n.H 0.N o.H o.H 0.N C.H 0.H 0.H 0.H 0.0 N... 330m
H. o H N H 0 n 0 H H 0 H c
0.0 H.N 0.H o.H H.H 0.N 0.H 1N 0.H 0.N .H.N 0.N 0.H HHHHm
N H N H H N N
0.0 0.0 o.H 0.0 0.0 o.H o.H Loo<
c 0 N 4 n H H H 0 0 ¢ 0 0
H.H 0.N o.H 0.N 0H 0.0 0.0 0.0 0.N TH 0.N 0.0\0.N «HHH...
H. n c n 0 c H n n 0 N H ,
0.0 ¢.H 0.N 0.H N.N 0.N 0.0 1H 1H 0.N o.H 0.0 232:.
t 0H 0 {H OH 0 HH .HH HH 0N .HH 0H
0. 0.0 0.: 0.0 0.0 0.N 0.0 0.0 0.: N.0 N.NH o.H. H.H. HHLHH
{n NN NN 0H NH HN 0N H.N H.N H.N On HN HHH
{JH 0.0 0.0 0.0 0.mH 0.0H 0.HH H.NH 0.0H 0.NH 0.1. 0.0H v.0 .365
.HH HH HH «« H H.H «H .3 «H HH HH 8 H.H J
H..NH 0.0 TH. o.HH 0.0N 0.HN o.HH 0.0N 0.0N 0.0N H.0H 0.0N o.HH Baum
0HH 0HH HNH 00H N0 00 0H. H0 NH no NH. 00 2.
0.01 H.N0 0.00 0.00 0.0d 0.04 0.Nn 0.00 0. 00 0.0N H.0n 0.0N H.H.n 9.6.330
N 0 c N . H n n 0 t 4 H.
0.0 ¢.H 0.N 0.0 0.0 H.H 0.H 0.N 0.N 0.N 0.0 H032.
0 NH N H H N H N H
H.« H.H o.H 10 H.o o.H H.o o.H H.o x23
H.H 0N 0H NN 0N {H 0H 0 0 0H 4H 0H NN
N.H. 0.0 1H. o.HH v.0 o.H. 0.0 N; 0.0 o.H. 0.0 0.N 0.0H 05::
n H H 0 0H HH 0 NH 2
n.H 0.0 0.0 1N 0.N. N.0 «.0 0.0 0.: Hoon

ooHn<
2:
HH NH HH 0H 0 o H. 0 0 c H N H 333502

IHIHHHHI 0 0:35.00 QHHH‘ 03333 Eon. 2:. $3300 00.5.02. 00 3:58 me
H

.HoboH so: an EH!» OON no 0980 H33 153:: n no con-p IuoHHon 00.5 00 :03350933 00303.80 on: 3:33 08358

mom on: on: .30 33 5:3 .> 502.

 

ROOM USE 'ouu

 

WW5
Mafia ,, 3443.61 E“ '
‘0’ 1961 ’8‘ {WW

 

 

 

 

Lelee

Ptxww
- .L.

 

 

 

 

MICHIGAN STQTE UNIV. LIBRQRIES

31293101563702