'^■v.

X27 'Lils

Bulletin

OF THE

Illinois State Laboratory

OF

Natural History

Urbana, Illinois, U. S. A.

VOLUME XI
1915, 1917, 1918

Contributions to the Natural History Survey of Illinois

MADE under the DIRECTION OF

Stephen A. Forbes

1918

Bulletin

OF THE

Illinois State Laboratory

OF

Natural History

Urbana, Illinois, U. S. A.

VOLUME XI
1915, 1917, 1918

Contributions to the Natural History Survey of Illinois

made under the direction of

Stephen A. Forbes

1918

/^/^-/g

ScHXEPP & Barnes. State Printers

Springfiei,d, Ii-l.

1918.

8040—600

CONTENTS.

ARTICLE I. AN OUTLINE OF THE RELATIONS OF ANIMALS TO page
THEIR INLAND ENVIRONMENTS. BY CHARLES C. ADAMS,

Ph.D. July, 1915 3-32

The dynamic relations of animals 3_17

1. Introductory note 3

2. The relations of animals to their environment 3'

3. Optima and limiting factors 8

4. Determination of dynamic status 9

5. Animal responses 10

6. The interrelations of animals 12

7. Ecological units for study 14

8. The animal association 15

9. Associational succession 16

The dynamic relations of the environment , 17-31

1. Introductory 17

2. The dynamic and genetic standpoint IS

3. Dynamic and genetic classification of environments 21

References to literature 31-32

ARTICLE II. AN ECOLOGICAL STUDY OF PRAIRIE AND FOREST
INVERTEBRATES. BY CHARLES C. ADAMS, Pii.D. (63 Platics)

September, 1915 33-280

Introductory 33

General description of the region and location of the ecological stations 35-40

I. General description of the region 35

II. The ecological stations 38

Description of the prairie habitats and animals ■ . 40-56

I. Prairie area north of Charleston, Station 1 40

1. Colony of swamp grasses (Spartina and 'Elymus), Sta-

tion I, a 41

2. Colony of wild rye {Elymus virr/inicus suhmuticus) , Sta-

tion I, c 43

3. Wet area of swamp milkweed (As'Tpnirifi Incarnata).

Station I, d 44

4. Cone-flower and rosin-weed colony, Station I, c 48

5. Colony of blue stem (Andropogon) and drop-seed (Sporob-

olus), bordered by swamp milkweed, Station 1, g 49

6. Supplementary collections from Station 1 52

IV

PAGE

II. Prairie area near Loxa, Illinois, Station II 52

III. Prairie area east of Charleston, Station III 55

Description of the forest habitats and animals 56-66

1. The Bates woods, Station IV 56

2. The upland oak-hickory forest, Station IV, a 57

3. Embarras valley and ravine slopes, forested by the oak-hickory

association. Station IV, b 59

4. Lowland or "second bottom," red oak-elm-sugar-maple-woodland

association. Station IV, c 62

5. Supplementary collections from the Bates woods. Station IV. . . . 65

6. Small temporary stream in the south ravine. Station IV, a 65

General characteristics of the gross environment 66-102

1. Topography and soils of the State 66

2. Climatic conditions 67

3. Climatic centers of influence 69

4. Relative humidity and evaporating power of the air 71

5. Temperature relations in the open and in forests 83

6. Soil moisture and its relation to vegetation 86

7. Ventilation of land habitats 88

8. The tree trunk as a habitat 91

9. Prairie and Forest vegetation and animal life 91

10. Sources and role of water used by prairie and forest animals.. 98

Animal associations of the prairie and the forest 102-158

I. Introduction 102

11. The prairie association 103

1. Swamp prairie association 103

2. The Cottonwood community 105

3. Swamp-grass association 107

4. Low prairie association 108

5. Upland prairie association 10j9

6. The tiolidago community 109

7. Dry prairie grass association -. Ill

8. A milkweed community 112

III. Relation of prairie animals to their environment 113

1. The black soil prairie community 114

2. The prairie vegetation community 117

3. Interrelations within the prairie association 119

IV. The forest associations 122

1. Introduction 122

2. Dry upland (Quercus ana Cnrya) forest association 124

3. Artificial glade community in lowland forest 125

4. Humid lowland (hard maple and red oak) forest asso-

ciation 126

5. Animal association of a temporary stream 127

V. Relation of the deciduous forest invertebrates to their environ- page
• ment 128

1. Forest soil community 129

2. The forest fungus community 135

3. The forest undergrowth community 138

4. The forest crown community 139

5. The tree-trunk community 142

6. The decaying wood community 148

7. Interrelations within the forest association 157

Ecologically annotated list: —

I. Prairie invertebrates 158-201

II. Forest invertebrates 201-238

Bibliography 239-264

Index - 265-280

ARTICLE III. THE VERTEBRATE LIFE OF CERTAIN PRAIRIE
AND FOREST REGIONS NEAR CHARLESTON, ILLINOIS. BY

T. L. HANKINSON. (16 Plates) Skptkmiser. 1915 281-303

Introduction 281

The prairie area, Station 1 282

Amphibians and reptiles 284

Birds 284

Mammals 288

Relations of the prairie vertebrates to their environment 289

The forest area, Station II 291

Fish, amphibians, and reptiles 293

Birds 294

Mammals ' 297

Supplementary list of birds 298

Relation of the woodland vertebrates to their environment 299

Summary and conclusions 301

ARTICLE IV. SOME ADDITIONAL RECORDS OF CHIRONOMID^
FOR ILLINOIS AND NOTES ON OTHER ILLINOIS DIPTERA.
BY JOHN R. MALLOCH. (5 Pi.atks) DF.n-MiiKU. 1915 305-363

Notes on blood-sucking Ceratopogonin;p.-. . 306-309

Additions to list of Illinois Chironomidge: —

Ceratopogonidge 310

Tanypinte 317

Descriptions of males of Ceratopogoninae previously unknown 317-319

Immature stages of some Illinois Diptera (Scij.ridae and Mycetophil-

idas ) and biological notes 319-324

Predaceous and Parasite Orthorrhapha 324-342

Key to pupae 325

Bombyliidae 327

VI

PAGE

Therevidffi 334

Mydaidse 336

Asilidae 337

Cyrtidae 341

Phytophagous and other Cyclorrhapha 342-352

Syrphidse 342

Ephydridse 345

Drosophilid£e 346

Agromyzidse 348

Descriptions of new Illinois Diptera 352-363

Phoridee 353

Anthomyiida^ 356

Geomyzidse 357

Agromyzidae 359

Chloropidae 360

ARTICLE V. PHYLLOPHAGA HARRIS (LACHNOSTERNA HOPE) : A
REVISION OF THE SYNONYMY, AND ONE NEW NAME. BY

ROBERT D. GLASGOW, Ph.D. February, 1916 365-377

Synonymy of the Phyllophaga of the United States and Canada 370

Alphabetical list of foregoing names 374

PTtyllophaga fordesi, n. sp 378

ARTICLE VI. AN EXPERIMENTAL STUDY OF THE EFFECTS OF
GAS Y/ASTE UPON PISHES, WITH ESPECIAL REFERENCE
TO STREAM POLLUTION. BY VICTOR E. SHELFORD. Ph.D.

(1 Figure. 4 Charts) March. 1917 381-412

I. Introduction 381

II. Statement of the fish and gas-waste problem 381

III. Material and methods 383-388

1. The character of University of Illinois water 383

2. Treatment for keeping fishes alive 384

3. Difficulties to be guarded against in fish experiments.... 385

4. Fishes used 387

IV. Gas waste — its character and constituents 388

V. Toxicity of waste 389-394

1. Methods of experimenting 390

2. Toxicity of waste and tar 391

3. Toxicity of illuminating gas and constituent gas-mixtures. 392

4. Reactions of fishes to waste 392

VI. The toxicity of illuminating gas waste constituents 394

VII. General discussion 406

1. Toxicity and size 406

2. Toxicity and species 40S

Vll

PAGE

3. Fish reactions to polluting substances .• 408

4. Treatment of by-products of the manufacture of coal gas.. 409

VIII. Summary 409

IX. Acknowledgments 410

X. Literature consulted 410

*

ARTICLE VII. SOME EDIBLE AND POISONOUS MUSHROOMS. BY
WALTER B. McDOUGALL, Ph.D. (2 Figurks, 59 Plates) Novem-
ber, 1917 413-555

Introduction 413

Mushrooms and toadstools 414

The mushroom plant 414

Life history and development 414

Structure 417

Spore production and liberation 417

Other types of mushrooms 419

The ecology of mushrooms 420

Dissemination 420

Gravity 421

Air 421

Heat 421

Light 422

Substratum 422

Water 422

Parasites and saprophytes 423

Mycorrhizas 424

Animal relations 425

Diseases 427

Fairy rings 427

Luminosity 427

Mushroom-growing 429

Food value of mushrooms 430

Poisonous properties of mushrooms 430

Collecting wild mushrooms 431

Preparation of mushrooms for the table ■ • 433

Classification of mushrooms 435

Use of the key 436

Key to genera of gill fungi • • • ■ 437

Descriptions and illustrations of species 440-552

Reference to literature 554

Index to described or briefly characterizetl species 555

Vlll

ARTICLE VIII. THE REACTIONS AND RESISTANCE OF FISHES

TO CARBON DIOXIDE AND CARBON MONOXIDE. BY MORRIS page
M. WELLS, Ph.D. (1 Fkiuke, 1 Chart) May, 1918 557-571

Introduction 557

Properties of the gases 559

Methods and materials 559

Ggneral resistance of fishes 565

Summary 568

Bibliography 569

ARTICLE IX. EQUIPMENT FOR MAINTAINING A FLOW OF
OXYGEN-FREE WATER. AND FOR CONTROLLING GAS CON-
TENT. BY VICTOR E. SHELFORD, Ph.D. CI Figure) May, 1918.573-575

ARTICLE X. A COLLECTING BOTTLE ESPECIALLY ADAPTED
FOR THE QUANTITATIVE AND QUALITATIVE DETERMINA-
TION OF DISSOLVED GASES, PARTICULARLY VERY SMALL
QUANTITIES OF OXYGEN. BY EDWIN B. POWERS, M. A.
(1 Figure) May, 1918 577^578

ERRATA AND ADDENDA.

Page 50, second column, line 13 from bottom, for Danais archippus read Anosia
plexipj)us ; line 8 from bottom, for mellifica read mellifera.

Page 51, line 11 from bottom, for Danais read Anosia.
Page 159, at right of diagram, for Bracon agrilli read Bracon agrili.
Page 289, second column, last line but one, for Scalops real Scalopus.
Page 294, line 3, for catesbeana read catesbiana.
Pages 327 and 330, line 12, for oreus read oreas.
Page 347, line 4, for Cecidomyidse road Cecidomyiidas.
Page 356, line 7, for Anthomyidas read Anthomyiid^.
Page 368, line 18, dele second word.

Page 373, after line 10 insert as follows: 53a, suhpruinosa Casey, 1884, p. 38.
Page 375, after sudviucida Le Conte, 48, Insert suhpruinosa Casey, 53a.
Page 377, after line 7, insert as follows: —
1884. Casey, Thomas L.

Contributions to the Descriptive and Systematic Coleopterology of
North America. Part I.
Page 379, line 11 from bottom, for sensu lata read sensu lato.
Page 382, line 12, for VII read VIII.

Page 408, line 2, for the next article in read Article VIII of.
Page 410, line 6 from bottom, for = 4 read '11.
Page 412, line 7, for 31 read 30.
Page 421, line 17 from bottom, insert it before grows.

Bulletin

OF THE

Illinois State Laboratory

Olf

Natural History

Urbana, Illinois, U. S. A.

STEPHEN A. FORBES, Ph.D., L-L.D.,
Director

Vol. XI. July, 1915 Article I.

AN OUTLINE OP THE RELATIONS OF ANIMALS
TO THEIR INLAND ENVIRONMENTS

BY

Charles C. Adams, Ph.D.

Bulletin

OK THE

Illinois State Laboratory

OF

Natural History

Urbana, Illinois, U. S. A.

STEPHEN A. FORBES, Ph.D., L.L.D.,
Director

Vol. XL July, 1915 Article L

AN OUTLINE OF THE RELATIONS OF ANIMALS
TO THEIR INL,AND ENVIRONMENTS

BY

Charles C. Adams, Ph.D.

CONTENTS

PAGE

The (Ijnaniic relations of animals 1-17

1 . 1 iitroduetory note 1

2. The relations of animals to their environment 1

3. Optima ami limiting factors 8

4. Determination of dynamic status 9

5. Animal responses 5

fi. The interrelations of animals 12

7. Ecological units for study 14

<S. The animal association 15

9. Associational succession IG

The dynamic relations of the environment 17-31

1. Introductory 17

2. The dynamic and genetic standpoint 18

3. Dynamic and genetic classification of environments 21

References to literature 31-32

Article I. — An Outline of the Relations of Animals to their In-
land Bnz'iromnents. By Charles C. Adams, Ph.D.

The Dynamic Relations oe Animals
I. introductory note

As creatures of habit, the attitude of mind with which we approach
a scientific problem has much influence upon what we see in it or get
from it. Although the essence of life is activity — the response of the
changing organism to its changing environment — yet this dynamic
conception of animal relations, and all that it implies, has not become
as prevalent a mental habit among biologists as one might expect.
While some naturalists view the animal from a more or less dynamic
standpoint, they do not include a similar conception of the relation of
an animal to its environment. Still others view the environment more
or less dynamically but do not extend this conception to the animal,
and thus both of these conceptions lack completeness and are not thor-
oughgoing and consistent. The study of activities, or in other words
the study of processes, has made great progress in the allied sciences,
much to their advantage, and undoubtedly the prevalence of similar
conceptions will lead to similar advances in biology.

In the present brief paper I have attempted to discuss only certain
phases of the problem with the idea of emphasizing the general prin-
ciples involved, and in the hope that it may aid in making these con-
ceptions of more practical value in investigation, and also facilitate an
understanding of the discussion contained in a report on the inver-
tebrates of the Charleston (Illinois) region, to appear in a subsequent
paper of this volume of the Laboratory Bulletin.

2. THE relations OE ANIMALS TO THEIR ENVIRONMENT

The study of animal ecology may be taken up from many sides and
in many ways. One of the most interesting and fundamental of these
is that which considers the dependence of the animal upon its environ-
ment, and at the same time orients it in the gamut of energies and
substances. Many phases of this discussion, though elementary and
for this reason easily overlooked, are yet of fundamental importance.
Every boy who has kept pets in confinement, and who has had the re-

sponsibility of caring for them, and every one who has cared for
domestic animals, knows what constant attention must be given to keep
them suppHed with food, water, sheher, and other "necessities of hfe."
And who can overlook the fact that it requires attention to maintain
his own physical health? In the laboratory this dependence upon the
environment is readily tested experimentally by any method of isola-
tion which will prevent an animal from securing any "vital necessity" :
as air — when sealed in a vessel ; or food — when locked up without it ;
or a favorable temperature. No animal can survive such isolation
from its normal environment. Every student of animals in nature
must also realize that similar supplies and conditions determine and
control the existence and welfare of all zuild animals. The animal is
not self-sustaining, but requires a constant intake of energy and sub-
stance from its environment. Chemical methods will readily show the
source from which the materials composing the animal body have been
derived. The ash came from the soil or rock, and shows the animal's
dependence upon the solid earth; the liquids came from the water ot
the earth and constitute from fifty to ninety-five per cent, of the bulk
of the animal's body, showing that a relatively large quantity of this
substance is essential to all living animals ; the abundant gaseous ele-
ment was derived from the atmosphere, to which it will again return.
The substance composing the animal body is thus derived mainly from
the water and the air rather than from the relatively inert and stable
earth. It will be profitable for us to imagine these proportions so
changed that the solids instead of the relatively mobile Hquids and
gases form the principal mass of the body, keeping in mind meanwhile
the slow rate of chemical change in S(jlids compared with the change
in substances in a finely divided condition, such as liquids and gases.
If the solids predominated, the rate of the chemical change, upon
which the active life of animals depends, would be greatly retarded,
and animals, including man, would be stolid beyond comprehension.
Furthermore, we must not overlook the fact that animals are not main-
tained solely by substance, because substances are also carriers of en-
ergy, substance and energy never being separated. The living animal
is not a producer: it can make neither substance nor energy, nor is it
a kind of energy; it is solely a transfonncr, a chemical engine which
changes the form of suljstance and chemical energy and produces new
combinations from the old. The living plant transforms energy and
inorganic substance, from the air, water, and earth, into complex
chemical compounds, and thus concentrates powerful chemical energy
in such a form that the animal, by a further change, is able to set it
free and to utilize it. Sugar, starch, and gluten are familiar examples

of this "tablet" or "cartridge" form of chemical energy which animals
explode or set free and then use in maintenance. During this trans-
formation, in which chemical energy is set free, waste products — inert
chemical substances — are formed which if not eliminated from the ani-
mal system will prevent its operation, just as ashes if not removed will
check a furnace. Respiration aids in the removal of carbonic acid
gas — a waste product — from the body, but we often forget that the
chemical energy derived from the oxygen is an important feature in
respiration. By another process the liquid and the solid waste is re-
moved. Thus gases, liquids, and solids are taken into the body and
later returned to the environment in a different chemical condition,
thus completing a cycle of transft^rmation. That the animal body is
so largely made up of solutions and gaseous substances is an important
factor in its relatively unstable chemical condition, a condition of uii-
stablc equilibrium, which determines the active and dynamic character
of the animal. Since, then, chemical activity is one of the essential
characteristics of a living organism, its influence forms one of the main
problems of the zoologist when studying the changes in animal activi-
ties ; their orderly secjuence and the laws which govern them.

On account of the fact that the animal is a chemical engine, it is
able to use chemical energy to the fullest extent. If we assume a hier-
archy in the forms of energy, chemical energy seems to belong to the
upper class; for though some forms of energy are not readily trans-
formed into chemical energy, chemical energy can be transformed into
all others. As a result the animal, being a chemical engine, has, as it
were, an "inside track" to the main sources of energy, and thus by
transformation is able to utilize chemical energy to form light, as in
the firefly, or electricity, as in the electric eel ; and other forms of en-
ergy useful to the animal are similarly derived. This study of the
activities of living animals, as contrasted with the study of dead ones,
is a phase of the general science of en'^rgetics, a science which fur-
nishes the basis for the correlation of many diverse branches of knowl-
edge.

The activities and transformations within the animal body show
us very clearly how an animal is dependent upon environmental condi-
tions. The animal transforms air, water, and rock, and all animal
habitats and environments must contain these elements. In nature
these are combined in a multitude of ways. The interrelations of these
fundamental environmental units have been strikitigly expressed by
Powell ('95: 22-23) -'^ follows:

"The envelopes of air, water, and rock are so distinct that they can
be clearly distinguished ; and yet, when they are carefully studied, it is

6

discovered that every one encroaches upon the territory of the others,
not only by interaction, but also by interpenetration. It has already
been shown that the water penetrates deep into the rock. Every spring
that falls from the hillside gives proof that the rocks above its level
hold water, which they yield slowly as a perennial supply; and the in-
numerable hills of the continents and islands have their innumerable
springs. Every well proves that there is water below; every artesian
fountain shows the existence of underground waters ; and every boring
in the crust of the earth, and every excavation in underground min-
ing, discovers the presence of water.

"Wherever water flows, air flows with it, and all natural waters
are permeated with air.

"The acjueous envelope is everywhere permeated with rock, which
it holds in solution or suspension, and there is no natural water abso-
lutely pure. The sea is full of salt. Salt lakes are more than full of
salt, and so they must throw it upon the bottom ; and the waters hold
lime and many other substances. Not a drop of pure water can be
found in the sea ; not a drop can be found in a lake ; not a drop of pure
water can be found in any river, creek, brook, or spring; and not a
drop of pure water can be found underground : it is all mixed to some
degree with rock.

"All natural waters are aerated. No drop of water unmixed with
rock and air can be found, except by the process of artificial purifica-
tion.

"But surely there is pure air? Nav, not so. There is no natural
air unmixed with rock and water. All the air that circulates above the
land and sea, within the ken of man, and all the air which circulates
underground, is mixed with rock and water.

"Pure air is invisible: it will not reflect light; it is transparent, but
will not convey light. Light is conveyed through the atmosphere by
ether, and is reflected and refracted by rock and water; and it seems
to be largely afifected in this manner by rock. If the ambient air of
the earth were pure, there would be no color in the sky, no rainbow
in the heavens, no gray, no purple, no crimson, no gold, in the clouds.
All these are due largelv to the dust in the air. The purple cloud is
painted with dust, and the sapphire skv is adamant on wings.

"Land plants live on underground waters : were there no subter-
ranean circulation of water, there would be no land plants. Fishes
live on under-water air: were there no circulation of subaqueous air,
there would be no fishes in the sea. The clouds are formed bv par-
ticles of dust in the air, which gather the vapor : were there no dust in
the air. there would be no clouds; were there no clouds, there would
be no rain."

Up to this point we have considered mainly the processes of main-
tenance of the animal body, but there are other processes as well which
must be called to mind, such as growth, development, multiplication,
and behavior. Physiologically considered, none of these activities are
essentially different from the fundamental phases of metabolism and
all are dependent upon it; they are special forms of the transformation
of substances and energy within the animal. As the individual animal
grows and develops in its life cycle, its metabolism, form, and behavior
change in an orderly manner, and this transformation is in the main
a continuous process like the other transformations of matter and en-
ergy. The changes which take place during entogeny are often greater
than the differences which exist between very distantlv related adults,
and these differences result in very different roles which the animal
often plays in the economy of nature.

