THE

BOGS AND BOG FLORA

OF T H E

HURON R I V E R V A L L E Y

A THESIS S U B M I T T E D TO T H E FACULTY OF T H E UNIVERSITY OF
MICHIGAN FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY
By E D G A R N E L S O N T R A N S E A U
Ferry Fellow in Botany

PRINTED AT THE UNIVERSITY OF CHICAGO PRESS

[Reprinted from the BOTANICAL

GAZETTE 40: 351—375, 418-448. 1905, and 41:
17-42. 1906.]

TABLE OF CONTENTS
Vol.

40

I. The Huron River valley
351
Physiographic features
• • 351
Physiographic history
353
Forests
. .
• 355
Meteorological conditions
356
II. The bogs: their development and ecological conditions
. . . .
360
Physiographic origin of the lake and bog basins
360
Bog and lake vegetation
362
The processes involved in peat formation
367
The physical and chemical properties of peat
372
The bog as a habitat for plants
418
A. 'Physical factors
418
1. Wind
418
2. Temperature
. . . . . 419
3. Texture
4
4. Mechanical properties
4 3
5. Diffusion properties
4 3
6. Water capacity
4 4
7. Osmotic.pressure
4 4
B. Chemical factors
4 5
1. Ground water . . . .
4 5
2. Acidity
4 ^
3. Food material
427
C. Biotic factors
4&
III. The bog - plant societies
. 4 9
West lake
430
First Sister lake
43
Bog north of Delhi
. 436
Bog near Oxford, Oakland county
439
The Delhi muskeags
441
Bog on Carpenter road
441
The Chelsea bog
444
General consideration of the bog flora
447
2 2

2

2

2

2

2

2

2

2

2

2

V o l . 41

IV. The ecological characteristics of the bog flora and their causes .
Experiments
Water cultures
V. Summary
Bibliography

.
.

17
22
24
34
40

THE

BOGS AND

BOG FLORA

OF THE HURON

RIVER

VALLEY,
EDGAR

NELSON

TRANSEAU.

(WITH SIXTEEN FIGURES)
I.

The Huron River valley.
PHYSIOGRAPHIC FEATURES.

T H E H u r o n R i v e r valley, to the b o t a n i c a l survey of which the
present paper forms the sixth contribution, is located in the south­
eastern part of M i c h i g a n .

A s indicated in jig. i, the valley e m b r a c e s

parts of five counties.
T h r o u g h o u t , its surface forms are of glacial origin and, with the
exception of the i m m e d i a t e borders of the river, have undergone b u t
slight modification since glacial times.

P e r h a p s its most

striking

topographic features are the rough m o r a i n i c hills of its upper and
middle courses, and the gently undulating plain of its lower course.
The

river h a s its source in west-central O a k l a n d County in B i g

L a k e , 9 miles ( 1 4 . 5
km

(64 )

northwest

k m

) southeast of Holly and approximately 40 miles

of D e t r o i t .

Starting with an elevation of

111

feet (290 ), after a course, extending for 50 miles ( 8 o
southwestward and then for another 50 miles ( 8 o

k m

)

k m

950

) generally

southeastward,
m

it empties into L a k e E r i e at an altitude of 573 feet ( i 7 S ) above tide.
As is c o m m o n i n areas of glacial deposition, the topography of the
drainage b a s i n of t h e . H u r o n

h a s little of the appearance usually

suggested b y the term " v a l l e y . "

T h e upper two-thirds of its course

is a winding depression a m o n g m o r a i n i c k n o b s , l a k e basins, abandoned
glacial drainage channels, and sand plains.

H e r e the river is char­

acterized b y long reaches and occasional slight riffles.

At intervals

it broadens into stretches of lake-like character, as is illustrated b y
such bodies of water as C o m m e r c e , T a y l o r , Strawberry, Whitewood,
and B a s s L a k e s , each with an a r e a of one-fourth to one-half a square
mile (65-130 h e c t a r e s ) .

T h e river m a r g i n is usually low and swampy.

I t s tributaries enter it at every angle, and bring to it the drainage
of hundreds of l a k e s and swamps.
1905]

351

M o s t of these lakes are small,

occupying areas of an acre (half a hectare) or more, but there are
several of considerable size.

1

P o r t a g e and W h i t m o r e L a k e s occupy

one and one-fourth to one and one-half square miles (325-390 hec­
t a r e s ) , while Union, Straits, F o u r - M i l e , O r e , I n d e p e n d e n c e ,

etc.

FIG. I.—Map of the Huron drainage basin. T h e boundaries of the interlobate
moraine are shown by the lines — — . T h e boundary between the clay morainic
belt and the lake plain is marked by the line
.
cover a fourth to half a square mile (65-130 h e c t a r e s ) .
percentage of the tributaries lie in

flat-bottomed

A very large

depressions whose

surface approximates the ground-water level, consequently producing
thousands of acres of swamp and m a r s h land.

Everywhere occur

small undrained depressions, some well above the average ground­
water level, others containing lakes and bogs.
1

I t is also worthy of

Not connected with the Huron River by surface drainage.

note that a large part of the surface drained b y the H u r o n and its
tributaries, before it m a k e s the great b e n d to the southeast below
P o r t a g e lake, is m a d e up of sand and gravel, composing and a c c o m ­
panying the S a g i n a w - E r i e interlobate m o r a i n e .

I t is a region of

steep hills, with occasional dry plains, everywhere penetrated

by

lakes and swamps.
The

country which the river next crosses, beyond the great bend,
km

for a distance of 20 miles ( 3 2 ) is composed of glacial till plains
and clay m o r a i n e s — a belt extending N E - S W , approximately parallel
to the interlobate moraine.

H e r e , although the hills are well m a r k e d ,

the slopes are m o r e gradual

and the basins broader.

T h e river

is bordered b y b a n k s several feet in height, and seldom attains a
width of 150 feet

m

(So ).

T h e last 30 miles ( S o

k m

) of the H u r o n R i v e r traverses a meander­
m

m

ing course sunken from 50 feet ( i 5 ) at Y p s i l a n t i to 25 feet ( 7 . 5 )
at R o c k w o o d below the surface of a glacial l a k e plain sloping gently
southeastward from the m o r a i n i c belt j u s t described, to the western
shore of L a k e E r i e .

T h e soil is here composed of sand, sandy loam,

and—-in the vicinity of the l a k e — c l a y ; the only topographic features
aside from the sunken water courses being the several b e a c h ridges
and dunes m a r k i n g the successive stages in the lowering of the glacial
lakes, forerunners of the present L a k e E r i e .
T h e r e are, then, three natural divisions of the H u r o n
basin:

(1) the loose-textured rough interlobate m o r a i n e ;

drainage
(2)

the

clay m o r a i n i c belt lying to the southeast of it; (3) and the low-lying
plain extending to L a k e E r i e .

E a c h implies important differences

in the way of bog formation and provides edaphic factors which
determine to a large extent the nature of the dominant forest covering.
PHYSIOGRAPHIC HISTORY.
The

history of these topographic features is for the most part

b o u n d up with the retreat of the ice at the close of the last ( W i s c o n s i n )
glacial epoch.

A topographic m a p of the region lying between L a k e s

M i c h i g a n and E r i e shows that the m o r a i n i c hills so characteristic
of the H u r o n b a s i n are part of a belt of similar physiography extending
from northern I n d i a n a well up into the " t h u m b " of lower M i c h i g a n
(fig. 2).

T h i s belt of glacial deposits is directly connected with the

development of reentrant angles along its crest, as the great con­
tinental ice s h e e t

2

b e c a m e m o r e and m o r e differentiated into lobes

during its retreat (52, 13).

I n northern I n d i a n a it m a r k s the first

areas uncovered, as the m a s s of ice, pushed forward from the basin
of L a k e M i c h i g a n , separated from that originating in the L a k e H u r o n
and L a k e E r i e basins.
W h e n the H u r o n R i v e r b a s i n was reached, the S a g i n a w lobe h a d
been developed and lay over the northwestern part, while the HuronE r i e lobe covered all of the territory southeast of the interlobate

FIG. 2.—Map of southern Michigan, northern Indiana, and northern Ohio,
showing "moraines with strong expression." After LEVERETT, U. S. Geol. Surv.
Mon. 41, plate 2. T h e irregular dotted lines mark the 1000-foot (300™) contour.
moraine.

T h e first portion of our area to b e uncovered is the triangu­

lar gravel outwash apron extending southwestward from Sugarloaf
Knob.

T h i s was the beginning of the H u r o n R i v e r .

Kavanaugh

L a k e then lay j u s t under the edge of the E r i e ice, and Crooked L a k e
occupied a similar position on the southern border of the Saginaw
lobe.

A s h a s been recently determined b y M r . F R A N K L E V E R E T T ,

of the U . S. Geological Survey, the subsequent history of the H u r o n
drainage is most r e m a r k a b l e .
The

waters from the glacial drainage at first flowed generally

westward, reaching the K a l a m a z o o R i v e r n e a r Albion, thence to the
S t . J o s e p h at T h r e e R i v e r s .
2

A t South B e n d , I n d i a n a , it crossed

For general map see no. 55, p. 4 1 1 . (Bibliography at close of this paper.)

to the K a n k a k e e R i v e r , and reached the Mississippi b y way of the
Illinois.
As the reentrant extended itself further to the northeast,

another

channel was opened for the H u r o n drainage westward past P i n c k n e y
into the G r a n d R i v e r , and from there to B a t t l e Creek and the K a l a ­
mazoo R i v e r .

B e l o w the city of K a l a m a z o o it cut across to the P a w

P a w R i v e r , and reached the Mississippi b y way of L a k e Chicago.
W h e n the ice of the E r i e lobe h a d retreated as far eastward as
Ann

Arbor, and all of the interlobate moraine h a d been uncovered,

a third outlet for the waters of the H u r o n was opened b y way of
Clinton and the R a i s i n R i v e r , which at that time emptied into glacial
L a k e M a u m e e at Adrian (32, pi. 20).

T h i s l a k e was drained b y

the W a b a s h R i v e r into the Mississippi.
As soon as the ice m a r g i n passed the clay morainic belt already
described, the H u r o n reached L a k e M a u m e e at Y p s i l a n t i b y way of
its present channel.

B u t L a k e M a u m e e h a d meanwhile

changed

its outlet to the northward, its drainage going b y way of I m l a y

(53)

to the G r a n d R i v e r , L a k e Chicago, and the Mississippi (32,

pis.

21, 23, 26).
L a t e r the E r i e basin was entirely freed of ice, and its water for
the first time flowed eastward into the O n t a r i o basin (glacial L a k e
I r o q u o i s ) , and thence b y way of the M o h a w k to the Hudson.

With

the clearing of the S t . L a w r e n c e channel the present system was
inaugurated.
Aside from the physiographic interest connected with their early
history, these glacial drainage
interest.

T h e y furnish

all directions.

channels are of distinct biological

continuous lowland h a b i t a t s

extending

I n so far as they are represented b y broad,

valleys, and connect with tributaries of the northern

in

open

O h i o valley,

they provide important highways for the dispersal of southern rivervalley species.
FORESTS.
T h e three topographic divisions already described exhibit m a r k e d
differences in their forest aspect.
richest and

O n the l a k e plain we find the

most mesophytic of the forest

types.

This

lowland

habitat is a continuation of the northern W a b a s h valley, and it is not
surprising that its flora should b e of m u c h the same character.

Here

we find the greatest variety of tree species, among which are Fagits

atropurpureus, Quercus rubra, Ulmus americana, Platanus occidentalis,
Acer saccharum, Tilia americana, Acer saccharinum, Fraxinus
americana, Gleditsia triacanthos, Liriodendron tulipifera, Gymnocladus
dioica, Cercis canadensis, Asimina triloba, and Celtis occidentalis.
T h e clay morainic area is dominated b y Quercus rubra, Q. alba,

Q. velutina, Hicoria ovata, H. glabra, Acer rubrum, Ulmus americana,
and Quercus macrocarpa.
In

the region of t h e interlobate m o r a i n e the disappearance of

the m o r e mesophytic forms is quite m a r k e d .

T h e forest is there

largely composed of Quercus coccinea, Q. macrocarpa, Q. velutina,
Q. alba; a n d as we go northeastward these b e c o m e associated with

Pinus

strobus. Quercus prinoides forms a characteristic shrubby

growth along the roadsides a n d in waste places.
S u c h is the forest b a c k g r o u n d in which are set the thousands of
acres of b o g a n d swamp, a n d to which the groves of Larix
exhibit a m a r k e d contrast.

laricina

T h e s e t a m a r a c k areas are to b e seen

on all sides in the region of the interlobate m o r a i n e ; they are quite
c o m m o n in the clay m o r a i n i c belt, b u t are practically wanting on the
l a k e plain.
As one follows along the morainic country from northern I n d i a n a
into t h e

" t h u m b " o f M i c h i g a n , h e passes from a region dominated

b y a rich mesophytic broad-leaved forest to one of conifer a n d xerophytic broad-leaved ascendency; from a region whose low grounds
are characterized b y a swamp flora to one in whose depressions the
bog flora reaches a high state of development.

I n this connection

it is interesting to note that one finds this gradual change epitomized
in the H u r o n valley as h e goes from its m o u t h to its source.
METEOROLOGICAL CONDITIONS.
U n d e r this h e a d we shall consider t h e general meteorological
conditions of t h e H u r o n b a s i n , a n d compare t h e m with t h e meteor­
ological conditions found about the center of the distribution of b o g
plants

(55, p . 406).

I n general, this center extends from

Lake

W i n n i p e g through the upper G r e a t L a k e region down the valley of
the S t . L a w r e n c e to the A t l a n t i c coast.
5

I t is in the coast provinces,

however, that the bogs reach their highest development, in the form

of the ' ' r a i s e d b o g . "

Certain temperature

phenomena

associated

with the bog habitat will b e discussed in connection with the analysis
of the life conditions obtaining in bogs.
Rainfall.—In

the following table is given the m e a n monthly and

annual precipitation for seven stations located within or near the
H u r o n basin.

A s their individual variation is b u t small, it is probable

that the average for the stations gives a fair estimate of the rainfall
and its distribution.

Appended are the corresponding records for

the maritime region of eastern C a n a d a :
MEAN PRECIPITATION I N INCHES.
Alt Rec­
feet ord Jan.
a.t. for
Yis.

Station
Ann Arbor
Ypsilanti. .
Jackson.. .
Annpere..
BallMt..
Birm 'gham

930
736
927
924
932
860

Feb.

Mar. Apr. May June July Aug. Sept

1.99
1 97
2 06
1.98
i-73
1.91

2.19
2 61
2 21
2.19
1.76
2 00

2 12
2 48
3-14
2 42
2 07
2 32

2.88 3-72 3 39
2.24 3 98 4 02
1 26 3 22 3.02
2 60 3 38 2.44
2.12 3-52 3 . 1 3
2 53 3 36 3 15

2 82
3 09
2 47
2 66
2 47
2.56

2.45
2.24
1 76
2.85
2 59
2 38

2 58 2.82
2 63 2.70
1.74 3 61
1 93 2.04
2 62 2 69
2 33 2 51

2.77
3-29
2.87
2.64
2.71
3-02

2.35
2 32
1.56
1.77
2.30
1.88

i-94

2.16

2.42

2 27

3 53

3-19

2 68

2.38

2 30

2 73

2 88

2 03

?

5 55

3-93

3 80

2.50

3 66

2.72

3 29

4 64

3.08

4 13

4 7i

5 16

22

5 63

4-94

5 15

4 00 4 43

3 68

3-43

3 96

3 53

5 21

5 26

5-52

23
18
6
11
13
IS

Average
St. John,
N. B , ( i 8 )
Halifax,
N. S.(5i)

Oct.

Nov. Dec.

I t will b e noticed from the above data that the precipitation is
q u i t e e v e n l y distributed throughout the year.

I t reaches its m a x i m u m

during the months of M a y and J u n e , when the vegetative processes
of the b o g plants are most active.
mum

during J u l y and August,

attains its greatest height.

I t approaches its s u m m e r mini­

when the temperature

commonly

T h e former implies that the water level

in the bogs is kept at or above the surface of the substratum for weeks
at a time.

T h e latter involves strong transpiration on the part of the

vegetation, when the water supply must b e drawn for the most part
from the substratum.

T h e average n u m b e r of rainy days during the

past five years is one hundred per annum.
T h e average snowfall in this region during the five years, 1898 to
1902, amounts to 38.4 inches (975

m

m

).

I n t h e case of the bogs this

thickness is usually increased b y the drifting of snow from the sur­
rounding hills.

Observations during the past two winters show that

the bogs are covered b y ice to a thickness of a foot ( 3 0

c m

) or m o r e .

Consequently, low shrubs, a n d herbs which pass the winter b y means
of underground stems, are well protected from low temperatures and
sudden temperature changes.

T h e ice further results in lowering the

temperature in spring a n d in retarding the beginning of favorable
growth conditions.
The

percentage o f sunshine is not published b y the several sta­

tions, b u t the n u m b e r of clear and partly cloudy days is stated.

The

n u m b e r s from the various stations show m a r k e d differences, due to
different standards established b y the observers; b u t perhaps these
are largely eliminated in the average.

I f we t a k e the average num­

ber of clear days, add to it one-half the n u m b e r of partly cloudy days,
a n d divide b y the n u m b e r of days in a year, we obtain a percentage
of forty-six.

T h i s probably approximates the percentage of sunshine.

I n comparison with the rainfall data for H a l i f a x a n d S t . J o h n , it
is notable that in the latter localities the m e a n rainfall, both monthly
and

annual,

is

considerably

larger.

The

annual

exceeds that of the H u r o n valley b y fully 20 inches (50
40 per cent.

precipitation
c m

) , or about

F i n a l l y , the sunshine percentage is slightly lower, being

39 for H a l i f a x a n d 42 for S t . J o h n .
Temperature.—The

following

table

exhibits the monthly and

annual m e a n s for the several stations already c i t e d :
MEAN TEMPERATURE I N ° F .
Jan.

. .

Average
St. John, N. B
Halifax, N. S . . . .

The

Feb.

22 2
25-9
26 1
24.7
22 6
23.1

23 1 30 6
24 2 32.3
20 3 32 7
21 7 31 3
2 1 . 4 27 6
22 8 30 4

45 4
46.7
47 5
50 2
45 3
46 5

24.1

Ann Arbor
Ypsilanti
Jackson
Annpere
Ball Mt
Birmingham.

Mar. April May

22 2

30 8

18 6
22.0

18.7
22.7

26 3
28 7

June July Aug.

Nov.

62 5
61.8
64.8
63.0
62 2
61 5

36.4
36.7
37-7
35-9
35-9
36.7

27.1
27.6
26.5
26.7
27.1
27.0

46.6
47-3
48.8
47 4
46. c
47-1

67.1

71.9

69.1

62.6

50.6

36.5

27.0

47-2

56.3
57-6

61.0
64.2

61.3
64.8

55-6
58 2

44-7
48 0

36.1
38.2

23-7
27.0

40 8
43-2

67 3
63.5
69 4
68 4
66 9
69 1

46.9

56 7

38 6
38.2

48.8
48.7

49 9
49 9
54-7
49 8
49-5
49-7

Dec.

Ann.

Sept. Oct.

72 0 69-3
71 1 69 7
74.8 71 8
67 6
71.8
70 0 68 1
72.2 68 4

53 2
58.1
59-4
56 5
56.0
57 3

table shows that the temperature conditions are compara­

tively uniform throughout the basin.
peratures occur in J u l y a n d August.

T h e m a x i m u m average tem­
B u t the significance of the data

b e c o m e s m o r e apparent, in so far as the b o g vegetation is concerned,
when they are compared with those of S t . J o h n a n d H a l i f a x (fig. 3).
I t is t o b e noted that, although the average temperature for J u n e ,

0

J u l y , and August of the H u r o n b a s i n is 8.6° F . (4.8 C . ) higher, its
rainfall during the same period is less b y 2.6 inches (66

m

m

).

There

c a n b e little doubt as to the effect of such differences upon the growth
of the bog species, especially the sphagnum whose moisture supply
is m o r e directly dependent upon atmospheric water than upon the
Indies
at-Rain.
1

- 'it*

\

\
\

60
r
/

SO

f

/

/

40

/ /
/

/

/

30

/
Hu

*°h V
/

Jan

?eS

Tftar Apr

77?ay Juru

July

Jay

Sept

Oct

Tloir 2>ec

FIG. 3.—Curves of rainfall and temperature conditions in' the Huron basin
compared with those of the maritime region of Canada.
soil

solution.

Again,

the

occurrence of high

temperature

with

decreased precipitation m e a n s the production of conditions impos­
sible for the development of the " r a i s e d b o g , " if not unfavorable to
the highest development of the " f l a t b o g . "
Since bogs attain their m a x i m u m development in a region of
great rainfall and comparatively low temperatures, it is reasonable
to infer that the extremes of s u m m e r heat b e c o m e peculiarly signifi­
cant in this region.

E x a m i n a t i o n of the weather records shows that

temperatures

0

0

of 97-100 F . ( 3 6 - 3 8 C . ) are likely to occur every

year, and that temperatures approximating these m a y b e prevalent
for

several days in succession each season,

W h e n these extremes

coincide with periods of drought, they must act as important checks
on the growth of the bog plants, especially the sphagnum.

As we

pass from northern I n d i a n a along the moraine into M i c h i g a n , the
gradual increase of bog development, of the variety of bog species,
a n d of the areas covered b y sphagnum is very m a r k e d .

Although

other factors are involved, this increase m a y b e correlated with a
decrease in summer temperature extremes.
II.

The bogs: their development and ecological conditions.
PHYSIOGRAPHIC ORIGIN OF T H E LAKE AND BOG BASINS.

I n connection with the special consideration of the bog flora, it is
of interest to note the origin of the depressions in which this flora h a s
developed and

flourished.

Indeed, in the morainic belt of the Huron

basin it would seem that among the agencies which have produced
important topographic changes since glacial times, the bog plants
stand near the head of the list.