Comparable to the responses of the animal to its environment, and
indeed essentially of the same kind, are the responses of any part of
an animal to all its other parts, the entire organism, in this case, being
considered as a unit. The environment of an internal parasite is
formed by the body of its host, and in a similar sense the different
parts of the body are parts of the environment of the other
parts. The different parts of the animal body are what they are
on account of three conditions. The first is determined by its
relative position and responses as a member of a series of
successive generations. In this way the hereditary potential-
ities are determined. Ecologically considered heredity may be
regarded both as the response of individuals (unicellular) and
germs to the conditions of life, and as the mutual responses of
different germs to one another. The crossing and intermingling of
germinal elements is as trulv a response as are other forms of activity.
Secondly, there is considerable evidence which indicates that at some
stage in the development of an animal any part is potentially capable
of developing into any other part. The character of development,
then, is conditioned by the character of the cell-environment — its rela-
tive position, and all that implies with regard to environment. A frag-
ment of a regenerating animal develops differently according to its po-
sition, and this is a response to its relative position in the cell commu-
nity. Thirdly, the development of an animal is conditioned by its ex-
ternal environment. The external conditions influence animals by
changing their internal activities. The internal changes modify the
cell community and change development. In this manner every part
of the animal is influenced by the conditions of its existence.

The processes of metabolism are continuous as long as life lasts.
Thus, as an animal respires there is a gaseous exchange, from the

8

earliest stages of its existence until its maturity and death. Eggs re-
spire as surely as larvse and adults, and the chemical, physical, and
physiological changes within them vary with their growth and develop-
ment. Some of these changes are primarily dependent on the orderly
course of development during the life cycle, and are therefore irrever-
sihle processes, because no higher animal which is mature may reverse
its development and become young again. At different stages of de-
velopment different enzymes and harmones appear v.hich modify the
physiological conditions of growth, development, and behavior. Envi-
ronmental changes, persistent and uniform, or periodic in character,
tend to modify and alter these internal processes, and are an additional
source of change, which is particularly shown in behavior.

It is interesting to observe in this connection that certain factors
are important as they hasten or retard other processes. Thus enzymes
hasten chemical changes which without them would take place at a
very slow rate, and they set free much energy in a relatively short
time. Temperature is another hastener of chemical reaction. Not
only is it a condition which sets limitations on the chemical reaction in
animals, but it also influences their optimum, and with increasing tem-
perature chemical changes take place within the animal irrespective of
the control of the animal, except in the warm-blooded animals, where
a mechanism exists which regulates, within certain limits, temperature
conditions.

3. OPTIMA AND LIMITING FACTORS

We have seen that the animal is dependent upon its environment
for both substance and energy. If. therefore, the environment does
not contain, in available form, both substance and energy, animals will
not be able to live in it permanently, although with energy stored in
their bodies they may be able to make more or less prolonged and suc-
cessful invasions into such an environment. The optimum is the most
favorable condition for any function. We may consider optima cor-
responding to units of different rank : a single cell or tissue in action,
an organ or system of organs, the animal as a whole, a taxonomic
imit — and so on, to an animal community or association. There are,
then, many kinds of optima, and the study of the conditions which pro-
duce them is a complex subject. The optima for different functions
may differ much ; for example, that for growth is often different from
that for reproduction, and the optima may also change greatly with the
development of the animal. Optima, therefore, are not fixed condi-
tions, even though they do represent a condition of physiological rela-
tive equilihriiim. The amount or intensity of substance and energy
which produces an optimum is limited above by the maximum and be-
low by the minimum. Thus departures from the optimum, toward an

9

increase or a decrease, are departures from the most favorable condi-
tions toward less favorable conditions, and hence toward liniiting^ con-
ditions. This form of expression is mainly that of the laboratory; it
is desirable therefore, in addition, to express it in terms of the normal
habitat. In nature we look upon the optimum as that complex of
habitat factors which is the most favorable, and departure in any di-
rection from this optimum intensity is in the direction of a less favor-
able degree of intensity or into unfavorable conditions. From this
standpoint any unfavorable condition is a Uniting factor and may re-
tard, hasten, or prevent vital and ecological activities. Optima arc
thus almost ideal conditions, and are probably realized in nature only
to a limited degree ; in other words only approximately. Here also,
as in the laboratory, they represent a condition of relative equilibrium.
The laws of the transformation and development of optima are of
great ecological importance, as I pointed out several years ago ('04).
In field study probably the most valuable criterion to be used in the
recognition of ecological optima is the normal relative abundance and
influence of animals in their breeding environment.

In the preceding discussion no special emphasis has been placed
upon the time element, or the rate at which changes may take place.
Natural environments are complexes, in the composition of which sev-
eral factors are involved. This being true, it is desirable to recall the
fact that the rate of change is determined by the pace of the slowest
factor, or, as Blackman ('05:289) has expressed it: "When a proc-
ess is conditioned as to its rapidity by a number of separate factors,
the rate of the process is limited by the pace of the 'slowest' factor."
This is a general law and applies to all changes, internal as well as
environmental.

In closing this section, I wish to call attention to another conclu-
sion of the English plant physiologists Blackman and Smith. They
state ('11) that from experimental study of the assimilation of water
plants, the conception of the optima is untenable, and that the phe-
nomena are better explained as the result of "interacting limiting
factors than by the conception of optima" (p. 412). This princijjle is
formulated as follows (p. 397) : "When several factors arc possibly
controlling a function, a small increase or decrease of the factor that
is limiting, and of that factor only, will bring about an alternation of
the magnitude of the functional activity." It will be of much impor-
tance to test the application of this idea to animal responses.

4. DETERMINATION OF DYNAMIC STATUS

In any study of the energetics of organisms it is desirable to have
clearly in mind one of the fundamental conceptions of this science —

10

the dynamic status. The law of conservation of energy teaches us
that energy can not be destroyed ; that it is transformed only, and thus
undergoes a cycle of changes. The animal or an animal community,
as a unit and as an agent or transformer, is constantly transforming
energy, setting it free. In this sense it originates, but not at a uniform
rate. At one time much energy may be transformed, and at another
very little. When a great amount of energy is being set free, when
the animal or community is exerting much influence, we may look upon
it as producing pressure or strain. A condition of stress is not a per-
manent one, because the pressure tends to cause such changes as will
equalize or relieve this condition. This is considered as the process
of adjustment to strain, and is called Bancroft's law ('ii). An ani-
mal in an unfavorable condition is stimulated, its normal activities are
interfered with, and a physiological condition of stress is produced
which lasts until by repeated responses or "trials" the animal escapes
stimulation or succumbs and a relative equilibrium is established. An
area may become overpopulated and consequcntlv there may be estab-
lished a condition of stress, which results in an adjvistment bv a reduc-
tion (through many causes) in the excess of population and a restora-
tion of the normal, or a condition of relative equilibrium. From these
examples it may be seen that the dynamic status means the condition
of a unit or system with regard to its degree of relative cqnilihriiini.
The cycle of change may be considered to begin at any point. I have
taken as the initial stage of the cycle the condition of stress or pres-
sure, and have indicated how this condition tends to change in re-
sponse to pressure, bringing about the process of adjustment to strain,
and leading to the condition of adjustment to strain, or that of relative
equilibrium. The activity of the agent produces the condition of
stress, the process of adjustment to the strain follows, and this leads
to the product — the establishment of the condition of adjustment or of
relative equilibrium.

These conceptions are very suggestive when applied to various
phases of organic activity, and aid greatly in utilizing the dynamic con-
ceptions which are in constant use in many of the physical sciences.
But we can not assume that these ideas will take definite form unless
the student makes some special effort to master the principles involved.

5. ANIMAI, RESPONSES

The general character of the changes within the animal, which re-
sult in the transformations of energy and substance, or the process of
metabolism in its broadest sense, is the basis of all animal responses.
It is well known that growth, development, and behavior are condi-

11

tioned by certain metabolic processes, the rate of which are further
conditioned by the presence of certain substances, as enzymes (from
Hver, etc.), and internal secretions (from thyroid, testes, adrenals,
etc.). The influence of certain physiological conditions or processes
is thus well known to affect the behavior of animals. The changes
of instinct through the removal of the testes or ovaries, may be cited
as examples of this influence. An animal whose metabolic processes
have reached a certain stage is said to be satiated ; later it is in the con-
dition of incipient hunger; and still later, in the physiological condi-
tion of intense hunger. These internal changes cause the animal to
react very differently to any food which is in its immediate vicinity.
These changes in physiological conditions are strictly comparable to
the change which an animal passes through in its ontogeny; to the life
cycle of an insect, for example, in wdiich the physiological conditions
and behavior of a caterpillar are very different from those of the pupa
and of the adult or moth. One of the higher animals, a dog, for in-
stance, will undergo internal changes which will completely alter its
responses at the sight of an old rival or enemy. Such considerations
as those just cited show^ clearly that extensive internal physiological
changes take place in animals, and that while some of them are very
gradual others are exceedingly rapid. These internal conditions or
changes have been well characterized by Jennings ('06:289) as fol-
lows : "The 'physiological state' is evidently to be looked upon as a
dynamic condition, not as a static one. It is a certain way in which
bodily processes are taking place, and tends directly to the production
of some change. In this respect the 'law of dynamogenesis,' pro-
pounded for ideas of movement in man, applies to it directly (Bald-
win, '97: 167) ; ideas must indeed be considered so far as their objec-
tive accompaniments are concerned, as certain physiological states
in higher organisms. The changes toward which the physiological
state tends are of two kinds. First, the physiological state (like the
idea) tends to produce movement. This movement often results in
such a change of conditions as destroys the physiological state under
consideration. But in case it does not, then the second tendency of the
physiological state shows itself. It tends to resolve itself into another
and different state."

I may thus summarize the relation of metabolic processes to
physiological conditions and processes of behavior by the following
table.

12

Table ]. — The Dynamic Eelations of Animal Activities

The Animal as an Agent
(Activitij of an Agent)

The animal as an agent transforms
energy and substance by its metab-
olic processes. These are accom-
panied by physiological conditions
or states; they constitute a condi-
tion of unstable equilbriuni. The
transformations take place as —

(1) Continuous and irreversible
processes, as development, differen-
tiation, etc.; or are —

(2) Periodic or rhythmic pioc
esses, as digestion sexual activity,
etc.

Processes of Activity

This unstable internal
condition tends towards
change, resulting in —

(1) New conditions;

(2) Movement;

(3) The processes of
behavior: trial, experi-
ment, investigation, etc.

Products of
Activity

New states.
Movement.
Response.
Regulation.
Adjustment.
Relative equilib-
rium.
Learning.
Orientation.
Data.
Concepts.
Explanation.
Theory.
Hypotheses.
Ideals.

Changes in the internal conditions
are produced also by external stim-
uli.

The responses of animals to the conditions in which they live are
of a composite character. Certain responses, such as the chirping re-
sponse of a coot within the egg, are inherited and are relatively auto-
matic in character; others are greatly modified by experience, as when
an animal "learns," or forms a habit by repeated responses.

The responses of animals to the conditions of existence are the
basis for any study of their relations, not only to other members of
their own species, but to all elements, living or otherwise, of their com-
plete environment. It is from this standpoint that animals must be
considered in estimating their place in the economy of nature; that is,
in estimating how they influence one another in an association of ani-
mals living together in the same habitat, and in judging of their rela-
tion to the succession of animal communities, and even to man him-
self.

6. the; interrelations of animals

"A group or association of animals or plants is like a single organ-
ism in the fact that it brings to bear upon the outer world only the
surplus of forces remaining after all conflicts interior to itself have
been adjusted. Whatever expenditure of energy is necessary to main-
tain the existing internal balance amounts to so much power locked
up, and rendered unavailable for external use." — S. A. Forbes.

We have now seen the dependence of the animal upon its environ-
ment, as this forms the basis for an understanding of conditions in-
volved in the problem of maintenance or the upkeep of the animal.
The optimum conditions for prolonged maintenance produce the vital

• 13

and ecological optima. These conditions imply more than mere main-
tenance ; they mean as well, a degree of favorable conditions which
permits the animal to exert an influence or stress upon its environment.
As Forbes has said, if all the energy available to the animal is utilized
internally there will be nothing left to influence the environment.
Metabolic changes show that large amounts of energy and substance
are used in maintenance. Under optimum conditions even greater
amounts must exist. An animal must not only be able to maintain
itself against other kinds of animals but even against its own kind, for
the overproduction of its own race wMl be practically self-destructive.
A good example of this kind of influence is seen in the hordes of lem-
mings which migrate, even into the sea, when overproduction becomes
extreme.

The vital and ecological optima are thus to be looked upon as in-
ternally balanced, but externally, not as a state of balance or poise, but
as a condition in which the animal is exerting stress, pressure, or in-
fluence upon its environment, instead of being passive or inert. A
group of animals living together in any given condition such as an
association, is an assemblage of interacting organisms. The active,
free-moving animals collide with each other, with other kinds of ani-
mals, especially the relatively sedentary kinds, and with their environ-
ment of plants and the inorganic factors. The relatively sedentary
animals are correspondingly bombarded by all elements of their en-
vironment. The association, as a whole, is thus in a continuous proc-
ess of bombardment and response frf)m every possible angle, and just
as the individual animal is stimulated and responds, so all the mem-
bers of any association are stimulated and respond in a similar man-
ner. It is by this form of activity that animals not only maintain
themselves but exert a radiating influence.

It will assist in realizing the constant pressure exerted by animals
if we compare their activity to the flow of a stream. The pressure ex-
erted by the stream may be realized if by a dam or similar means the
current is resisted. Think for a moment of the amount of energy
which would be transformed in an effort to prevent animals (or
plants) from taking possession of a favorable habitat. Imagine an
area lo feet square and think of the effort it would reciuire to prevent
animals permanently from invading and establishing themselves in this
habitat if no barriers were interposed, and if the means of destruction
of the invaders were not so drastic that they materially changed the
character of the habitat. Increase the size of the area and the diffi-
culties will increase in geometrical ratio, and the utter futility of such
an undertaking will soon be realized. The si)reading processes of the
gypsy moth in Massachusetts, and of the San Jose scale and the cotton

14

boll weevil, show us in terms of human experience something of the
energy expended by these radiating animal activities even when there
are strong human economic inducements against such invasions.

When a balanced condition, or relative equilibrium, in nature is
referred to, we must not assume that all balances are alike, for some
are disturbed with little effort and others are exceedingly difficult to
change. This distinction is an important one. Once the balance is
disturbed, the process of readjustment begins. This is a phase of the
balancing of a complex of forces. Just what stages this process will
pass through will depend, to an important degree, upon the extent of
the disturbance. Slight disturbances are taking place all the time and
grade imperceptibly into the normal process of maintenance, as when
a tree dies in the forest and its neighbors or suppressed trees expand
and take possession of the vacancy thus formed. Disturbances of a
greater degree, on the other hand, may only be adjusted by a long
cumulative process. This change can progress no faster than the rate
at which its slowest member can advance. Thus a forest association
of animals may be destroyed by a fire so severe that all the litter and
humus of the forest floor is burned. The animals which live in the
moist humic layer as a habitat, such as many land snails, diplopods,
and certain insects, can not maintain themselves upon a mineral soil,
rock, or clay. As such a forest area becomes reforested, these animals
can only find the optimum conditions when the slow process of
humus formation reaches a certain degree of cumulative development.
Under such circumstances this later stage must be preceded by ante-
cedent processes, and restoration of the balance is long delayed. Some
adjustments take place so quickly that little can be learned of the
stages through which they pass. There are, however, many slow proc-
esses which afford an abundance of time for study; in fact some are
too slow to study during a lifetime. The processes which are moder-
ately slow are often particularly illuminating because all stages are
frequently so well preserved that comparison is a very useful method
of study ; the slowness of a process has a certain resolving power, as
it were, recalling the influence of a prism upon a beam of white light,
which reveals many characteristics obscure to direct vision. A study
of the processes of adjustment among animals is a study of an im-
portant phase of the problem of maintenance. The continued process
of response will, if circumstances permit, lead to a condition of rela-
tive adjustment, or to a balancing among all the factors in operation.

7. ECOLOGICAL, UNITS FOR STUDY

In the study of animal responses many different units are avail-
able, and a brief consideration of these will aid in an vmderstanding of

15

the methods which are useful. Because the animal body has been
found to be composed of a single cell or a multitude of cells, a com-
mon belief has grown up that the cell is the natural unit for study.
This opinion seems to be due to overlooking the fact that there is just
as much reason for considering the zvhole animal as the unit. The
unicellular animals are zvhole animals as truly as they are cells, and
in multicellular animals the activity of single cells means little inde-
pendently of the animal as a whole. It thus seems that ecologically at
least the smallest valuable unit for study is the individual animal. The
responses of the individual, as a kind of animal, to its condition of
existence form the basis for what may be called individual ecology.
Animals which are related by descent from common ancestors, as a
community of social animals (e. g., an ant colony), or taxonomic
units, such as genera, families, orders, etc. (e. g., fish, birds, catfishes,
and salamanders), are also units which may be studied ecologically.
Some of these hereditary units are, ecologically, fairly homogeneous,
as, for instance, when a taxonomic unit is equally distinct ecologically :
e. g., the woodpeckers with their arboreal habits. In other cases the
taxonomic unit contains animals of great ecological diversity, as in the
case of beetles, which possess almost unlimited ecological diversity, in-
cluding littoral, aquatic, subterranean, and arboreal habitats, and para-
sitic, herbivorous, and predaceous habits. The study of ecology, upon
the basis of such a unit, may be called aggregate ecology. Still another
unit is available, based upon the animals which live together in a given
combination of environmental conditions, as in a pond, on the shore
of the sea, in a cave, within the bodies of animals, on the floor of the
forest, or in the tree tops, etc. The animals found living together in
such conditions form an animal association or a social community, and
the study of the responses of such a community is the province of
associational ecology.

8. the; animal, association

In the study of the animal association as a unit, we consider it as
an agent, whose modes of activity, or responses, are of primary inter-
est. We desire to know the kinds of animals which compose the com-
munity, the optimum and limiting influences which control its activity,
the character of its responses, and the orderly sequence of changes in
the environment to which it is responding.

The maintenance of an association depends upon the maintenance
of the individual members which compose it, just as the maintenance
of the entire animal depends upon the activities of the cells. There is
the same basis for speaking of the responses of the association as there

16

is for speaking of the responses of the individual. The association can
continue to exist indefinitely only in such environments as possess, in
available form, substance and energy for its individual members. The
activities of the individuals transform energy and substance, produc-
ing growth, development, multiplication, and behavior. The persist-
ence of an association in a given habitat brings about the formation of
certain waste products, which if not changed or transformed at a cer-
tain rate, or transported from the environment in some way, tend to
limit the optimum activity of the individuals and of the association.
In the association, as in the individual, there must be an internal rela-
tive balance before there can be such a surplus of energy that the asso-
ciation can radiate or exert outward stress or pressure. An association
which is only maintaining itself is not at an optimum, for in this latter
condition there is a surplus of energy, and the activity, rate of multi-
plication, and favorable development under normal conditions are fa-
vorable to the extension of the association. The pressure which such
an association exerts is shown by the progressive extension of its range
of influence. By the active movements of the animals, by the activity
of the environment, or by both together, they tend to invade other
habitats and areas, and in such of these as afford favorable conditions
they tend to survive and extend the area of the association. From the
standpoint of the association the behavior of these active pioneering
animals corresponds to the trial activities in the behavior of the indi-
vidual animal. These activities are not different in kind from those
which are invohed in normal maintenance. They are those which
form the initial stages in the establishment and extension of the asso-
ciation in a new locality or the re-establishment in an old one, and thus
lead to a sequence or succession of associations. Ecological succession
thus consists in an orderly sequence or series of associations which
occur successively and form a genetic series.

9. ASSOCIATIONAI, SUCCESSION

A succession of associations takes place either through the trans-
formation of older ones, or through the origin of a new one on a
surface which has been newly formed and has had no population. A
favorable habitat without a population of animals is comparable in
some respects to a vacuum ; it exists as a condition of unstable equi-
librium which tends to change toward a more stable state. The active
life of animals tends to lead them into all possible habitats, and where
they find the conditions favorable for existence they tend to survive
and thus bring about the establishment of an association. Each asso-
ciation, like the individual animal, has a certain amount of unity and

17

tends to maintain or perpetuate itself. But the stability of associations
is only relative, and some are much more stable than others. Naturally
the unstable ones are those which show succession most readily. Thus
if we destroy a few trees in a hardwood forest and produce a glade, a
large number of the characteristic animals of the dense forest will dis-
appear and be replaced by animals which normally frequent open
places; then in a few years sprout-growth and young and suppressed
trees will change the conditions so much that tlie kind of forest ani-
mals which were eliminated for a time will begin to return ; and when
the new growth is replaced by the mature forest the animals of the
mature forest will return and a new e(juilibrium will be formed. In
such a forested region the glade is to be looked upon as an unstable
condition, which through a succession of associations will later arrive
at a relatively stable condition, which is able to perpetuate itself indefi-
nitely under existing conditions. Such an association is considered a
climax, or the culmination of a series of successions under existing
conditions. The succession of associations leading to a climax repre-
sents the process of adjustment to the conditions of stress, and the
climax represents a condition of relative equilibrium. Climax associa-
tions are large units, and are the resultants of certain climatic, geolog-
ical, physiographic, and biological conditions.