S t r e a m erosion and deposition have

been slight, while l a k e basins have been

filled

and the level of

depressions generally raised b y the deposition of plant debris.
As no attempt has as yet b e e n m a d e at the mapping of peat
deposits and m u c k soils, no reliable estimate of the total amount of
aggradation accomplished b y plant agencies can b e made.

Y e t the

frequency with which in field work one encounters peat soils, in
various stages of m a k i n g or decay, suggests that in the aggregate
such deposition has been most effective in this region.

T h e northwest

quarter o f the A n n A r b o r topographic m a p , which e m b r a c e s an area
of about 215 square miles (55,700 h e c t a r e s ) , located in the morainic
portion of this basin, indicates approximately 43 square miles (11,500
hectares)—20 per cent.—as swamp land.

I t is probable that at an

early time this area was very m u c h larger, but with the settlement of
the l a n d m a n y extensive areas have b e e n drained and only the dark
humous soil remain's to suggest its past history.
T h e most frequent source of l a k e and bog basins is here found in
connection with the deposits m a d e by glacial drainage.
vicissitudes attending

A m o n g the

the retreat of a glacier are the occasional

detachment of b l o c k s a n d masses of ice through differential melting
(19).

I f these detached masses happened to b e in the line of the

overloaded glacial drainage, they b e c a m e covered to a greater or less
extent b y sand a n d gravel.

Owing to the poor conduction of h e a t b y

such deposits, they melted with extreme slowness.

W h e r e this latter

process was prolonged until the drainage line h a d been abandoned
or the stream h a d ceased depositing, subsequent melting
about a settling of t h e deposits a n d t h e production of basins.

brought
Sister,

K a v a n a u g h , a n d Crooked L a k e s are examples o f this type.
I n the case of the chain of lakes which form a part of the H u r o n
T

R i v e r in northwestern W ashtenaw County, a n d such lakes as P o r t a g e ,
T a m a r a c k , O r e , a n d B a s s , according to L E V E R E T T , there was a n
additional

settling of t h e

fluvio-glacial

deposit itself.

T h i s latter

process h a s been of the greatest importance in the development of
extensive b o g areas.

I n the P o r t a g e L a k e region this settling h a s
m

amounted to as m u c h as 40 feet (12 ) in certain places, a n d h a s
resulted in reducing m a n y hundreds of acres of land to the ground
water level.
T h r o u g h o u t t h e belt of till plains occur shallow marshes, some­
times drained, b u t usually b y a sluggish meandering stream, itself
impeded b y the growth of swamp plants.
natural

expression of t h e unequal

T h e s e basins are t h e

deposition of glacial material.

T i l l plains result from a comparatively rapid retreat o f the i c e ; hence
the depressions are usually shallow, a n d have been mostly filled with
peat to the level of the present drainage.

T h e several small lakes

lying to the west of D e x t e r are examples of basins not yet obliterated.
W h e r e the retreat of the glacier is slow a n d deposits are m a d e to
a great thickness about the edge of the ice, k a m e or " k n o b a n d k e t t l e "
topography

results.

T h e basins of such areas are characterized

usually b y high margins

a n d comparatively steep

slopes.

West,

Silver, North, Island, a n d South L a k e s m a y b e cited as examples.
As we know from remains discovered in peat deposits, among t h e
animals inhabiting this region in early postglacial times were the
mammoth,

mastodon,

bison,

peccary

(Platygonus

compressus

L e C o n t e ) (57), elk, and " b i g b e a v e r " (Castoroides ohioensis F o s t e r ) .
The

last n a m e d is not a b e a v e r (34, p. 256), b u t is more nearly

related to the Coypu rat of South A m e r i c a .

T h e c o m m o n beaver

(Castor canadensis K u h l ) h a s b e e n an important factor in the creation
of b o g areas (37), a n d in t h e extension of areas already existing, b y
the building of dams.

T h e b e a v e r was found in this section when it

was first settled, b u t the last known specimen was killed sixty-nine
years ago.

T h e occurrence of peat deposits several feet in thickness

and covering quite large areas, bordering streams, whose channels
lie deeply sunken in the deposits, seems to find its best explanation
in this m a n n e r .

B u t little field work h a s been done on the relation

of beavers to the peat deposits, and examples are still too hypothetical
to cite in this connection.
BOG AND LAKE

VEGETATION.

O f the plants which might come into a new land area containing
basins, such as was laid b a r e on the retreat of the glaciers, none is
better adapted to rapid migration than the group of a q u a t i c plants.
W h e t h e r we have in m i n d the smaller submerged varieties or the
partially submerged littoral species, their wide geographic distri­
bution a n d uniform associations b e s p e a k their evident solution of the
problems of dispersal.

T h e fact that deposits of peat and m a r l have

been found in northern I n d i a n a a n d lower M i c h i g a n to a thickness
m

of 40 feet (12 ) would indicate that in these particular basins the
vegetation must have obtained an early foothold.
Concerning the deposition of marl, it is of interest to us only in so
far

as it b e c o m e s an agent of aggradation in the basins.

I n the

reports (5, 42, 21) on the m a r l deposits o f I n d i a n a and M i c h i g a n ,
m a n y examples are cited where t h e m a r l forms the underlying sub­
stratum of peat deposits.

T h a t its deposition to a large extent is due

to plant life h a s b e e n shown b y D A V I S (9, 10). T h e plants most
concerned with this process are the C h a r a c e a e and Cyanophyceae
(Schizothrix,

Zonotrichia).

T h e y are probably

aided

mollusks, and perhaps also b y c h e m i c a l precipitation.

b y certain
A s for the

C h a r a c e a e and Cyanophyceae, they have a wide range of habitat in
different lakes, and m a y occur in deep or shallow water and on
various r o c k substrata.

W h e r e they c o m e into competition

with

shore species, t h e r a n k growth of the latter usually precludes their
existence in sufficient amount to b e of importance in m a r l formation.
W h e r e wave action is strong, the c h a r a is confined to deeper water,

but the blue-green algae m a y b e present up to the water's edge, in
such situations frequently forming m a r l pebbles.

T h e lower limit

of existence is largely determined b y the transparency of the water,
m

and m a y lie between 20 and 30 feet ( 6 - 9 ) .

O f the littoral plant

associations there are commonly two quite distinct divisions, the
outer m a d e up largely of submerged or floating pondweeds and waterlilies, the inner of half-submerged rushes and sedges.
cerned in the process of peat formation.

B o t h are con­

U n d e r such conditions

there naturally develop, in regions of calcareous underground waters,
an outer zone of c h a r a dominance and m a r l deposition, and an inner
zone of pondweed-sedge

dominance and peat deposition.

Varia­

tions in the slope of the bottom, in the amount of wave action, in the
presence of shore currents, and in the color of the water, determine
whether one or both of these processes shall go on, and to what
extent these activities are kept distinct or grade into one another.
I n the case of the peat, however, the process is not
upon water species alone.

dependent

T h e y act merely as forerunners of a

denser and more luxuriant vegetation which frequently is of greater
quantitative importance.

Briefly, we m a y note here that in the case

of the bogs, unlike that of the swamps, the plants which develop on
the margin, especially Car ex filiformis and forms of Eriophorum,
are able to secure all of their food materials from the water and air
and build their own substratum.

T h i s tangle of roots and rhizomes

usually attains a thickness of several inches, and on account of its
low specific gravity floats on the surface of the water.
foundation

the sphagnum

quota to the debris.
the t a m a r a c k .

U p o n this

and bog shrubs advance, adding their

L a t e r , these are followed b y such tree forms as

Coincident with this increased weight and augmented

rate of deposition, comes the progressive submergence of the

floating

substratum, and its gradual disintegration and humification.

The

accompanying fig. 4 will serve to illustrate this process.
W i t h i n the last two years m u c h has been promised toward the
utilization of the peat deposits in this region for fuel purposes.

Com­

panies have been organized, and the machinery necessary for the
drying and consolidating of the peat has been m u c h improved.

At

C a p a c and Chelsea, factories have been erected, and attempts are
being m a d e to place the industry on an economic basis.

I f these

ventures prove successful, we m a y hope for an interesting body o f
scientific information to come from the study o f b o g sections.

The

work o f A N D E R S O N , L A G E R H E I M , SERNANDER, W E B E R , a n d others
in Sweden a n d G e r m a n y , gives indication o f the d a t a concerning
postglacial migrations of plants a n d animals, a n d climatic changes,
which will b e o b t a i n a b l e when our b o g deposits b e c o m e of economic
importance.

'/// / /

7

/ //dwai/Tlrt,

/ / / / / / / / / / / / / /

/ / / / ///////7/7777/ // / / / / / /
r

/ / / / / A

FIG. 4.—Diagrams illustrating three stages in the development of peat and mar
deposits in lake basins. In drawing the figures it has been assumed that the rates of
marl and peat deposition are approximately equal. The peat accumulates most rapidly
on the western side of the basin. O n the east side a common effect of wave action is
illustrated. T h e process of peat formation is hindered, while that of marl deposition
goes on until the aggradation of the bottom reduces the force of the waves sufficiently
to allow the bog plants a foothold. A represents conditions in early postglacial times
when these basins acquired their first flora. T h e several plant societies represented
are (1) conifer (2) bog shrub, (3) bog sedge, (4) aquatic, the outermost division of
which is the chara association. In B the conditions for the growth of plants belonging
to the northeastern conifer forest formation have reached their optimum. C repre­
sents present conditions in southern Michigan. T h e plants belonging to the south­
eastern broad-leaved forest formation, being climatically favored, occupy the areas of
mineral soils, while the conifers are almost restricted to bog areas.

THE GEOGRAPHIC DISTRIBUTION OF PEAT DEPOSITS.
I n N o r t h A m e r i c a the distribution of recent peat deposits m a y b e
conveniently summarized

under two heads, genetically unrelated:

(1) those of glaciated regions; (2) those of the coastal plain.
T h e peat of the glaciated area constitutes the great b u l k of these
American deposits.

T h e southern boundary of this region is marked

by a line passing westward from central N e w J e r s e y through northern
Pennsylvania and Ohio, central I n d i a n a and Illinois, thence north­
ward through southern Wisconsin, northwestward to the M i n n e s o t a
valley and the R e d R i v e r of the N o r t h in M a n i t o b a , westward through
northern Assiniboia and southern Alberta to the R o c k i e s .

Here

the boundary is deflected southward into M o n t a n a , but in crossing
toward the coast it is again carried northward into British Columbia,
and finally southward among the Cascades of Washington to the
Pacific O c e a n .
Along this southern border the peat deposits are exceedingly
scattered and m a k e up a small fraction of the total land surface.
T h e y have accumulated under water in depressions among
recessional moraines.

the

As we go northward, the relative proportion of

peat bogs and peat deposits regularly increases, and there is a notable
tendency toward the accumulation of pure humus in situations other
than depressions containing water.

W h e n the tundra or

"barren

g r o u n d " is reached, the accumulation of humus is almost universal.
T h e contrast with our own region is well brought out in R U S S E L L ' S
account of the tundra (43, p. 129).

T h e vegetation

grows rapidly during the long, hot, summer days, dies below and partially decays,
but becomes frozen and has its complete destruction arrested, while the dense
mat of roots and stems continues to thrive. In this way an accumulation of
partially decayed vegetable matter is formed, which increases in thickness from
year to year by additions to its surface. The process is similar to that by which
peat bogs are formed in temperate latitudes, except that the partially decom­
posed vegetation becomes solidly frozen. It is in reality an example of cold
storage on a grand scale.
Under existing climatic conditions there does not seem to be any limit to the
depth such deposits may attain. The amount of carbonaceous material already
accumulated in the tundras of America and Asia must equal that of the most
extensive coal field known.
South of the boundary above described, peat deposits of consider-

a b l e extent are occasionally m e t with.

I n the region of the great

plains they are sometimes found b e n e a t h a surface covering of sand
and wind-blown deposits.

T O D D (54, p. 121) has mentioned the

occurrence of such peat deposits in eastern South D a k o t a .

BARBOUR

also reports such deposits from central and eastern N e b r a s k a

(2).

O n the basis of their field relations and certain fossils which they
contain, they are believed to b e of G l a c i a l and early Pleistocene age.
I f the plant materials of these deposits could b e carefully worked
over with reference to their successive floras, we might hope for some
new light on glacial climate, since a part of the deposits are beyond
the margin of the W i s c o n s i n ice sheet.

B u t even their location and

existence give evidence of climatic change, and plant and
migration.

Although

now

widely

separated

from

the

animal

region of

active bog formation, they are historically connected with this division.
A m o n g the mountains of b o t h the eastern and western U n i t e d
States, bogs and swamps are to b e found in association with mountain
lakes.

M o r e frequently t h a n otherwise these depressions are con­

nected

with

former

local glaciation, perhaps the most

frequent

situations being those afforded b y the d a m m i n g b a c k of water b y
terminal and lateral moraines.

B a s i n s for peat accumulation are

also found in solid r o c k m a d e b y glacial erosion.

T h e conditions

here are quite similar to those of the north, the altitude bringing
about the same general effect as the latitude.

T h e analogy is still

further shown on mountains in moist regions where alpine meadows
are strongly developed.

N o t only are the plants related to those

of the tundra, b u t the deposition of peat or h u m u s is again irrespective
of basins.
I n m a n y places east of the great plains there is another type of
situation not directly connected with glaciation, but in which vegetable
debris m a y a c c u m u l a t e to considerable thickness, viz., about
debouchure of cold springs.

the

T o w a r d the north these springs m a y

bring about h u m u s accumulation on slopes, but further south peat
is usually associated with pools and small lakes.
The

second group of situations in which peat accumulation takes

place on a grand scale, are those associated with coastal plain phe­
n o m e n a , such as the rising and sinking of the land, the irregular
deposition of alluvial materials in deltas, and the extension of the

land through reef building.

T h e s e swamps have b e e n described b y

SHALER, K E A R N E Y , JULIEN, and others (46, 26, 24, 7).
their

greatest

development

Florida, and the Mississippi

in

eastern

T h e y reach

Virginia, N o r t h

floodplain.

Carolina,

T h e y m a y contain either

salt or fresh water, and their vegetation is noted for its density and
luxuriance.
The

geographical distribution of peat deposits is of interest in

this connection because it points to certain factors which contribute
to the preservation of h u m u s materials.

Certainly in a r c t i c latitudes

the most significant factor is the low temperature, for h u m u s accu­
mulates to great thickness even with a scant vegetation.

In

the

northern states and southern provinces of C a n a d a , peat is associ­
ated with basins containing stagnant water or cold springs.

The

a n n u a l increment from the vegetation is greatly increased over that
of the tundra.

M i l d temperatures

to preserve the plant debris.

and stagnant waters c o m b i n e

W h e n we come to the coastal plain

swamps of the southern states, this process t a k e s place only where a
luxuriant vegetation is c o m b i n e d with areas of stagnant water of
considerable depth.
T o put it sharply, we m a y say that, in spite of the scant vegetation,
the cold of the tundra results in peat accumulation.

I n temperate

latitudes, mild temperatures and stagnant water combine to prevent
the complete disintegration of a vigorous vegetation.

I n the south,

in spite of the high temperature, the luxuriance of the vegetation
and stagnant water unite to m a k e peat formation possible.
THE PROCESSES INVOLVED I N PEAT FORMATION.
W h e n for any reason the living protoplasm in a plant or any of
its organs is brought to the condition of death rigor, the continuance
of this state for a prolonged period inaugurates

certain chemical

and physical processes which result in the b r e a k i n g down of the
exceedingly complex structures and compounds m a k i n g up the living
plasma.

A m o n g the first outward

is the loss of water.

signs of such

disorganization

T h e cells of soft tissues lose their n o r m a l form,

and in any case the tissue b e c o m e s more or less filled with gases.
The

protoplasts

as such disappear,

and

in their place

carbohydrate and proteid bodies are to b e found.

granular

Aside from the mineral substances composing the ash of such
bodies, the organic compounds are m a d e up for the most part of
carbon, hydrogen, and oxygen.

I n the case of the proteids, there

are added to these nitrogen, sulfur, and

phosphorus.

As to the

exact nature of the compounds existing in the dead material, aside
from the carbohydrates, very little is known.

T h e same statement

holds as to the nature of the decomposition which goes on without
the intervention of saprophytic organisms.
that oxidation does occur.

B u t it seems probable

T h i s action, then, is the beginning of

the more comprehensive process known as peat formation.
W h e n plants or their organs die, under ordinary circumstances they
are at once attacked b y fungi and bacteria.

T h e progress of disso­

lution is then greatly hastened, and the final disintegration is more
complete.

According to the operation of certain external factors,

the destruction m a y involve two very different groups of organisms
and result in bodies of very different chemical and physical properties.
T h e s e two processes are known as eremacausis and

putrefaction

(61, 39).
W h e r e access to oxygen is accompanied b y favorable temperature
and moisture conditions, the first of these processes, eremacausis,
takes place.

T h e formation of ordinary soil humus m a y be cited as

an example.

T h a t oxygen plays the important role has been demon­

strated both b y experiment, and b y the analysis of the gaseous and
solid products.

I t has been shown, for example, that soils in which

eremacausis is in progress contain C 0

2

and O in inverse proportion

to one another.

U n d e r constant volume, as the one increases the

other decreases.

I t has been also shown b y experiment that

the

process is wholly dependent upon the activities of certain lower
plants.

A m o n g these m e m b e r s of the genera M u c o r , Aspergillus,

Penicillium,

Saccharomyces, Micrococcus,

Bacterium,

Spirillum,

Crenothrix, and B e g g i a t o a are most important.
T h e carbohydrates are b y this m e a n s broken down to C 0
H 0.
2

2

and

T h e albuminoids and amides constitute the principal forms

of the nitrogenous materials.

U n d e r the influence of these organisms,

especially their katabolic processes, the oxygen unites with the carbon
to form C 0 , the S is oxidized to H S 0 , the P to H P 0 , and the
2

H to H 0 .
2

2

4

3

4

T h e first form in which the nitrogen reappears is that

of a m m o n i a .

T h i s is at once attacked b y the nitrifying bacteria,

and changed successively to the form of a nitrite and a nitrate.

The

two latter changes again involve the addition of oxygen.
I f we consider only the temperatures occurring in nature, we m a y
say that these activities increase regularly with the temperature.

As

to water conditions, it has been shown that in air-dry soil eremacausis
is practically wanting, and that when the soil is filled with water it is
reduced to a m i n i m u m .

B e t w e e n these two extremes lies an optimum

at which there is sufficient moisture for the life of the organisms, and
yet not enough to interfere with the diffusion of oxygen.

A n acid

condition impedes, and a slight alkalinity favors, the production of
both the carbon and the nitrogen compounds.
E r e m a c a u s i s is then essentially a process of oxidation,

brought

about b y lower organisms, whose activities are favored b y a high
temperature, a slightly alkaline medium, a n d free access to the air.
I t s products are simple compounds which m a y furnish food materials
for the higher plants living on the substratum in which they are formed.
is m e a n t that process of disintegration

which

occurs when organic m a t t e r decays in the absence of oxygen.

B y putrefaction

Here

again organisms are involved, but they belong for the most part to
the anaerobes, and are wholly forms of bacteria.

T h e process is

essentially one of reduction.
Carbon dioxid is again the principal gaseous product,
relative amount is greatly reduced.

Along with it C H , H, H S ,
4

H P , N 0 , and N are produced in small quantities.
3

but its

2

2

I n the manu­

facture of the carbon dioxid the oxygen is not only derived from the
organic matter, but also from nitrous oxid, nitrites and nitrates which
m a y b e present.

I n the decomposition of cellulose, carbon dioxid

and m e t h a n e result from the hydrolysis of the cellulose molecule.
Albumins at first b r e a k up into amido-acids, nitrogenous compounds
of the aromatic series, and other little-known bodies.
position

continues, the

compounds

amido-acids in turn form

of the fatty-acid

series.

I f the decom­
ammonia

and

T h e latter substances

may

still further disintegrate to carbon dioxid, hydrogen, and methane.
Depending upon the stage in the progress of decomposition, we m a y
find complex organic compounds, organic acids, and their salts, or
comparatively simple substances.

A s to the influence of external factors, high temperatures increase
the rate of disintegration, while the presence of acids prevents its
continuance, due to the killing of the b a c t e r i a involved.

I t is to b e

noted that the products of putrefaction, both intermediate and final,
can b e of little use in furnishing food materials for the higher plants.
W i t h these two processes in mind, we m a y now consider the
m a t t e r of peat formation as it occurs in this region.

W e have already

seen how the substratum is being extended at the edge and renewed
at the surface b y the plants forming the outer zone of the bog vege­
tation.
phorum.

I t consists of sedges, especially forms of C a r e x and E r i o E a c h year these plants send up stems and leaves from the

m a t t e d rhizomes.

A t the approach of winter these are killed, and

the snow later on aids in bringing them down to the water level.

In

the spring the water covers almost the whole of this zone to the depth
of several inches.

W i t h the gradual lowering of the water level and

t h e coming of warmer temperatures, the conditions for eremacausis
are m a d e favorable.

I f the water is approximately neutral in its

chemical reaction, the fungi and b a c t e r i a begin the work of disintegra­
tion, which if continued would result in the complete destruction of
the vegetable debris.

However, on account of the great demand for

oxygen, the process can b e carried on only near the surface of the
water.

E v e n at a depth of a few centimeters the rate of oxygen

diffusion is so small, as compared with the demand for it, that practi­
cally all aerobic bacterial action is prevented. All of the surface waters
which I have examined have been found to b e teeming with bacteria.
Close upon

the

sphagnum-heath

extension of the

zone.

bog-sedge zone comes

the

H e r e the surface is characterized b y hol­

lows and elevations, the latter frequently due to the upward growth
of the sphagnum beneath the shade of the heath plants, but in some
cases due to the building of mounds b y ants.

I n the hollows the

water stands above the substratum throughout a large part of the
year and even during dry periods lies j u s t at the surface.
sedges,

the

principal

plants

of

this

zone

are

U n l i k e the

evergreen.

The

sphagnum forms a continuous m a t of living plants several centi­
meters in thickness, through which all of the oxygen must diffuse
before it can b e available for the eremacausis of the dead plantmaterial beneath.

T h e cassandra, cranberry, and andromeda which

compose the b u l k of the shrubby vegetation add to the debris largely

by their leaves a n d underground stems.

T h e former fall to t h e

substratum as they die, b u t not at t h e close of each vegetative period.
Consequently they a r e soon lost among the sphagnum, a n d there is
no distinct annual layer added.
B u t b e n e a t h this layer of possible aerobic activity, the material
would seem to b e subject to putrefactive agencies.

A n d there c a n b e

no doubt that such destructive processes are carried on in those situ­
ations in which t h e acidity of the soil solution does not preclude
the existence of the anaerobic bacteria.
A m o n g t h e taller shrubs a n d trees, such as Vaccinium corymbosum

y

Aronia nigra, and Larix laricina, the defoliation takes place each
autumn.

A s these plants are of relatively large size, the b u l k o f the

material forms a noteworthy

annual addition

to the substratum.

W h e n - t o this is added the twigs a n d small branches which fall each
season, we c a n understand the fact that the substratum is almost
entirely free of surface water.
5-10

c

m

below.

Usually the ground-water level lies

B u t the substratum h a s a high water-capacity a n d

is kept constantly moist.

W h e r e the sphagnum covering is wanting

for one reason or another, the dark color of the surface peat shows
how m u c h more complete is its disintegration as compared with that
of the other zones.

T h i s condition is m a d e possible b y its position

relative to t h e ground water.

O n t h e other hand, as will b e shown

later, the temperature conditions are more favorable in the zones of
herbaceous and shrubby vegetation.
M o s t of the basins in which peat formation is going on actively,
are subject to considerable variation in water level, b o t h seasonal
and annual.

During t h e last two years the rainfall h a s been con­

siderably above the normal in lower M i c h i g a n , a n d m a n y o f these b o g
areas were

flooded.

A t W e s t L a k e , for example, a large part o f t h e

t a m a r a c k area was covered with water to a height of several inches
above the level of t h e roots.

M o s t o f the basins are also subject to

higher water level in the spring and during prolonged rainy periods.
Accompanying such changes there are great differences in the rate
and m a n n e r of decay.

H i g h water, in so far as it excludes oxygen,

favors putrefaction; if it comes as a result of heavy rains, it decreases
the acidity of the soil solution, increases its oxygen content, and a t
least for a short time favors the growth of the saprophytes causing
eremacausis.

L o w water level exposes a m u c h greater b u l k of the

substratum to disintegration, and favors the carrying away of the
products of decomposition;

in general, it favors eremacausis.

In

the samples of water which I have examined at various times from
the same depressions, there have been m a r k e d variations within short
periods of time in the color of the water and in the presence of such
animals as D a p h n i a and Cyclops.

N o attempt has been m a d e to

count or even separate the b a c t e r i a present, but it is probable that
they too vary with the color of the water and the animal life.
W h e n the bog land has been cleared and ditched, the m a r k e d
increase in the rate of decay is apparent.

Eremacausis becomes

exceedingly active, and in the course of a few years the substratum
is reduced to a brownish-black, pulp-like mass.

I f continued, this

goes to form " m u c k , " a substance which when dry is powdery and
somewhat resembles soot.

During these processes o f . decay there

occurs a succession among the organisms present.

T h e accumulation

of disintegration products m a k e s the medium unfavorable for the con­
tinued existence of the organism involved in their production.

At the

same time it m a y furnish optimum conditions for the development of
other forms.

An acid medium favors the growth of the Phycomycetes,

while alkalinity favors the bacteria.

I n such regions as this, where

the underground waters are alkaline, the latter fact, together with
fluctuations in the ground-water level, m a y have an important bearing
upon the presence of more thoroughly decayed peat and of a distinct
depression about the margins of m a n y of the bogs.
I f to the factors of relative scarcity of oxygen and the acidity of
the soil solution is added the occurrence of temperatures considerably
lower t h a n those of the surrounding uplands, it is not difficult to
understand why a large part of each year's vegetative products should
escape complete destruction.

I n our estimate of the bog substratum

as a habitat for higher plants, the strong competition with the micro­
scopic plants to which the former are subject in the acquisition of
oxygen for their underground parts, must b e emphasized.
THE PHYSICAL A N D CHEMICAL PROPERTIES

OF PEAT.

T h e peat formed through the agency of the bog sedges and their
attendant plants has a fibrous and m a t t e d appearance.

T h e structures

of the various dead stems, roots, and leaves have suffered but slight

alteration.

T h e y were originally strongly cuticularized, and this has

aided in their preservation.
brown.

T h e color is commonly a pale yellowish-

During life these plant materials b e c o m e strongly matted

and interwoven, and this structure frequently persists.

I t is this

structure that gives to the C a r e x - E r i o p h o r u m zone in m a n y lakes its
strength to support heavy bodies.

A m a n ' s weight will carry the

substratum a foot beneath the surface of the water, but it seldom
b r e a k s under the strain.

I n the case of lakes where this zone is

unusually developed, it m a y cover a large part of the lake surface and
b e of great importance in the filling in of peat.

I n such cases the

deposition takes place largely b y the gradual falling of material from
the under side of the floating substratum.

O n account of the slight

weight of the material, it does not descend and produce a compact
deposit on the bottom, but forms a sort of thick liquid peat.
T h e sphagnum-shrub zone, where well developed, usually shows a
brown peat beneath it.

I t is composed largely of sphagnum and the

semi-decayed twigs, rhizomes, and leaves of the other plants.

I t is

distinctly fibrous, but of a type different from that of the sedge zone;
the fibers are short, and the material is not nearly so tenacious.
U n d e r the t a m a r a c k s a large part of the annual peat increment is
m a d e up of the t a m a r a c k needles, though mosses (Hypnum, Sphag­
num, and Polytrichum) usually are of importance in this connection.
T h e color is reddish-brown and darker than that of the shrubby
zone.

T h e fibrous structure is still less apparent, though present.

W h e n these bogs have been burned over and partially drained,
there frequently comes in a dense ground covering of moss (Poly­
trichum).

I n such cases the peat continues to accumulate, largely

through the agency of this plant.

I n such situations the peat is a

reddish-brown, and the plant structures have practically disappeared
through decay.

B e l o w the upper layer, the peat when moist has the

sticky, clayey properties of well-decomposed peat.
O n e other well-marked stage is shown in the areas of muck land
now under cultivation to onions and celery.

Under the influence of

drainage and tillage, the disintegration is nearly complete.

All plant

structures have disappeared, the humous acids have been largely
neutralized or washed out, and there is left only a fine, powdery,
brownish-black " m u c k . "

T h e following table shows some other physical properties of these
several varieties of peat.
Eriophorum Zone
Peat

H 0-capacity %
by volume
H 0-capacity %
by weight
Air dry H 0-con­
tent %

Fresh
Sphagnum

CassandraSphagnum
Zone Peat

Tamarack
Zone Peat

Chelsea
Bog Peat

Onion
Marsh Muck

2

78.0*

87.0

91.0

84.0

82 6

75-o

2

892.0

I550 0

960. 0

530-0

477.0

283.0

8.5

10 6

10.0

10.0

10.0

10.0

2

*Low volume percentage due to air present in tissues.
T h e s e measurements were m a d e b y placing the peat samples in a
zinc cylinder of 600

c c

capacity.

closed with a wire gauze cap.

T h e b o t t o m of the cylinder was

T h e moist peat was tamped into the

cylinders with as nearly uniform a stroke as possible.

T h e cylinders

were then set in a dish of water for eighteen hours, after which the
cylinder was removed and allowed to drip.
ceased, the cylinder was weighed.

W h e n all dripping h a d

T h e peat was then removed and

allowed to dry at room temperature, and again weighed.

F i n a l l y it

was dried a t n o ° C , and the absolute weight determined.

A s usual

in such measurements, considerable irregularity was shown b y the
different samples, owing to the difficulty of removing the air, and of
p a c k i n g to the same degree.

However, the figures bring out clearly

t h e fact that sphagnum m o r e t h a n any other plant influences the
water-capacity of a peat containing it.

T h e eriophorum peat h a s a

lower capacity, owing to its coarse fibrous structure.

O f the series

examined, the highest water-capacity was found in the cassandra
zone.

T h e effect of further decay and destruction of the plant tissue

is shown b y the reduction in water-capacity of the last three m e m b e r s
of the series.

T h e percentages are of interest in connection with the

utilization of such lands for agricultural purposes, in showing the
difficulty of proper drainage.

I t is the experience of the m e n who

ditch these bogs that until the peat has reached the condition termed
" m u c k " the ditches act only with extreme slowness.
Chemically, peat or h u m u s is m a d e up of varying quantities of
several substances of a rather indefinite character, which are com­
monly

classified

among

the

dehydration

products

of the carbo-

hydrates.
ficially

T h e s e bodies not only occur in nature, but m a y b e arti­

produced b y the action of strong acids on starch, sugar, and

cellulose.

T h e relation of nitrogen to these bodies is still unknown.

Principally on the basis of color and solubility in alkalies and acids,
there are several substances distinguished.

U l m i n and ulmic acid

are brown, and are early products of decomposition.

Humin

and

h u m i c acid are b l a c k , and occur m o r e abundantly where eremacausis
has been active for a long time.

C r e n i c and apocrenic acids appear

to b e further oxidation products;
latter varies from yellow to brown.

the former is colorless, and the
M A Y E R believes these bodies to

b e organic nitrogen compounds (36), and on this basis STOCKBRIDGE
(50, p. 135) explains the insolubility of peat soils and the presence of
the unavailable nitrogen in peat.

B e s i d e these substances

saccharic, and glucinic acids have been recognized.

xylic,

Although great

advances have been m a d e in soil chemics, it seems strange that the
only suggestion of formulae for these substances was m a d e b y M U L D E R

1861 (38).

in

H u m i c acid forms water-soluble compounds with the alkalies, and
to these are due largely the brown colors of the bog waters.

T h e color

m a y b e produced b y the presence of free h u m i c acid.

W i t h the

alkaline earths h u m i c acid forms insoluble or difficultly soluble com­
pounds.

H e n c e there is slight c h a n c e of lime and magnesia pene­

trating from the surrounding soil into the peat deposits.
D u r i n g the changes which the plant material undergoes in the
process of peat-making there are alterations in the relative amounts
of volatile hydrocarbons, fixed carbon, and ash—using these terms
as in ordinary coal analyses.
Eriophorum Stems
and Leaves

Volatile combustible
Fixed carbon
Ash.
H O
a

68.2
21.0
3-8
7.0

Sphagnum

67.0
17.9
4-5
10.6

Eriophorum Zone
Peat

62 .0
21.8
7-4
8.8

Cassandra
. Zone Peat

54-0
22 .9
13.8
9-3

Tamarack
Zone Peat

53-o
23.4
13.6
10.0

Onion
Marsh
Muck

45-7
21.8
22.5
10.0

T h e proportion of volatile combustible m a t t e r decreases regularly
as the humification proceeds.

T h e ash regularly increases, while the

air-dry water content shows b u t slight modification.
(To be continued.)

THE

BOGS

AND B O G F L O R A

OF THE HURON

RIVER

VALLEY.
EDGAR

NELSON

TRANSEATJ.

(WITH SIXTEEN FIGURES)
[Continued

from p. 375.]

THE BOG AS A HABITAT FOR PLANTS.
W H E N we consider the b o g as a h a b i t a t for plants, there is at once
brought to m i n d the m a r k e d contrast between its characteristics and
those of the other plant habitats of its vicinity.
pheric and edaphic conditions it is unique.

I n both its atmos­
T h e various factors

entering into the plant environment will b e discussed as physical,
chemical, and biotic agents.
A.

PHYSICAL F A C T O R S . — 1 . Wind.—Because

of the fact that so

large a n u m b e r of our bogs lie in depressions surrounded b y hills,
the influence of the wind is somewhat lessened.

I t is only in the

case of the larger basins that its effects b e c o m e m a r k e d .

I t has been

noted b y several students of bogs (41, 5, p. 3 7 ; 59, 47) that in the
region of prevailing westerly winds the greatest development of bog
areas and peat deposits occurs on the western sides of l a k e basins.
W h e r e the deposition has taken place in a large l a k e basin, which
is now only partially filled, we c o m m o n l y find open water occurring
toward the eastern side.

T h e peat deposits at Portage, P a r k s , and

W e s t L a k e s in the vicinity of Ann A r b o r are massed on the western
shores, while the eastern margins exhibit an ordinary l a k e b e a c h .

At

the bogs north of D e l h i , although nine-tenths of the original ba,sin has
been filled, the two small lakes are near the eastern margin.

The

facts noted in this region all favor the idea of the bog plants being
unable to gain a foothold on the eastern side in the presence of wave
action.

T h e shoreward thrust of the ice is of importance at times in

this connection.
F a r t h e r north in M i c h i g a n the wind frequently shows its extreme
effect in these bog areas in the presence of " w i n d f a l l s . "

Owing to

the c h a r a c t e r of the substratum, such areas are more readily affected
418

[DECEMBER

than the forests o f mineral soils.

T h e s e p h e n o m e n a have not been

observed in a n y o f the bogs in this vicinity.
T h e s a m e statement holds for t h e presence of loose floating bogs
which are driven about on lakes b y winds (35).
2.

Temperature.—In its temperature relations both the topography

and the c h a r a c t e r o f the substratum c o m b i n e to influence the b o g
habitat.

I t h a s long been noted b y agricultural writers that reclaimed

b o g areas are particularly subject to late frosts in the spring.

One

of the causes of this peculiarity lies in the fact that on clear a n d quiet
nights t h e cooled air overlying elevations drains into the depressions
(11).

S o m e recent observations m a d e b y S E E L E Y (45) near C h i c a g o

show how effective such atmospheric drainage m a y b e even in dis­
111

tricts whose range of elevations amounts to b u t 15 feet (4.5 ).

He

found that the hilltop averaged, on the night of the observations,
2.5

0

0

F . ( 1 . 4 C . ) higher t h a n that of t h e depression while a ther­

m o m e t e r placed 30 feet (9™) above the hilltop averaged
(5

0

C . ) above that of the " s w a l e . "

8.8° F ,

O n comparing the temperatures

of atmospherically undrained a n d drained depressions with that o f
0

the hilltop, h e found that the hilltop temperature was 3 6 . 3 F . when
0

that of the drained depression was 3 6 F . a n d that of undrained
31.8° F .

H e r e is a particular instance in which frost occurred in

the undrained depression, b u t not in the other situations.
nights low grounds in general are subject to lower

O n quiet

temperatures

than the adjoining highlands, a n d it is p r o b a b l e that these effects
are m o r e pronounced in the case of undrained depressions.
A second factor in t h e production of low temperatures in bogs
is found in the nature of the substratum.

I n the spring the ice which

h a s formed b e n e a t h the cassandra a n d t a m a r a c k areas melts with
extreme slowness, when once the surface of t h e soil h a s been reached.
T h i s is explained b y the low conductivity of the loose, partially
decayed, vegetable covering, a n d b y the shading of the plants above.
F o r example, at F i r s t Sister L a k e , in 1904, the ice h a d disappeared
from the water surface on April 10.

O n April 17, with a n air tem­

perature o f i o ° C , the temperature of the substratum in t h e bog
sedge zone averaged i o °
t a m a r a c k zone 3

0

in the C a s s a n d r a zone 6° C , in the

C , a n d the area of willows a n d sedges 8° C.

Ice

was found at several points a m o n g the t a m a r a c k s , an inch below the

surface.

mm

T h e sedge zone was covered with i to 3 inches ( 2 5 ~ 7 5 )

of dark colored water.

T h e other soils were wet, but their loose

texture was effective in preventing a rise of temperature.
I t follows that of the various situations in bog areas those most
liable to extreme low temperatures in the spring are in the cassandra
and t a m a r a c k zone.

Since their m a x i m u m temperatures are con­

siderably below those of neighboring areas, on quiet nights the plants
there are but little protected b y radiation from the soil as compared
with plants of other situations.
I n the following table it is shown that the soil temperatures of
the several plant societies formed about a bog are different, and that
each society has a characteristic temperature range.
were m a d e at F i r s t Sister L a k e .

are averages of readings m a d e in the second inch (2 5
surface.

T h e records

T h e temperatures, given in ° C ,
m m

) below the

T h e " w i l l o w - s e d g e " conditions correspond to those of the

ordinary swamp.

T h e " m a p l e - p o p l a r " is an area of these trees on

the peat substratum.

T h e " u p l a n d " is a sandy, sod-covered area

111

3 feet (0.9 ) above the surface of the bog.

T h e temperatures for

the most p a r t were taken on clear afternoons about 3 P. M. when the
differences are at their m a x i m a .

Date-^
Air temperature
Upland
Willow-sedge
Cassandra
Tamarack
Bog-sedge
Maple-poplar

April April April April April May May Mav May May June June
4

10.5
11.0
7.0
1 5
0 0
9 0
7 0

12

2.0
7 0
8 0
2 0
0 0
8 0
8 0

17

25

29

10 0
10 0
8.0
6 0
3 0
10 0
8 0

8-5
10 0
9.0
7 5
4 5
9.0
8.0

18.0
17.0
14.5
11.0
9 9
18 0

i5-o

3
24 0
20 0
17.7
14.7
11.7
19.0
18 0

6

16

27.0
23.0
iQ-5
15 5
15 0
22.0
19 0

21

27

15.0
16.0
15.0
13-0
10.6
16.0
15-0

26 0
20 5
20 0
16 5
15-0
20 0
16 0

21 0
21.5
22.0
19.0
17.0
23.0
15.0

6
26 0
25 5
22 0
20 0
18.0
24 0
17.0

15

26.0
26 0
20 5
19.5
18.0
23.0
16 5

I n the accompanying diagram (fig. 5 ) it will b e seen that the
upland, bog-sedge, and willow-sedge soil temperatures do not deviate
widely from those of the air, while the temperatures of the cassandra
and t a m a r a c k areas range considerably lower.

T h e high temper­

ature of the bog-sedge zone finds its explanation in that the brown
bog water overlying its surface absorbs heat.

I have tested this point

m a n y times in various bogs and have always found such bog water to
have a higher temperature than that of the saturated
adjoining it.
drained sand.

substratum

I n its ability to absorb heat rays it approaches that of
I t s range, however, is m u c h less and it retains its

heat for a longer time.

Consequently on cloudy days and following

a sudden lowering of the air temperature,

the surface

bog-water

temperature stands above that of the drained and undrained soil.
W h e n we compare the effects of loss of heat from a free water
surface and a saturated h u m u s soil surface due to evaporation, there
30

Temp.

T&mp

?A

Air

26
//'
il
//
\

21

\

J ' ' /

? 4 *

ft

/

II

It

I $

Y

M

y

/

w\

C\\ //
\
>

//

/

f..

/
A

17 /

' / /

\ \ /

/

It A

ffl

/

/

y

J

<• /
e

/ /

l

z
0

^ April

/

17

w

*9

arruu/ «

/ e

«*'

*?

oJcm &

FIG. 5.—Diagram showing temperatures of the air and the substrata in the
several plant societies.
is a m a r k e d difference due to their specific heat.

T h e h u m u s will

b e cooled m o r e rapidly b y the evaporation of a given amount of
water.

W h e r e so large an evaporating surface is exposed to the air

as in the case of a sphagnum-covered area the loss of heat b y this
process is most effective in preventing the penetration of heat below
the surface.
I n the case of drained soils, the most effective agent in raising

the temperature of the subsoil is that of percolating water which has
been warmed at the surface of the soil.

B e c a u s e of the high water-

table and the stagnant condition of the underground water in bog
areas, this source of heat is relatively unimportant.
T h e effects of these factors, resulting in low soil temperatures,
are far-reaching.

As compared with well-drained soils, chemical

action is retarded, the rate of diffusion, solution, and osmosis is
greatly reduced, and the conditions for the existence of soil bacteria
m a d e unfavorable.

P l a n t s which can successfully compete for the

occupancy of such areas must b e able to withstand low temperatures
and late frosts.

T h e difference between the temperature of the air

and that of the substratum favors plants having a low transpiration
ratio.
However, in so far as the region of southern M i c h i g a n is concerned,
the temperatures prevailing in bog areas do not seem to b e adequate
to account for the presence of the bog plants or their xerophilous
structures.

I t is to b e noted that with the leafmg-out of the trees,

about M a y 27, the temperature of the maple-poplar substratum falls
below that of the t a m a r a c k .

B u t that the soil temperature is one

of the factors entering into the problem of competition between
species there can b e little doubt.

I t is probable also that in the region

of optimum conditions for bog plants the conditions which occur
here only in the spring are prolonged through the summer.

That

is, the difference between air and substratum temperatures is more
m a r k e d , and is a powerful factor in the selection of plants for bog
areas and in the production of xerophilous structures.
3. Texture.—This

property of the substratum has already been

referred to in connection with the genetic changes in peat.

T h e sedge

zone is developed upon a raft of interwoven rhizomes and roots.

It

is a coarse m e s h w o r k ; but since it lies at or below the surface of the
water, its texture is of slight importance except as a means of mechan­
ical support.

As the bog develops, the admixture of moss and shrub

debris brings about the formation of a rather compact peat, overlaid
b y a stratum of loose material.

I n some cases, as at D e l h i and

km

Oxford, 45 miles ( 7 2 ) northeast of A n n Arbor, the living sphagnum
m a k e s up the b u l k of this loose covering.
lies j u s t beneath it.

Usually the water level

As a consequence, this covering b e c o m e s the

principal seat of root activity.

T h e small, fibrous roots of cassandra,

andromeda, and the cranberry penetrate it in all directions, and it is
from the water which is held among this moss a n d debris that they
derive their water and mineral salts.
T h e substratum beneath the t a m a r a c k s is also covered b y a loose
litter of leaves and twigs, with more or less moss.