The Dynamic Rfxations of the Environment
i. introductory

In the preceding section we have seen that to understand animals
we must consider them as active living agents which are constantly
changing and responding to their environment. That the environment
of animals should also be studied as an actiz'cly changiiiy iiicdiiuii has
not been as clearly recognized by students of plants and animals as one
might anticipate from its importance. Some students feel that the
study and understanding of the environment is not a part of zoology,
or at least not an essential part. Furthermore, to some of these stu-
dents at least, the environment seems largely chaotic, a confused un-
wieldy mass with no evident favorable point of attack. This view is
quite natural to those who have had no training and practical experi-
ence in recognizing the "orderly sequence" or laws of environmental
changes, and particularly to those who do not feel that environmental
relations are an essential part of their subject. By many such students
the environment is viewed in a manner compar.'ible to the prevailing
chaotic views on weather before meteorology became a science, or on
taxonomy before Linnaeus, or on geology before Lyell. If one has seri-
ous doubts on this point, he need only turn to the standard treatises

18

on zoology and search for a comprehensive and adequate recognition
and utilization of the orderly and regulatory character of the environ-
ment as an essential part of the subject.

The fallacy of this position has been well expressed as follows by
Brooks ('99) : "I shall try to show that life is response to the order of
nature. . . . But if it be admitted, it follows that biology is the study
of response, and that the study of that order of nature to which re-
sponse is made is as well within its province as the study of the living
organism which responds, for all the knowledge we can get of both
these aspects of nature is needed as a preparation for the study of
that relation between them which constitute life." Later he says : "But
if we stop there, neglecting the relation of the living being to its en-
vironment, our study is not biology or the science of life." No one
seems to have attempted to refute this ; naturally an easier path is fol-
lowed — to ignore it. Perhaps up to the time of the present generation
there has been some excuse for this confusion; l)ut now the respon-
sibility does not rest upon students of the physical and vegetational en-
vironment but upon students of animals, because the former students
have arranged their scientific data in a manner which clearly shows
the orderly lawful sequence of changes in environmental activities.
This should form the basis for a study of the corresponding series of
changes which take place within the animal, and also be the basis for
a study of the reciprocal responses taking place between the animal and
the environment.

In this section an outline will be given of some of the most impor-
tant phases of environmental changes in inland areas viewed as lawful
and orderly, particularly those changes which influence animal hab-
itats.

2. THE DYNAMIC AND GENETIC STANDPOINT

Since Lyell taught the scientific world that a study of processes
now in operation is the key to an understanding of the present as well
as of the past, the process method has been slowly but inevitably pene-
trating to the utmost subdivisions of inquiry. With the progressive
appreciation and use of this method its efficiency has been increased.
Its progress has been the most rapid where the principles of its appli-
cation have been most clearly understood. As models become known
in each field of work others will find the method much easier to apply,
and for this reason it is desirable that such examples become fairly
numerous and wide-spread.

In the application of the process method to an imperfectly under-
stood subject, and particularly to a complex one, it is desirable to con-
sider the subject as a imit or entity. This unit may then be regarded
as an agent whose process of activity is to be studied, for the activity

19

of an agent gives us a process. Thus an organism, a plant society, or
an animal community is a very complex unit or agent, which largely
through chemical energy, under conditions of a normal environment,
responds in an orderly sequence or changes. The environment changes,
the internal conditions of the animal change, and so do the correspond-
ing responses on the part of the animal. When all of these changes
are studied as orderly processes we are able to see the advantage of
this method of study. It is desirable to investigate all phases of animal
responses in this manner, such as growth, development, heredity, etc.,
in order to determine the causes and conditions of this orderly se-
quence. As a rule our recognition of the orderly sequence or laws of
action or succession precedes our knowledge of the causes and condi-
tions of the sequence. This order of sequence is thus of fundamental
importance and must be recognized before it can be investigated or
explained. This method of studying the activity of agents, the char-
acter of their processes, constitutes the dynamic standpoint.

When the dynamic relations of an agent have been investigated, the
orderly sequence of its responses established, and the causes and con-
ditions of its activity determined, it is then possible to explain fully the
origin or genesis of its activities. The genetic method is the study of
origins in terms of the processes involved, and therefore the classifica-
tion of facts genetically implies a knowledge of the processes involved
in their origin. There are thus many degrees or stages in the develop-
ment of a genetic classification, the first step of which is to determine
the orderly sequence of changes. In a certain sense, in its broadest
application, the process method is universal and includes the genetic,
but until their mutual relations become clearly recognized and are gen-
erally understood both should be emphasized.

Particular attention should be called to the fact that the activity of
an agent results in a process, and processes give us the laivs of change.
Many processes are reversible ; that is a process may go forward in
one direction and then become reversed and proceed in the opposite
direction. Other processes are non-reversible, and operate in only one
direction, being in a sense orthogenetic, as in the later stages of the
ontogenetic process.

Let us summarize the main characteristics and principles involved
in the dynamic and genetic method. They have been well expressed
by Keyes ('98), and for my purpose are arranged as follows :

"A truly genetic scheme for the classification of natund phenomena
thus always has prominently presented its underlying principle of
cause and effect. ... To begin with, an adeciuate scheme should be
based directly upon . . . agencies. ... All products must find accu-
rate expression in terms of the agencies. . . . The primary groupmgs

'20

of the . . . processes must be based, therefore, upon the manner in
which these agencies affect the . . . materials. . . . Constructive and
destructive agencies can be recognized only when the phenomena are
made the basis for the scheme. Processes are merely operative. If
coupled with products at all, in classification, all must be regarded as
formative or constructive. The product's destruction, its loss of iden-
tity, is wholly immaterial. The action of agencies is merely to pro-
duce constant change."

Van Hise ('04) has formulated other principles of the process
method as follows :

"The agent is the substance containing energy which it expends in
doing work upon other substances. The substance upon which work is
done may thereby receive energy, and thus become an agent which
does work upon other sul)stances ; and so on indefinitely. Indeed, the
rule is that one process follows another in the sequence of events, until
the energy concerned becomes so dispersed as to be no longer trace-
able. Theoretically this goes on indefinitely. . . . We have seen that
the action of one or more agents through the exertion of force and the
expenditure of energy upon one or more substances is a geological
process. It is rare indeed, if it ever happens, that a single agent works
through a single force upon a single substance. ... If geology is to
be simplified, the proces.ses must be analyzed and classified in terms of
energies, agents, and results. Kach of the classes of energy and agent
should be taken up, and the different kinds of work done by it dis-
cussed. . . . The general work of each of the agents and the results
accomplished should be similarly considered. Not only so, but the
work of the dirt'erent forms that each of the agents takes should be
separately treated. Thus, besides considering the work of water gen-
erally, the work which it does both running and standing must be
treated. The first involves the work of streams; the second, the work
of lakes and oceans. This involves the treatment of streams as enti-
ties. . . . The treatment of the agents will be more satisfactory in pro-
jjortion as the work done by each of the forms of each (jf the agents
is explained under physical and chemical principles in the terms of
energy."

\'iewed fr<Mn this standpoint it is remarkable how many of our
current zoological conceptions are essentially static, and how confused
are our conceptions of the process method. Physiology is supposed
to be devoted solely to processes, yet physiologists use the terms anab-
olism and katabolism, constructive and destructive influences, and,
likewise, zoologists frequently use the expressions "the friends" or
"the enemies" of animals — a dual terminology which has a certain

()

21

utility but which exists mainly on account of the static conceptions of
organic relations.

The dynamic or process concept is a difficult one to attain, and t
apply in all cases, as any one will soon learn if he strives to do this
consistently ; and yet as a scientific ideal there can be no doubt that it
has the same superiority over the older static methods and point of
view that an explanation has over an empirical description.

3. DYNAMIC AND GE;ne;TIC CLASSIFICATION OF ENVIRONMENTS

In the natural history sciences w^e have two main sorts of classifi-
cations of phenomena, those which we call "natural" and those which
we call "artificial." Natural classifications are those in which
the basal criteria are of origin, the method of processes or gen-
esis. A classification of lakes upon the basis of the processes iv/iich
operated in their origin — crustal movements of the earth, the mean-
ders of streams, the work of an ice-sheet, volcanic activity, etc. —
would at the same time furnish an explanation of them in terms of
their origin. Artificial classifications are those in which the criteria
are arbitrarily chosen. Any character may be made the basis for an
artificial classification. Thus lakes may be classified upon the basis of
their size, depth, color of the water, distance from cities, number ot
boats upon them, etc., but such classification would not furnish the
basis for a scientific explanation of lakes. The artificial is often useful
or convenient for a special purpose; the genetic is illuminating from
the standpoint of scientific interpretation. This method may be ap-
plied to any kind of environment, physical, physical and biological
combined, or solely biological. To the degree that the environment i-^
dominated by the physical conditions the laws of physical change and
physical genesis will preponderate in the origin of such environments,
and corresponding relations apply to biological environment*^.

The dependence of the genetic method upon causes atid conditions
makes it impossible to divorce it from the local conditions. This is
at once the strength and weakness of this method, for it is particular,
and generalized averages mean little because origins are different un-
der different conditions ; this is the key to individuality. Thus streams
viewed as stages in the progressive transformation of a lidiiid mediiun
for life, may be formed in many diverse ways, and for this reason tlv:
general principles of the method of genesis mav be exi)ressed most
simply in an ideal case. Genetic series are unending, thcv extend into
the past and will continue in the future. The point of deDarturc for
study must therefore be arbitrarily chosen, and the more nearly a nat
ural basis can be approximated the simpler its application bcronies

90

For this reason a cycle will be followed here which begins with a con-
dition of stress, advances through the process of adjiistiuent to strain,
and reaches a condition of relative eqitilibriiini. The starting point in
such a cycle we will consider as the original conditions, and the later
activities as the derived ones. The original conditions we will assume
to be an uplifted undulating plain, composed of relatively homoge-
neous materials, in a humid climate, and covered by a varied vegetation
including trees. The elevated condition of the land produces a condi-
tion of unstable eqiiilibriuni or stress for the rain falling upon its sur-
face ; and. furthermore, the vegetation will tend to spread over the en-
tire surface, and thus exert a certain pressure also These original
C(MKlitions are, therefore, unstable and destined to change, and mu-
tually to influence and regulate one another.

If we now imagine the rain "turned on" under such conditions,
what are the main processes which will operate? The rain falling in a
depression will be supplemented by that which drains from the eleva-
tions; thus, through the agency of running water, a standing water
habitat will have its origin. With this concentration of water will
come also a burden of debris from the upland ; and in this way the
"constructive" and "destructive" processes will begin at the same time.
Plants will invade such a depression and add their remains. Some of
the depressions will overflow and the outflowing streams will cut down
the outlet to progressively lower levels, and ultimately drain the basin.
On the other hand, inwash and organic debris may tc^gether accumu-
late at such a rate as to raise the level of the basin above ground water
and thus transform the conditions to that of land. The progressive
stages of the process of degradation thus favor the transformation of
the depression and a progressive formation of lakes, which are con-
verted into ponds and swamps and ultimately, with drainage, to dry
land. For depressions we thus get a genetic series which we may call
the lake, pond, and swamp series. This does not classify the depres-
sion series according to size, depth, character of water, etc., as in an
artificial classification, but in the order of their development or genesis
through the agency of running water. Accompanying this sequence
there are of course changes in size, depth, etc., but these are subordi-
nated in the classification to the developmental sequence centering
about the process of the degradation of the land by the agency of run-
ning zvater. This is therefore a classification of environments, not on
the basis of the product, as it might appear from calling it a depression
or standing-water series, but upon the basis of the activity or proc-
esses of tJie dominant agent.

We will assume that all the lakes, ponds, and swamps, due to the
original relief of the land, become drained and constructively con-

23

verted into streams or dry land. Let tis consider the streams, particu-
larly those which did not develop from the lake, pond, and swamp se-
ries, in order to consider them in their simpler conditions of develop-
ment.

The first shower on the new land surface, or the begirininj^ of a
cycle, forms an extensive ramification of small streamlets, their den-
dritic branches flowing down all slopes. With the confluence of the
smaller branches the progressively larger trunks are formed, and with
their increase in volume, cutting progresses; but all traces of this
stream itself tend to vanish soon after the shower is over, although
some water may linger in pools in the deeper depressions. These con-
ditions form an initial stage in the development of the activity of run-
ning water as an animal habitat. These temporary streams are rain
waters intermingled with dust from the air and soil from the ground.
Since, viewed chemically, such waters have not existed as a liquid long
enough to dissolve much gaseous and solid material, they represent a
relatively original condition, or an initial stage in the chemical devel-
opment of the stream as a medium for living animals. Again and
again these showers are repeated, and where there is a slight variation
in the hardness of the substratum small pools are formed on the softer
materials, where erosion is more rapid. In these pools it is possible
for some aquatic or amphibious animals, of marked powers of disper-
sal, to become lodged, or even entrapped, as in the case of animals
which migrate up the stream during its temporary flow; such pools,
in fact, may be reached even by individuals from the ground water.

Finally these temporary streams cut down to ground-water level
and become permanent. Such a stream then, in additi(^n to the fresh
rain-water which it receives with each shower, has a permanent supply
of ground water. This water, having filtered through the soil, con-
tains both gas, particularly COo, and minerals, and thus as a solution
differs much from rain water. The composition of ground water
varies much with the chemical differences of the substratum. Such
water generally contains enough substance in solution to be a favor-
able medium for plant growth, such as algse — aquatic pioneers which
are comparable to the lichens in their invasion upon bare rock. But
the temporary flow of water is still dominant, and will remain so until
the supply of permanent ground water is of such a volume that, hav-
ing a good current, it rushes over the obstacles in its path ; then a per-
manent brook has been evolved, and a permanent rapid-water habitat
has originated.

As the erosion of the stream advances, organic debris not only
multiplies indigenously in the water, but it is also washed and blown
in, and through its decay the composition of the water is changed.

24

particularly in the amount of CO2 present. This gas causes the water
to take into solution a greater amount of lime ; and at the same time
the agitation to which it is subjected while dashing over obstacles or
flowing over falls increases the amount of oxygen present, a process
further aided by the oxygen set free in it by water plants. Carbonic
acid, moreover, is set free by the rapids and falls. It is thus very
evident that the chemical processes are undergoing an important devel-
opment as the stream progresses, since there are going on both the
process of gaseous equilibrium with the air, and an increase of the sol-
ids in solution. The stream is progressively becoming a more favor-
able or enriched culture medium for organisms. The rapidly flowing
water which characterizes the brook is the predominant physical fea-
ture of this environment, the stretches of relatively quiet water which
form the pools, between the more rapidly flowing parts, anticipating
the kind of conditions which are destined to increase with the trans-
formation of the brook conditions into those of a creek. With the
progress of development in drainage a brook is progressively trans-
formed by the processes of erosion into a creek. Here the rapid-water
conditions are more nearly equaled by a corresponding enlargement of
the pool or the quieter stretches of water, where the finer sediments
are deposited and the animals dwelling on the surface film or in the
mud and sand, find suitable conditions. The falls and rapids which
characterize the brook are exceptional in the creek, but may linger
where the rate of change has been very slow on account of the resist-
ance of the substratum. The alternation of rapid and slower water,
which characterizes the creek stage, with the preponderance of the
relativelv rapidlv flowing water, is gradually transformed into that of
a river, where the water flows at a slower rate and rapids and falls
have as a rule become extinct, and where a condition of relative chem-
ical equilibrium has also been reached. Here the burden of coarse
debris is at a minimum, and the surface, sides, and bottom of the
stream, have become differentiated as relatively distinct habitats. With
progressive approach toward baselevel all conditions of the environ-
ment tend to become more stable and equalized until the stream erodes
to tide level, becomes brackish and finally as salt as the sea itself, and
reaches an equilibrium determined by the dominant animal environ-
ment upon the earth — that of the sea.

We have now outlined the developmental sequence of wet depres-
sions, the lake-pond-swamp series, and the running water, the brook-
creek-river series, these two series including the main inland animal
environments in a liquid medium in a humid climate. We have yet to
consider the animal environments of land animals proper, those which
live in the gaseous medium of air. The complexity of conditions upon

25

land is much greater than that in water, either fresh or salt. In othe.-
words the land habitats are the most complex on earth. For simplicity
in handling this involved problem, an ideal series will also be followed,
and instead of attempting to discuss all the principles involved, only
such will be mentioned as may be illustrated by a single example. This
will serve to show the application of the method. We shall consider
the process of degradation of the land, such as might be developed
during a topographic cycle of erosion, and as applied to a snow-capped
conical mountain in a temperate humid region.

Let us consider the series of processes which operate upon such a
mountain. The snow and rain which fall upon it are in unstable equi-
librium, the snow creeps or plunges down the slopes, and the water
flows down. In the zone of ice and snow physical and mechanical
changes preponderate ; but at lower altitudes, with the melting of the
snow and ice, on account of the higher temperature, chemical changes
become more prominent and supplement the mechanical work of run-
ning water. Here, also, plants and animals become an important factor
in modifying the processes of change by hastening or retarding the
processes of degradation. We thus see that on different parts of a
mountain there are important modifications in the processes of degra-
dation. The same general processes which operate to form lakes, ponds,
swamps, brooks, creeks, and rivers, are also at the same time producing
changes in the land habitats. The entire surface of such a mountain is
undergoing change, but because of the concentration of degradativc
progress near its base, particularly on account of the concentration of
the drainage there, ravines and valleys develop here more rapidly and
converge toward the main divide, the mountain top. As these ravines
and valleys enlarge, the mountain is lowered ; and ultimately all is re-
duced to a plain, and to baselevel. The condition of stress which existed
upon the slopes of such a mountain as degradation progressed, became
relatively adjusted at that place, but where the degraded materials
were deposited a stress was becoming citmnlative, and it is this ever
changing adjustment of stresses which makes natural processes unend-
ing.

With the degradation of the mountain, progressively higher zones
are lowered; the snow cap disappears; the region above the tree limit,
and later the lower parts, are spread over a large area, and the moun-
tainous character is largely gone. In this manner and at the same time
as the land is degraded to a lowland by running water, in the water
itself a series of habitats is developing, and thus all the environment
is being transformed, along relati7'ely distinct but mutually interde-
pendent lines, tozvard the same general direction or condition — a rela-

26

tive equilibrium resulting from the balancing of all stresses near sea
level.

In the preceding discussion no emphasis has been placed on the fact
that degradation of the land is only a part of a large cycle of activity,
and that the deposition of the degraded materials may he a cause of so
much stress as to initiate an elevation of the land. If the heavy solu-
ble materials from the land are washed into the sea and only lighter
materials remain behind, the increased stress resulting between the sea
and the land will tend to elevate the lighter areas until an equilibrium
is established between the heavy sea and the lighter land ; therefore,
some crustal movements, at least, may be complementary phases of the
degradation of the land. The elevations and depressions of the sur-
face of the land with regard to the sea level may thus initiate new
cycles of transformation in all environments. These processes do not
need amplification here, although they should be noted ; but this lack
of amplification does not imply a minor influence of this factor. Still
another cycle may be initiated by the processes of vulcanism, a factor
the influence of which is easily overlooked in large parts of the world
but in others is very prominent. Only one more comprehensive physi-
cal factor will be mentioned ; that due to alterations in the atmos-
phere — climatic changes. Although the temperate humid climate has
been made the basis for the preceding discussion, it must be remem-
bered not only that there are other kinds of climates, but that these
undergo transformation or changes from such extremes as the cold
arctic deserts on the one hand, to the dry hot deserts on the other.
Within this great amplitude of climatic possibility is found one of the
greatest causes both of complexity in land environment and of many
local differences in the transformation of habitats.

To simplify this sketch of the operation of the physical features of
the environment the organic factors have been neglected, and these
should now be considered. On account of the ultimate dependence of
animals for food upon vegetation, many intimate relations exist be-
tween plants and animals ; furthermore, in addition to the food rela-
tions there are manv other important ones, such as the physical and
chemical influence of the vegetation upon the soil, its influence upon
the temperature and humidity of the air and on light ; and, finally,
there is qualification of these influences by the different kinds of vege-
tation. A vegetational cover of grass has a very different effect from
one of shrubs or a forest cover; conifers and hard-wood forests differ
in effect also; and the succession of plant societies varies, not only with
different kinds of vegetation but also in different climates, and with
different physiographic conditions. As Cowles ('ii) has shown, there
are several cycles or series of successions of vegetation. Many of

27

these changes are dependent upon physical conditions which are
equally potent in their influence upon animals. Thus physical and
vegetational changes in comhination influence animals directly and in-
directly, and in the conditions due to this fact we find the basis for the
important control which vegetation exerts upon animals.