Depending upon

the height above the ground water level, this surface layer is of greater
or less thickness.
rack.

I n it occur the wide-spreading roots of the t a m a ­

D u r i n g summer a n d autumn it furnishes admirable conditions

for the growth of fungi, a n d it is penetrated everywhere b y their
mycelia.
W h e n bog land is cleared, the decomposition of the surface layers
is very rapid, owing to exposure to sunlight and higher temperatures.
I f the water-table is maintained near the surface, sedges and willows
develop as the covering.

T h e annual increment of plant material

is often decreased, and in place of the fibrous and porous substratum
there is produced a b l a c k , close-textured, and plastic m u c k .
I f ditching and draining are added to clearing, the summer drought
dries the surface layer so thoroughly that it often becomes the habitat
for m a n y dry-ground weeds.

D e c a y progresses in moist weather

under the influences of the higher temperatures

resulting

from

increased absorption of the sun's energy b y the dark colored soil.

4. Mechanical properties.—Bog soils in general do not afford as
good a foothold for the development of tree species as do the mineral
soils.

O n account of the high water-table, the roots of the plants

are not able to penetrate to a depth of more t h a n a few inches.

The

roots of the t a m a r a c k s spread out in all directions from a flat trunk
base, and upon the size a n d strength of these horizontal roots depends
the tree's ability to withstand m e c h a n i c a l strains tending to displace
it.

T h e r e c a n b e no doubt b u t that, in the t h i c k groves in which the

t a m a r a c k occurs, the interweaving of the roots from adjacent trees
becomes of mutual advantage, in so far as the roots function as hold­
fast organs.
5. Diffusion properties.—A most important soil property relates
to the diffusion of mineral salts.

T h i s becomes of especial signifi­

cance in saturated stagnant substrata.

T h e mineral salts must b e

distributed to the roots mainly b y diffusion, for lateral

drainage

and percolation are at a m i n i m u m .

I t is well known that when salt

solutions are passed through soil, m u c h of the salt is retained b y
absorption.

T h e relative amount is greatly increased in the case

of humous bodies.

B L A N C K (4) has further found that the diffusion

of water in h u m u s soils is decreased b y the presence of acid h u m u s
compounds, and that this m a y b e corrected b y the addition o f a
neutralizing agent, such as lime.

All analyses of peat show how little

of this mineral m a t t e r h a s been derived from the adjacent soils.

It

is only in the case of samples taken from the b o t t o m or edge qf a bog
that the mineral salts cannot b e accounted for b y the amount derived
from the decay of the plant material, and that obtained from the
atmosphere.
6. Water-capacity.—The
been noted.

high water-capacity of peat has already

I n relation to plant growth, it is detrimental in that it

prevents proper aeration of the substratum (39, p. 346).

S o far as

the diffusion of gases is concerned, such substrata are less favorable
t h a n a free water surface.

K i n g (29, p. 161), in speaking of sand

and clay soils whose water-capacity is only 1 7 . 5 to 32.2
by weight, says that 30 to 40 per cent, of their saturation

per cent,
amounts

must drain away before the soil can contain air enough to m a i n t a i n
the respiration of roots and germinating seeds.

As compared with

a free water surface, saturated h u m u s cannot admit oxygen as freely,
owing to the large part of the surface actually occupied b y the h u m u s
(29, p. 239).

I n a chemical way it is still more effective, as will b e

noted later.
7. Osmotic pressure.—The

osmotic pressure of bog waters has

been found to b e about the same as that of ordinary lakes and rivers.

3

T h e y are approximately equivalent to a o . 1 to o . 5 per cent, normal
K n o p ' s solution.

T h e y indicate quite certainly that bog plants do

not owe their distribution and their peculiar structures to a high
osmotic pressure of the bog water.
3 Four samples of bog water from this vicinity were tested by Dr. B . E . LIVING­
STON, of the University of Chicago, and found to have the following pressures in milli
0

meters of mercury at 2 5 C :
First
First
West
West
Lake

See 33.

Sister Lake, Sample A
Sister Lake, Sample B
Lake, Sample A
Lake, Sample B
Michigan water

50 0742
40 0593
100.1484
150..2226
100.1484

B.

CHEMICAL

Ground water.—The ground water

FACTORS.—i.

of the H u r o n basin derives its m i n e r a l constituents from the glacial
drift.

T h e following analyses show the c h a r a c t e r of t h e solution.

Quantities a r e expressed in parts per million (31).
*
CaC0
Ann Arbor,
University well
Ann Arbor,
spring
Ypsilanti,
water works
Ypsilanti,
well
Ann Arbor,
creek
*.

3

CaS0

4

178 00

Fe 0
3

3

MgC0 K S0
3

4

Si0

3

NaCl' N a , C 0 N a S 0
2

3

7 30

4 48

1 52

5.31

9.20

4 88

0 42

100 00

14.00

6 43
21 00

Organic Total
and
mineral
volatile matter

5 07

3-85

267.72

9.71

25 00

353-31
498 00

35 00

60.58
89 36

4

71.00

6.78

3 99

228.00

2

289.00

39 00

156.00

223.00

Tr.

109 00

18.00

62.00

17 00

14.00

585.00

128 00

99 00

Tr.

83 00

25 00

15.00
(NaK)

25.00
(NaK)

14.00

375-00

I t is to b e noted that they are all high in calcium and magnesium
content, a n d under favorable drainage conditions contain sufficient
minerals for plant growth.

T h e ground water is of especial impor­

t a n c e in the early stages of b o g development, when t h e sedge a n d
a q u a t i c vegetation is dominant.

W i t h the further development of

the sedge zone a n d t h e formation of a thick peat deposit, its relation
to t h e vegetation b e c o m e s of less m o m e n t .

T h e r e is a notable

difference between t h e total mineral content of b o g water a n d that
of the soil waters adjoining.

I n the above table the total mineral

content of t h e ground water varies from 267.7 to 5^5 parts per
million.

I n three analyses of the b o g water at t h e F i r s t Sister L a k e

I found the total mineral content to vary from 89.9 to 219 parts p e r
million, t h e highest figure being that for t h e sample obtained n e a r the
margin of the t a m a r a c k s , i. e., nearest the mineral soil.
T h e absence of sphagnum from certain bogs h a s b e e n explained
by t h e presence of calcium salts ( 1 5 , p. 23, 16).

I n order to test

this point, I have cultivated the species found in this vicinity in tap
water and in a saturated solution of C a C 0 , a n d have found no
3

detrimental effects due to calcium.
cussed later.

T h e experiments will b e dis­

I further found that the ash of sphagnum growing at

F i r s t Sister L a k e contained 18 p e r cent, of C a O .

I t would seem,,

therefore, that, in so far as this vicinity is concerned, the presence of
calcareous waters will not explain the absence of species of sphagnum.

2. Acidity.—Much

stress h a s been

laid

b y various

authors,

following SCHIMPER (44, p p . 6, 18, 124), upon the acidity of the bog
water as a factor in the b o g habitat.

I n order to get a quantita­

tive statement of the acidity for the bogs of this vicinity, a num­
ber of 5 o

c c

samples have been titrated with a n n/100

solution of

potassium hydrate Phenolphthalein was used as a n indicator.

The

results show a n acidity varying from .00015 to .00258 n o r m a l a c i d .
The

lowest values are found in the areas occupied b y b o g sedges

and b y swamp plants, a n d they are practically the same.
occupied b y cassandra a n d sphagnum
acidity.
The

4

T h e areas

have a somewhat

greater

T h e highest percentages are found b e n e a t h t h e t a m a r a c k s .

explanation of these variations in acidity is suggested b y the

tests, m a d e
cultures.

from

time to time, of t h e water in m y experimental

I found that the acidity of t h e water increased slowly in

the undrained

peat

substratum

cultures

(see experiments).

The

increase was small in the case of the w a r m cultures, b u t quite notable
in the case of t h e cold undrained substratum.

O n exposure to a i r

in t h e water cultures, a n d in bottles, t h e acidity very slowly decreased,
the decrease being greatest in the case of t h e water which was kept
warm.

T h i s is probably due to increased oxidation.

T h e s e relative

amounts of acid, it will b e seen, m a y b e correlated with the tempera­
tures in t h e several plant societies of t h e bog, t h e lowest temperatures
corresponding to the highest percentages of acid.

T h i s suggests

the probability that t h e acidity of the b o g substratum

increases

farther north.
O n allowing open dishes of b o g water to stand for some time,
I found that the evaporation was not sufficient to raise the acidity of
the water, oxidation apparently being more rapid than concentration
of t h e solution.
T h e r e is n o apparent relation between color a n d acidity, although
the lightest colored solutions usually show b u t slight acidity.

This

seems to indicate t h a t , o n l y a part of t h e color is produced b y free
h u m u s acids, the remainder b y humates of the alkalies.
4 Following are the determinations expressed in fractions of a normal acid solution:
First Sister Lake: sedge zone, .00066, .00094; cassandra zone, .00152, .00119;
tamarack area, .00165, .00179, -00227, -00258; willow-sedge area, .00089, -00072.
Chelsea: ditches, .00086, .00015, • 4 3 J -00019, d .00029.
Delhi: tamarack area, .00146, cassandra zone, .00117.
Oxford: cassandra zone, .00094.
0 0 0

a n

T h e effect of acidity upon cultivated plants h a s been investigated
in this country especially at t h e R h o d e I s l a n d Agricultural E x p e r i m e n t
Station, under t h e direction of Professor H . J . W H E E L E R .

The

experiments have been conducted upon " a c i d upland s o i l s ' ' (60),
and

numerous

reports have been published.

T h e s e experiments

involved a great variety of plants a n d were carried on under natural
field conditions.

T h e areas planted for comparison h a d their acidity

neutralized b y the addition

of C a C 0 .
3

T h e plants which

were

favored b y the liming include the orange quince, b l a c k T a r t a r i a n

cherry, J a p a n plum, Tilia americana, Ulmus americana, rhubarb,
Australian

salt-bush,

hemp,

barley,

oats, onions,

Anthoxanthum

odoratum, Poa pratensis, Festuca ovina, Holcus lanatus, Festuca
elatior, Alopecurus pratensis, etc.

Plants which*appear to b e adapted

to t h e acid soil conditions include cranberry, b l a c k b e r r y , raspberry,
sheep sorrel, cow-pea, flax, corn, lupine, a n d soja b e a n .

I t would

appear, then, that the acidity of the soil solution is unfavorable for
the growth of some plants, a n d that it is a factor in the selection o f
species for acid soil conditions.
3. Food material.—As to t h e presence of plant food materials in
the b o g soil there is a n agreement among all the analyses that have
been m a d e .

5

T h e soils are unusually rich in nitrogenous materials,

some analyses showing three times as m u c h as good upland soils.
B u t in the slow decay of t h e vegetable m a t t e r the nitrogen remains
b o u n d up in organic compounds a n d is unavailable for the growing
plants.

T h i s is confirmed b y experimental tests in which nitrogen

was directly applied, a n d b y tests in which the conditions were
modified so as to permit the action of nitrifying b a c t e r i a .
cases crops were produced when t h e untreated h u m u s

I n such
produced

none.
U n d e r natural conditions the growth of the nitrifying b a c t e r i a
in b o g soils is almost impossible.
activity:

T h r e e factors work against their

(1) the acidity of the soil solution;

due to high water content;

(2) the l a c k of oxygen

(3) the lower temperature.

I t h a s been
0

found that t h e optimum temperature for these b a c t e r i a is 9 8 F .
0

(36.6° C ) , a n d that their activity is very slight at 50 F . ( i o ° C . )
s Analyses of Wisconsin soils. Ann. Rept. Wis. Agric. Exper. Sta. 1 3 : 304. 1896.
See also 27, p. 12; 23; 22, p. 276; 30; 48, p. 234; 12, p. 39; 1 4 ,

(3).

F u r t h e r m o r e , it has been shown that when soil rich in nitrogen

is saturated with water so as to exclude free oxygen, denitrification
t a k e s place and nitrogen gas is set free (29, p . 115).
T h e phosphoric acid content is c o m p a r a b l e with that of the best
soils, and it is at least partially in a condition for plant use.
The

potassium content is very low.

Analyses and the results

of agricultural experiments show that in order to produce crops this
substance must b e added, and preferably in an alkaline form.

Inquiry

a m o n g the owners of onion marshes in this vicinity confirms the need
for potassium in local bog soils.
The

amount of calcium present is reported as equal to that of

the best upland soils.

B u t -it is p r o b a b l e that as it exists under

natural conditions in bogs it is bound up largely in insoluble h u m a t e s .
U n d e r the influence of oxidizing processes it would b e c o m e available
to the plants at the surface.
W h e n we consider the conditions under which the various plant
societies in our bogs exist and their competition with one another,
there can b e little doubt b u t that the substratum varies in each case
as to its c h e m i c a l composition.

T h a t the societies m a y b e classified

on a physiographic basis is certain, but how to determine the chemical
factors

a c c o m p a n y i n g each physiographic

problem.

change is an

unsolved

T h e ordinary methods of analysis give us the minerals

present, b u t tell us little about their form and availability for plant
assimilation.

T h e colorimetric methods for determining the quantity

of m i n e r a l salts present in bog water are mostly open to objection.
T h e ease with which the humous bodies of the bog water are decom­
posed render their quantitative estimation b y present methods of
little value.

Y e t it seems p r o b a b l e that work upon the chemistry

of h u m u s and humous compounds must result in data valuable alike
to the ecologist, the forester, and the agriculturist.
C.
be

BIOTIC F A C T O R S . — T h e interrelations of the b o g species will

discussed in connection with their other ecological characters.

I t will b e sufficient to mention here that they are with a few excep­
tions light-demanding

forms.

Consequently, size and

ability to

produce shade are the important factors in their competition with one
another.
A second element enters into this problem of the struggle between

species near the borders of the area of geographic distribution of
the bog plants, viz., climate.

T h e bog plants of this vicinity come

into conflict with species whose range is either more nearly continental
or more southern.

T h a t the climatic and edaphic conditions of

this region are at present unfavorable to the successful competition
of the bog species with swamp species is evidenced wherever the bog
conditions have been disturbed.

T h a t the reverse is the rule in

eastern C a n a d a has been shown b y GANONG (18, p. 178).
tenacity with

which

species, whose multiplication

is

The

principally

accomplished b y vegetative m e a n s , hold an area under complete
control is apparent to any who have studied the vegetation of l a k e
shores.

I t is j u s t as strongly m a r k e d in the case of the herbaceous

and shrubby bog vegetation.

W h e n we examine the chemical and

physical data, now at hand, concerning the soils occupied b y bog
and swamp plants respectively, the conclusion must b e that they are
wholly inadequate to account for the difference in vegetation.

The

forester lays stress upon the fact that trees cannot gain a foothold
on areas now covered with a grass turf because of the difficulty of
the seedlings getting started.

T h e bog societies form an equally

compact plant growth, and their preservation in this region would
seem to b e dependent upon analogous factors.
III. The bog-plant societies.
T h e following descriptions of local bog areas occurring in the
H u r o n valley aim not only to present lists of plants found in this
vicinity, but to show their natural associations.

T h e order in which

the areas are described corresponds to the relative amount of filling
which has occurred in the several basins.

T o a certain extent this

order is genetic, yet there can b e little doubt but that m a n y arctic
plants which were concerned in the pioneer stages of our mature bogs
are now extinct.

I f we accept the areas at W e s t and F i r s t Sister

L a k e s as representing bogs in youth, maturity m a y b e illustrated b y
the original vegetation of the b o g on Carpenter's road.
area defines that
inaugurated

stage beyond the climax, when

b y cutting,

firing,

and

T h e Chelsea
the

conditions

ditching have destroyed

the

original t a m a r a c k forest, and in its place has come a rude mixture
of bog relicts and arborescent weeds.

WEST L A K E .
T h i s lake, situated three miles north of Chelsea ( S e c . 30, D e x t e r
T p . ) , is also known locally as J o h n s o n ' s L a k e .

I n area it is slightly

m o r e than a fourth of a square mile (65 h e c t a r e s ) .
the lake originally extended a half mile ( o . 8
southwest.

k m

T h e margin of

) farther west a n d

T h i s part is n o w occupied b y a partially floating bog.

T h e north, south, a n d east shores are sandy a n d low.

Patches of

bulrushes a n d water-lilies occur here a n d there over t h e l a k e a n d show
its generally shallow character.

T o w a r d the east there is a narrow

swampy outlet b y which its water after a long a n d circuitous route
reaches the H u r o n R i v e r .
lake.

T h e r e are n o streams tributary to the

T h e basin lies near the southeastern margin of the interlobate

moraine, a n d is bounded on the north a n d south b y hills 60 to 80 feet
(18-24™) in height.

N o t all of the original extension to the southwest

has been filled b y p e a t ; two small areas o f open water still remain.
T h e shores, with the exception o f the western side, support a
vegetation

similar to that

of m a n y

lakes in this region.

Three

societies of plants m a y b e distinguished.
Aquatics.—The

most

abundant

plants

are Scirpus

lacustris,

Castalia tuberosa, a n d Sagittaria rigida.

T h e s e occur not only along

shore, b u t in shallow water throughout

the lake.

Associated with

these are N a i a s flexilis, B r a s e n i a purpurea, Potamogeton heterophyllus, C h a r a

(sp.), Spirodela polyrhiza, Vallisneria spiralis, Scirpus

americanus, a n d D e c o d o n verticillatus.
Sedge-grass society.—Very n e a r the north, south, a n d east shores
occur a great n u m b e r of species of grass-like plants.
vary greatly at different parts of the shore line.

T h e i r associations

T h e dominant forms

are C a r e x filiformis, P a n i c u l a r i a nervata, E l e o c h a r i s palustris, C a r e x
teretiuscula, C. Muskingumensis, D u l i c h i u m arundinaceum,

Panicu­

laria Canadensis, Dryopteris Thelypteris, and Scutellaria galericulata.
A m o n g t h e species of secondary importance are O n o c l e a sensibilis,
C a r e x riparia, C. stipata, C. hystricina, C. interior, Spartina cynosuroides, T y p h a latifolia, I r i s versicolor, L o b e l i a K a l m i i ,
palustre, L y c o p u s americanus, a n d E u p a t o r i u m m a c u l a t u m .

Comarum
Closely

associated with these plants are the seedlings of the shrubs a n d trees
which m a k e up the next society.
Willow-maple society.—The

shrub a n d tree border is composed,

for the most part, of Salix B e b b i a n a , S. discolor, S. sericea, Cornus
candidissima,

A c e r rubrum,

and

Ulmus

americana.

B e s i d e the

m a n y plants of the sedge-grass society which remain as relicts, the
accessory species include R o s a Carolina, I m p a t i e n s biflora, S a m b u c u s
pubens,

Spiraea

velutina,

and

salicifolia,

Opulaster

Prunus

opulifolius.

serotina,

Quercus

alba,

Q.

into

T h e s e trees grade

the

forests of the upland and establish a natural order of succession.
A n interesting comparison is afforded when we note the species
dominant along the western or bog margin.

H e r e the outer zone of

aquatics is m a d e up of the s a m e species, but this substratum is a
floating

raft constructed b y the plants themselves.

W i t h o u t again

enumerating the species, we pass to the society which closely follows
their development.

Bog-sedge and shrub society.—This

society forms a very complex

m

growth, averaging 50 feet (15 ) in width.

O n the lake ward side are

the a q u a t i c s ; on the other, the growth of t a m a r a c k s .

T h e sedges and

shrubs are not separable, as in m a n y other localities.
formis is b y far the most important plant in the society.

C a r e x filiI t s vigorous

production of rhizomes and roots especially fit it for the position which
it occupies.
consequence.

Certain other plants are locally abundant and of great
T h e s e include

Dryopteris

thelypteris,

Menyanthes

trifoliata, E l e o c h a r i s palustris, C o m a r u m palustre, Sagittaria latifolia,
E r i o p h o r u m polystachyon, C a r e x teretiuscula, T y p h a latifolia, Salix
myrtilloides, S. Candida, B e t u l a glandulosa,
pus, a n d A n d r o m e d a polifolia.
tioned

6

Oxycoccus macrocar-

A s accessory species m a y b e men­

Salix discolor, S. B e b b i a n a , C i c u t a bulbifera,

Cardamine

pratensis, C h a m a e d a p h n e calyculata, C a m p a n u l a aparinoides, R u m e x
B r i t a n n i c a , E p i l o b i u m adenocaulon, Asclepias incarnata,

Pogonia

ophioglossoides, Blephariglottis blephariglottis, L i m o d o r u m

tuber­

osum, M a r c h a n t i a polymorpha, A u l a c o m n i u m palustre, S a r r a c e n i a
purpurea, D r o s e r a rotundifolia, B o e h m e r i a cylindrica, C a r e x comosa,
C. hystricina, Cornus stolonifera, P a r n a s s i a caroliniana, Viola blanda,
and P e n t h o r u m sedoides.

H e r e and there occur young t a m a r a c k s

which b y their growth inaugurate the next society.

Tamarack society.—As development proceeds, the shrubs and
6

The form found here and at Delhi corresponds more closely to this species than
to B. pumila, but its characters are intermediate.

herbs gradually are superseded b y a growth of L a r i x .

T h i s society

h a s b e e n m u c h disturbed b y lumbering, and a large part of the
original area has been cleared.

B u t there is good evidence to show

that the part of the basin filled with peat formerly supported a dense
covering of t a m a r a c k s .

W h e r e best developed and least disturbed,

it shows an undergrowth of V a c c i n i u m corymbosum, Aronia nigra,
etc.

A s the other species are practically the same as at the lake to

b e described next, they need not b e enumerated here.
with most of the areas studied,
sphagnum is worthy of note.

I n contrast

the almost complete absence of

I t is also important that the absence

of any gradation between the forest societies of the upland and of the
bog b e kept in mind.
O n this lake, then, there are two divergent series of plant societies.
Starting with practically the same species, the one series leads us
on mineral soil through willows, maples, and elms to the oaks of the
surrounding forests; the other, owing to the development of a

floating

substratum, involves a very different set of shrubs and ends with the
tamarack.