Animals themselves form an important part of their own environ-
ment, not only in their relation to their own kind, as mates or as prog-
eny, but also as members of an animal community whose members
must adjust their activities to one another through symbiotic, competi-
tive, or predatory relations. If any animal becomes abnormally abun-
dant, that is, more numerous than the conditions can support, this
number in itself becomes a weakness, through the positive attraction
of the organisms (plant and animal) which are able to prey upon it,
and soon the normal abundance is restored. For example, in a conif-
erous forest, bark-beetles (Scolytoidea) may increase to such an extent
that the forest is largely destroyed, and a succession is produced in
the vegetation as the conifers are replaced by a growth of aspen and
birch. As a result of this destruction of the kind of food and habitat
essential for the next generation of beetles, a proper habitat is lacking,
and the restoration of the normal number of beetles is hastened. This
same example also shows how one kind of animal may influence the
character of a whole community by its control over the vegetation.

The influence of man must be looked on from the same standpoint
as one views the activitv of anv other animal ; as that of a member of
an animal community. He hastens and retards the changes in his en-
vironment as do other animals. In general his early methods are pred-
atory ; he reaps where he does not sow ; but later tlie milder competi-
tive and symbiotic relations and the constructive or productive aspect
become more prominent. Civilization is an attempt to make the en-
vironment "to order." but as yet man has not learned how to produce
a permanent "optimum" along the lines of an ecological commiuiity.
As has already been said, to understand man we should view him as
an integral part of an ecological community, as one member of a biotic
community of plants and animals, or at least of an animal community
which includes all animals that are influenced by man — and not con-
sider him, as some students do. as a distinct entity with little regard
to his animal and plant associates.

The main features of the preceding discussion may be summarized
as in the following table.

28

Table 2. — The Genesis and Fokmation of Inland Habitats in a Humid Climate
AND THE Dynamic Status of the Processes

Dynamic status.

Phases in the formation of inland environ-
ments.

I. Unstable equilibrium — condition
of stress or pressure

Original conditions; elevated land area, or
new land surface, or beginning of new
cvrde.

II. Process of adjustment to stress
or strain.

Process of formation of habitats; all hab-
its are constructive.

bd

re

(yj

^

f^

•^

CD

s s a

5^5

que
(de
seri

§1

que
prei

sibl

t-J

ti iT3 B

^ tJ

CD ^ B

55 S3

tn ►I (-S

ri-

■ '£..n,

,— ^ rn

a> (t>

m fX)

O CD

CO

2.0

ft
2.0

B
o

<%%

m ^

«

o si

P

(The following are examples of the

B - »

3 » ^.
3 » B

mding wat
series) ; r

nd habital
reversible

ream habi
es; parti

major processes; :

<3 "-J

^^

-^^

£.1

H< cn

1. The processes of degradation of
the land.

r-f-

CD (D
CD CO

CD 2
en -1
OQ W

CD ■*

7 ?

2. The processes of adjustment to

climate.

B-5.

t-'

r:;

H

y

•2^

rx 2

pr

-*• 3

3. The process of the establish-

1 1

o

P3
B

11

3 P

ment of biotic (plant or animal)

•~^

dominance.

1— i

^

a

a

o

o

"1

B

o
o

era

B

* 5-

r^

a>

r.

O

CD
PT

O

tJ

a

S

^2

pS

— ^

5'

3 —

:i^ t)

(D

S 5-

^§

S.?^

•-i

«•§

Qj

"^ 2.

rt-

I

III. Relative equilibrium.

-^

Derived conditions; lowland area

, old land

surface (baseleveleil to the ni

arine en-

vironnient), end of a cycle.

or domi-

nance; under relatively stable

ponditions.

29

The preceding discussion is based upon the conditions of a humid
cUmate, but the semi-arid and the arid cHmates should also be touched
on. In time, as ecological studies are extended to all kinds of land
areas, it will be possible to formulate all of the general principles ot
the origin or process of development of land habitats; but at present
vast areas of the land have never been observed by a zoologist from a
modern ecological standpoint. Most of the ecological studies of ani-
mals have been carried on in a humid climate, only slight attention
having been given to the ecological relations existing in -an arid cli-
mate, and still less to those in alpine and polar regions. After the
humid regions have been better studied, the arid regions will probably
be the next to be carefully investigated. The plant ecologists, by their
studies in these regions, have already furnished important facts pre-
paring the way for the animal ecologist, because they investigate both
the physical and vegetational conditions upon the prairies and plains
of the West. If the regions of progressively increasing aridity are
examined, there will be found to be a corresponding series of changes
in the animal habitats. The standing- water series of habitats found in
such a series, in contrast with those of humid regions with fresh-water
lakes, ponds, and swamps in addition to the temporary fresh waters,
are alkaline and salt waters, and we find an extensive series ranging
from Great Salt Lake, Salton Sea, and Devil's Lake, to strong briny
pools and alkaline mud flats. These are, of course, as capable of a
genetic treatment as are the corresponding fresh-water bodies of hu-
mid areas. The stream series is also present in the arid region, but it
exists under conditions quite different from those in humid areas. The
through-flowing streams are relatively independent of local conditions
because their main supply of water is from the mountain ; but they are
nevertheless much modified by the character and amount of the burden
which they carry during the time of high water, and they tend to be-
come clogged at low water stages. The chemical composition of such
waters is quite different from that of regions continually leached by
rains. The small streams flowing from the mountains, whose dimin-
ishing volume does not allow them to traverse the arid regions, suc-
cumb, and disappear in the dry earth — examples of a second degree of
dominance of the desert or plains. But the truly characteristic streams
of the arid regions are those primarily dependent upon the desert con-
ditions. Such streams are well within the arid regions and are domi
nated wholly by them. They are solely of a temporary character, and
correspond to the initial stage of stream development, the temporary
stream, in a humid climate. In an arid climate, however, development
does not proceed beyond this early stage, and the degradation and

30

baseleveling of the land is due to the combined influence of water and
the wind.

On land, the movements of the soil by the wind, as in the sand-
dune regions of true deserts, show us a characteristic condition; in a
more humid climate, however, the dunes would tend to become an-
chored by vegetation. Other soils than sand are also blown about. The
extreme of dry desert conditions must be looked upon as the ultimate
or climax condition, a condition of relative equilibrium, under present
climatic conditions, for certain regions. A slight departure from these
extreme conditions is seen in such localities as receive most abundant
showers during the growing season for vegetation. These are able to
influence the development of the drainage only in a minor way, but
they moisten a shallow surface layer of soil and permit the growtii
of short grasses, such as the buffalo-grass (Schantz, 'ii 140). Very
recently another important source of water in the arid regions has
come to be recognized. This, McGee has shown to be the subsurface
or artesian waters which come up from below ; and this is an important
supplementary source of moisture in extensive areas in the arid West
(McGee, '13), where the evaporation is large. It is not unlikely that
even in humid regions where the soils are very sandy, as upon the
Coastal Plain, and where the strata dip in such a manner as to favor
an underflow of water, this supply may be of considerable importance
to the biota. With a greater rainfall during the growing season, per-
mitting a relative humidity greater than on the short-grass area of the
plains, a deeper-rooted vegetational cover gives us the long prairie
grasses of the eastern prairie.

As soon as the physical conditions permit a growth of vegetation
this material becomes an environmental factor which reflexly modifies
the physical conditions of the air, the soil, and the animal habitat. This
is shown to a marked degree in the humid area of the southeastern
United States, where the rainfall, greater than that on the arid plains
and prairies, favors the development of a forest cover. Such a forest
not only tends to retard evaporation but also acts as a sponge and by
its vegetable debris and loose soil retards the run-off. In this manner
not only are land habitats influenced, but this conservation of moisture
tends to prolong the duration of temporary streams, and to stabilize
the flow of permanent ones ; and, further, through the same influence,
the ground-water level declines slowly, and bodies of standing water
are also influenced. Thus all the more important habitats are to some
degree regulated and made more stable by a forest cover.

The foregoing discussion and examples, selected from the activi-
ties of animals and changes in their environments, are varied enough to

31

show how diverse are the appHcations of the process method to inves-
tigation. The general idea is easily grasped, but to make the dynamic
method a regular habitual procedure in investigation is truly difficult,
so difficult, indeed, that there is reasonable ground for doubting if this
method can be mastered without a practical application of it to a con-
crete problem, at the same time giving special attention to the method
of procedure.

REFERENCES TO LITERATURE

Adams, C. C.

'04. On the analogy between the departure from optimum vital

conditions and departure from geographical life centers.

Science, n. s., 19:210-211.

'13. Guide to the study of animal ecology. 183 pp. New York.

(This book contains numerous references to the literature bearing upon the
subject of this article.)

Bancroft, W. D.

'11. A universal law. Science, n. s., 23: 159-179.

Blackman, F. F.

'05. Optima and limiting factors. Ann. Bot. 19: 281-295.

Blackman, F. F. and Smith, A. M.

'11. Experimental researches on vegetable assimilation and res-
piration. IX. On assimilation in submerged water-plants,
and its relation to the concentration of carbon dioxide and
other factors. Proc. Royal Society, B., 83: 389-412, 1910.

Brooks, W. K.

'99. The foundations of zoology. 339 pp. New York.

Cowles, H. C.

'11. The causes of vegetative cycles. Bot. Gaz., 51: 161-183;
also, Ann. Assoc. Amer. Geogr., i : 1-20. 1912.

32

Jennings, H. S.

'06. Behavior of the lower organisms. 366 pp. New York.

Keyes, C. R.

'98. The genetic classification of geological phenomena. Journ.
Geol., 6:809-815.

McGee, W J

'13. Field records relating to subsoil water. U. S. Dept. Agr.,
Bur. Soils, Bull. 93. 40 pp.

Powell, J. W.

'95. Physiographic processes. Nat. Geogr. Monographs, i : 1-32.

Schantz, H. L.

'11. Natural vegetation as an indicator of the capabilities of land
for crop production in the Great Plains area. U. S. Dept.
Agr., Bur. Plant Industry, Bull. No. 201. 100 pp.

Van Hise, C. R.

'04. The problems of geology. Journ. Geol., 12 : 589-616.

The New York State College of Forestry, Syracuse, N. Y.

Bulletin

OP THE

Illinois State Laboratory

Olf

Natural History

Urbana, Illinois, U. S. A.

STEPHEN A. FORBES, Ph.D., L.E.D.,
Director

Vol. XL September, 1915 Article IL

AN ECOLOGICAL STUDY OF PRAIRIE AND FOREST
INVERTEBRATES

BY

Charles C. Adams, Ph.D.

ORIGINAL FOREST AND PRAIRIE AREA IN ILLINOIS
(after brendel anb barrows )

Bulletin

OK THB

Illinois State Laboratory

OK

Natural History

Urbana, Illinois, U. S. A.

STEPHEN A. FORBES, Ph.D., L.L.D.,
Director

Vol. XI. September, 1915 Article H.

AN ECOLOGICAL STUDY OF PRAIRIE AND FOREST
INVERTEBRATES

BY

Charles C. Adams, Ph.D.

CONTENTS

Page

Introductory 33

General description of the region and location of the ecological stations 35

I. General description of the region 35

II. The ecological stations 38

Description of the prairie habitats and animals 40-56

I. Prairie area north of Charleston, Station 1 40

1. Colony of swamp grasses (Spartina and Elymus), Station I, a. . 41

2. Colony of wild rye, Elymus virginicus submuticus, Station I, c. . 43

3. "Wet area of swamp milkweed (Asclepias incarnata), Station l,d 44

4. Cone-flower and rosin-weed colony, Station I, e 48

5. Colony of blue stem (Andropogon) and drop-seed (Sporohulus),

bordered by swamp milkweed, Station 1, g 49

6. Supplementary collections from Station 1 52

II. Prairie area near Loxa, Illinois, Station II 52

III. Prairie area east of Charleston, Station III 55

Description of the forest habitats and animals 56-66

1. The Bates woods, Station IV 56

2. The upland oak-hiekory forest, Station IV, a 57

3. Embarras valley and ravine slopes, forested by the oak-hickory

association. Station IV, fc 59

4. Lowland or * ' second bottom, ' ' red oak-elm-sugar maple wood-

land association, Station IV, c 62

5. Supplementary collections from the Bates woods. Station IV. ... 65

6. Small temporary stream in the south ravine, Station IV, d 65

General characteristics of the gross environment 66-102

1. Topography and soils of the State 66

2. Climatic conditions 67

3. Climatic centers of intluenee 69

4. Eelative humidity and evaporating power of the air 71

5. Temperature relations in the open and in forests 83

6. Soil moisture and its relation to vegetation 86

7. Ventilation of land habitats 88

8. The tree trunk as a habitat 91

9. Prairie and forest vegetation and animal life 91

10. Sources and role of water used by prairie and forest animals. ... 98

Animal associations of the prairie and the forest 102-158

I. Introduction 102

II. The prairie association 103

1. Swamp prairie association 103

2. The Cottonwood community 105

3. Swamp-grass association 107

4. Low prairie association 108

5. Upland prairie association 109

6. The Solidago community 109

7. Dry prairie grass association Ill

8. A milkweed community 112

III. Relation of prairie animals to their environment 113

1. The black soil jarairie community 114

2. The prairie vegetation community 117

4. Interrelations within the prairie association 119

Page
IV. The forest associations 122

1. Introduction 122

2. Dry upland ( Querents and Carya) forest association 124

3. Artificial glade community in lowland forest 125

4. Humid lowland (hard maple and red oak) forest association. . . . 126

5. Animal association of a temporary stream 127

V. Eelation of the deeiiluous forest invertebrates to their environment. . . . 128

1. Forest soil community 129

2. The forest fungus community 135

3. The forest undergrowth community 138

4. The forest crown community 139

5. The tree-trunk community 142

6. The decaying wood community 148

7. Interrelations within the forest association 157

Ecologically annotated list: —

I. Prairie invertebrates 158-201

II. Forest invertebrates 201-238

Bibliography 239-264

Article II. — An Bcological Study of Prairie and Forest Inverte-
brates. By Charles C. Adams, Ph.D.

Introductory

In four generations a true wilderness has been transformed into
the present prosperous State of IlHnois. This transformation has been
so complete that in many parts of the state nearly all of the plant and
animal life of the original prairie- and forest has been completely ex-
terminated. Between the degree of change which has taken place in
any given area and the suitability of that area for agriculture there has
been an almost direct relation. Fortunately, however, for the preser-
vation of prairie and forest animals, the state is not homogeneous,
some areas being too hilly, rocky, or sandy for prosperous agriculture.

The character and mode of transformation w^hich has taken place
in the past is instructive in several particulars because it serves to
guide our anticipations as to the future of our fauna. The forested
southern part of the state (see frontispiece) was first invaded by trap-
pers and hunters, who began the extermination of the larger animals.
These invaders were in turn followed by others who, with the round
of the season, were hunters or farmers, and continued this exterminat-
ing process, particularly in the clearings, which began to replace the
forest. These pioneers, men of little wealth, possessed a combination
of mental and economic habits which was the result of life in a for-
ested country, and naturally they settled in those places most like their
former homes — within the forest or near the forest margin. From
these settlements they looked out upon the prairies as vast wastes to
be dreaded and avoided. As a result of this attitude toward the prai-
ries, it required some time, even a new generation, some economic
pressure, and a change of habits before the prairies were settled. Mean-
while the northern part of the state was yet a wilderness; but through
the influence of the Great Lakes, as a route of communication with
the populous East, a rapid invasion of settlers set in from that direc-
tion. Though these settlers also came from a wooded country, they
were more wealthy, settled upon a very fertile soil which was favorably
located with regard to eastward communication, and they therefore
progressed more rapidly than the less favored, more isolated southern
invaders on the poorer soil ; consequently they spread from the forest

34

to the prairie more rapidly than did the settlers in the South. There
thus developed two active centers of influence, each of which trans-
formed the primeval conditions in the same manner and in the same
direction toward an environment suitable for man.

The forests and the: upland prairie were first changed. Then the
fertile wet prairie was drained, so that today it has largely become
either the hilly and rocky areas that survive as forests or the low
periodically flooded tracts, and the undesirable sand areas which simi-
larly preserve patches of sand prairie. All the changes are more
rapid and complete upon fertile soil than upon the poorer soils in the
southern part of the state.

Such considerations as these will aid one in estimating the probable
rate of future changes in different parts of the state, and will serve to
show in what parts there is urgent need of local studies if ecological
records are to be made before extinction of some forms is complete.

A study has been made with the idea of reporting upon represen-
tative patches of prairie and forest in a manner which would aid others
in making similar local studies, and would at the same time preserve
some records of the present condition of the prairie and forest. When
this work was planned, we had no general or comprehensive discussion
of the conditions of life upon the prairie and in the forest. For this
reason a general summary of these conditions and a sketch of the gen-
eral principles involved are given, so that the reader may gain some
conception of the relation of the local problems to those of a broader
and more general character.

A section for this report was prepared giving general directions
for making such local studies, but later it was decided to publish this
separatelv, in somewhat extended form, as a "Guide to the Study of
Animal Ecology."* This volume should be regarded as intimately re-
lated to this paper, and this report should at the same time be consid-
ered as a concrete example of the procedure suggested in that "Guide"
for ecological surveys. It will be observed that the study of the
Charleston area here referred to has been conducted in much the same
way as was my cooperative study of Isle Royale, Lake Superior, en-
titled "An Ecological Survey of Isle Royale. Lake Superior" ('09),
although certain aspects have been elaborated here which, for lack of
time, were not treated there. The time devoted to the study of the
Charleston area was also limited, but in the preparation of the report
upon it use has been made of many years' experience and a general
knowledge of the prairie and forest. Without such a background

*The Maemillan Co. 1913.

35

much greater caution would have been necessary In discussing many
phases of the problem.

ACKNOWLEDGMENTS

The study of the Charleston area was carried out with the coop-
eration of the Illinois State Laboratory of Natural History, through
its director, Prof. Stephen A. Forbes, and with the further coopera-
tion of Professors E. N. Transeau and T. L. Hankinson, of the East-
ern Illinois State Normal School, located at Charleston. Personally I
am indebted to Professor Forbes for the opportunity of taking part in
this study as the State Laboratory representative, and for the aid he
has given in the illustration of the report. To Professor Transeau I
am particularly indebted for the plant determinations, for lists of the
plants, and for evaporation data. To Professor Hankinson I am
under especial obligation for many specimens, which materially added
to my lists, and for a large number of photographs. I am indebted
likewise to my associates in this study for their hearty cooperation
throughout the progress of the work.

For the determination of entomological specimens I am indebted
primarily to Mr. C. A. Hart, Systematic Entomologist of the State
Laboratory of Natural History, who named most of the insects col-
lected. For the names of certain flies I am indebted to Mr. J. R. Mal-
loch, of this Laboratory. Others who have determined specimens are
as follows: N. Banks (Phalaugiida), J. H. Emerton (spiders), R. V.
Chamberlain (myriapods), F. C. Baker (Molkisca), Dr. W. T. M.
Forbes (lepidopterous larvae), Dr. M. C. Tanquary (ants). Dr. M. T.
Cook (plant galls), J. J. Davis {Aphididcc) , and Dr. A. E. Ortmann
(crawfishes collected by T. L. Hankinson). I am indebted to the U.
S. Geological Survey for photographs. Acknowledgments for illus-
trations are made under text figures and in explanations of plates.

GENERAL DESCRIPTION OF THE REGION AND LOCATION
OF THE ECOLOGICAL STATIONS

I. General Description of the Region

The town of Charleston, Coles county, Illinois, in the vicinity of
which these ecologic studies were made, is situated on the Shelbyville
moraine which bounds the southern extension of the older Wisconsin
ice-sheet. To the south of this moraine lie the poorer soils which char-
acterize so much of southern Illinois ; to the north, upon the older Wis-
consin drift, are some of the most productive soils found in the upper

36

Mississippi Valley. The economic, sociologic, political, and historical
significance of the difference in the soils of these regions is funda-
mental to any adecjiiate understandng of man's response to his ecolog-
ical environment within this area. Some of the results of this differ-
ence have long been known, but it is only in recent years that their
general bearing has been adequately interpreted in terms of the en-
vironment. Hubbard ('04) was the first, I believe, to show the sig-
nificance of this difference in soils and its influence upon local eco-
nomic problems. That such an important influence should affect one
animal (man) and not others seems very doubtful, and yet in only one
other case do we know that the lower animals respond to this ecologic
influence. Forbes ('07b) has shown that certain kinds of fish found
in streams on the fertile soils are wanting in streams on the poorer
soil. To what degree the land fauna and the native vegetation respond
to this distinction is not known, as this subject has not been investi-
gated except agriculturally. Here, then, is a factor in the physical
surroundings which should be reckoned with in any comprehensive
study of the biotic environment. In this portion of the state, on ac-
count of the differences in the soil, the physical environment is prob-
ably more favorable to certain organisms and less favorable to others,
and consequently, to a certain degree, the environment selects, or fa-
vors, some organisms. Through their activities and through other
agencies of dispersal, the animals along the borders between the two
soil types transgress these boundaries, and are therefore forced to
respond to the new conditions and to adjust themselves, if possible.

But the soil is not the only environmental influence which has pro-
duced an unstable zone or tension line in this area. A second factor is
the difference in the vegetation — the difference between the forest and
the prairie. In all probability. Coles county was at one time all prairie,
but the Kaskaskia and Embarras rivers, as they cut their valleys
through the moraine and developed their bottoms, have led forests
within the morainic border from farther south. The forests about
Charleston have extended from the Wabash River bottoms. On account
of the southerly flow of the Embarras through this county, the forest
and prairie tension line is about at right angles to that produced by the
differences in the soil. The forests have tended to spread east and west
from the streams and to encroach upon the prairie, and thus to restrict
its area more and more. The fundamental significance of the tension
between the forest and the prairie has long been known within the
state. It influenced its economic, social, political, and historic develop-
ment as much as any other single factor during its early settlement.
And just as Hubbard ('04) has shown the influence of soil upon man

37

within the state, so also has Barrows ('lo) shown the influence of the
forests and prairie upon the state's development. While the influence
of the soil upon the animal life of the state is not so well known or es-
tablished, the influence of prairie and forest upon the animals is univer-
sally recognized, even though the subject has been given relatively
little study by naturalists.

A third leading agency is the influence of man, who has trans-
formed the prairie and forest to make his own habitat. There are thus
recognized in the Charleston region three primary environmental in-
fluences : first, the relative fertility of the soil (this depending on the
geological history) ; second, the kind of vegetable covering, whether
prairie or forest (this probably depending largely on climatic condi-
tions) ; and third, the agency of man. The general background of the
Charleston region, then, ecologically considered, depends on the com-
bined influence of five primary and secondarv agencies, four of which
we may call natural and one artificial. All these are different in kind
and so independent that they tend toward different equilibria or dif-
ferent systems of unity. Two of these are due to differences in the
soil, two others to the character of the vegetation (whether prairie or
forest), and the fifth, or artificial one, is due to man. Though the
present report does not undertake to include all the problems centered
here, as any complete study would, it is desirable to see the relation of
our special study to the general problems of the region as a whole.

The undulating plain about Charleston, formed as a terminal mo-
raine, is broken along the small streams by ravines, which have cut a
few hundred feet below the general level of the region as they ap-
proached the larger drainage lines. The main drainage feature is the
Embarras River, which flows southwest about two to three miles east
of Charleston, in a narrow valley partly cut in rock. The wooded
areas are mainly near the streams; the remainder of the area is under
intensive cultivation.

During the preliminarv examination of the region, which was made
to aid in selecting representative areas for study, it soon became evi-
dent that the only samples of prairie which could give any adequate
idea of the original conditions were those found along the different
railway rights-of-way. Other situations, vastly inferior to these and
yet a valuable aid in the determination of the original boundaries of
the prairies, were the small patches or strips along the country roads.
Most of the patches of prairie along the railway tracks represent the
"black soil" type of prairie, which is extensively developed in this part
of the state upon the "brown silt loam" soil" (see Hopkins and Pettit,
'08 : 224-231). Much of the region studied was originally wet prairie

38

(which has since been drained), but some of the higher ground,
formed by the undulation of the surface and surrounded by the black
soil, is lighter in color and is well drained. Thus in the black soil areas
there are both wet and well-drained tracts, and corresponding differ-
ences in the habitats.

The originally wooded and the present wooded areas east of
Charleston, in the vicinity of the Embarras River, are in a region quite
different from the prairie both in topography and in soil. Here the re-
lief is much more pronounced, on account of both the proximity of the
river and the greater development of the drainage lines, which have cut
a few hundred feet below the general level of the country. The tribu-
tary valleys and ravines are numerous and steep-sided, and in general
are wooded, the density varying with the amount of clearing done.
Most of the soil of the wooded areas and along the bluffs is distinctly
lighter in color than that of the black soil prairie, and is presumably
"gray silt loam" (Hopkins and Pettit, '08 : 238-242), though along the
flood-plain and the river bottom the soils are mixed in character.

II. The Ecological Stations

In the study of an area or an animal association of any considera-
ble size two methods are available. One is to examine as much of the
area as is possible and secure data from a very wide range of condi-
tions. This method is useful in obtaining the general or broad features
of a region or an association, though to a corresponding degree it must
ignore local influences and details, and by it most of the previous stud-
ies upon prairie animals have been made. It seemed, therefore, that in
the present study a somewhat more intensive method was desirable,
particularly in view of the fact that the extinction of prairie and for-
est is rapidly progressing. The method followed was to examine a
large area in order to select a representative sample, and upon the
basis of this sample to make as intensive a study as time and circum-
stances would permit. This method has the advant^ige of making it
possible to preserve at least some record of the local details ; and at the
same time, to the degree that the selected area is a true sample, it also
gives the results a much wider application.

The prairie samples examined were all along the rights-of-w'ay,
and the forest was a second-growth woods on the bottoms and bluff
of the Embarras River, on a farm belonging, at that time, to Mr. J. I.
Bates. Practically all of the observations here reported upon were
made during August, 1910. The forest is a modified one, but it ap-
pears to have been cut over so gradually that its continuity as a forest
habitat was not completely interrupted, although the cutting has prob-

39

ably seriously influenced many animals, particularly those which fre-
quent mature forests, abounding in dead and dying trees and with an
abundance of logs upon the ground in all stages of decay. Such con-
ditions are the cumulative product of a fully mature climax forest. Of
course the conditions have also been influenced by the extinction, or
reduction in the number, of the original vertebrate population of the
forest.

The different localities or regions examined are, for brevity and
precision, indicated by Roman numerals; the particular minor condi-
tions, situations, or habitats, by italic letters. An effort has been made
to indicate the location of the place studied with enough precision to
enable students to re-examine the habitats at any future time (PI. I).
The photographs which accompany this report may also aid in locat-
ing the places studied. Had similar photographic records been
made fifty years ago, they would have been of much value and inter-
est to us in this study, in much the same way as fifty years hence this
report will form a part of the very limited record of the conditions
found at the present time.

List of Ecological Stations, Charleston, Illinois, August, ipio

Station I. Prairie along the right-of-way of the Toledo, St. Louis and
Western, or "Clover Leaf" R. R., between one and two miles north
of Charleston: Section 2, Township 12 N., Range 9 E., and S. 35,
T. 13N., R. 9 E. (PL I.)

a. Cord or Slough Grass (Spartina) and Wild Rye (Elymus) Asso-
ciation. At mile-post marked ' ' Toledo 318 miles and St. Louis 133
miles": S. 2, T. 12 N., R. 9 E.

b. Couch Grass (Agropyron smithii) Association. The distance of
two telegraph poles north of Station I, a, and west of the railway
track: S. 2, T. 12 E., R. 9 E.

c. Wild Rve (Elymus) Association. East and north of the "Yard
Limits"^ sign: S. 2, T. 12 N., R. 9 E. (PI. II, Fig. 1.)

d. Swamp Milkweed (Asclepias incarnata) Association. North of
first east-and-west cross-road north of Charleston ; east of railway
track : S. 35, T. 13 N., R. 9 E. A wet area. (PI. II, Fig. 2 ; PI. Ill,

Fig. 1.)

e. Cone-flower (Lepacliys pinnata) and Rosni-weed (Silplimm tere-
U7itliinaceu7n) Association. Just north of the preceding Station;
east of railway track: S. 35, T. 13 N., R. 9 E. (PI. V.)

/. Couch Grass (Agropyron smitliii) Association. West of radway
track : S. 35, T. 13 N., R. 9 E. Moist area.

g. Prairie Grass (Andropocjon furcatus and A. virgimcus and Si)0-
roholus cryptandrus) Association, bordered by Swamp Milkweed
(Asclepias incarnata) and Mountain Mint (Pycnantliemum flex-

40

uosum). This formed the north boundary of the area studied:
S. 35, T. 13 N., R. 9 E. (PI. Ill, Fig. 2 ; PI. IV, Fig. 1 and 2.)

Station II. Prairie area west of Loxa, Illinois. Right-of-way along the
Cleveland, Cincinnati, Chicago and St. Louis, or "Big Four,"
R. R.: Sections 10 and 11, Township 12 N., Range 8 E.

a. From one half mile west of Loxa west to near Anderson Road, to
telegraph pole No. 12330: S. 11, T. 12 N., R. 8 E. (PI. VI. and
VII.)

h. Prairie at Shea's: S. 17, T.12 N., R. 8 E.

c. Cord Grass (Spartina) Association. East of Shea's: S. 17, T.12
N., R.8 E.

Station III. Prairie east of Charleston. Right of way along the C. C. C.
& St. L. R. R. : S. 12, T. 12 N., R. 9 E. ; S. 5, 6, and 7, T. 12 N.,
R. 10 E.

a. Rosin-weed (SUpliium terehintJiinaceum) Association. Just west
of the place where the Ashmore Road crosses the Big Four track;
about one mile east of Charleston : S. 12, T. 12 N., R. 9 E.

h. Blue Stem (Andropogon) and Rosin-weed (SUpliium. terehintJiina-
ceum) Association. Three fourths of a mile east of the crossing of
the Ashmore Road and the Big Four track : S. 6 and 5, T. 12 N.,
R. 10 E. An area which grades from prairie into transitional for-
est conditions. (PI. VIII and IX.)

Station IV. Bates Woods. On the east bluffs and bottom of the Embar-
ras River, north of where the Cleveland, Cincinnati, Chicago and
St. Louis, or Big Four, R. R. crosses the river. On the farm of
J. I. Bates: S. 5, T. 12 N., R. 10 E. (PI. X, Fii,'. 1 ; PI. XI, XII,
and XIII.)

a. Upland Oak-Hickory Association (Quercus alha and Q. velutina,
and Carya alha, C. glahra, and C. ovata.) Second-growth forest.
(PI. XII and XIII.)

h. Embarras Valley and Ravine Slopes, with Oak-Hickory Associa-
tion.

c. Red Oak (Quercus ruhra), Elm (TJlmus americana), and Sugar
Maple (Acer saccliarum) Association. Lowland or "second bot-
tom, ' ' Embarras Valley. (PI. XIV ; XV ; and XVI, Fig. 1 and 2.)

d. Small streamlet in South Ravine. This formed the southern bor-
der of the area examined. A temporary stream. (PI. XVII, Fig.
1 and 2.)

DESCRIPTION OF THE PRAIRIE HABITATS AND ANIMALS
I. Prairie Area North of Charleston, Station I

This area includes patches or islands of prairie vegetation oc-
curring along the right-of-way of the Toledo, St. Louis and West-

41

ern, or "Clover Leaf," Railway, north of Charleston. The south-
ern border began just beyond the area of numerous side tracks and ex-
tended north of the first east and west cross-road for a distance of
about one mile, to the place where the right-of-way is much narrowed
and fenced off for cultivation. This is a strip of land through the level
black soil area, which was originally composed of dry and wet prairie.
The higher portions have a lighter colored soil, and the lower parts
have the black and often wet soil which characterized the original
swamp or wet prairie. The railway embankment and the side drain-
age ditches have favored the perpetuation of patches or strips of these
wet habitats ; the excavations for the road-bed, on the other hand, have
accelerated drainage of the higher grounds. The soil taken from these
cuts and heaped up on the sides of the tracks reinforces the surface
relief noticeably in a region which is so nearly level. Through the
depressions fillings have been made in building the railway embank-
ment, and as a result the drainage has been interfered with in some
places.

The disturbances brought about by railway construction and main-
tenance have greatly modified the original conditions, so that the
prairie vegetation persists usually only in very irregular areas, some-
times reaching a maximum length equal to the combined distance be-
tween three or four consecutive telegraph poles — these poles are gen-
erally about 200 feet apart. In breadth the area is usually less than
the space between the ditch bordering and parallel to the road-bed or
embankment and the adjacent fence which bounds the right-of-way, or
about 40 feet. This entire right-of-way is about 100 feet wide. In
addition to these changes in the physical conditions, a large number of
weeds not native to the prairie have been introduced, opportunities for
this introduction being favorable, as railways traverse the entire area.
In general, attention was devoted solely to the areas or colonies of
prairie vegetation and their associated invertebrate animals, the areas
of non-prairie vegetation being ignored, not as unworthy of study, but
because the vanishing prairie colonies required all the time available.

I. Colony of Szvamp Grasses (Spartina mid Blymus) , Station I, a

This colony of slough grass (Spartina tnichauxiana) and wild rye
(Blymus) is located a short distance north of the "Clover Leaf" switch
tracks and just south of the telegraph pole marked "Toledo 318 miles
and St. Louis 133 miles." The length of this colony was about 40
paces.

During August, 19 10, it was dry, but probably in the spring and
early summer, rains make this area a habitat for swamp grasses.

Though it was an ahiiost pure stand of slough grass, with this were
mixed a few plants of wild rye (Blynius virginicus sitbninticiis and B.
canadensis). These grasses reach a height of abou: four feet. The
ground was very hard and dry, and there were large cracks in it.
A single collection of animals was made here, No. 179.

Common Names

Common Garden Spider

Ambush Spider

Differential Grasshopper, adult

and nymphs
Red-legged Grasshopper, adult

and nymphs
Texan Katydid
Meadow Grasshopper

Dorsal-striped Grasshopper
Black-horned Meadow Cricket
Four-spott-ed White Cricket
Ground-beetle
Sciomyzid fly

Scientific Names

Argiope aurantia
Misiimena aleatoria

Melanoplus differ entialis

Melanoplus femur-rubrum

Scudderia texensis

Orchelimum vulgare, adult, and

nymphs of vidgare or glaherri-

mum.
Xiphidium sirichnn
Qlcanthiis nigricornis
QEcanfJius quadripimctatus
Leptotrachelus dorsalis
Tctanoccra plumosa

The basic food-supply in such a habitat is of course the grasses, and
this fact fully accounts for the presence of large numbers of individ-
uals which feed upon grasses, as do the Orthoptera in general. But
the Orthoptera listed are not exclusively vegetable feeders, for Forbes
('05: 147) has shown that XipJiidiimi strictum feeds mainly upon in-
sects, chiefly plant-lice, as well as upon vegetable tissues, including fun-
gi and pollen; Orchelimum vidgarc (p. 144), largely upon plant-lice
and other insects; and CEcanthus qiiadripunctatus (p. 220), upon plant
tissues, pollen, fungi, and plant-lice. These observations were based
upon a study of the contents of the digestive tract. The food of the
sciomyzid fly is unknown. The garden spider lives exclusively upon ani-
mal food; and being abundant, it must exert considerable influence
upon other small animals. It not only destroys animals for its food, but
many others are ensnared in its web and thus killed. In one of the
webs I found a large differential grasshopper. The rank growth of
vegetation furnishes the necessary support for the webs of this spider.

Some of the insects, as Melanoplus differentialis and M. femur-
ruhriim, oviposit in the soil, but others — Scudderia texensis, Xiphid-
ium strictum, Orchelimum vidgare, and CEcanthus — deposit their

43

eggs in stems of plants or under the leaf-sheaths of grasses (Forbes,
'05: 143, 145, 148, 216). The mode of oviposition in these Orthop-
tera raises the question whether or not they are able to pass their com-
plete life cycle within this habitat. Are the species which oviposit in
the soil able to endure submergence during the wet season of the year,
or must they each year re-invade this habitat from the more favorable
adjacent regions? The sciomyzid fly is a regular inhabitant of such
situations, for an allied species, Tetanocera pictipcs Loew, has been
found by Needham ('01 : 580) to be aquatic, breeding on colonies of
bur reed (Spargmmim) , and Shelford ('13a: 188, 284) also finds
pUiuiosa in wet places.

The flower spider, Misumena, captures its prey direct, frequenting
flowers where its prey comes to sip nectar,

With more perfect drainage the character of this habitat would
change ; a more varied growth of vegetation would probably devel-
op ; and the relative abundance of the various kinds of animals would
also change. The present imperfect drainage is more favorable to the
accumulation of vegetable debris than if the habitat was connected
with a stream which could float it away. The periodical drying hastens
decay, and the deep cracks in the soil become burial places for various
kinds of organic debris.

2. Colony of Wild Rye, Blymus zirginiciis submitticus, Station I, c*

Wild rye is a swamp grass. This colony was located about half a
mile north of the colony of slough grass (Station I, a) and about 222
feet south of the first east and west cross-road north of Charleston.
For a/ general view of this grassy habitat see Figure i, Plate II. In
length this habitat extends about one third the distance between two
consecutive telegraph poles, or about 65 feet. The conditions of the
habitat are in general similar to those in the colony of Spartina. The
black soil was very dry and much cracked when examined, late in Au-
gust. Though a few plants of Asclepias siillivantii grew here among
the grass, it was a dense, almost pure stand of wild rye, which reached
a height of about three and a half feet.

Only a very few collections were made here, and these were for
the sole purpose of determining the general composition of the asso-
ciation.

These collections, Nos. 153, 180, and 181, were as follows:

*Aiiimals were not studied at Station I, 6, and therefore the location will not be
discussed here.

No.

1 80

No.

180

No.

180

No.

181

44

Common Garden Spider Argiope aurantia No. 153

Differential Grasshopper Mela no plus differentialis

Red-legged Grasshopper Melanoplus femur-rnhrum

Dorsal-striped Grasshopper Xiphidiiim strictum
Meadow Grasshopper Orchelimum vnlgare, adult,

and nymphs of vnlgare

or glaherrimum
Texan Katydid Scudderia texensis

These are all abundant species. O. vnlgare, by its persistent fid-
dling, is noticeable in all such grass spots during hot sunny weather.
A live differential grasshopper was found in the web of the garden
spider. A comparison of the two colonies of swamp grasses, Spartina
and Blymns, will probably help to give one a general idea of the kind
of invertebrates which were abundant in the original swamp-grass
area of this vicinity. It will be noticed that grass and grass eaters are
the dominant species, and that upon these a smaller number of preda-
ceous animals depend. The characteristic species are the Ortlioptera
and the garden spider. This spider, on account of its predaceous hab-
its, is able to live in a great variety of open situations, but does not
normally live in dense woodlands.

3. Wet Area of Szvamp Milkzveed (Asclepias incarnata), Station I, d

This colony of swamp milkweed was about one eighth of a mile
north of the east and west cross-road. This flat, poorly drained black-
soil area, about 80 feet long, was wet throughout August, crawfish
holes being abundant (PI. IIIA, fig. 2; PI. IIIB, figs, i, 2). To
the east, beyond the boundary fence, in the adjoining corn field, stood
a pool of water surrounded by a zone of yellowish weakened corn,
visited occasionally by a few shore birds. Along the east side of the
newly formed railway embankment (PI. Ill, fig. i) is a shallow
trench containing water and a growth of young willows (Salix) and
cottonwoods (Popnlns deltoides), also blue flags flris versicolor),
bulrush (Scirpns), and sedge (Carex). The characteristic plants
over this area were the abundant swamp milkweed (Asclepias incar-
nata, PI. IIIA, fig. I ) and Bidens. A few plants of water horehound
(Lycopiis) and dogbane (Apocyniim medium) were present, and many
individuals of a low plant with a winged stem (Lythrnni alatuni).

The collections (Nos. i, 12, 13, 14, 15, 18, 32, 37, 45, 156, and
157) of animals taken here were as follows:

45

Pond snail

Prairie Crawfish

Garden Spider

Ambush Spider

Chigger

Nine-spot Dragon-fly

Stink-bug

Small Milkweed-bug

Large Milkweed-bug

Ambush Bug

Tarnished Plant-bug

Soldier-beetle

Black Flower-beetle

Four-eyed Milkweed-beetle

Milkweed-beetle

Leaf-beetle

Dogbane Beetle

Celery Butterfly

Philodice Butterfly

Idalia Butterfly

Milkweed Butterfly

Honeysuckle Sphinx

Giant Mosquito

Giant Fly

Honey-bee

Pennsylvania Bumblebee

Bumblebee

Bumblebee

Carpenter-bee

Rusty Digger-wasp

Galba iiinhilicata

i8

Cambarns gracilis

Argiope aurantia

Misiimena aleatoria

157

Tromhidium sp.

%j f

Libclliila pulchella

Biischistiis variolarius

12

Lygccus kalmii

12

Oncopeltiis fascia tits

I

Phymata fascia ta

12

Lygiis pratcnsis

12

Chauliognathus pcnnsylvaniciis

156

Hnphoria scpiilchralis

156

Tetraopes tetraoph thalmus

12

Tetraopes femoratiis (?)

I

Diahrotica atrip ennis

I

Chrysochus auratiis

14

Papilio polyxenes

15. 45

Bitrymus philodice

12

Argynnis idalia ■

33

Anosia plexippiis

Hemaris diffinis

32

PsoropJiora ciliata

13

Mydas clavatus

12

Apis mellifera

Bonihus pennsylvanicus

155

Bonibus f rat emus

12

Bombus separatus

12, 157

Xylocopa virginica

1,156

Chlorion iclinciinwncutn

12

The soft, wet, black soil contained large numbers of crawfish holes,
and from several of them T. L. Hankinson dug specimens of Cambarns
gracilis. Frogs (Rana) were seen but none were secured. A Caro-
lina rail was flushed from the ditch along the track, and on the mar-
gins of the water in the adjacent corn field Mr. Hankinson recognized
some shore birds. The dragon-fly Libclliila pulchella was abundant on
the wing and resting on the vegetation, and two examples were found
in the webs of Argiope aurantia. No nymphs were found, but doubt-
less eggs were laid by some of the numerous adults. It was interest-
ing to observe the fresh burrows of the crawfish which had traversed
the fresh firm yellow clay of the recently reinforced railway embank-

46

ment (shown in PI. II, fig. 2) and appeared upon its surface. The
occurrence here of a small snail, Galba iiinbilicata, is of interest. A
very large species of mosquito with conspicuously banded legs, Psoro-
phora cilia ta, was found here. Though these aquatics and the ground
forms did not receive much attention, they are representative of wet
places.