T h e former series therefore more closely approximates

the climatic type, while the latter is dependent upon edaphic factors.
FIRST SISTER L A K E .
T h i s lake and its accompanying bog are located three miles west
of A n n A r b o r in a glacial drainage valley.

I t s origin is probably

connected with the melting of a m a s s of stagnant
a b a n d o n m e n t of the valley b y glacial drainage.
and underlying soil is a sandy gravel.

ice after

the

T h e surrounding

At least a part of the western

side presents an original t a m a r a c k bog vegetation, and it is particu­
larly interesting in showing the results of competition between bog
plants and those of other habitats (fig. 6).

T h e vegetation in general

presents a different phase of the bog societies, as compared

with

W e s t L a k e . . Especially to b e noted are the dominance of cassandra
and sphagnum in the shrub zone, the absence of cattails and swamp
loosestrife as important m e m b e r s of the outer margin.

T h e tamarack

zone is also raised somewhat more above the water level.
Aquatics.—With

the exception of the shallow-water forms,

lake is almost free of higher vegetation.
P.

zosteraefolius

occur

sparingly.

About

the

P o t a m o g e t o n lucens and
the

margin,

however,

N y m p h a e a advena is o f great importance.
tinuous zone 10 to 25 feet (3-7.5

m

I t forms an almost con­

) in width.

tuberosa a n d B r a s e n i a purpurea occur.

P a t c h e s of Castalia

T h i s arrangement in groups

seems to b e connected with their rapid multiplication b y rhizomes.
T y p h a latifolia occurs in a small area at the north end of the l a k e .
Ceratophyllum
Naias

flexilis

demersum

and

^

^

occur as secondary

species.

Bog-sedge society.—Carex

fili-

formis, C. oligosperma, E l e o c h a r i s
palustris

glaucescens, a n d

Erio­

phorum polystachyon are the pri­
m a r y factors in t h e formation of
this zone. C a r e x riparia has gained
a foothold at the north end o f the
lake, where muskrats

have been

active in destroying the original
sedge zone. Dryopteris thelypteris,
O n o c l e a sensibilis, J u n c u s effusus,
J . canadensis, C o m a r u m palustre,
Salix myrtilloides, D u l i c h i u m arundinaceum,
Bidens

Equisetum

trichosperma

fluviatile,
tenuiloba,

Scale.

Menyanthes t r i f o l i a t a , V i o l a
blanda, a n d E r i o p h o r u m

FIG. 6.—-First Sister Lake.

virgini-

cum occur as accessory plants.

T h e great majority of these plants

aid in t h e construction o f t h e substratum b y their roots and rootstocks.
H e r e a n d there among t h e sedges occur the forerunners of t h e
shrub society.
sphagnums.

A m o n g t h e very first to gain a foothold are t h e

T h e s e build small tufts

gradually overcome the sedges.

of great compactness, a n d

T h e rootstocks of the cassandra

also send up shoots a n d prepare the way for another vegetation form.
O x y c o c c u s m a c r o c a r p u s a n d O . O x y c o c c u s b o t h occur at intervals
in this zone.,
Cassandra-sphagnum society.—Beyond
tion is no longer arranged zonally.

the sedge zone t h e vegeta­

Conditions have been so m u c h

disturbed that on the western side the area of cassandra-sphagnum
dominance is very irregular.

O n the eastern side this plant society

is in the last stage of its existence.
Chamaedaphne

T h e intimate association of

calyculata, S p h a g n u m

cymbifolium,

dum, and S. recurvum is well illustrated here.
the whole of the territory where they
are decidedly secondary.

S.

subsecun-

T h e plants occupy

flourish.

T h e other species

I t is to b e further noted that in the com­

petition with the sedge species these plants actually override them,
and

only

an

occasional E r i o p h o r u m

virginicum

survives.

The

water-conserving properties of the sphagnum are too well known to
need description here.

B u t the mutual advantage of the cassandra-

sphagnum combination is worthy especial note.

T h e former b y its

numerous b r a n c h e s furnishes a framework which aids in the upbuild­
ing of the moss and in shading.

T h e sphagnum, on the other hand,

furnishes a moist cover in which the conditions for the shrub are most
favorable.
T h e accessory species include the moss, Aulacomnium

palustre;

the herbs, D r o s e r a rotundifolia, Arethusa bulbosa, H a b e n a r i a lacera,
S a r r a c e n i a purpurea, P o g o n i a ophioglossoides, L i m o d o r u m tubero­
sum,

Viola

Scutellaria
Betula

blanda,

Osmunda

galericulata;

pumila,

and

regalis,
the

Campanula

shrubs,

O x y c o c c u s macrocarpus,

aparinoides,

Andromeda
O.

polifolia,

Oxycoccus,

Aronia

occur m a n y

young

nigra, and Ilicioides mucronata.
Tamarack

society.—Among

the

cassandra

t a m a r a c k s , and these b y their development come to overshade the
shrubs and form the tree society of the bog.

T h e dead remnants of

the cassandra mounds m a k e up a large part of the floor beneath them.
The

species of secondary importance

Aronia nigra,

Chamaedaphne

are

Ilicioides

calyculata, O s m u n d a

mucronata,
cinnamomea,

O . regalis, Dryopteris spinulosa intermedia, D . cristata, Polytrichum
juniperinum,

Plagiothecium denticulatum,

Thuidium

recognitum,

A u l a c o m n i u m palustre, M a r c h a n t i a polymorpha, Sphagnum cymbi­
folium, Boletinus porosus, and T h e l e p h o r a intybacea.
T h e t a m a r a c k zone has been m u c h disturbed b y clearing and
burning.

At the present time a large part of the area on the south­

west side is dominated b y other tree species.

S o m e of the plants of

the clearing have spread into the pure t a m a r a c k growth.

Poplar-willow-maple society.—Where

the original conditions have

been disturbed a n d a second growth allowed to come in, Populus
tremuloides, Salix sericea, Salix discolor, a n d A c e r rubrum have
obtained dominance.

W h e r e groups of the more mature

occur there is scarcely any undergrowth.
plants

occur:

Ilicioides mucronata,

poplars

Elsewhere the following

Salix

Bebbiana,

Sambucus

pubens, A m e l a n c h i e r oligocarpa, A r o n i a nigra, R u b u s nigrobaccus,
Cornus stolonifera, a n d R u b u s strigosus.

T h e s e form a dense mixed

association, with b u t slight reference to substratum conditions.
smaller species present are Adicea pumila, O s m u n d a

The

cinnamomea,

R o s a Carolina, O n o c l e a sensibilis, E p i l o b i u m adenocaulon, Spiraea
salicifolia, Dryopteris thelypteris, V e r b e n a hastata,
c a m a r a , Polygonum sagittatum,

S o l a n u m dul­

Spiraea tomentosa, G e u m rivale,

Polygonum hydropiperoides, R i b e s floridum, R i b e s oxyacanthoides,
Rumex

Britannica,

Impatiens

biflora,

Viola

blanda,

Osmunda

regalis.
O n t h e southeast side of the lake a n d on the north, conditions
have b e e n still more interfered with, a n d there is now a mixed growth
of b o g a n d low-ground plants, which represent stages in the decline
of the b o g flora a n d the advent of swamp plants.
are willows and clumps of mountain holly.

T h e tallest forms

F o r convenience only,

the plants m a y b e enumerated together under the following title:

Mixed

low-ground society.—The

dominant

plants are Salix

sericea, S. discolor, S p i r a e a salicifolia, P o a flava, Solidago serotina,
Chamaedaphne

calyculata, O x y c o c c u s macrocarpus, Aster Novae-

Angliae, a n d R o s a Carolina, E p i l o b i u m adenocaulon, Aronia nigra,
Andromeda

polifolia,

Rubus

strigosus,

Dryopteris

thelypteris,

Scutellaria galericulata, J u n c u s effusus, K o e l l i a virginiana, S a m b u c u s
canadensis,

Geum

rivale,

Osmunda

regalis,

Scirpus

cyperinus,

G a l i u m aparine, H o m a l o c e n c h r u s oryzoides, J u n c u s tenuis, Asclepias
incarnata,

Salix

Bebbiana,

Andrewsii,

Lycopus

stolonifera,

Carex

Eupatorium

virginicus,

riparia,

Viola

Osmunda
blanda,

perfoliatum,

Gentiana

cinnamomea,
Sarracenia

Cornus

purpurea,

Dryopteris cristata, D . spinulosa intermedia, and T r i a d e n u m

vir­

gin icum also occur.
T h e last two societies are found upon a b l a c k peat

substratum

which is more thoroughly decayed than in other parts of the b o g .

Acidity tests show that the relative acidity is less t h a n in the case of
the cassandra-sphagnum and t a m a r a c k societies.

T h e soil tempera­

ture also runs somewhat higher as noted elsewhere.
T h e F i r s t Sister L a k e m a y b e said to b e dominated b y three wellm a r k e d bog and two mixed societies in which bog and swamp species
are brought into competition.

T h e result can b e

foretold with considerable certainty.
tation

will

sooner

or

later

T h e bog vege­

b e replaced by

the

swamp species.
BOG NORTH OF D E L H I .
Two

miles north of D e l h i occurs an extensive

bog which was formerly a mile and a quarter (2
long b y a half mile wide (0.8
FIG. 7. —Delhi
bog and .adjacenttopography. Scale
1:95,000 (f inch
= 1 mile).

(fig- 7)'

k

m

k m

)

) at its broadest part

T h e southwestern third has been cleared

and is in part under cultivation.

T h e eastern and

northern parts have been somewhat interfered with
by the cutting of timber, but areas occur which have
been but little disturbed b y these influences.

Near

the eastern margin are two small lakes, the last remnants of the larger
lake which must have occupied this territory in early postglacial
times.

T h e basin is located in a clay moraine of the E r i e ice-lobe,

and probably owes its origin to unequal deposition b y the glacier.
T h e plant societies found about the southeastern l a k e will give an
idea of the whole vegetation (fig. 8).
Aquatic society.—The

a q u a t i c vegetation is represented

wholly b y the yellow water-lily, N y m p h a e a

advena.

forms a broader zone completely encircling the l a k e and
m

from 5 to 10 feet (1.5-3 )
senia

purpurea,

m

width.

Ceratophyllum

almost

This

plant

varying

Accompanying it occur B r a -

demersum,

Lemna

minor,

and

Spirodela polyrhiza.

Typha-cassandra-sphagnum society.—On the floating margin of
the bog substratum occurs a zone which partially encircles the lake.
N e a r its outer edge T y p h a latifolia is the characteristic plant, but in
certain places it is wanting or extends the full width of the zone.
C h a m a e d a p h n e calyculata, S p h a g n u m cymbifolium, S. subsecundum,
S. recurvum, C a r e x filiformis, E r i o p h o r u m polystachyon, and Salix

myrtilloides are t h e most frequent

plants.

T h e accessory species

include C a r e x oligosperma, M e n y a n t h e s trifoliata, C o m a r u m palustre,
T r i a d e n u m virginicum, O s m u n d a regalis, O n o c l e a sensibilis, R u m e x
B r i t a n n i c a , Asclepias

incarnata,

Viola

blanda,

Cicuta

bulbifera,

G a l i u m Aparine, Scutellaria galericulata, R h u s V e r n i x , D u l i c h i u m
arundinaceum,

Oxycoccus

macrocarpus,

H ypnu m

cordifolium,
Schreberi,

Hypnum

Aulacomnium

palustre, a n d M n i u m .

Vaccinium-aronia society.
— F o r m i n g a narrow tran­
sition society between t h e
low shrub zone

j u s t de­

scribed a n d the tree society,
occurs a dense line of tall
shrubs.

The d o m i n a n t

species are V a c c i n i u m corymbosum, Gaylussacia resinosa, A r o n i a nigra, Ilici­
oides

mucronata,

glandulosa,
serotina.
present

and

Betula
Prunus

T h e other species
are A c e r rubrum,

Sc&le ut ftel

°

Y

'f

FIG. 8—Portion of Delhi bog.

Sambucus p u b e n s , O s ­
m u n d a c i n n a m o m e a , Salix discolor, S. B e b b i a n a , Spiraea salicifolia,
I l e x verticillata, R o s a Carolina,

S a r r a c e n i a purpurea,

Andromeda

polifolia, Calamagrostis canadensis, a n d E l e o c h a r i s palustris glaucescens.

T h e s e shrubs border t h e t a m a r a c k s a n d to varying distances

extend b a c k among them.
Tamarack-birch society.—Larix

laricina a n d B e t u l a lutea

must

have m a d e up the great b u l k of the original forest which occupied
this area.

T h e relative abundance of t h e latter h a s probably been

increased b y t h e cutting of the t a m a r a c k .

T h e next most important

tree is A c e r rubrum, which occurs scattered throughout, b u t is espe­
cially a b u n d a n t near the northeast side.

W h e r e isolated trees have

b e e n removed, t h e shrubs which occur among t h e undergrowth have

m a d e a rapid growth.

T h r o u g h o u t the forest area are patches in

which Aronia nigra, V a c c i n i u m corymbosum, and Ilicioides m u c r o n a t a
stand so thickly as to b e almost impenetrable.
been but slightly disturbed

W h e r e the forest has

and the t a m a r a c k s are more or less

scattered, one finds a deep carpet of sphagnum with slender stems
of cassandra, andromeda, and E r i o p h o r u m virginicum rising through
it.

Clusters of S a r r a c e n i a purpurea are common.

T h e other plants

found in this society are T r i e n t a l i s americana, Unifolium canadense,
Coptis

trifolia,

Rumex

Acetosella, R u b u s

strigosus,

Dryopteris

spinulosa intermedia, O s m u n d a cinnamomea, V i o l a blanda, I m p a tiens

biflora,

trichum

Solanum

dulcamara,

Thelephora

intybacea,

Poly-

juniperinum, S a m b u c u s pubens, Agrostis alba, B l e p h a r i

glottis lacera, Cornus candidissima, and Cicuta m a c u l a t a .
Clearing society.—Surrounding

the forest on the east, south, and

west sides is a large area, in part dominated b y sedges and grasses,
and in part b y a typical " s l a s h i n g . "

I t is impossible to characterize

this plant association b y any p a r t i c u l a r species.

All that have been

thus f a r mentioned occur in scattered clusters, the proportions and
dominant plants varying from one locality to another.

T h e notable

facts are that on the east side C a r e x teretiuscula, C. vulpinoidea,
C. riparia, C. filiformis, Scirpus cyperinus, Calamagrostis canadensis,
Aster Novae-Angliae, Eupiatorium

perfoliatum, and Aster junceus

have b e c o m e the most a b u n d a n t forms.

T o the west of the l a k e

these plants are present, b u t the taller shrubs are in control.

Salix

discolor, Cornus stolonifera, Salix B e b b i a n a , S. sericea, a n d m a n y
others already mentioned as occurring among the t a m a r a c k s

are

present.
T h e second l a k e and the more northerly one is bordered b y an
exceedingly narrow

zone of low-growing

species are Decodon verticillatus and
daphne

calyculata,

Carex

riparia,

The

dominant

T y p h a latifolia.

Chamae­

Panicularia

B r o m u s K a l m i i are of secondary importance.
to the water's edge.

plants.

T h e proportion

canadensis,

and

T h e trees come almost

of red maples among

the

t a m a r a c k s and birches is considerably greater than in the vicinity
of the other lake.

Otherwise the tree society is essentially the same.

W e have illustrated, then, in the bogs at W e s t L a k e , F i r s t Sister
L a k e , and Delhi, three stages in the filling of old lake basins.

We

have seen that, although there are minor variations in the species
present, all of the bogs show a series of bog-sedge, shrub, and conifer
societies which are genetically related.
is almost completed.

I n the D e l h i bog the filling

I n the bog about to b e described w e find this

process finished, and what was formerly a ring of bog-sedges sur­
rounding an open

lake has b e c o m e an irregular disk forming the

central plant society of the area.

FIG. 9.—Bog near Oxford, Oakland county.
BOG

NEAR

OXFORD,

OAKLAND

COUNTY.

N e a r the northeast corner of S e c . 31, Oxford T p .
bog (fig. 9) covering about 4.5 acres (1.8 h e c t a r e s ) .

? >

there is a

Although it lies

a few miles beyond the real boundary of the H u r o n R i v e r basin, it
is included because it exhibits a flora somewhat different from the
other areas, and m a y b e considered as a near approach to the type of
bogs occurring farther north.

T h e basin is a depression in

out wash sands and gravels of the interlobate moraine.

the

I t is sur-

m

rounded b y hills 25 to 30 feet (7.5—9 ) in height above the bog level.
D u r i n g wet weather it has a shallow outlet to the southwest.

The

land surrounding it has all been cleared and is now under cultiva­
tion.

As shown b y other timber areas in the vicinity, it is probable

that the original upland t i m b e r was m a d e up in part of Pinus strobus,
Quercus coccinea, and B e t u l a papyrifera.

Bog-sedge society.—Toward

the center of the bog is a considerable

area in which the water level lies j u s t at the surface.

T h e sphagnum

is for the most part submerged, and the dominant plants are C a r e x
oligosperma and Scheuchzeria palustris.
following society are scattered
Bog-shrub society.—While

O c c a s i o n a l plants of the

throughout.

this zone is characterized b y C h a m a e ­

daphne calyculata, S p h a g n u m

cymbifolium, S. recurvum,

and S.

subsecundum, young and dwarfed specimens of the spruce, t a m a r a c k ,
and pine are present in large numbers.

T h e surface formed b y the

sphagnum is exceedingly rough and m a r k e d b y h u m m o c k s .
the depressions E r i o p h o r u m virginicum, E . vaginatum,
polifolia,

S a r r a c e n i a purpurea,

and

Among

Andromeda

O x y c o c c u s macrocarpus

are

abundant.
Tamarack-spruce society.—This

society forms a zone completely

surrounding the shrub society, and is dominated b y trees of L a r i x
laricina and P i c e a M a r i a n a .

O c c a s i o n a l specimens of Pinus Strobus

are found, especially toward the southwest corner, where the sub­
stratum is somewhat higher than elsewhere.

B e n e a t h the trees is

an

especially V a c c i n i u m

almost

corymbosum

impenetrable * tangle
and

of shrubs,

Ilicioides mucronata.

tically b a r e of lower vegetation.

T h e substratum is prac­

A n occasional m a t of Aulacomnium

palustre m a y b e found at the tree bases.

T h a t this society will come

into possession of the central bog area is certainly indicated b y the
great numbers of young trees among the bog shrubs.

Willow-sedge society.—As usual in the clearing of the adjacent
land, the larger trees of the bog margin were also removed, and in
their stead has come up a growth of willows.

T h e dominant plants

of this zone are Salix sericea, Cornus stolonifera, Spiraea salicifolia,
Salix discolor, C a r e x riparia, and C. stipata.

Associated with these

plants are S a m b u c u s pubens, Salix nigra, I r i s versicolor, Populus
monilifera, Dryopteris spinulosa intermedia, O s m u n d a

cinnamomea,

Equisetum

limosum,

Cornus

candidissima,

Aronia

nigra,

Rosa

Carolina, J u n c u s effusus, Calamagrostis canadensis, R u b u s strigosus,
Ilicioides m u c r o n a t a , C o m a r u m

palustre, C a r e x

laria canadensis, and P o a

F o r m i n g a high border about the

t a m a r a c k s and

flava.

spruces are numerous

filiformis,

large plants

Panicu-

of V a c c i n i u m

corymbosum and Ilicioides m u c r o n a t a .
T h e very m a r k e d difference between the vegetation of the central
a n d marginal parts of the bog are worthy of especial note.
former represents the original vegetation of the bog.

The

T h e latter

illustrates most forcibly that under present conditions a very different
set of plants springs up and b e c o m e s dominant, in spite of the fact
that the true bog plants were near at h a n d
occurred.

when

the

clearing

T h i s bog also illustrates that stage in the filling of a

depression immediately following the disappearance of the lake.
I n other bogs near Oxford, D a s y p h o r a fruticosa and Chiogenes
hispidula occur among the shrubby growth.
THE DELHI MUSKEAGS.
I n the bog north of D e l h i which h a s already been described
occur two areas, somewhat to the west of the lakes, which seem to
represent a later stage in the history of a bog than that shown b y the
lakes.

T h e s e areas, if they were found in northern M i c h i g a n , would

be termed " m u s k e a g s . "

T h e y are surrounded b y large t a m a r a c k s ,

and small t a m a r a c k s occur throughout, the smallest specimens toward
the center.

I f the bog at Oxford were to continue its work of filling

until the central society disappeared, we should have a bog a r e a of
m u c h the s a m e appearance.

T h e small t a m a r a c k s stand far apart,

and between them is a most luxuriant growth of cassandra
sphagnum.

and

T h e h u m m o c k s rise between 3 and 4 feet (0.9-1.2™)

above the substratum.

A s one attempts to traverse these areas, he

sinks knee-deep in the long, fibrous, peat moss.
The

total n u m b e r of species is very small, and includes, besides

those already mentioned, A n d r o m e d a polifolia, S a r r a c e n i a purpurea,
O x y c o c c u s macrocarpus, and a few specimens of V a c c i n i u m corym­
bosum.
BOG ON CARPENTER ROAD
T h i s bog is situated in the S W . % S e c . 36, A n n A r b o r T p .

Its

basin is a small depression in the glacial moraine occupying about

one-tenth of an acre (fig. id).

O n the south, west, and north sides
m

it is bordered by clay hills which rise 25 to 40 feet ( 7 . 5 - 1 2 ) above
the bog level.

T h e vegetation of the hills is dominated by Q u e r c u s

velutina, Q . alba, and Q . rubra.