The presence of certain plants in this habitat bus determined the
occurrence of several species of animals. Thus the dogbane Apocy-
niim medium accounts for the brilliantly colored leaf-beetle Chry-
sochiis auratus, which feeds upon its leaves and roots. But the most
conspicuous feature of this habitat in August is the variety of insects
which are attracted by the flowers of the swamp milkweed. These
flowers may be regarded as so much insect pasture. A few butterflies
were observed, Papilio polyxenes being found in an Argiope web; and
on the flowers of the swamp milkweed were Papilio crcsphontcs, Bury-
mus philodice, Argynnis idalia, Anosia plexippus, and the honeysuckle
sphinx (Heinaris diffiiiis). Among the most abundant Hynieuoptera
were the honey-bee (Apis mellifcra) and the common rusty digger-
wasp (Chlorion ichnciiiuoneuni). Others were the carpenter-bee
(Xylocopa virginica)- and the bumblebees Bonibus fratermis and sep-
arafiis. On the flowers of the thistle (Cirsiinn) near this station, Bom-
btts pe^insylvanicus was also taken The giant fly (Mydas davatus)
was taken on the flowers of the swamp milkweed. Beetles from these
flowers were the spotted milkweed-beetles (Tctraopes tetraophthaUiius
and fenioratiis?) the flower-beetle BupJioria sepulcliralis, and, late
in August, great numbers of the soldier-beetle CJiaidiognatlius penn-
syhanicus. The Heiniptera found are equally characteristic, and in-
clude both of the common milkweed-bugs (Oncopeltus fasciatus and
Lygccus kalmii) and Lygus pratensis. Still other insects were present
on the milkweeds, preying not upon the plant, but upon its guests.
These were the ambush bug (P/iyniafa fasciafa) and the ambush
spider (Mismnena aleatoria), the latter being captured with a large
bumblebee (Bonibus separatus) in its grasp. It is thus quite evident
that this milkweed has an important controlling influence upon the in-
sects of this habitat at this season. Another abundant animal was the
chigger, a larval mite of the genus Tronibidium, whicli is brushed from
the vegetation by one's arms and legs. These irritating pests were so
abundant that to work with comfort in this region it was necessary
to powder one's clothes and body with flowers of sulphur. These
young six-legged mites are supposed to prey upon insects, as do the
adults. According to Chittenden ('06 14) chiggers are most abun-
dant in damp places and forest margins, and among shrubs, grass,

47

and herbage. The adults are known to eat plant-Hcc, small caterpil-
lars, and grasshoppers' eggs. This mite is thus an important preda-
ceous member of the association. The dragon-flies are well known to
feed upon small insects, which they capture on the wing, and on ac-
count of their abundance they are influential insects here.

An examination of the list of animals secured at this station
shows that there is considerable diversity in the conditions under which
their breeding takes place. Indeed the breeding habits and places are
almost as diverse as are the feeding relations. Thus the snail Galba
breeds in the water; and the crawfish, Cambarns gracilis, lives as a bur-
rower except for a brief period in spring, when it is found in streams.
It is distinctly a subterranean species. The garden spider, in the fall,
leaves its eggs in its web. The Hfe history of the ambush spider is not
known. It seems probable that the sexes meet upon flowers, and as
the flowers fade they migrate to fresh ones — a response wdiich Han-
cock has observed ('ii : 182-186) in the allied species Misumcna
vatia. The ambush bug, when found on flowers, is in a large number
of cases copulating, but where the eggs are laid and the young devel-
oped is unknown to me. Though this bug also must migrate with the
fading of the flowers, after the habit of Misumena, it is winged and
does not have to go "on foot" as the spider probably does. When dis-
turbed these bugs do not as a rule seek to escape by flight, and it is not
unlikely that they often crawl from one flower to another when the
distance is short. The soldier-beetle is similar to the ambush bug in
its propensity to copulate on flowers. The milkweed beetles and the
dogbane beetle are commonly seen copulating upon the leaves and
stems of the plants on which they live. The larva of the milkweed
beetles bore into the roots and stems of plants ; the dogbane beetle has
similar habits. Of the butterflies, Anosia was observed copulating on
the willows, one sex with the wings spread, the fore ones overlapping
in part the hinder pair, the other sex with the wdngs folded together
vertically, the heads of the insects being turned in opposite directions.
The eggs of the mosquito are laid near the surface of the water. The
honey-bee and bumblebees are social, and the breeding and care of
the young are quite different from those of the other animals found
in this habitat. Xylocopa cuts the nest for its brood in solid wood,
and seems rather foreign upon the prairie, although posts and ties
are now to be found there. The rusty digger-wasp provisions its nest,
which is dug in the ground, with various grasshoppers ; upon these the
tgg is laid and the young larva feeds. This wasp probably did not
breed in this moist habitat. The wet substratum here is probably un-
favorable for the breeding of those Orthoptcra which deposit their
eggs in the soil.

48

4- Cone-flower and Rosin-zi'ecd Colony, Station I, e

This station was continuous with and just north of the swamp
milkweed area (Station l,d) just described. The surface of the
ground sloped gently upward toward the north, but none of it was free
from crawfish holes, and the ground-w^ater level was not far below.
The soil is very dark in color.

The general appearance of this habitat is shown in Plate V. The
large-leaved plants are SilpJiium terebinfliinaccum, and the heads of
the numerous cone-flowers (Lepachys pinnata) show as black points in
the picture. The cone-flower was the dominant plant at this time.
There were a few scattered plants of Silphium intcgrifoliiim and of
wild lettuce (La'ctiica canadensis). At the time the collecting was done
in this area Silphium was not in blossom, and all the flower-collecting
was from Lepachys.

The collections of animals taken here (Nos. 8, 40, and 158) are
as follows :

Crawfish

Garden Spider

Sordid Grasshopper

Differential Grasshopper

Red-legged Grasshopper

Texan Katydid

Dorsal-striped Grasshopper

Black-horned ^leadow Cricket

Membracid bug

Jassid

Lygseid

Ambush Bug

Chrysomelid beetle

Southern Corn Root-worm

Beetle
Robber-fly
Trypetid fly
Eucerid bee
Eucerid bee
Xomadid bee
Social wasp

Canibarus sp. (Burrows observed)

Argiope aurantia 40

Bncoptolophus sordidus 158

Melanopliis differentialis 40

Melanoplus fenwr-ruhrum 40

Scudderia texensis 40

Xiphidiitni strictum 40

CEcanthus nigricornis 40

Canipylcnchia curvata 40

Platymetopius frontalis 40

Ligyrocoris sylvestris 40

Phymata fascia ta 40

Xodonota convexa 40

Diabrotica 12-punctata 40

A si li dee —

Euaresta ceqnalis 40

Melissodes biniaculata 8

McUssodes obliqua 8

Epeolus concolor 8

Polistes sp. —

Collection No. 40 was made by sweeping the vegetation with an in-
sect net. No. 8 is a collection made from the flowers of Lepachys pin-
nata. The nest of Polistes was across the railway track from this
station. The abundance of Melissodes obliqua and of the pretty

49

Hpeolus concolor on the flowers of Lcpacltys indicates the attractive
power of this plant. The coarser plants furnish support for the webs
of Argiope; the flowers serve as drinking cups in which Phymata lies
in ambush ; and the varied vegetation affords food for the numerous
Orthoptcra. The proximity of ground-water accounts for the pres-
ence of Cainbarus, and an adjacent corn field explains the presence
of Diabrotica. A robber-fly (Asilidcr) was seen but not captured. It is
interesting to see Melissodes obliqua as it hurries round and round the
heads of cone-flowers and sweeps up the great masses of yellow poilcn.
The hind pair of legs, when loaded with pollen, have nearly the bulk
of the abdomen. Robertson ('94; 468) says that this is the most
abundant visitor to the cone-flower, and more abundant on this flower
than on any other.

It is probable that the conditions within this habitat were suitable
for the breeding of most of the species listed. Huaresta cequalis has
been bred from the seed pods of the cocklebur (Xanthium) and prob-
ably came from the adjacent corn field. It is most likely on flowers
that the strepsipterid parasitic insects find many of their hosts (Pierce
'09 b : 116). These insects are found on the following prairie insects :
Polistes, Odyneriis, CJdorion icJincuinoncum, C. pennsylvanicum, and
C. atratum. Robertson ('10) records many important observations on
the hosts of Illinois Strepsiptera.

5. Colony of Blue Stem (Aiidropogon) and Drop-seed (Sporobolus) ,
bordered by Szvanip Milkweed, Station I, g*

This colony formed the extreme northern part of the prairie area
examined along the "Clover Leaf" track. It extended along the track
for a distance of about 200 feet. The area is level black soil prairie.
Its general appearance and location are indicated in Figure 2, Plate
II, and in Figure 2, Plate III, photographs taken at the time of our
study, and in Figure 2, Plate IV, a photograph taken by T. L. Hankin-
son April 23, 191 1. This latter view clearly shows the character of the
drainage during the spring wet season. During the late summer, the
dry season, the ditch along the railway track concentrates the drainage
so that a colony of swamp milkweed (Asclepias incarnata) and small
willows flourish in it. Upon the well-drained part of this area there is
a rather rich growth of Andropogon ftircatus, A. virginiciis, and
Sporobolus cryptandrus, and many plants of the dogbane Apocynnm
medium and a few plants of Asclepias sullivantii. This was the larg-
est and best colony of the upland prairie grasses seen along the Clover
Leaf tracks ; and yet when it is compared with the patches of such

*No collections were made at Station I, /.

50

grass east of Charleston (Station III) it is a meager colony. Just
south of this grassy colony was a large one of the mountain mint.
Pycnanthemum flexuosum. This is shown in Figure i, Plate IV.

The collections of animals (Nos. i, 2, 3, 4, 6, 7, 19, 28a, 36, 39, 44,
157, and 159) are as follows:

Pond snail

Crawfish

Harvest-man

Garden Spider

Ambush Spider

Red-tailed Dragon-fly

Nine-spot Dragon-fly

Prairie Ant-lion

Lace-wing Fly

Grasshopper

Sordid Grasshopper

Differential Grasshopper

Red-legged Grasshopper

Texan Katydid

]\Ieadow Grasshopper

Cone-nosed Katydid

Four-spotted White Cricket

Stink-bug

Small Milkweed-bug

Large Milkweed-bug

Rapacious Soldier-bug

Ambush Bug

Four-eyed Milkweed Beetle

Rhipiphorid beetle

Bill-bug

Milkweed Butterfly

Giant Mosquito

Mycetophilid fly

Giant Bee-fly

Vertebrated Robber-fly

Honey-bee

Bumblebee

Bumblebee

Eucerid bee

Nomadid bee

Leaf-cutting bee

Rusty Digger-wasp

Myzinid wasp

rliysa gynna

19

Cambarus sp.

Liobunum politiimf

7

Argiope aurantia

6, 39

Misiunena aleatoria 6,

157. 159

Sympetrnm rubicundiiliim

7

Libcllula piilchella

B rachynemurus abdo minalis

36

Chrysopa oculata

44

Syrbula admirabilis

3

BncoptolopJnis sordidus

44

Melanoplus differcntialis

39

Mclanopliis femiir-rubrum

3. 39

Scudderia texensis

2, 44

Orchelimmn vidgare

. 3

ConocepJialus sp.

159

Qlcanthiis 4-piinctatits

3

Buschistiis variolariiis

39

Lygcrus kalmii

1,6

Oncopeltus fascia tiis

I

Sinca diadema

6

Phymata fasciata

I

Tctraopcs tctraophthaUmis

I

Rhipiphoriis dimidiatus

6

Sphenophoriis venatiis

39

Danais archippus

Psorophora cilia ta

44

Sciara sp.

6

Bxoprosopa fasciata

6

Promachits vertebratus

39. 44

Apis mclUfica

I

Boinbus fratcrnits

I

Bombus separatus

I

Melissodes hiniacidata

6

Bpeolus concolor

6

Megachile mendica

I

Chlorion ichneumoneum

I

Myziiie sexcincta

I, 6

51

Physa and Camharus were found among the milkweeds on account
of the wet ground, and the presence of the giant mosquito was prob-
ably due to the same condition. The majority of the other animals
were attracted to this habitat by the milkweed, particularly by its flow-
ers. Among these were the milkweed bugs and beetles, the milkweed
butterfly, the honey-bee, and the rusty digger-wasp. The dense growth
of the milkweeds does not appear to be so favorable to the garden
spider as is the more open and irregular growth of vegetation else-
where. The ambush spider frequented the milkweed flowers for prey
and also the flower masses of the mountain mint, on which it was in
active competition with the ambush bug and the rapacious soldier-bug,
which have similar food habits. The mountain mint, whose flowers
are frequented by the predaceous animals just mentioned, is also vis-
ited by rhipiphorid beetles, the bee-fly (Bxoprosopa fasciata) , the bees
Mclissodes himacnlata and Bpeolus concolor, and the myzinid wasp
Myzine sexcincta. The prairie grasses were frequented by a large
variety of Orthoptera, which showed a decided preference for them,
their abundance being evident in the list. The wide-ranging predators
and parasites, such as Liohunnm, Libellnla, Sympefrum, Chrysopa,
Brachynemunis, Promachiis, Clilorion, and Myzine, probably forage
over extensive areas compared with the relatively sedentary kinds,
such as Misuniena, Argiope, Phymata, and Sinea. Pliymata was cap-
tured on a milkweed flower with a honey-bee ; Promachiis vertehratiis
was taken on a grass stem with a stink-bug (Buscliistus variolarius) ;
and Misiimena aleatoria was taken with a large, nearly mature female
nymph of Conoccplialus.

The conditions which permit an animal to breed in a habitat have
an important influence upon the character of its population. It is evi-
dent that manv of the animals taken do not breed here. Some of the
relatively sedentary kinds, such as Physa, Camhams, and Argiope, and
probably Misumeiia, do not cover long distances. Good examples of
the wider ranging forms are Synipetruni, Lihcllula, Danais, Prouia-
chus, Apis, Bonihus, and Chlorion. Several of the animals, as the
snails, crawfish, and the dragon-flies, require an aquatic habitat.
Chrysopa places its eggs among colonies of plant-lice, and Brachyne-
murus probably spends its larval life in dry or sandy places, feeding
upon ants and other small insects, as do other ant-lions. Several of
the Orthoptera deposit their eggs in the soil ; and some of the locustids,
among grasses and herbaceous stems. Others are found copulating
upon the plants on which the young feed, as Tetraopes, Chrysochus,
Lygceits, and Oncopeltns; and still others copulate in the flowers
mainly, as Phymata. It is probable that on the flowers some of the para-

52

sitic species find their hosts, as Pierce ('04) has shown' to be the case
I'n the rhipiphorid genus Myodites. Rhipiphorus is probably parasitic.

6. Supplementary Collections from Station I

In addition to the specimens given in the preceding lists for Station
I there are others, general collections from this area, which should be
listed for this prairie. For details concerning each species of the fol-
lowing consult the annotated list.

Garden Spider

Argiope aiirantia

26

Ambush Spider

Mismnena aleatoria

31

Chigger

Trombidium sp.

Dorsal-striped Grasshopper

Xip Indium s trie turn

35

Coreid bug

Harnwstes reflexulus

27

Ambush Bug

Phyniata fasciata

24, 26, 43

Ladybird

Hippodaniia parenthesis

Hankinson

Leaf-beetle

Trirhabda tomentosa

Hankinson

Four-eyed Milkweed Beetle

Tetraopes 4-ophthahniis

35

Old-fashioned Potato Beetle

Bpicauta vittata

Hankinson

Margined Blister-beetle

Bpicauta marginata

Hankinson

Black Blister-beetle

Bpicauta pcnnsylvamca

26, 152

Snout-beetle

Centrimis peniccllus

41

Snout-beetle

Centrinus scutcllum-albmn

Hankinson

Giant Bee-fly

Bxoprosopa fasciata

24, 31

American Syrphid

Syrphus americaniis

II

Tachinid fly

Trichopoda ruficauda

38

Bumblebee

Bondnis separalus

22

False Bumblebee

Psithyrus varia bills

22

Eucerid bee

Melissodes obliqua

24, 48

Short Leaf-cutting Bee

Megachile brevis

Hankinson

Halictid bee

Halictiis fascia tits

26

Halictid bee

Halictns z'irescens

23

Stizid wasp

SticHs brevipennis 35,

Hankinson

Rusty Digger-wasp

Chlorion ichneumoneum

6

Harris Digger-wasp

Chlorion harrisi

24

Digger-wasp

A mmophila nigricans

24

Solitary wasp

Odynerus vagus

46

II. Prairie Area near Loxa, Illinois, Station II

This station includes patches of prairie along the Cleveland, Cin-
cinnati. Chicago and St. Louis (Big Four) railroad right-of-way be-
tween Charleston and Mattoon, 111., and about one mile west of

53

the small station of Loxa. Along this track the telegraph-pole num-
bers were used in locating our substations. This is a rather level black
soil area, originally poorly drained and wet, but now considerably
modified by the ditching and grading occasioned by railway construc-
tion and maintenance. The changes have been similar to those on the
prairie north of Charleston, but the ditching has been a few feet deeper
and the embankment is higher. The most abundant and characteristic
kinds of vegetation are the tall prairie grasses — blue stem (Andropo-
gon furcatiis), drop-seed (Sporobolus cryptandrus) , and beard grass
(Andropogon virgimcus) — a rosin-weed (SilpJiium laciniatiini), the
flowering spurge (BupJiorhia corollata), wild lettuce (Lactuca can-
adensis), rattlesnake-master (Bryngiinn yucci folium), and beggar-
ticks (Desnwdinni). Many other kinds of plants were also present.
The general appearance of this habitat is shown in plates VI and VII.
Our collections from this prairie (Nos. 47-57 and 176-178) are as
follows :

Garden Spider
Ambush Spider
Sordid Grasshopper
Two-lined Grasshopper
Differential Grasshopper
Meadow Grasshopper
Lance-tailed Grasshopper
Dorsal-striped Grasshopper
Stink-bug
Ambush Bug

Dusky Leaf-bug

Soldier-beetle

Southern Corn Root-worm

Margined Blister-beetle

Black Blister-beetle

Rhipiphorid beetle

Rhipiphorid beetle

Snout-beetle

Thoe Butterfly

Dogbane Caterpillar

Giant Bee-fly

Robber-fly

Vertebrated Robber-fly

Corn Syrphid

Syrphid fly

Argiope aiirantia 49,179

Misiinicna aleatoria 47, 178

Bncoptolophus sordidiis 48

Mclanoplus bivittatus 55

Mclanopliis differentialis 48

OrcJicliuium. vulgare 178

Xipliidinm attcmiatiim 48

Xiphidiiun strictum 48, 50, 57

Bnschistus variolariiis 50, 52, 178
Phymata fasciata

48, 52, 54, 55, 57, 178

Adclpliocoris rapidus 55
CJiauliognathus pennsylvanicus 178

Diabrotica 12- punctata
Bpicauta marginata
Bpicauta pennsylvanica
Rhipipliorus dimidiatns
Rhipiplwrus limbatiis
Rhynchitcs ccncus
Chrysophanes tJioe
Amnialo eglenensis or tenera

Bxoprosopa fasciata
Deromyia sp.
Proniachus vertcbratus
Mesogramma politimi
Allograpta obliqua

55
48

48, 178

52

178

48

55
53
176

51

56

177

177

47. 57.

54

Tachinid fly Cisfogaster immaciilata 55

Pennsylvania Bumblebee Bomhus pennsylvanicus 50, 52, 55, 176

False Bumblebee PsitJiyrns variabilis 176

Eucerid bee Mclissodcs himacnlata 48

Nomadid bee EpcoUis concolor 48, 52

Halictid bee Halictus ohscnnis 55

Halictid bee Halictus fasciatus 48, 52

Black Digger-wasp Clilorion atratum 55

Pennsylvania Digger-wasp Chlorion pennsylvanicum 55

Alyzinid wasp My due sexcincta 52, 55

Ant Formica pallid c-fidva schaufussi

incerta 52

The general conditions of this prairie appear to have been less dis-
turbed than at Station I ; at least the prairie vegetation is more exten-
sive and uniform. The change in the vegetation is i^pparently greater
than the change in the kinds of animals. Their feeding and breeding
relations appear to be much like those at the prairie stations previously
discussed.

In the flowers of the cup-leaved rosin-weed (Silphimn integri-
foliuui) was found a giant bee-fly (Bxoprosopa fasciafa) which had
been captured by the ambush spider (Misiimcna alcatoria) , and on
webs in colonies of this same plant the garden spider (Argiope aiiran-
tia) was observed, with a grasshopper (Melanoplus diffcrentialis) en-
tangled in the web. From the flowers of this SUpJiiuui the following
insects were taken : Epicauta marginata and B. pcnnsylvanica, Rhyn-
chitcs (uncus, Phyniata fasciafa, Encoptolopluis sordidus, Melanoplus
diffcrentialis (nymph), Xiphidiuni strictuni (adult and nymph), A'.
attcnuatuni, Mclissodcs hiniaculata and obliqua, Bpcolus concolor, and
Halictus fasciatus. The margined blister-beetle (Bpicauta marginata)
was found both upon the flowers and the leaves of the plant. On the
flowers of the purple prairie clover {Petalosteniuni purpurewni), Bom
bus pcnnsyh'anicus, Xiphidium strictum, and Buscliistus variolarius
were taken. Collection 176 was taken from the flowers of Liatris
scariosa, and Nos. 55 and 178 from the flowers of Bryngium yiicci-
folium.

Swarms of the small corn syrphid, Mesogramma politum, were
present, on one day settling by dozens on my hands and clothes, where
they were easily grasped by the wing. It had been a warm day, and
this swarming was in the sunshine at about 4 :30 p. m. The flies came
from a large corn field a few feet away.

55

III, Prairie Area East of Charleston, Station III

This prairie area is about two miles east of Charleston along the
"Big Four" railway track. There were two colonies here. One, sub-
station a, was on low black-soil prairie just west of the first north and
south road crossing the railway track east of Charleston. This was
largely a colony of the large-leaved rosin-weed, Silphiimi terebinthi-
naceuni. The second colony, substation h, was a mile and a half di-
rectly east of substation a, and half a mile east of the second north and
south road east of Charleston.

Substation or "station" a was originally far out upon the black soil
prairie ; h, on the other hand, is of special interest because it was origi-
nally wooded, has been cleared and maintained as a railroad right-of-
way, and contains today, therefore, a practically unique mixture of for-
est and prairie plants and animals, with the prairie kinds dominating.
The soil, lighter in color than the black soil prairie, is representative of
the wooded regions. This colony has every appearance of a cleared
forest area invaded by prairie organisms.

The animals at station a were not studied, and the only record is
that of the black blister-beetle, Bpicauta pennsyhanica (No. 119),
which was abundant on the flowers of Silphiiim terebinthinaceitm.

At station b excavation was necessary to' lower the road-bed, and
upon the disturbed soil thus thrown up along the track the prairie veg-
etation had become established. The general appearance of this region
is shown in plates VIII and IX. Here grew large quantities of rosin-
weed (Silphiiim terebinthinacciun) and blue stem (Andropogon) ; in
places upon high ground, indeed, this prairie grass was dominant.
Associated with it was the flowering spurge, Buphorhia coroUata, as
seen in Plate VIII. The forest near by is shown in the background.
This same forest and grass area is shown in the background and mid-
dle of Plate IX, and in the foreground of the same picture is shown
the mixture of prairie and forest plants. Here are hickory sprouts,
crab-apple, grape, sumac, and smilax, intermingled with Silphiiim,
blue stem, and Lactiica canadensis. Not all of these appear in the
photograph, but they were present in some parts of the colony.

The collections here (Nos. 58-62 and 175) are as follows:

Leather-colored Grasshopper Schistocera alntacea 59

Black-horned Meadow Cricket Q^cauthiis nigricornis 62

Meadow Grasshopper Orchelimiim vulgar e 175

Soldier-beetle Chanliognathus pcnnsylv aniens 175

Spotted Grape-beetle Pclidnota punctata 58

Black Blister-beetle Bpicauta pcnnsvlvanica

(Sta. Ill, a) 119

56

Cabbage Butterfly Pontia rapce 6i

Vertebrated i^obber-fly Proniachits vertehratus 62

Pennsylvania Bumblebee Bomhus pennsylvanicus 175

Impatient Bumblebee Bomhus impatiens 175

Bumblebee Bombns auricomus 175

(Rose-gall) Rhodites nebulosus 60

No animals were taken here which were dependent upon the sumac,
hickory, crab-apple, or smilax. Pclidnota lives upon the grape, and
grapes are primarily woodland or forest-margin rather than prairie
plants. ScJiistoccrca is also probably a marginal species. On the flow-
ers of Silphiuni terchinthinaccum were taken Orchelimum vidgare,
Chauliognathus pennsylvanicus, and Bonibus pennsylvanicus, auri-
comus, and impatiens.

The persistence of woodland vegetation in this locality, in spite
of the repeated mowings and burnings, shows that it has much vigor,
and would, if undisturbed, in a few years shade out the prairie vege-
tation and restore the dominance of the forest. With such a change in
the vegetation there would of course be a corresponding change in the
animals.

DESCRIPTION OF THE FOREST HABITATS AND ANIMALS

/. The Bates Woods, Station IV

The Bates woodland area is located about three and a half miles
northeast of Charleston on the farm that was owned by Mr. J. I. Bates,
and consists of about 160 acres. It includes a bottom-land area near
the Embarras River, and extends up the valley slope on to the upland.
It is isolated from the trees bordering the river (PI. X, fig. i) by a
narrow clearing, and from those on the northeast, north, and north-
west by another clearing (PI. XI) ; on the south and southwest it is
continuous with partially cleared areas, which extend south to the Big
Four railway track.

The river bottom-land is undulating and rises rather gradually
toward the base of the bluffs. The bluff line is irregular on account of
the ravines which have been etched in it, the largest of which forms
the southern boundary of the region examined. The upland is rela-
tively level. The soils on the bottom are darker colored, except in
places near the base of the bluff, and at the mouths of the ravines
where the upland soil has been washed down. The upland soil is pre-
sumably the "light gray silt loam" of the State Soil Survey (Moultrie
County Soils, 111. Exper. Sta. Soil Rep., 191 1, No. 2, p. 23). All of

57

the area examined was well drained, and all was forested. ^The region
is not homogeneous physically or in its vegetation, and for this reason
' the area is divided into substations in order that the influences of the
local conditions within the forest might be preserved, and their indi-
viduality recognized.

2. The Upland Oak-Hickory Forest, Station IV, a

The general appearance of this forest is shown in plates XII and
XIII. This is an open second-growth forest composed of oaks and
hickories — such as white oak (Quercus alba), black oak (Q. vclutina),
shag-bark hickory (Carya ovata) , bitternut (C. cordiforniis) , pignut
(C. glabra), and scattered individual trees of red oak (Q. rubra), wal-
nut (Jnglans nigra), and mulberry (Morus rubra). The shrubs are
sassafras (Sassafras variifoliiim), sumac (Rhus glabra), Virginia
creeper (Psedcra quinquefolia) , poison ivy (Rhus toxicodendron),
rose (Rosa), raspberry (Rubus), moonseed (Menisperniuni cana-
dense) , and tree seedlings. The average diameter of the largest trees
is 8-IO inches. Most of the small growth consists of the sprouts from
stumps, and many of these are 2-3 inches in diameter. The forest
crown is not complete, and as a consequence there are more or less open
patches in which most of the herbaceous growth is found, such as
horse mint (Monarda bradburiana), pennyroyal (Hcdcoma pide-
gioides), everlasting (Antenuaria plantaginifolia), tick-trefoil (Des-
modium nudifloruui) , and other, less abundant kinds. Even a plant
quite characteristic of the prairie, the dogbane Apocynuin, was found
here in one of the open glades.

The forest floor has an unequal covering of dead leaves, largely
oak, most of which lie in the low vegetation and in slight depressions.
Occasionally there is but little cover and the light-colored soil is ex-
posed. There are few stumps and logs in this part of the forest, and
no thick layer of vegetable mold, so that one would not expect to find
any animals which normally frequent moist soil and vegetable debris.
As this is a second-growth forest it lacks the conditions which abound
in an original growth, where are old, dead and decaying trees, and
numerous decaying logs and stumps. In this respect the woods is not
fully representative of an original upland forest on well-drained bluff
land.

The relative evaporating power of the air of this substation was 54
per cent, of that of the standard instrument in the open garden at the
Normal School, a fact which indicates a relative evaporation com-
parable to that of the ordinary black-soil prairie ; in producing this con-
dition, the glade-like, open character of this forest is undoubtedly
an important factor.

58

The characteristics of this habitat may be summed up as follows :
upland, open, relatively dry second-growth oak-hickory forest, with
little undergrowth of shrubs and herbs, and with a small amount of
litter and humus ; soil dry and firm ; and few decaying stumps and tree
trunks.

The collections of animals made here (Nos. 64-67, 69, 71, 74-83,
88, 91-93, 102, 103, 107, 109, 118. 120-123, 127, 135, 136, 142, 145,
147, 150, 151, 162, 163. 166, 169, 170, 171, and 183) are as follows:

Land snail

Predaceous snail

Land snail

Carolina slug

Land snail

Harvest-spider

Harvest-spider

Stout Harvest-spider

Island Spider

White-triangle Spider

Rugose Spider

Ground Spider

\Miite Ant

Ant-lion

Dog-day Harvest-fly

Periodical Cicada

Forest Walking-stick

Grouse Locust

Short-winged Grouse Locust

Green Short- winged

Grasshopper
Sprinkled Grasshopper
Boll's Grasshopper
Lesser Grasshopper
Acridiid grasshopper
Acridiid grasshopper
Forked Katydid
Angle-winged Katydid
Common Katydid
Meadow Grasshopper
Meadow Grasshopper
Striped Cricket
Spotted Cricket
Woodland Cricket

Polygyra alholahris 91

Circinaria concava 71

Zoiiitoides arhorea 71

PJiilomycus carolinensis 71

Pyramidula perspectiva 71, 88

Liohunum vittatmn 82, 123

Liohumim ventricosinn 123b

Liohunum grande 82

Bpcira insularis 70

Epe'ira verrucosa 70

Acrosoma rugosa 70, 147

Lycosa sp. 142, 150

Vermes flavipes "^2, 76, 79
Myrmeleonidcv (Forest border) 183

Cicada linnei 162

Tibicen septendecim —

Diapheromera femora ta 64, 93

Tcttigidea lateralis 109

Tettigidea pari'ipennis 122
DicJiro nwrpJia z'iridis

67, 92, 93, 121, 123

Chloealtis conspersa 67, 93, 122

Spharagemon holli
Melanoplus atlanis
Melanoplus amplectens
Melanoplus ohovatipennis
Scudderia furcata
Microceiitrmn laiirifoliiim
Cyrtophyllus perspicilla his
Orchelimum cuticulare
Xiphidium nemorale
Nemohius fasciatus
Nemohius maculatus
A pi thus agitator

^7^ 150
67
67

93

109

135

145

67. 93

93» 103

67, 93. 122

122
93

59

Woodland Tiger-beetle
Caterpillar-hunter
Carabid beetle
Ladybird

Splendid Dung-beetle
Dogbane Beetle
Tenebrionid larva
Philenor Butterfly
Turnus Butterfly
Troilus Butterfly
Sphingid larva
Arctiid moth
Notodontid moth
Notodontid moth
Notodontid moth
Geometrid moth
Gelechiid moth

(Cecidomyiid gall)
(Cecidomyiid gall)
(Cecidomyiid gall)
Syrphid fly
Corn Syrphid

Vespa-like Syrphid
Pigeon Tremex
(Oak Bullet-gall)
(White Oak Club-gall)
(Oak Wool-gall)
Formicid ant
Formicid ant
Formicid ant
Mutillid ant
Short Caterpillar-wasp

Cicindcla nnipimctata

136

Calosoma scrutator

64

Galerita janus

171

Coccinellidtr

81

Geotriipes splendidiis

120

Chrysochiis atiratus

103

Meracantha contracta

83

Papilio philenor

^9,

166

Papilio turnus

Papilio troilus

163

Cressonia juglandis

102

Halisidota tessellaris

168

D at ana angusii

65,

162

Nadata gibbosa

169

Heterocampa guttivittaf

127

Bustroma diversilineata

163

Ypsolophus ligulellus?

. 76. 78,

Hankinson

Cecidomyia holotricha

107,

I/O

C ecidomyia tubicola

107

Cecidomyia carycecola

107,

170

Chrysotoxum ventricosum

163

Mesogramma politum

?(>, 78,

Hankinson

Milesia ornata

103

Tremex columba

66

Holcaspis globulus

170

Andricus clavida

170

Andricus lana

170

Cremastogaster lincolata

118

Aphccnogaster fidva

74-80

Formica fusca subsericea

163

Sphccroplithalma

151

Ammophila abbreviata

127

J. Bmbarras Valley and Ravine Slopes, forested by the Oak-Hickory

Association, Station IV, b

This station included the slope of the valley from the river bottom
(Station IV, c) to the upland forest (Station IV, a) and the side of
the south ravine, the bottom of which forms Station IV, d. This sub-
station is not as homogeneous physically as the upland or lowland for-
est, because the part along the south ravine is relatively open, is well
drained, and has a south exposure, and the southeast slope to the low-

60

land forest on the other hand, is well wooded and shaded, and much
more humid. The substation also has a considerable amount of litter,
leaves, and humus. This region may be considered as transitional be-
tween the upland and lowland forest, but it represents, not one but two
transitional stages, the south slope approaching the upland forest type,
and the southeast slope approaching that of the lowland forest.
Thus, if one walked from the upland forest down the slope of the
south ravine, and eastward to the southeast valley slope to the bottom-
land forest, he would traverse all the main degrees of conditions found
at Station IV.

The forest cover consists primarily of the following trees : white
oak (Qucrcns alba), black oak (Q. velutiiia), walnut (Juglans nigra),
pignut (Carya glabra), and, in smaller numbers, mulberry (Morns
rubra), red oak (Quercus rubra), shag-bark hickory (Carya ovata),
bitternut (C. cordiformis) ; and of the following shrubs: redbud (Cer-
cis canadensis), sassafras (Sassafras varii folium), moonseed (Menis-
perinuni canadcnsc), five-leaved ivy (Pscdcra qmnquefolia) , grape
(Vitis cinerea), prickly ash (Zanthoxylum amcricamim), and sumac
(Rhus glabra), the latter growing in large colonies on the open south
ravine-slope. On the more moist and shaded southeast slope lived the
clearweed (Pilca pumila), a plant quite characteristic of moist deep-
shaded woods. Thus sumac and clearweed may be considered as in-
dex plants to the physical conditions in different parts of these two
slopes, one shaded and the other rather open.

The atmometer, located on the upper part of the south ravine slope,
gave a relative humidity of 31 per cent, of the standard in the garden
of the Normal School. It will be recalled that in the upland forest
(Station IV, a) the atmometer gave 54 per cent., the comparison
showing how much less the evaporating power of the air is on the
south ravine slope than it is in the upland forest. The relative evap-
oration was not determined for the southeast slopes, but the presence
of Pilea clearly indicates that it is less than on the south ravine slope,
where the instrument was located. On the lower parts of the valley
slope, where this substation grades into the lowland, the layers of dead
matted leaves and humus reached to a considerable depth, and looked
as if they had been pressed down by drifting snows. Such places were
found to contain very few animals.

This habitat is characterized by a sloping surface, by relative open-
ness on the ravine side and dense shade on the valley slope, by rela-
tively humid air, by second-growth forest somewhat transitional be-
tween that of the uplands (Station IV, a) and the river bottoms (Sta-
tion IV, c), by a relatively large amount of shrubbery, by considerable

61

humus and litter, by moist soil, and by more logs and stumps than are
in the upland forest.

The collections of animals made at this substation (Nos. 68, 84, 85,
87, 89, 90, 94, 100, 104, 105, 106, 108, no, III, 124, 125, 131, 132,
133, 140, 149, 161, 164, 165, 166, and 168) are as follows:

Land snail

Polygyra claiisa

164

Land snail

Vitrea indentata

140,

164

Land snail

Vitrea rhoadsi

164

Land snail

Zonitoides arborea

84

Carolina Slug

PlUlomycus carolinensis

89,

125

Land snail

Pyramidida perspcctiva

84,

164

Milliped

Cleidogona ccesioanmdata

140

MilHped

Polydesmus sp.

125

Stout Harvest-spider

Liobunum grande

III

White Ant

Vermes flavipes

125

Woodland Cockroach

Ischnoptera sp.

140

Green Short-winged

DichromorpJia viridis

no

Grasshopper

Boll's Grasshopper

Spharagemon bolli

133

Scudder's Grasshopper

Melanoplus scitdderi

124

Woodland Cricket

Apithes agitator

124

Caterpillar-hunter

Calosoma scrutator

100,

149

Wireworm

Melanotus sp.

125

Horned Passalus

Passalus cornutiis

85

Tenebrionid larva

Meracantha contracta

140

Troilus Butterfly

Papilio troilus

161

Philenor Butterfly

Papilio philenor

166

Lyca^nid butterfly

Bveres coniyntas

161

American Silkworm

Tclea polyphemus

163

Hickory Horned-devil

atheroma regalis

68,

108

Arctiid caterpillar

Halisidota tessellaris

163,

168

Rotten-log Caterpillar

Scolecocampa lihurna

125

Notodontid

Datana angusii

104

Notodontid larva

Nadata gibbosa

94

Geometrid

Caberodcs confusaria

161

Slug Caterpillar

Cochlidion or Lithacodes

165

Pigeon Tremex

Tremex columba

132

(Acorn Plum-gall)

Amphibolips primus

131

Old-fashioned Ant

Stigmato mma pallipes

140

Tennessee Ant

Aphccnogaster tennesaeensis

87

Formicid ant

Myrmica rubra scabrinodis

schnecki

140

62

Carpenter-ant Canipouotus hcrculeanns penn-

sylvanicus 84, 85

Rusty Carpenter-ant Canipouotus Jierculeanus penn-

syh'anicus ferriigineus 90

Short Caterpillar-wasp Ammophila ahhreviata 124

4. Lowland or "Second Bottom," Red Oak-Bhu-Sugar Maple Wood-
laud Association, Station IV, c

This station includes the part of the forest located upon the upper
or higher part of the river bottom. This area is sometimes called the
"second bottom" because it is above the present flood-plain. The gen-
eral position of the forest is shown in Figure i, Plate X. The fringe
of willows along the river bank is shown at a; the flood-plain area is
cleared at b; the substation forest is at c; and part of the forest of the
valley slope is seen at d. Other views of this station are shown in
plates XIV, XV, and XVI (figures i and 2). The general slope is
toward the river; minor inequalities are due to the action of the tem-
porary streams which are etching into the uplands and depositing their
burdens of debris at the mouths of the ravines. Soil, leaves, and other
organic debris are washed from the upland, the ravines, and the val-
ley slopes, and are deposited upon the bottoms, forming low alluvial
fans, which have been built up in successive layers or sorted again and
again as the temporary streams have wandered over the surface of
the fan on account of the overloading and deposition which filled up
their channels. In this manner the soil in general is not only supplied
with moisture, drained from the upland, but the various soils are both
mixed as successive layers of organic debris are buried by storms and
also mulched by the large amount of this debris which is washed and
blown to the lowland. No springs were found upon the southeast
valley slope, but in the south ravine pools of water were present dur-
ing August, 1 9 10, when my observations were made.

The forest, characterized by hard maple (Acer saccharum), red
oak (Qucrcus rubra), and elm (Ulunts americana), forms a dense
canopy which shuts out the light and winds, thus conserving the mois-
ture which falls and drains into it, and making conditions very favor-
able to a rich mesophytic hardwood forest. That the relative humid-
ity is high is shown by the moisture found in the humus of the forest
floor, and, further, not only by the presence of clearweed (Pilea pu-
mila) and the nettle Lap or tea canadensis, which characterize such
moist shady woods, but also by the presence of the scorpion-flies (Bit-
tacus). These organisms are permanent residents where such condi-

63

ditions prevail, and their presence is as clearly indicative of certain
physical conditions as that of aquatic animals would be indicative of
other physical conditions. In addition to these evidences we have
the readings of our atmometer, which showed the evaporating power
of the air to be 26 per cent, of the standard in the garden at the Normal
School. This shows that the relative evaporation is very low, and
that conditions for the preservation of the moisture which falls and
drains into this area are very favorable. The general character of this
forest is shown in plates XIV, XV, and XVI, Figure i.

The vegetational cover on the lowland is quite different in its com-
position from that on the upland. This is shown mainly by the pres-
ence of the elm (Ulmus americana), hard maple (Acer saccharum),
and red oak (Quercus rubra), and secondarily, by the presence, in
smaller numbers, of the black cherry (Priimts serotina), slipperv elm
(Uhmis fiilva), shingle oakf (Quercus imhricaria) , nnd the Kentucky
cofifee-tree (Gyninocladus dioica). Other trees present are walnut
(Juglaus nigra), mulberry (Moms rubra) , and bitternut (Carya cor-
diformis). The shrubs and vines are gooseberry (Ribes cynosbati),
prickly ash (Zanthoxylimi americanum) , redbud (Ccrcis canadensis) ,
buck-brush (Symphoricarpos orbiculatus), green brier (Smilax),
five-leaved ivy (Psedcra quinquefolia) , moonseed (Menispermum
cana dense), bittersweet (^ Celastrus scandens), and grape (Vitis cine-
rea). The characteristic herbaceous vegetation is nettle (Laportea
canadensis) , clearweed (Pilea puniila) , bellflower (Campanula ameri-
cana), Indian tobacco (Lobelia infiata) , tick trefoil (Desmodimn
grandifloruni), Actinomeris alternifolia, maiden hair fern ( Adiantuni
pedatuni), beech fern (Phegopteris hexagonoptera) , the rattlesnake
fern (Botrychiiun z'irginianuni), and Galium circcczans and G. tri-
foliuni.

Although the forest is generally dense and therefore deeply shaded,
there are some places which are comparatively open. Attention, how-
ever was devoted mainly to the denser parts. At one place, near the
base of the eastern slope of the valley, a few trees had been cut within
a few years, and in this glade the conditions and plants and animals
were different from those in the dense forest. (See PI. XVI, figs, i
and 2.)