W i t h these trees occur H i c o r i a

ovata, H a m a m e l i s virginiana, etc.
O n the north side the upland h a s been cleared, and the land is
now

under

tion.

cultiva­

F r o m time to

time t a m a r a c k s have
been

removed

from

the bog, until at the
present time only the
central area remains
to indicate the origi­
nal covering. A c c o m ­
panying the
there has

clearing

grown

up

about the t a m a r a c k s
^

FIG. 10.—Bog on Carpenter road.

u s u

^l

shrubs
trees.

thicket of
and

young

As elsewhere, the peat is more thoroughly decayed and the

substratum level somewhat lower about the margin than toward the
center.

T h i s fact is of importance in differentiating the willow-sedge

society.
Tamarack society.—This

society is dominated b y the group of

rather m a t u r e t a m a r a c k s .

T h e substratum has the characteristic

h u m m o c k y surface, m a r k e d b y large exposed roots, common to such
areas.

I t is overlaid b y a loose covering of vegetable matter, m a d e

up principally of t a m a r a c k needles.

T h e undergrowth is sparse,

b u t most of the bog shrubs and herbs are represented.

T h e more

important species are C h a m a e d a p h n e calyculata, S p h a g n u m cymbifolium,

S. recurvum,

S. subsecundum,

and L y c o p u s virginicus.

Eriophorum

virginianum,

A very noticeable growth about the bases

of most of the shrubs is produced by the fungus, T h e l e p h o r a intybacea.

T h e mycelium in developing its sporophores rises about
cm

the stems, frequently to a height of a foot ( 2 5 ) .

F r o m the cylinder

thus formed, irregular fan-shaped pilei are developed, which gives

the appearance of an elongated brown rosette about the stem bases.
Clitocybe l a c c a t a and Boletinus porosus are also abundant in the
autumn.

The

partially

decayed

stumps

bear

Peltigera

canina.

O t h e r species occur in this area, but reach their dominance in the
next society.
Poplar-maple society.—Here

are brought together the

remnant

of the bog species, and those more characteristic of swamps and
clearings.

T h e trees are mainly Populus tremuloides, with a scat­

tering of A c e r rubrum.

E l m seedlings occur.

however, m a k e up the bulk of the vegetation.

T h e shrubby plants,
Ilicioides mucronata,

I l e x verticillata, Aronia nigra, and V a c c i n i u m corymbosum

have

almost complete possession, and are struggling with one another for
space.

All these forms send up stems from the underground parts,

so that among them the struggle is largely a m e c h a n i c a l one.

How­

ever, where the red maple overtops them, the factor of shade enters,
and the b l a c k choke-cherry and high-bush blueberry are the most
tolerant.

T h e mountain holly and b l a c k alder prevail elsewhere.

T h e next most important plants are the willows, Salix sericea and
S. discolor.

M i x e d with

these are

Cornus

candidissima,

Rubus

nigrobaccus, R o s a Carolina, Cornus stolonifera, Spiraea salicifolia,
and R u b u s strigosus.
Willow-sedge society.—The

area dominated b y these plants is

covered with water in the spring and during moist weather.

Although

this society is fast being crowded out b y the next preceding, it is
probable that only a small part of that area was ever occupied b y these
plants.

T h e s e plants require a more moist substratum.

T h e domi­

nant species are Salix sericea, C a r e x riparia, C. stipata,
stolonifera, and O s m u n d a c i n n a m o m e a .

Cornus

I n the case of the cinna­

m o n fern found in this bog there is a r e m a r k a b l e development of
aerial roots.

T h e y are about an inch long and extend outward from

the thick rootstock in all directions, forming a dense covering.

The

roots are thickly covered with root-hairs which have been persistent
at least through one winter.
in color.

T h e root-hairs are large and brown

T h e appearance of these rootstocks, as a whole, is very

suggestive of certain tropical tree ferns.

T h e other species present

are R a n u n c u l u s abortivus, Polygonum sagittatum, Cicuta bulbifera,
Prunella vulgaris, R u b u s americanus, R h u s V e r n i x , S o l a n u m dulca-

mara,

I m p a t i e n s biflora, E u p a t o r i u m

perfoliatum,

Calamagrostis

canadensis, Dryopteris thelypteris, D . spinulosa intermedia,

Doel-

lingeria umbellata, L a c t u c a spicata, Coptis trifolia, B o e h m e r i a cylindrica, O n o c l e a sensibilis, M a r c h a n t i a polymorpha, a n d R o s a Carolina.
T h e further development of these societies under present condi­
tions will bring about a complete change.

T h e r e c a n b e no doubt

that t h e poplars and red maples are the coming trees, with elm a
close third.

W h e n these have b e c o m e sufficiently large a n d numer­

ous to overshade t h e shrubs, t h e latter will b e killed out, a n d we
shall have in their place the maple-elm forest c o m m o n to t h e low
grounds.

T h e shrubs, however, are c a p a b l e of persisting for a great

length of time, because of t h e difficulty of tree seedlings obtaining a
start b e n e a t h them.
THE CHELSEA BOG.
O f the bogs which have been subjected to clearing, burning, a n d
ditching, b y far t h e most interesting in this region is located j u s t to
the southeast of the town of Chelsea.

I t covers a n area of about
m

50 acres (20 h e c t a r e s ) , a n d t h e peat is reported to b e 40 feet ( i 2 )
thick at the deepest places.

T h e divisions into societies, as indicated

on the m a p (fig. 1 1 ) , are b a s e d on t h e most general characters of t h e
vegetation.

T h e r e are gradations between all of t h e societies, a n d

these a r e so gradual that it is difficult to determine' definitely t h e
boundaries.

F u r t h e r , owing to t h e tendency of m a n y of t h e shrub

species to form dense local growths b y the development of stems
from underground shoots, t h e smaller associations are very diverse
in different parts of the same society.
Birch-vaccinium society.—This mixed society of bog shrubs occupies
about one-fourth the area of t h e bog.

I t s substratum consists of

peat standing about a foot above the average water level.

The

dominant plants are B e t u l a pumila, V a c c i n i u m corymbosum, R u b u s
frondosus,
aquilinum.

Aronia

nigra,

Vaccinium

J u s t as c o m m o n perhaps,

canadense,

and

Pteridium

b u t of lower growth, a r e

R u b u s hispidus, S p i r a e a salicifolia, S. tomentosa, Aralia hispida,
Chamaedaphne

calyculata, a n d R u m e x Acetosella.

T h e ground

covering, except beneath the dense shade o f the shrubs, is m a d e up
of P o l y t r i c h u m juniperinum.

T h e r e are m a n y small areas of which

this plant n o w holds exclusive control, a n d forms a rich carpet of

green, yellow, and red, depending

upon the season of the year.

W h e r e the moss is disturbed b y the uprooting of plants, the substra­
t u m b e c o m e s exceedingly dry.

T h e moss dies out, and in place of

it there springs up a growth of Cladonia rangiferina, C. pyxidata,

FIG. 11.—Chelsea bog.
C.

gracilis, C. verticillata, C . cristatella, and frequently

admixture of R u m e x Acetosella.

a small

T h e s e plants gradually close over

the surface a n d aid in the conservation of the moisture.

As the

conditions b e c o m e m o r e favorable, the Polytrichum again closes over
the area, driving out the lichens.

About the borders of the shrubs

the Polytrichum is killed out b y the shade.

R u m e x Acetosella is

better fitted to withstand such conditions, and consequently forms
an inner border about each group of shrubs.

W h e r e depressions

occur and are flooded for any length of time, the Polytrichum is
replaced b y E r i o p h o r u m virginicum and Scirpus cyperinus.

Along

the northwestern border R u b u s nigrobaccus is m a k i n g inroads upon
this society.

T o the north of the railroad, however, the most impor­

tant changes are being wrought b y the development of Populus
tremuloides and Q u e r c u s velutina.

Y o u n g trees of the former are

now scattered throughout, while the latter is present in small number.
T h e plants of m i n o r importance are I l e x verticillata, V i b u r n u m lentago,

Ilicioides mucronata,

Amelanchier

Botryapium,

Euthamia

graminifolia, Doellingeria umbellata, B i d e n s trichosperma tenuiloba,
Dulichium arundinaceum, P o a flava, and S p h a g n u m cymbifolium.
Chokeberry society.—Aronia

nigra forms the most dense

exclusive growth that occurs on the bog.

and

Usually the substratum

is somewhat lower and more subject to overflow than in the last
society.

I t would seem from observation that this condition is in

part due to the chokeberry itself.

Owing to its dense growth, it

protects the surface of the peat from drought and favors the processes
of decay.

At the same time it adds very little to the substratum in

the way of debris.

W h e r e it attains its best development it is prac­

tically without undergrowth.

About the borders it is mixed with

V a c c i n i u m corymbosum, B e t u l a pumila, and I l e x verticillata.

Of

the smaller plants, Pteridium aquilinum penetrates to the greatest
distance.

O t h e r species occurring about the borders are mentioned

among the other societies.

Poplar-willow society.—About the borders of the bog, and extend­
ing to a greater or less extent into its interior, is a dense zone composed
of Populus tremuloides, Salix discolor, Quercus velutina,
grandidentata, and S a l i x nigra.
the trembling aspen.

Populus

B y far the most abundant form is

T h e substratum varies from areas well above

the water level to areas which are constantly submerged.

T h e aspen

is also the most important of the plants which are invading the shrub
societies.

I n the relative proportion of the individual species there

is the greatest variation at different places in this border zone.

Of

the more enduring species, Q u e r c u s velutina is the most a b u n d a n t .
T h e other species present are S a l i x B e b b i a n a , S. sericea, S. lucida,

P r u n u s serotina, Q u e r c u s a l b a , Q . m a c r o c a r p a , A c e r rubrum, B e t u l a
lutea,

Amelanchier

salicifolia,

Botryapium,

S. tomentosa,

Viburnum

Corylus

pubescens,

Spiraea

Sambucus

americana,

pubens,

Cornus candidissima, C. stolonifera, C i c u t a m a c u l a t a , Aster lateriflorus, Carduus altissimus, G a l i u m asprellum, O s m u n d a c i n n a m o m e a ,
O. regalis, R a n u n c u l u s pennsylvanicus, Calamagrostis canadensis,
V i o l a blanda, E u t h a m i a graminifolia, B i d e n s frondosa, and Aster
Novae-Angliae.
Sedge society.—On the northeast side of the bog is an area domi­
nated b y sedges.

I n the fall of the year it appears to b e a uniform

a r e a of Scirpus cyperinus, but there are m a n y other species mixed
with it.

T h e substratum

is low and is m a i n l y characterized b y

tussocks formed b y the sedges.

T h r o u g h o u t , occur small clumps

of the willows already mentioned.
species

are

teretiuscula,

Isnardia
C.

palustris,

stipata,

C.

T h e most a b u n d a n t accessory

Calamagrostis

filiformis,

canadensis,

C. fusca,

C.

Carex

oligosperma,

C. riparia, and A u l a c o m n i u m palustris.
The
growth

future flora of this bog appears to be indicated by the rapid
of the poplars,

willows, and

oaks.

T h e few

tamaracks

remaining are approaching maturity and are not being reproduced.
The

m e a n s by which these tree species c o m b a t the shrubs is mainly

by shading, while the latter in the same way interfere with the develop­
m e n t of the tree-seedlings.

T h e time involved in this struggle must

be very great, b u t the ultimate outcome will b e an o a k forest, the
intervening stages being filled in by poplar and willow growths.

If,

however, the decay of the peat b e n e a t h these trees brings the surface
to the water level, the poplar-willow stage will be indefinitely pro­
longed.
GENERAL CONSIDERATION OF T H E BOG FLORA.
Beside the trees mentioned

in the preceding descriptions, note

should be m a d e of the occasional occurrence of the b l a c k ash, F r a x i n u s
nigra, and swamp white oak, Quercus platanoides, in bog areas.

It

frequently happens, when the t a m a r a c k s are cut, that the b l a c k ash
becomes a b u n d a n t , as in the area one-half mile southeast of K a v a naugh L a k e , where it is now associated with U l m u s a m e r i c a n a and
A c e r rubrum.

A n o t h e r example occurs about a mile north of Chelsea

in the N E . % S e c . 1, Sylvan T p .

H e r e in a small area from which

the t a m a r a c k s were removed, F r a x i n u s nigra, Q u e r c u s platanoides,

Fraxinus

a m e r i c a n a , F . pennsylvanica, A c e r rubrum, O s t r y a vir-

giniana, T i l i a a m e r i c a n a , and Liriodendron tulipifera are associated.
The

undergrowth

consists of Solidago p a t u l a ,

S. neglecta, Aster

lateriflorus, M i t e l l a diphylla, E u o n y m u s obovatus, Viola pubescens,
Agrimonia hirsuta,

Cornus

florida,

C. candidissima,

Eupatorium

perfoliatum, R o s a C a r o l i n a , V i b u r n u m L e n t a g o , J u n i p e r u s communis,
a n d S p i r a e a salicifolia.

T h e substratum is almost entirely occupied

by mosses, including H y p n u m

fluitans,

H . Schreberi, H . Blandovii,

H . roseum, T h u i d i u m recognitum, and C l i m a c i u m a m e r i c a n u m .
O n the farm of J a m e s B a r t o n ( S W . % S e c . 2, L y n d o n T p . ) the
b l a c k ash, red m a p l e , and A m e r i c a n elm have replaced a former
growth of t a m a r a c k s a n d b l a c k a s h . '
I n a previous publication (55: p. 403) the writer called attention
to the absence of a genetic relationship between the bog
and

the surrounding

vegetation in southern M i c h i g a n .

plants

T h i s was

explained on the b a s i s that the bog vegetation is a relict of former
climatic conditions;

that it h a s a genetic relationship

with

the

conifer forest formation of northeastern N o r t h A m e r i c a , as shown b y
studies in northern

M i c h i g a n and Pennsylvania, and

that in this

region it h a s been surrounded by a more southern flora whose center
of distribution is the southeastern U n i t e d S t a t e s .

Consequently no

order of succession between the t a m a r a c k and the o a k floras is to
be expected.
W h e n , however, bog areas are cleared or their n o r m a l development
disturbed, such trees as the b l a c k ash, white ash, red maple, and elm
replace the t a m a r a c k , and a definite order of succession is established.
I t was also m a i n t a i n e d that present bog habitats are continuations
of similar h a b i t a t s which c a m e into existence when a colder climate
prevailed than at present.

M o r e recent observations tend to confirm

a n d strengthen this statement.
T h e dominance of bog and swamp plants respectively in adjoining
areas is to b e explained largely by the time when the areas c a m e to
support their present ground vegetation.

I f the h a b i t a t h a s existed

undisturbed since the time when a colder climate prevailed, the bog
plants will b e dominant.

I f it c a m e into existence in recent times,

or h a s been disturbed, it will b e dominated b y swamp species.
(To be concluded.)

THE

BOGS AND B O G FLORA

OF T H E HURON

RIVER

VALLEY.
EDGAR

NELSON

TRANSEAU.

(WITH SIXTEEN FIGURES)
[Concluded
IV.

from p. 448.]

The ecological characteristics of the bog flora and their causes.
T h e plants occurring in the b o g h a b i t a t a r e almost all perennials.

I n the case o f the herbaceous vegetation, the winter is passed b y
m e a n s of subterranean rootstocks.
a n d in p a r t deciduous.

T h e shrubs a r e in p a r t evergreen

T h e t a m a r a c k s a n d the two birches a r e

deciduous, a n d the b l a c k spruce a n d pine a r e evergreen.
M o s t o f the herbaceous a n d shrubby forms multiply abundantly
by vegetative shoots of one form or another.

T h e length of the

underground stems of the shrubs is proverbial, b u t is best appreciated
by one who h a s attempted to dig up one of t h e m entire.

I n con­

nection with the competition between species for s p a c e in the habitat,
this is of the greatest importance.

A luxuriant growth of cassandra

furnishes the most :favorable situation for the development of sphag­
n u m in this vicinity.

I t s profuse b r a n c h i n g affords a framework

for the upbuilding of the sphagnous layer, its shade properties do not
interfere with the photosynthetic work of the moss, and it protects
it from the drying effects of wind a n d direct insolation.

W h e r e such

associations occur, the difficulties presented for the germination for
most seeds, a n d the efficiency with which competition is combated,
are evidenced b y the fact thal^among the tree species only the t a m a ­
r a c k , spruce, a n d pine a r e successful invaders.

AH o f these plants

send out adventitious roots from the stems a n d branches, a n d so keep
p a c e with the upward development of the moss.

T h e absence of

poplars, willows, red maples, a n d elms in such undisturbed situations
must b e in p a r t attributed to the completeness with which such terri­
tory is controlled by the cassandra-sphagnum association.
ECOLOGICAL ANATOMY.
Aside from the purely a q u a t i c forms which have received m u c h
Botanical Gazette, vol. 41.]

[17

ecological attention, it is of interest to look at the a n a t o m i c a l char­
acteristics of certain of these plants.
Eriophorum virginicum m a y b e taken as a type of this group, a n d
also of t h e sedge zone vegetation in general.

T h e culm is very

slender a n d erect, leaves flat, a n d very narrow, perennial b y root
stocks.

Stem:

epidermis very thick-walled a n d cuticularized.

development proceeds, certain radial rows of the primary

As

cortex

cells have their walls thickened, a n d served to connect the tissues of
the central cylinder with those of the three or four outer layers o f
hypodermal cells which also become thick-walled.

Between these

radial groups of cells lysigenic a i r cavities are formed.

Root: epi­

dermal cells in part thin-walled a n d in p a r t secondarily thickened,
no definite arrangement of the thick-walled cells apparent;

internal

structures closely resemble those of t h e stem; n o mycorhiza present.
Leaf: outer epidermal cell walls strongly thickened a n d cuticularized,
radial a n d inner walls less s o ; lysigenic a i r spaces traverse t h e leaf
longitudinally;

a very thick layer of stereome adjoins the leptome,

decreasing to o n e or two cell layers on the hadrome side of the
bundle;

chloroplasts massed among the outer layers of the cortex, b u t

occur throughout.

Sarracenia
pitchers.

purpurea.—Well

known

for its insect-capturing

Stem: epidermis a n d first hypodermal layer thick-walled;

lysigenic a i r cavities throughout

pith a n d cortex;

resin

deposits

confined to the epidermis a n d one or two hypodermal cell layers, b u t
where wounded heavy deposits of resin take place in the exposed
a n d underlying cells.

Root: cell walls firm, resinous bodies present

throughout, b u t especially prominent in the two outer cortical layers,
in which t h e cell walls a r e also strongly thickened.

Leaf: epidermis

thick-walled a n d slightly cuticularized;

stomata on both sides of

the l a m i n a , with

cuticularized

protuberant;

guard cells strongly

resinous deposits throughout;

a n d slightly

inner face of l a m i n a

with strong downward pointing bristles.
Oxycoccus macrocarpus.—Stem: pith thick-walled, with resinous
bodies; a thick layer of broad-celled b a s t forms a complete cylinder
within
without

the epidermis.

Leaf:

margins

revolute,

upper

epidermis

stomata, heavily cuticularized, radial walls thick,

wavy;

hypodermal collenchyma of two or three cell layers on leptome side

of midvein, one or two cell layers on t h e side o f the hadrome, develop­
ment of the stereome cells also smaller on hadrome side;

palisade

of two cell layers; lower epidermis covered with wax, especially at
the stomata, guard cells slightly elevated.

M y c o r h i z a present in the

larger roots, wanting in t h e hairlike branches, no root hairs.
Andromeda polifolia.—Leaf:

margins revolute, upper epidermal

cells thick-walled, radial walls undulate, n o s t o m a t a ; lower epidermis
supplied with unicellular short stiff hairs, a n d covered with w a x ,
stomata slightly protuberant,

strongly cuticularized beneath m i d-

vein; palisade of three layers of long narrow cells; stereome strongly
developed above a n d below vascular b u n d l e ;

on the ventral side

this adjoining three layers of large thin-walled a i r cells and a onelayered hypoderma.

Root: resinous deposits throughout, no mycor-

hizal fungi found.
Chamaedaphne calyculata.—Lea}: margin slightly revolute, epi­
dermis thick-walled, heavily cuticularized, cuticle rough, n o s t o m a t a
on upper surface; ventral epidermis covered b y shield-shaped multi­
cellular hairs, a n d a deposit o f w a x ; cuticle unusually

thickened

beneath the midvein, guard cells sunken, subsidiary cells protuberantr
dalisade tissue o f four or five layers.

Root: inner a n d radial walls

thickened, cortical tissues thick-walled;

resin deposits in vascular

bundle a n d cortex; no mycorhizal fungi found.
Chiogenes

hispidula.—Leaf:

margin

revolute, epidermal

walls

very thick, cuticle present, papillate, palisade not strongly developed;
mesophyll cells in part thick-walled a n d in part thin-walled;
bodies in the epidermis; s t o m a t a slightly protuberant.

resinous

Stem: resin

present in cortex; mycorhizal fungi in t h e epidermis o f t h e s m a l l e ;
roots a n d throughout the cortex of t h e larger.
Vaccinium corymbosum.—Leaf:

cuticle present, epidermal walls

not thickened, palisade of one layer, mesophyll tissues with resinous
bodies, cuticle o f ventral surface papillate;
hairs on lower epidermis few on upper;

a b u n d a n t unicellular

leptome side of mid-vein

adjoined b y three layers of stereome a n d two or three layers o f hypo­
dermal collenchyma, on the hadrome side reduced to two of stereome
a n d two of collenchyma, cuticular papilli usually developed beneath
the midvein a n d at edge o f leaf.
mycorhiza present.

Root: cortical tissue with resin,

N o resin deposits found in stem.

Salix

1

sericea. —Lea):

cuticularized;

upper

epidermal

cells

small,

strongly

mesophyll compact, palisade of two layers o f long

narrow cells; stomata on under surface, guard cells sunken beneath
the slightly protuberant

companion cells;

hypoderma

o f five- or

six-cell layers on h a d r o m e side, a n d eight layers on leptome side o f
midvein.

Root:

resinous

bodies

present in medullary

rays a n d

cortex, the latter consisting o f thick-walled cells; no mycorhiza.
Ledum

groenlandicum.—Leaf:

upper

epidermis

rugose,

with

scattered unicellular hairs, margins strongly revolute, cuticle present,
cell walls thickened, the radial walls being broadly undulate;

lower

epidermis covered with a thick cuticle a n d a felt of long multicellular
a n d short unicellular hairs, glandular hairs usually present near the
small veins, stomata protuberant;
of broadly oblong cells;

palisade of three or four layers

beneath vascular tissue of midvein a n d

between the mestome bundles occur large a i r cells which m a y form
lysigenic a i r cavities in the older leaves.
Larix

laricina.—Leaf:

bifacial,

Root: mycorhizal.

deciduous;

epidermis

thick-

walled, slightly cuticularized, guard cells sunken beneath the com­
panion cells;

palisade tissue developed toward the dorsal surface,

two layers thick showing a radial tendency, stereome reduced to a
few cells beneath the leptome; two resin ducts near edges of leaf.
Root: composed o f mycorhiza, resinous deposits throughout, cortical
tissues early destroyed b y fungus.

W h e n grown in culture solutions

a n d well aerated, n o r m a l roots with root hairs a r e produced.

Picea Mariana.—Plants

in bogs a r e stunted.

cells thick-walled, cuticle present,

Leaf: epidermal

guard cells sunk beneath the

companion cells; mesophyll cells compact, of a more or less radial
palisade type.

Root: mycorhizal, resin deposits throughout, cortical

tissues destroyed b y fungus.

N o r m a l roots a r e developed under

culture conditions.
Pinus Strobus.—Plants
shorter and thicker.

very m u c h stunted in the bogs, leaves

Leaf,

epidermal walls so greatly thickened

that scarcely a lumen remains, beneath this a hypodermal layer of
thick-walled cells; mesophyll cells compact a n d of t h e usual lobate
type.

Root:

hyphae;

mycorhizal, cortical tissues traversed b y the fungus

resinous deposits throughout.

Stem: annual rings narrow

dan

distorted,

resin

bodies throughout

cortex and

meristematic

tissues of the wood.
T o summarize these characteristics, it is evident ( i ) that epidermal
and hypodermal tissues are thick-walled; (2) that for the conserva­
tion of water these are reinforced outwardly b y a heavy cuticle, by
coverings of wax and air containing hairs; (3) that resinous bodies are
found in the roots and leaves of m a n y of the plants;

(4) that there

is a general reduction in the size of the leaves, and that these are
frequently revolute-margined;
formly developed;

(5) that palisade tissue is quite uni­

(6) that mycorhizal fungi are present in the

roots of most of the plants; (7) that, when compared with the xerophytes of dry sand plains (25, 6), they show a similarity in respect
to the reduction in size of the foliage, in the development of external
protective coverings of the sub-aerial parts, and in the presence of
palisade tissues, but are very different in the matter of root develop­
ment and character of root structures.
To

account for the peculiarities of the bog vegetation various

theories have been brought forward.

K I H L M A N (28), in accounting

for the xerophilous character of the plants of arctic swamps, which
include several species c o m m o n to American bogs, lays stress upon
two factors: (1) the low temperature of the moist substratum, and
(2) the presence of drying winds.

T h e former influences the plants

by decreasing the power of absorption, the latter increases the rate
of transpiration.

T h e plants of such habitats must therefore b e

protected against the loss of water by the subaerial parts.
SCHIMPER (44, p. n ) in classifying the natural habitats in which
xerophytes occur mentions among others " p e a t bogs, because of the
humous acids in the soil."

O n page 18 he says:

The xerophilous character of the vegetation of peat moors has hitherto been
considered an incomprehensible anomaly, and yet the rich supply of humous
acids in the soil furnishes a condition for its occurrence as comprehensible as it
is necessary. The presence of Scotch pine and heather on both dry sand and on
wet peat is thus not more remarkable than is that of Ledum palustre, Vaccinium
uliginosum, and other peat-plants on the cold dry soil in the polar zones.
F u r t h e r (p. 124) the statement occurs that " o n the very acid h u m u s
of moors the vegetation assumes a decidedly xerophilous character,
because the humous acids impede the absorption of water b y the

roots."

However, in describing the arctic vegetation (44,

pp. 1 1 ,

715), he follows the suggestion of K I H L M A N , a conclusion to which
he h a d come independently.

G A N O N G (16) also accepts K I H L M A N ' S

explanation for the xerophilous nature

of the raised-bog flora of

New Brunswick.
I n the study of the structural adaptations of these plants and the
causes of their occurrence in bog areas, several questions arise.
these two factors, cold substratum
xerophily ?

Are

and acidity, efficient causes of

D o they act, in the case of the bogs of this region, with

sufficient strength to cause xerophilous modifications in the plants
there found, or to permit the growth of only such forms as are xero­
philous ?
The

last

question

may

b e answered

from

field

observations.

T h e y indicate that most low-ground plants grow quite as well on the
bog substratum as on the ordinary swamp soils, and that the swamp
species of this vicinity m a y all b e found at one place or another grow­
ing on bog soils.

I t would seem that here the bog substratum is no

more efficient as a selective agent than are the swamp soils.
T h e only cases which have c o m e under m y observation in south­
ern M i c h i g a n which will throw light upon the question of the effecttiveness of the temperatures and acidity in the production of xero­
7

philous adaptations is in the case of Picea Mariana
Strobus.

and Pinus

T h e s e two plants both show reduced size of stem

and

leaf, in the Oxford bog, when compared with plants growing on the
margin.

B u t to what extent this m a y b e due to sterility of the

bog substratum rather than to temperature and acidity is indeter­
m i n a b l e at this time.
EXPERIMENTS.
To
and

answer the question of the efficiency of a cold

soil acidity to produce

substratum

xerophily, experiments have been

progress for approximately two years.

in

T h e difficulties in the way

of experimentation along these lines are numerous.

T h e m e a n s for

controlling soil temperatures in bodies of soil sufficiently large for
experimentation with the larger bog plants are practically beyond
the possibility of a university laboratory.

W h e n it is further realized

7 T h e so-called P. brevifolia Pk. This form is certainly no more deserving of a
distinctive name^than is the bog form of the white pine.

that the experiments should extend over a series of years in the
case of the shrubby forms, the problem b e c o m e s still more com­
plicated.

Cold bog water

W a r m nutrient solution

Cold nutrient solution

W a r m bog water

FIG. 12.—Average plants from the several cultures of Indian corn.
graphs.

From photo­

I n order to test the relative effects of humous acids (of the con­
centration found in the bogs of this vicinity) and low substratum
temperatures, experiments were made in the form of water cultures

and with a peat substratum.

All of t h e bog water used was brought

to t h e plant house from t h e F i r s t Sister L a k e .

T h e acidity of t h e

water varied from .0005 t o .0023 n o r m a l acid, as measured b y n /100
KOH

solution.

W A T E R C U L T U R E S . — ( I ) T h e plants were grown in four-liter battery
j a r s covered with a plaster of P a r i s plate, having five one-inch open­
ings for the passage of the' plants a n d one of smaller size for a ther­
mometer.

F o u r such j a r s were employed in each experiment, two

containing a 0.2 p e r cent. K n o p ' s solution, a n d the others bog water.
One

of each was further maintained at a lower temperature.

The

cooling was accomplished b y passing t a p water through 15 feet of
quarter-inch

m

(4.5 X7

m m

)

glass tubing, arranged in a coil within

the j a r , somewhat below the surface of t h e liquid.

T h e sides a n d

b o t t o m s of t h e j a r s were covered with b l a c k paper, a n d those which
were to b e cooled were further
sphagnum.

surrounded b y white paper a n d

D a i l y readings of the temperatures of the air, warm-water

solutions a n d cold-water solutions during t h e warmest period of t h e
day were recorded.

I n this way t h e m a x i m u m differences between

substrata a n d a i r were obtained.

A s these temperatures were n o t

constant they exaggerate, to a slight degree, t h e average differences
in temperature.
comparable:

T h u s , four conditions were obtained which a r e

(1) warm nutrient solution (temperature approximat­

ing that o f t h e a i r o f t h e plant-house), (2)

warm b o g solution, (3)

cold nutrient solution, a n d (4) cold b o g solution.
Fig.
The

12 shows t h e results of one of these experiments with corn.

photograph was taken eighteen days after t h e experiment w a s

started.

W h e n t h e cultures were set up, t h e plumule h a d developed
c m

to a length of 2 inches ( 5 ) .

T h e air temperatures during the period
0

of experimentation averaged 18.8 C , that of the warm cultures 1 8 . 7
C,

0

0

a n d of t h e cold cultures 10.8 C .
I t is to b e noted that under these conditions t h e best growth of

the leaves a n d roots occurred in t h e b o g water.

B u t a reduction of

8° in t h e substratum temperatures caused a diminution in t h e devel­
opment of b o t h leaves a n d roots; t h e plants in t h e nutrient solution
and t h e b o g water being equally affected.

W h e n all of t h e plants

h a d developed five leaves, it was noted that in t h e case of t h e cold
cultures t h e two lower leaves h a d withered.

T h i s experiment was

repeated with corn, white lupine, a n d b e a n under similar conditions,
with similar results.

T h e greater development of roots in the case

of t h e warm bog water m a y b e due to t h e presence of a poison in very
minute quantities; b u t this I have been unable to prove.
(2) A third culture was then m a d e in which five plants of corn
were grown in each of t h e four water culture conditions, a n d in
addition in four similar conditions, using a mixture of sphagnum
and peat for t h e substratum.

W o o d e n boxes 2 feet long, 1 foot
c m

wide a n d a half foot deep ( 6 o X 3 o X i 5 ) were constructed, a n d two
were lined with galvanized iron.

T h e bottoms of t h e unlined•ones

were perforated so as to allow of easy drainage.
served for t h e undrained conditions.

T h e lined boxes

F u r t h e r , in one of t h e drained
m

a n d in one of t h e undrained boxes, 40 feet ( i 2 ) of glass tubing,
bent into coils, t h e j o i n t s being connected b y rubber

tubing,were

arranged so that a constant flow o f cold water, for lowering t h e
temperature, could b e maintained.

T h e water level in t h e undrained

bog substratum was kept j u s t below t h e surface.

T h e water was

obtained from t h e bog at F i r s t Sister L a k e , b u t occasionally all were
watered with distilled water.
practically t h e s a m e .

, T h e amount added to each b o x was

I n order to keep t h e solutions in t h e water

culture j a r s a t t h e same acidity a s in t h e undrained boxes, t h e water
was siphoned off and transferred once a week.

C a r e was taken in this

transfer to aerate t h e water in t h e boxes as little as possible, while
that of t h e j a r s was aerated at irregular intervals b y m e a n s of a
bulb.

T h e r e were thus produced eight conditions, in which it was

possible to test t h e effect o f t h e acidity o f t h e b o g water, of aeration
(drainage) o f t h e substratum, a n d of low temperatures.

A s a result,

it was found that t h e growth of roots a n d leaves was best in t h e
w a r m b o g water, in t h e w a r m nutrient solution, a n d in t h e drained
warm peat substratum.

R e d u c t i o n in size of both roots a n d leaves

occurred in t h e cold b o g a n d nutrient solutions, a n d in t h e drained
cold a n d undrained w a r m a n d cold peat substrata.

B u t the plants

in t h e undrained cold peat showed t h e most m a r k e d reduction in
size.

T h e conclusion was reached (1) that humous acids (acidity

varying from .0005 to .0023 n o r m a l acid) have n o effect upon corn
in t h e m a t t e r of leaf a n d root development; (2) that low temperature
a n d l a c k of aeration of t h e substratum both cause reduction in size;

and (3) that when low temperature is combined with poor aeration
the effect is very marked.
T h i s experiment was repeated with peas, and the same result
was obtained, although the effects were not so m a r k e d (fig. 1 3 ) .
T h e roots in the undrained substrata were killed when they attained
a depth of a half inch ( i 2

m m

) below the surface.

(3) I n order to test the effects of drainage and of low temperature
on bog species, another set of cultures in peat-sphagnum
was m a d e .

substrata

T h e apparatus used consisted of two flower-pots and

two glass dishes aproximately a foot in diameter and

three inches

FIG. 13.—Effect of the several conditions upon the development of pea seedlings.
All are average specimens. From photographs.
cm

deep ( 3 o X 7 . 5 ) .

A flower-pot and a glass dish were kept cool by

passing cold water through fifteen feet of glass tubing arranged in
coils, a s in previous experiments.

T h r e e species were tested in these

conditions:

laricina,

two-year-old

Prunella vulgaris.

Larix

acetosella,

and

T h e first cultures were made in the spring of

1903 with the R u m e x and Prunella.
about eighteen degrees.
ten degrees lower.

Rumex

T h e air temperature averaged

T h e cold substratum was maintained about

I n the case of R u m e x it was found that

the

largest leaves were produced in the drained peat-sphagnum substra­
tum.

L a c k of drainage and low temperature both caused a reduction

in leaf area, and when combined produced leaves which were less
than half as large as those of the drained warm

substratum.

T h e Prunella under the s a m e conditions showed the same results.

C. Undrained w a r m bog substratum

E . D r y sand

D. Undrained cold bog substratum

F.

FIG. 14.—A, B, C, D, E, camera drawings of leaf sections resulting from cultures
in the five conditions named. X 1 3 5 . F, diagrams showing average length and
breadth of leaves.

Fifteen plants were grown in each condition.
experiment each h a d produced

At the end of the

six to eight mature leaves.

leaves were measured as to length and breadth.

The

A n index was

obtained b y multiplying these two numbers together and averaging
for each culture.

Following are the indexes of leaf area thus derived:

drained warm substratum

1268.3, drained

cold 682.6,

undrained

warm 518.5, undrained cold 421.8.
I n the spring of 1904 the experiment with R u m e x was repeated.
The

results correspond with those of the preceding year.

The

structure of the leaves, resulting in the several cultures, was investi­
gated, and found to show m a r k e d variations (s6).
the

cross-sections and

average leaf

Fig. 14 represents

areas produced

leaves being measured in each case).

(seventy-five

W h e n grown on a

warm

drained substratum, the leaves are large, and the cells are exceedingly
loose and turgid.

T h e epidermis is composed of large thin-walled

cells, having a thin cuticle outside.

T h e mesophyll consists of a

single layer of palisade and three layers of spongy tissue.
bodies are present.

T h e plants grown in the undrained

N o resin

substratum,

whose temperature was reduced about 8° C. below that of the air,
show m a r k e d xerophilous characters.

T h e leaf is reduced in area,

increased in relative thickness, and the margins b e c o m e revolute;
the epidermal cells are smaller and outwardly thick-walled; a wellm a r k e d cuticle is present; the mesophyll is very compact and m a d e
up of two or three layers of well-developed palisade cells and three
layers of spongy tissue; and in the epidermal cells and those adja­
cent to the bundles there are m a r k e d accumulations of resinous
bodies.
F o r the purpose of comparison, a corresponding set of plants
were grown on sand kept just sufficiently moist to allow the plants
to live.

As will be seen in fig. 14, the xerophily is not more m a r k e d

than that of the undrained cold bog substratum.

Fig. 1 5 shows the

relative appearance of the plants produced b y the different con­
ditions.
I n the case of the plants grown in the undrained warm and the
drained cold substrata, these same effects were noticeable, but to a
less m a r k e d degree.

T h a t , in the case of the undrained

cultures,

these effects are not due to the acidity of the bog water is shown by

the fact that plants grown in bog-water cultures develop normally.
T h e light conditions in the several cultures were the same, direct
sunlight being avoided b y a cloth screen.

I t is evident that in this

case there is no response to strong light in the development of the
palisade tissue (49).

I t would seem rather to b e a response called

forth b y a reduced transpiration current (44, p. 7 ) .

As to function,

it m a y aid in the transfer of food materials as suggested b y H A B E R LANDT (20, p . 260).

FIG. 15.—Average plants showing effect of surrounding
photographs.

conditions.

From

T h i s plant proved t o b e t h e most plastic of all o f the species used
in the experimentation, and was the only one which showed marked
variation in the internal structures.

Ecologically the results indicate

(1) that an undrained peat substratum m a y cause xerophilous struc­
tures, but that the effect is to b e correlated with l a c k of aeration of
the substratum rather t h a n with the acidity; (2) that the same effect
m a y b e induced b y lowering the substratum temperature

(the air

temperature r e m a i n i n g the s a m e ) , and thus impeding the rate of

root growth and absorption;

(3) that a cold undrained

bog sub­

stratum is analogous to a dry warm soil in that both produce physio­
logical drought;

(4) that resin bodies, which are characteristic of \he

bog plants, m a y b e produced by this environment in a plant which
under favorable conditions is without them.

Drained w a r m bog substratum

Drained cold

Undrained cold

Undrained w a r m

FIG. 16.—Relative effects of drainage and reduced substratum temperature, on
Larix. From photographs.
T h e seedling t a m a r a c k s , ten o f which were cultivated in each of
the four conditions j u s t described for the R u m e x , also showed con­
siderable variation.

T h e i r relative development at the end of forty-

four days is shown in fig. 16.

T h e leaves of the drained warm sub­

stratum have an average length of 12.6
of the undrained warm n . 4

m m

m

m

, of the drained cold 10

, and of the undrained cold 6.3

m

m

m

m

,
.

Internally, the leaves show a reduction in the intercellular spaces and
in the size of the cells in the case of the plants grown on the undrained
cold substratum, when compared with those of the warm

drained

condition.
(4) I n another series of experiments with plants of L a r i x four to
five years old practically the s a m e results were obtained.

There

were the greatest n u m b e r and length of leaves and b r a n c h e s produced
in the case of the drained w a r m substratum.

T h e smallest and

shortest leaves and b r a n c h e s were produced b y the undrained cold
substratum.

Experiments with Ledum Groenlandicum, Chamaedaphne caly­
culata, Andromeda Polifolia, Betula pumila, and Oxycoccus macrocarpus have failed to produce satisfactory results.

T h i s is believed

to b e due to the shortness of t h e . t i m e under which they were under
cultivation.

T h e plants were brought from the bogs in the late

autumn and placed in cold frames over the winter.

About

the

beginning of M a r c h they were brought into the greenhouse,

and

after a few days planted in the warm and cold, drained and undrained
boxes, previously described.

T h e y have grown vigorously, but the

differences noticeable m a y not b e correlated with the four conditions.
The

cranberry has shown the greatest amount of plasticity, but this

could not in all cases b e correlated with the environment.

I f these

plants can b e kept under known conditions for two or more years, it
is probable that they will yield valuable results.
(5)

I n order to test the effect of mineral soils, and the ability to

withstand the presence of large quantities of calcium and magnesium,
specimens of andromeda, cassandra, and cranberry were grown in
sandy l o a m and sand.
The

T h e y were watered daily with tap water.

cultures were started in the autumn of 1902, and

vigorous vegetative shoots during the s u m m e r of 1903.

produced

T h e y failed

to bloom, however, and although they are growing well at this time
( J u n e 1904), they have again failed to bloom.
due

to the w a r m

plant-house

conditions.

T h i s m a y b e in part
The

experiment

was

originally started to observe the changes in the roots, and in so far

have been of value.

I n a sphagnum substratum all three of the

plants produced hairlike roots which attain a length of 5 - 7

c m

.

The

roots are commonly several times branched, very little difference in
thickness being shown b y t h e several branches.

W h e n grown in

sand the roots are still slender, b u t t h e frequency of branching is
enormously increased.
the growing tip.
develops.

Usually the branching occurs j u s t b a c k of

T h e older root ceases growth as the lateral root

T h e b r a n c h continues for 2-3

m

m

, and it also stops growth

with the formation of a second lateral root.

T h e result of this pro­

cess is a zigzag root showing root branches which have been succes­
sively the m a i n root tip.

Occasionally several lateral roots develop

and the m a i n axis is divided.
(6) T h e statement that waters containing lime and other mineral
salts are unfavorable to the growth of sphagnum h a s gained wide
circulation in ecological literature.

B e c a u s e of the great abundance

of lime and magnesia in the waters of this vicinity, I was led to test
this fact b y growing the sphagnum in tap water a n d solutions of
CaC0 .

I n one experiment t h e water in a battery j a r was saturated

3

with C O , C a C 0
2

3

was added in excess, and the C 0

allowed to pass through the water for thirty minutes.

2

was again

I n this solu­

tion sphagnum was placed, and it h a s been growing vigorously for
three months, although watered daily with water containing over 100
parts of C a C 0

3

to the million.

S o m e of the sphagnum cultures have

been running for ten months, and show no signs of deterioration.
W h e t h e r the sphagnum of this vicinity h a s b e c o m e accustomed to
the presence of lime, owing to the nature of the soil waters, or whether
sphagnum is generally able to withstand such conditions, remains to
b e proved.

Since the above experiments were performed, I have

found a n account of somewhat similar experiments b y W E B E R (58),
the results of which are the same.

I t would seem, therefore, that the

presence or absence of sphagnum is not to b e correlated with the
presence or absence of lime.
(7) Among the plants growing in the bogs of this vicinity the fol­
lowing have been found to possess mycorhiza: Larix laricina, Pinus

Strobus, Picea Mariana, Betula lutea, Betula pumila, Oxycoccus
macrocarpus, O. Oxycoccus, Chiogenes hispidula, Vaccinium corymbosum, Ledum Groenlandicum, Populus tremuloides.

I n order to get at the conditions which favor or cause the develop­
m e n t of mycorhiza, cultures of L a r i x were m a d e in loose sphagnum,
sand, undrained sphagnum, etc.

T h e roots in the m a n y other cul­

tures previously noted were also carefully watched.

I t has been

found without exception that where the plants were grown under
properly aerated soil conditions, n o r m a l roots developed in place of
the mycorhiza.

T h a t the acidity of the bog water has nothing to do

with the production of mycorhiza is shown by the fact that in water
cultures of the s a m e acidity as the solution in the undrained peat, the
plants develop n o r m a l roots.

I n the case of roots developed in loose

sphagnum, sand, and moist air, an a b u n d a n c e of root hairs were pro­
duced.

T h e n o r m a l roots in L a r i x have a diameter about

three

times that of the mycorhiza, so that when they begin to develop they
appear like white pendants from the dark brown mycorhiza.

That

m y c h o r i z a will not develop in a well-aerated substratum was further
tested b y the following experiment: T w o 3 0
upright, and 8
each.

c

m

c m

test tubes were set

of glass beads were poured into the b o t t o m of

I n t o one a glass tube, at whose end were several small open­

ings, was passed to the b o t t o m .
connected with a gasometer.

T h e upper part of the tube was

U p o n this foundation of beads, three

plants of L a r i x were planted in a j

c

m

layer of peat in each tube.

T h e water level in the two tubes was kept j u s t at the surface, bog
water being used throughout.

A i r was then forced from the gas­

ometer to the bottom of the one tube and allowed to pass slowly
through the beads and peat.

W h e n the experiment was started, all

of the plants possessed only mycorhiza.

I n the course of a week the

aerated plants b e g a n to develop n o r m a l roots.
continued for six weeks.

T h e unaerated

T h e experiment was
plants

developed

only

mycorhiza, while those which were aerated developed normal r o o t s .

8

T h e growth of mycorhiza is exceedingly slow, and the fungus grows
with the root.

T h e development of the above ground parts cor­

responds to the root development.

T h e plants which produce normal

roots have longer shoots, and longer, thicker leaves.
I t seems evident, in the case of L a r i x at least, that ( i ) the mycorhizas develop only in poorly aerated substrata;
8

(2) their growth is

In the case of a number of the plants of Larix grown in the undrained peat
in previous experiments, one or two normal roots were developed just at the surface
of the substratum.

exceedingly slow, the fungus developing along with the root; (3) the
acidity of the substratum is not a factor in their development; (4) in
a naturally well-aerated soil or in an artifically aerated substratum
normal roots develop;

(5) when the roots are not surrounded b y

water, root hairs develop abundantly.
to b e an a b n o r m a l root condition.

M y c o r h i z a therefore appears

W h e t h e r the fungus is of advan­

tage to the root under these poorly aerated conditions cannot as yet
be stated.
(8) I n order to determine whether the zone of t a m a r a c k s follows
the shrub zone because of the occasional submergence of the sedge
zone, the following test was m a d e :
7

c m

T e n L a r i x seedlings averaging

in height were placed in a crystallizing dish with the

imbedded in 2

c m

of sphagnum.

roots

O v e r this a layer of bog water 4

in depth was maintained for six weeks.
well as those in a peat substratum.

c m

T h e plants grew quite as

S t e m and root submergence is

therefore not a factor in preventing the growth of seedlings t a m a r a c k
in the sedge zone.

T h e liability to submergence in the bogs I have

studied would not extend over nearly so long a period of time.
V.

Summary.

T h e H u r o n R i v e r basin shows three well-marked physiographic
divisions which differ in forest covering and the possibilities for bog
development.

T h e s e are (1) the region of the S a g i n a w - E r i e inter­

lobate m o r a i n e ; (2) the E r i e morainic b e l t ; and (3) the lake plain.
I n discussing the meteorological conditions of a region as affecting
the flora, attention is called to the fact that the significance of the
data is not apparent unless the temperature and rainfall phenomena
are compared with those of the optimum region of dispersal of the
plant societies involved.

I n the case of the b o g plant societies the

temperature of the region under discussion averages several degrees
higher during the summer months t h a n the eastern maritime prov­
inces of C a n a d a (the optimum region of dispersal for the bog p l a n t s ) ,
while the rainfall during the same period averages about threefourths as m u c h .

T h i s is believed to account for the general differ­

ence in character and development of bog societies in the two regions.
B o g and lake basins are here associated with deposits of glacial
drift.

T h e most frequent causes of these basins are (1) the melt-

ing of stagnant bodies of ice in old glacial drainage channels after
their a b a n d o n m e n t ;

(2)

the

differential settling of

fluvio-glacial

deposits; and (3) unequal deposition of glacial material in moraines
and till plains.
M a r l and peat deposits are commonly associated.

T h e former

are of interest in so far as they aid in the filling of the lake basins.
B o t h are formed through plant agencies.
P e a t deposits m a y b e classified under two general h e a d s : (1) those
connected with glaciation, and (2) those associated with coastal plain
phenomena.
the first head.

I n N o r t h A m e r i c a the b u l k of the deposits come under
T h e i r geographic distribution approximates that of

the Pleistocene glaciers.

N e a r the southern border the peat areas are

scattered, but they b e c o m e more nearly continuous and more inde­
pendent of depressions as we go northward.

T h e same effect is

brought about in mountainous regions b y increased altitude.

In

the tundra, peat accumulates because of the low temperature and in
spite of the scant vegetation.

I n temperate regions a vigorous vege­

tation and areas of stagnant water render peat accumulation possible.
I n the southern coastal plain swamps, peat is formed in stagnant
water because of the luxuriant vegetation and in spite of the high
temperature.
D u r i n g peat formation two processes are involved:
causis and (2) putrefaction.

(1) erema­

T h e former is essentially an oxidizing

process, brought about in the presence of air b y certain fungi and
bacteria.
plants.

I t s products are of direct value as food materials for
Putrefaction is carried on in the absence of oxygen and is

essentially reduction; the organisms involved are anaerobic bacteria,
and the products are of no value to the higher plants as food materials.
T h e accumulation of peat depends upon the scarcity of oxygen below
the water level, the acidity of the ground water, and the occurrence
of low temperatures.
P e a t varies in color b e n e a t h the various plant societies, being light
brown in the youngest (bog sedge) and dark brown in the oldest, the
darkest and most thoroughly decayed form being known as " m u c k . "
As disintregration proceeds it brings about a decrease in water capac­
ity, a decrease in volatile combustible matter, and an increase in
the amount of ash.

T h e bog as a habitat for plants differs widely from the other plant
habitats of the region in that its substratum h a s been built b y fore­
runners of the present vegetation.

Owing to the influence of the

wind in the production of waves, the bogs are largely wanting on the
eastern shores of lakes, and in the case of basins which have been
almost completely filled with peat, the open water lies toward the
eastern margin.
I t is well known that bog areas are more liable to late spring
frosts than adjoining uplands.

T h i s is due to the topography as it

affects air drainage, and to the low conductivity of the substratum
covering.

U n d e r natural conditions it has been found that the areas

of cassandra and t a m a r a c k dominance are more exposed to late frosts
t h a n other societies.

,

Observations in bog areas show that the soil temperatures beneath
the several plant societies differ markedly in range.

T h e records

indicate that the areas of bog sedges have temperatures correspond­
ing closely with those of the upland and approximating those of the
atmosphere.

T h e willow-sedge (swamp) and maple-poplar

have slightly lower temperatures during early spring.

areas

W h e n the

trees leaf out, however, the shade produced causes the maple-poplar
area to have the lowest temperatures recorded.

T h e bog shrub and

t a m a r a c k societies show the lowest average temperature

throughout

the spring months.
L o w soil temperatures retard chemical action, diffusion, solution,
a n d osmosis, and render the substratum unsuited to soil bacteria.
W h e n coincident with higher air temperatures, plants having a low
transpiration ratio are favored in the competition between species.
I n so far as southern M i c h i g a n is concerned, the

substratum

temperatures prevailing in bog areas do not seem to b e adequate to
account for the presence or absence of bog plants or their xerophilous
structures.

E x p e r i m e n t s suggest, however, that farther north this

factor is of prime importance.
I n texture the bog substratum shows every gradation from the
coarse fibrous peat of the bog-sedge zone to the b l a c k powdery m u c k
of cleared land.

B o g soils in general do not afford as good a foothold

for trees as do the mineral soils.
P e a t i s very resistant to the diffusion of mineral salts, hence bog

areas have a very different soil solution from that of the mineral
soils adjoining.

T h e high water capacity of peat is detrimental to

plants, in so far as it prevents proper aeration of t h e substratum.
B o g waters have n o higher osmotic pressure

t h a n ordinary

soil

waters.
T h e absence of sphagnum from local bogs cannot b e explained
by t h e presence o f calcium salts, as shown b y observation, chemical
analyses, a n d experiments.
T h e acidity o f local b o g water varies from
normal acid.

.00015 to .00258

T h e lowest values are found in areas covered b y b o g

sedges a n d swamp plants, a n d they are approximately t h e s a m e .
T h e highest occur under t h e t a m a r a c k s .

T h e variations in acidity

are related inversely to t h e temperature.

A s shown b y experiment,

this is because o f increased oxidation at the higher

temperatures.

I t is suggested that we should find increased acidity as we go north.
T h e r e is n o apparent relation between color a n d acidity, except that
light colored waters usually show slight acidity.

T h e acid nature

of t h e soil solution is a factor in the competition between different
species for t h e occupancy of b o g areas.
B o g soils are notably deficient in potassium and available nitrogen.
Nitrifying b a c t e r i a a r e prevented from carrying on their n o r m a l activ­
ities b y the acidity o f the soil solution, b y the l a c k of oxygen, and b y
the lower temperature of the substratum.
W i t h few exceptions b o g plants are light-demanding forms; hence,
in their competition

with

one another, size and shading ability

are prime factors.
T h a t the conditions in the H u r o n valley are at present n o t as favor­
able to t h e b o g plants as to t h e swamp plants, is shown wherever the
two societies c o m e into competition. T h i s fact must b e contrasted with
the situation in t h e optimum region of the distribution o f b o g plants,
where the opposite relation h a s been shown to exist.
An examination of all the physical and chemical data n o w avail­
able fails to account for the differences in flora of b o g a n d swamp
areas in this region.

T h e most important factor is believed to b e

their physiographic history.

W h e r e the habitat dates b a c k to Pleis­

tocene times a n d h a s remained undisturbed, we find today the b o g
flora.

W h e r e the h a b i t a t is o f recent origin or h a s been recently dis-

turbed, we find the swamp flora, or mixtures of swamp and

bog

species.
T h e nature of the bog plant societies of the H u r o n basin is shown
b y the description of several local bogs, selected to show both the
local bog flora and the variation in societies, and arranged to present
the genetic changes in a bog flora as a basin filled b y peat accumu­
lation.

I t is shown that during the early stages of bog development,

bog sedge, bog shrub, and conifer societies follow each other in the
invasion of the basin.

T h e s e several societies m a y vary considerably

in composition, but they are closely related and show every gradation
in a definite order of succession.

T h e bog conifers, however, show

no relationship to the surrounding broad-leaved forests of the upland.
O n the other hand, where clearing has occurred, swamp sedges,
swamp shrubs, and swamp trees gain the ascendency, and these not
only show an order of succession among themselves, but are genetically
related to the broad-leaved trees of the region.

T h e bog societies

are part of the northeastern conifer forest formation, while the swamp
societies are related to the southeastern broad-leaved forests.
A n a n a t o m i c a l study of the bog plants shows that epidermal and
hypodermal tissues are thick-walled, that a heavy cuticle is present,
frequently supplemented b y w a x and hairs.

R e s i n o u s bodies are

to b e found in the roots and leaves of m a n y of the plants.
are usually small and revolute-margined.
up a large part of the mesophyll.
of the plants.

T h e leaves

Palisade tissue m a k e s

M y c o r h i z a s are present in most

B o g plants resemble the plants of dry sand plains

in reduction of foliage area, in development of protective coverings
for above-ground parts, and in palisade tissues, but differ from the
latter in the m a t t e r of root development and root structures.
E x p e r i m e n t s indicate that the local bog water itself has no tendency
toward

the

production

of xerophilous

modifications.

L o w soil

temperatures and l a c k of soil aeration, however, cause a reduction
in the development of the several plant organs.

W h e n these two

factors are combined, the effect is very m a r k e d .
E x p e r i m e n t s with Rumex acetosella are of especial interest in
that nearly all of the characteristics of bog plants m a y b e developed
either b y lowering the soil temperature, as compared with the air
temperature, b y preventing proper soil aeration, or b y growing in

dry sand.

Palisade tissue was developed in the leaves of these plants

in diffuse light, and it is shown that palisade tissue is to b e correlated
with physiological drought.

A n analogy between the bog h a b i t a t

and the dry sand h a b i t a t is established.
E x p e r i m e n t s with L a r i x indicate that mycorhizas develop only in
poorly aerated

substrata;

their growth

is exceedingly slow;

the

acidity of the substratum is not a factor in their development;

a

naturally or artificially aerated substratum favors the development
of normal roots, and these roots when not surrounded b y water
develop

root

prolonged

hairs

abundantly.

submergence.

Larix

seedlings

can

withstand

W h e n exposed to low substratum

tem­

peratures and poorly aerated soil conditions, L a r i x produces m o r e
xerophilous leaves.
F u r t h e r field work on the bog plant societies needs to b e carried
on in the region extending from W i n n i p e g to N e w B r u n s w i c k .

Data

on the soil and air temperatures, the acidity, the chemical composition
of the soil solution, and the plants associated in bog areas throughout
this region will go far toward solving the problems of the distribution
of bog plants.

E x p e r i m e n t a t i o n on the production of xerophilous

structures b y bog conditions should b e continued on a larger scale
t h a n is possible in the ordinary university plant-house.
T o Professor V . M . SPALDING and Professor F . C . NEWCOMBE,
of the University of M i c h i g a n , under whose direction this work was
planned and carried out, I desire to express m y sincere thanks both
for helpful suggestions and the facilities of the institution which were
freely placed at m y disposal.

M a n y thanks are also due Professor

I . C . R U S S E L L for criticism of the physiographic part of this paper.
I wish to acknowledge the kindness of M r . F R A N K L E V E R E T T , of
the U . S. Geological Survey, whose intimate knowledge of the glacial
geology of this region has been most helpful to m e in the prosecution
of m y own field work.

T o M r s . N . L . B R I T T O N I a m indebted for

the determination of the mosses.

F i n a l l y I t a k e this

opportunity

to express m y appreciation of m y friend and former instructor, D r .
H.

C . C O W L E S , to whose writings and lectures I owe m y interest

in ecological botany.
UNIVERSITY OF MICHIGAN.

BIBLIOGRAPHY.
1. BAILEY, V., Tamarack swamps as boreal islands.

Science, N. S. 3:250.

1896.

2. BARBOUR, E. H., Report State Geologist of Nebraska 1 : 1 9 3 - 1 9 8 . 1903.
3. BELZ, J. O., Influence of temperature on the rate of nitrification. Ann.
Rept. Wis. Agric. Exp. Sta. 18:224. 1901.
4. BLANCK, E., On the diffusion of water in humus soils. Landw. Vers. Sta.
5 8 : 1 4 5 . 1903. Review, Exper. Sta. Rec. 14:848. 1903.
5.

BLATCHLEY, W. S. and ASHLEY, G. H., The lakes of northern Indiana and

their associated marl deposits.

25th Ann. Rept. Dept. GeoL, Indiana.

1901.

6. BRITTON, W. E., Vegetation of the North Haven sand plains.

Bull. Torr.

Bot. Club 3 0 : 5 7 1 . 1903.

7. COULTER, S. M., An ecological comparison of some typical swamp areas.
Rept. Mo. Bot. Gard. 15:40. 1904.
8. COWLES, H. C , Physiographic ecology of Chicago and vicinity. B O T .
GAZ. 3 1 : 1 4 5 . 1901.

9. DAVIS, C. A., A contribution to the natural history of marl.

Jour. Geol.

8:485. 1900.

10.
, A second contribution, etc. Jour. Geol. 9:491. 1901.
1 1 . DAVIS W. M., Elementary meteorology. Ginn & Co. 1898.
12. DECORIOLIS, E. G., Investigation of the composition of soils rich in vege­
table matter. 29th Ann. Rept. Ont. Agric. Coll. 1903.
13. DRYER, C. R., Lessons in physical geography. Amer. Book Co. 1901.
14. DUMONT, M. J., Les causes d'infecondite des sols tourbeux. Compt. Rend.
Acad. Sci. Paris 133:1243. 1901.
15. FRITH, J. J., Ueber Torf und Dopplerit. Zurich. 1883.
16. GANONG, W. F., Raised peat bogs in the province of New Brunswick.
4

Proc. Roy. Soc. Can. II. 3 : i 3 i . 1897.

17.
18.

, Raised peat bogs in New Brunswick. BOT. GAZ. 16:123. 1891.
, The vegetation of the Bay of Fundy salt and diked marshes. BOT.

19.

GILBERT, G. K., Glaciers and glaciation. Harriman Alaska Expedition

GAZ. 3 6 : 1 6 1 . 1903.
3:1-231.

Review, Science, N . S. 19:783. 1904.

20. HABERLANDT, G., Physiologische Pflanzenanatomie. Leipzig, 1896.
21. HALE, D . J., Marl and its application to the manufacture of Portland
cement. Geol. Survey of Mich. 8 . 1903.
22. HOPKINS, C G., Soil treatment for peaty swamp lands. Bull, 93, 111.
Agric. Exper. Sta. 1904.
23. HUSTON, H. A., and BRYAN, A. H., Swamp muck. Rept. Ind. Agric.
3

Exper. Sta. 1900:73.

24.

JULIEN, A. A., On the geological action of humus acids. Proc. A. A. A. S.
2 8 : 3 1 1 . 1879.

25.

KEARNEY, T . H., The plant covering of Ocracoke island.

Contrib. U. S.

Nat. Herb. 5 : 2 6 1 . 1900.

26.

, Report on a botanical survey of the Dismal Swamp region.

Contrib-

U. S. Nat. Herb. 5 : 3 2 1 - 5 5 0 . 1901.

27. KEDZIE, R. C , Michigan soils. Bull. 99, Mich. Agric. Exper. Sta. 1893.
28. KIHLMAN, A. O., Pflanzenbiologische Studien aus russisch Lappland.
Act. Soc. pro Fauna et Flora Fennica 6 : 1 1 3 . 1890.
29. KING, F. H., The soil. Macmillan Co. 1895.
30.
, and JEFEERY, J. A., The character and treatment of swamp or humus
soil. Bull. 80, Wis. Agric. Exper. Sta. 1900.
31. L A N E , A. C , Lower Michigan mineral waters. Water Supply and Irriga­
tion Papers U. S. Geol. Survey, No. 31. 1899.
32. LEVERETT, F., Glacial formations and drainage features of the Erie and
Ohio basins. Mon. 41, U. S. Geol. Surv. 1902.
33.

LIVINGSTON, B. E., Physical properties of bog water.

B O T . G A Z . 37:383

1904.

34.
35.
36.
37.

LUCAS, F. A., Animals before man in North America. Appletons, New
York. 1902.
MACMILLAN, C , On the formation of circular muskeag in tamarack swamps.
Bull. Torr. Bot. Club 23:500. 1896.
MAYER, A., Agriculturchemie 2:69. Heidelberg. 1875.
MORGAN, L . H., The American beaver and his works. Lippincott & Co.
1868.

38. MULDER, G. J., Die chemie der Ackerkrume, pp. 308-364. Berlin 1861
39. RAMANN, E., Forstliche Bodenkunde und Standortslehre. Berlin. 1893.
40. RIES, H., Uses of peat and its occurrence in New York. 21st Rept. N. Y .
State Geologist. 1901. r 55.
41. ROWLEE, W. W., Swamps of Oswego county. Amer. Nat. 31:690. 1897.
42. RUSSELL, I. C , The Portland cement industry in Michigan. Ann. Rept.
U. S. Geol. Surv. 22:635. 1901.
43.
, Glaciers of North America. Ginn & Co. 1901.
44. SCHIMPER, A. F. W., Pflanzengeographie auf physiologischer Grundlage.
Gustav Fischer, Jena. 1898.
45. SEELEY, D . A., The temperature of the soil and surface of the ground.
Monthly Weather Rev. 29:501. 1901.
46. SHALER, N. S., The freshwater morasses of the United States. Ann. Rept.
U. S. Geol. Surv. 10:261. 1890.
47. SITENSKY, F., Ueber die Torfmoore Bohmens. Arch, der Naturw. Landesdurchforschung von Bohmen 6 :228. 1891.
48. SNYDER, H., Report on composition of muskeag soils. Bull. 81, Minn.
Agric. Exper. Sta. 1903.
49. START, E., Ueber den Einfluss des sonnigen oder schattigen Standortes auf
die Ausbildung der Laubblatter. Jenaische Zeitschrift fur Naturw.
I

1883:16.

50.
51.
52.

STOCKBRIDGE, H. E., Rocks and soils. J. Wiley & Sons, New York. 1895.
STUPART, R. F., The climate of Canada. Scot. Geog. Mag. 14:73. 1898.
TARR, R. S., The physical geography of New York state. Macmillan Co.
New York. 1902.
53. TAYLOR, F. B., Correlation of Erie-Huron beaches with outlets and moraines
in southern Michigan. Bull. Geol. Soc. Amer. 8:31. 1897.
54. TODD, J. E., The moraines of southeastern South Dakota and their
attendant deposits. Bull. 158, U. S. Geol. Surv. 1899.
55. TRANSEAU, E. N., On the geographic distribution and ecological relations
of the bog plant societies of northern North America. BOT. GAZ. 36:
401. 1903.
56.
, The development of palisade tissue and resinous deposits in leaves.
Science, N. S. 19:866. 1904.
57. WAGNER, G., Observations on Platygonus compressus LeConte. Journal
Geol. 11:777. 9°358. WEBER, C. A., Ueber die Moore, u. s. w. Jahresbericht der Manner vom
Morgenstern 3:1-23. Review, Bot. Cent. 88:17.
59. WELD, L. H., A peat bog and morainal lake. BOT. GAZ. 37:39. 1904.
60. WHEELER, H. J., Results of many experiments on "acid upland soils" are
to be found in the 6th, 10th, and 12th Ann. Rept. of the R. I. Agric. Exper.
Sta. 1893, 1897, and 1899. Also see bulletins no. 46, 47, 49, 66, 69, 71.
90 and 95 of the same station.
6T. WOLLNY, E., Die Zersetzung der organischen stoffe. Heidelberg. 1897.
I