This habitat may be characterized as follows : lowland densely cov-
ered by sugar maple-red oak forest (climax mesophytic) ; very humid
air ; a moist soil ; relatively few shrubs ; herbaceous plants — nettles and
clearweed — characteristic of damp, shady, rich woods ; and considera-
ble litter and humus in places.

64

The collections of animals made here (Nos. 113, 114,
137-139- 141- ^43, 144, 173, 1S2, and 184) are as follows,
cised numbers designating collections from the glade :

Predaceoiis Snail

Land snail

Slug eggs

Alternate Snail

Milliped

Ambush Spider

Tent Epeirid

Three-lined Epeirid

Spined Spider

Rugose Spider

Ground Spider

Cherry-leaf Gall-mite

Clear-winged Scorpion-fly

Leaf-hopper

Pentatomid

Coreid

Spined Stilt-bug

Short-winged Grasshopper

Acridiid grasshopper

Acridiid grasshopper

Scudder's Grasshopper

Round-winged Katydid

Nebraska Cone-nose

Meadow Grasshopper

Meadow Grasshopper

Meadow Grasshopper

Striped Cricket

Elaterid larva

Elaterid

Black-tipped Calopteron

Reticulate Calopteron

Horned Fungus-beetle

Common Skipper

Imperial Moth (larva)

Noctuid moth

Asilid fly

Vespa-like syrphid

Long-sting

Black Longtaii

Cocoanut Ant

Circinaria concava
Vitrca indentata
Philomycus (?) eggs
Pyramidiila alternata
Callipus lactarius
Misumena aleatoria
Bpcira domicilionim
Epeira trivittata
Aero soma spine a
Aerosoma rugosa
Lyeosa scutulata
A earn s serotince
Bittacus stigmatcriis
Aidacizes irrorata
Hymenareys nervosa
Aeanthocerus galeator
Jalysus spinosiis
Diehromorpha viridis
Mclanoplus ainplectens
Mclanoplus gracilis
Mclanoplus scudderi
A mblycorypha rotundifolia
Conoeephalus nehrascensis
Orclieliniuni euticulare
O rchclinm ni gla h crrimu m
Xiphidiuni nemorale
Nemohius fasciatus
Corymhites sp.
Asaphes memnonius
Calopteron terminale
Calopteron reticulatum
Boletotherus hifiircus
Bpargyreus tityrus
Basilona imperialis
Autographa precationis
Deromyia discolor
Milesia ornata
Thalessa lunator
Peleeinus polyturator
Tapinoma sessile

[I6,

ii7>

the itali-

113

113

114

173

113

184

131.

> 173

138

138,

172

172

144

116

141

ii7>

143

113

182

117

ii7>

143

117,

143

143

117

ii7>

143

117

143

ii7>

143

ii7>

143

143

113

113

173

143

173

173

106

143

117

i43>

184

143

ii7>

143

139

65

5- Supplementary Collections from the Bates Woods, Station IV

Tent Epeirid
White-triangle Spider
Spined Spider
Rugose Spider
Mealy Plata
Leaf -hopper
Pentatomid bug
Pentatomid bug
Tarnished Plant-bug
Coreid bug
Coreid bug

Rapacious Soldier-bug
Acridiid grasshopper
Pennsylvania Firefly
Margined Soldier-beetle
Soldier-beetle
Chrysomelid beetle
Clubbed Tortoise-beetle
Portlandia Butterfly
Eurytus Butterfly
Gelechiid moth
(Hairy Midge-gall)

Corn Syrphid Fly
(Horned-knot Oak-gall)
(Oak Wool-gall)
Ichneumon Wasp
Formicid ant
Rusty Carpenter-ant

Spider Wasp

Bpeira domiciliorum 167

Bpeira verrucosa 126

Acrosoma spinea 148

Acrosonia rugosa 126

Ormenis pruinosa Hankinson

Gypona pect oralis Hankinson

Buschistiis fissilis 124

Mormidea lugens Hankinson

Lygus pratensis Hankinson

Alydiis qninqiiespinosus Hankinson

Acanthoceros galeator Hankinson

Sinea diadema Hankinson

Melanoplus ohovatipennis 124

Photuris pennsylvanica Hankinson
ChaidiognatJiiis tnarginatus Hankinson

Tclephorus sp. Plankinson

Cryptoccphalus mutahilis Hankinson

Coptocycla clavata Hankinson

Bnodia portlandia 63

Cissia eurytus Hankinson

YpsolopJius ligulcllus Hankinson
Cccidoniyia holotricha

(Near collection No. 96)

Mesogramma politum Hankinson

Andricits cornigerus (Near 96)

Andricus lana (Near 96)

Tragus ohsidianator Hankinson

Aphccnogaster fidva 125
Camponotus hercideanus penn-

sylvanicus ferrugineus 97

Psammochares cuthiops Hankinson

6. Small Temporary Stream in the South Ravine, Station IV, d

This small temporary stream in a ravine formed the southern
boundary of the area examined (PI. XVH, figs, i and 2). At the sea-
son of our examination it was a series of small disconnected pools.
Very little attention was devoted to the collection and study of its life.
Most of the collections were secured by T. L. Hankinson. A few aquat-
ic animals were collected here. In a small pool were taken numerous
specimens of the creek chub (Semotilus atromacidatiis), and one stone-

66

roller (Campostoiua anonialuin). Frogs, toads, and salamanders were
also taken in the vicinity by Mr. Hankinson, who dug" from their bur-
rows specimens of Cambarus diogenes, and also secured immunis and
propinqnus. On the surface of the pools were numerous specimens
of a water-strider, Gerris remigis. The forest cover is undoubtedly an
important factor in the preservation of such pools, as it controls the
evaporating power of the air.

Mr. Hankinson tells me that during the summer of 191 2 this tem-
porary stream was completely dry, and that no fish have been taken
from it since the earlier collection mentioned above. From the mouth
of the ravine across the bottom to the river it is only a few hundred
feet, and in time of heavy or prolonged rains these pools are in direct
communication with the river. Such a stream is an excellent example
of an early stage in the development of the stream habitat, and shows
its precarious character, and the liability to frecjuent extermination
of these pioneer aquatic animals which invade it in its early stages.
This applies particularly to those animals which have no method of
tiding over dry periods. On the other hand, those animals which live
in the pools, those parts of temporary streams which persist longest
between showers, have better chances of survival, particularly bur-
rowing animals, like the crawfish and its associates. It seems prob-
able that crawfish burrows harbor a varied population; not only the
crawfish leeches (Branchiohdcllidce) but also the eggs of certain Cor-
ixidcc (Forbes, '76:4-5; '78, p. 820; Abbott, '12) may almost cover
the body of some crawfishes. By means of this burrow ground-water
is reached, and a subterranean pool is formed. For the elaboration of
the stream series see Adams ('01) and Shelford ('11 and 13a).

This temporary stream shows how, by the process of erosion, the
upland forest area is changed into ravine slopes, and, later, even into
the bed of a temporary stream. Thus progresses the endless transfor-
mation of the habitat.

GENERAL CHARACTERISTICS OF THE GROSS

ENVIRONMENT

I. Topography and Soils of the State

Illinois lies at the bottom of a large basin. This is indicated in
part bv the fact that so many large rivers flow toward it. The mean
elevation of the state is about 600 feet, and about a third of it lies be-
tween 600 and 700 feet above sea-level. Except Kentucky, the bor-
dering states are from 200 to 500 feet higher. Iowa and Wisconsin
are considerably higher, so that winds from the north and northwest

67

reach the state coming down grade. Taken as a whole the land sur-
face is a tilted plain sloping from the extreme northern part — where a
few elevations exceed a thousand feet — toward the south, bowed in
the central part by a broad crescentic undulation caused by a glacial
moraine, and then declining gradually to the lowland north of the
Ozark Ridge, near the extreme southern part of the state. This east
and west ridge occasionally exceeds i,ooo feet, but its average height
is between 700 and 800 feet. It is very narrow, only about 10 miles in
average width, and rises about 300 feet above the surrounding low-
land (Leverett, '96, '99). South of this ridge lie the bottoms of the
Ohio River. The largest river within the state is the Illinois.

The soils of the state are largely of glacial origin. Even the un-
glaciated extreme northwestern part and the Ozark Ridge region have
a surface layer of wind-blown loess. In some places considerable sand
was assorted by glacial water, forming extensive tracts of sandy soil,
and locally dune areas are active. Along the larger streams there are
extensive strips of swamp and bottom-land soils. The remaining soils,
which characterize most of the state, were either produced mainly by
the lowan or Illinoian ice-sheets, as in the case of the relatively poorer
soils, or by the Wisconsin sheet, which formed the foundation for the
better soil. The dark-colored prairie soils are due to organic debris.
Coffey ('12:42) has said: "Whether this accumulation of humus is
due to lime alone or to the lack of leaching, of which its presence is an
indication, has not been definitely determined. Neither do we know
whether it is due to chemical or bacteriological action ; most probably
the latter, an alkaline medium being necessary for the growth of those
bacteria or other microorganism which cause this form of decomposi-
tion."*

2. Climatic Conditions

The climatic features of a region are generally conceded to have a
fundamental influence upon its life. The controlling influences upon
climate are elevation above sea-level, latitude, relation to large bodies
of water — generally the sea — and the prevailing winds. The eleva-
tion and relief of Illinois have but a slight influence. In latitude
Illinois is practically bisected by the parallel 39>^° in the north tem-
perate zone. This position influences the seasons and the amount of
heat received from the sun. The sea is far distant, but the Great
Lakes are near by, and proximity to the interior of a large continent

*Consult Hopkins and Pettit ('08) and the County Soil Eeports of the State
Soil Survey for a detailed account of the chemical conditions of Illinois soils.
The bacterial, algal, and animal population have hardly been noticed by stu-
dents of Illinois soils.

68

brings the state within that influence. And, finally, it lies in the zone
of the prevailing westerly winds, and directly across the path of one
of the main storm tracks, along which travel in rapid alternation the
highs and lows which cause rapid changes of temperature, wind, and
precipitation, and thus produce the extremely variable weather condi-
tions.

The state is 385 miles long, and as latitude has much influence
upon climate, the climate of Illinois differs considerably in the extreme
north and south. This is clearly shown in the average annual tempera-
ture, which in the northern part is 48.9° F., in the central part is
52.70°, and in the southern part is 55.9° (Hosier, '03). These aver-
ages probably closely approximate the soil temperatures for these re-
gions. The average date of the last killing frost in the northern part
is April 29 ; in the central part, April 22 ; and in the southern part,
April 12. The average date of the first killing frost for the northern
part is October 9, central part, October 11, and the southern part is
October 18 (Henry). The growing season for vegetation in the
northern half of the state averages from 150 to 175 days and for the
southern half from 175 to 200 days (VVhitson and Baker, 12:28).
The precipitation shows similar differences, increasing from north to
south. The annual average for the northern part is 33.48 inches, in-
creasing to 38.01 in the central and to 42.10 inches in the southern
part (Hosier, '03:62). Hosier has shown that the Ozark Ridge,
with an average elevation of about 800 feet, condenses the moisture
on its south slope so that it has a precipitation of 7.15 inches more
than do the counties just north of the ridge. This same humid area
appears to extend up the Wabash Valley to Crawford county, and
gives the valley counties a rainfall 3 inches in excess of the adjacent
counties to the west. The average annual rainfall for the state is
37.39 inches — nearly one third of it during April, Hay, and June,
and if July is included, more than half. The heaviest precipitation,
8.23 inches, is in Hay and June.

As previously mentioned, the state lies in the zone of prevailing
westerly winds and across the path of storms. These have a dominant
influence upon the direction of the winds. In the northern part of the
state, they are, by a slight advantage, southerly — a tendency which
progressively increases toward the south, for in the central part the
southerly winds reach 55 per cent., and in the southern part 62 per
cent. During the winter the northwest winds predominate throughout
the state, to a marked degree in the central part, where they reach
60 per cent., and where also the velocity is greatest, reaching an av-
erage of 10.3 miles an hour. The velocity of the wind for the entire

69

state is highest during spring. During the summer, the southwest
winds predominate in the northern and central parts, and in the south-
ern part 82 per cent, of the winds are southerly. The velocity of the
wind is least during the summer, and the greatest stagnation occurs
in August. During autumn there is a falling off of the southerly
winds and an increased velocity as winter conditions develop. The
transition in the fall is in marked contrast with the vigor of the
spring transition. The cooler seasons are more strongly influenced
by northerly winds, and the warmer seasons by southerly winds.

J. Climatic Centers of Influence

In the preceding section the average conditions of temperature,
precipitation, and the direction and velocity of the winds have been
summarized, but little effort was made to indicate the mode of opera-
tion of the determining factors which produce and maintain these aver-
age conditions. It is often true that the main factors which explain
the conditions seen in some restricted locality can not be found within
it because the local sample is only a very small part of a much larger
problemi. Thus no one attempts to find an explanation of the through-
flowing upper Mississippi system within the state of Illinois; a larger
unit of study is necessary. The region examined must extend to the
headwaters. So, also, with most of the climatic features of Illinois;
their approximate sources must be sought elsewhere. Let us there-
fore consider some of the broader features which influence the climate
of North America, particularly that of the eastern part.

The climates of the world have been divided into two main kinds,
depending primarily upon the controlling influence of temperature.
This is due to the relative specific heat of land and water, that of water
being about four times that of land. The sea, which covers three
fourths of the earth's surface, is thus an immense reservoir of heat,
which is taken up and given off slowly, at a rate one fourth that of the
land. It is therefore relatively equable. The northern hemisphere
contains the largest amount of land, and is therefore less under the
control of the sea than the southern hemisphere ; yet the sea's influence
is very powerful, particularly near the shore. The large land masses,
on the other hand, on account of their lower specific heat, receive and
give off heat more rapidly to the air above. For this reason the tem-
perature changes, as between day and night or summer and winter,
are much more rapid and much more extreme over land than over
the sea. A climate dominated by the equable sea is oceanic; that
dominated by the changeable lands is continental. Illinois lies far

70

from the sea and is therefore strongly influenced by continental con-
ditions. To what degree is the marine influence shown?

Meteorologists (cf. Fassig, '99) have come to look upon the large
areas of permanent high and low barometric pressure as among the
most important factors in climatic control. There are five of these
powerful "centers of action" which influence our North American
climate (Fig. i), and four of these are at sea. A pair of loius are in
the far north, one in the north Pacific near Alaska, the other in the

Fig. 1. DiagTam showing the positions of the relatively stable areas of high and
low barometic pressure, and indicating their influences upon the evaporating power of
the air and upon the climate in general.

north Atlantic south of Greenland. A pair of highs are farther south,
one in the Pacific between California and the Hawaiian Islands, and
the other centering in the Atlantic near the Azores. The highs and
lows in each ocean seem to be paired and to have some reciprocal rela-
tion. The fifth center of action is upon the land. It is a high baromet-
ric area in the Mackenzie basin of Canada, where it becomes a pow-
erful center of influence through winter and spring, but with the prog-
ress of summer conditions weakens, and through the accumulation of
continental heat becomes converted into a lozv; thus there is a complete
seasonal inversion on the continent.

These large highs and lows, although relatively permanent, are con-
tinually changing in intensity and position. The highs are regions of
descending, diverging, warming, and drying air, producing clearing
and clear air on their western side, but the reverse on their eastern side.

71

The lows are regions of ascending, converging, cooling air, with in-
creasing moisture and clouds on their western side, but are the re-
verse on their eastern side (Moore, 'lo: 153). These same character-
istics apply to the small highs and lows which we are accustomed to
see on the daily weather maps.

If, now, we consider these large centers of action, such considera-
tion will do much toward giving us a graphic idea of our climate. Dur-
ing the winter, because of the small amount of heat received in the
Mackenzie basin, the temperature becomes very low, and a powerful
high barometric area is formed ; then the descending air blowing from
the eastern part of this high, or from small highs originating from the
larger one, produce the cold winters and cold waves in winter which
characterize the northeastern United States. If, however, the Atlantic
high wanders on the eastern coast of the United States in winter, the
zvestern part of this high, with its descending, diverging, warming, and
drying air, produces a mild winter. The climate of the eastern United
States is thus, in the cold season, under the alternate invasion of these
two powerful centers of action. During the warm season the conti-
nental winter high is replaced by a low, due to the accumulating warm
continental temperatures which thus have produced an inversion or
seasonal overturning. But the Atlantic high is permanent and exerts
its influence continuously. If the zvestern part of this high encroaches
upon the eastern United States during the summer, with its descend-
ing, drying, and clear air, it may produce drouth, this depending, of
course, on its degree of development. The continental low of sum-
mer, with the drying influence of its eastern side, has a similar ten-
dency. Thus the character of the summer is determined, to an im-
portant degree, by the interplay and relative balance between these twO
ivarming and drying centers. The activity of these centers has a pow-
erful influence upon the moisture-bearing winds, which influence hu-
midity and evaporation in Illinois, and in the eastern United States.

4. Relative Hnmidity and Evaporating Pozver of the Air

We are now in a position to examine the facts of relative humidity
and the relative evaporating power of the air in the eastern United
States. The relative aridity on the plains east of the Rocky Moun-
tains is due primarily to the removal of moisture from the prevailing
westerlies in their passage from the Pacific over the various western
mountain ranges which extend across their path, combined with the
excessive summer heating of the continental mass. Here, then, is the
influence of the continental summer low. Farther east the Atlantic
high tends to supplement the continental low and to cause the Gulf

winds to brings moisture inland,* and the Great Lakes region adds its
quota.

In the storm-track zone, where stagnation of the air is due largely
to the balance existing between the continental low and the oceanic
high, the aridity of the plains extends the farthest east, and as an arid
peninsula it crosses Illinois, giving during August a relative humidity
to the prairie area of 60-70 per cent, of saturation (Johnson, '07).
The reality of the arid peninsula across Illinois is further shown by
the rainfall-evaporation ratios computed and mapped by Transeau
('05). These ratios were determined by dividing the mean annual
rainfall at each place by the total mean annual evaporation. These
mapped percentages show that the prairie region is closely bounded
by the region with an evaporation ratio of between 60 and 70 per
cent, of the rainfall received. These conditions furnish a general
background or perspective for a profitable consideration of the local
and more detailed studies which have been made of the relative evap-
orating power of the air in different plant and animal habitats.

For our purpose it is not necessary to consider the history of meth-
ods of measuring relative evaporation. This measurement may be
made by evaporating water in open pans or by the porous porcelain-cup
method. Such cups have been devised by several students, but a modi-
fied form of the Livingston atmometer has been mainly used by plant
ecologists, and this was the kind we used at Charleston. Transeau
('08) was the first to use such an instrument and to show its value in
studying the relation of intensity of evaporation to plant societies.
His work on Long Island, N. Y., showed very clearly that evaporation
in open places was much greater than in dense forests. These obser-
vations were enough to show that evaporation is a factor related to the
physical conditions of life upon the prairie and in the forest, and there-
fore in our cooperative study of the Charleston area in 19 10 relative
evaporation was made a special feature in the study of representative
environments, in order to determine its relation to both the plants and
the animals. So far as is known this is the only study yet made in
which these determinations have been recorded from the same places
where the animals have been studied. Since our data were secured,
several papers have been published on relative evaporation in different
sorts of habitats in this state and in northern Indiana by plant ecolo-
gists Fuller ('11, '12a, '12b), McNuttand Fuller ('12), Fuller, Locke,

*Zon ('13) has recently asserted that the moisture from the sea does not
make a single overland flight inland, but rather is largely precipitated near the
sea, is evaporated and carried farther inland, is precipitated again, and this
process repeated again and again, so that its inland flight is a vertical revolv-
ing cycle of precipitation and evaporation. If this contention is valid, evapo-
ration from the land is a much more important climatic factor than it is usually
thought to be.

73

and McNutt ('14), Sherff ('12, '13a, '13b), and Gleason and Gates
('12). Shelford ('12, '13a, '13b, '14a), utilizing the evaporation
data of the plant ecologists, has applied the same to animal associa-
tions also, and he has further tested some of these ideas experiment-
ally in the laboratory. In Ohio, Dachnowski ('11) and Dickey ('09)
have made records of data obtained by the use of the porous cup, and
in Iowa Shimek ('10, '11) has used the open-pan method. Mention
should also be made of Yapp's observations ('09) on a marsh in Eng-
land. A very important summary of evaporation records, in the open
and in forests, is given by Harrington ('93). The effect of wind-
breaks upon evaporation has been studied by Bates ('11) and Card
('97). Finally, mention should be made of Hesselman's studies of
relative humidity in forest glades in Sweden ('04).

Our records from the Charleston region will be given first, and then
their significance will be discussed. The unglazed porcelain cups, with
a water reservoir, were placed so that the tops of the cups were about
six inches above the soil in the habitats examined, and at weekly in-
tervals the water loss was measured. The instruments were in opera-
tion simultaneously, so that the results are comparable. The standard
instru-ment was located in the open exposed garden of the Eastern
Illinois Normal School at Charleston, which was considered as unity,
or 100 per cent. For further details as to the conditions where the
atmometers were located consult the description of the stations and
the photographs.

An examination of the diagram (Fig. 2) will show that although
based upon a limited amount of data (for less than a month, from

Intensity of evaporation .

Standard, open garden, Normal School
Sta. Ill, b. Mixed prairie and young forest
Sta. II, a. Grassy area, Panicum
Sta. II, a. Grassy area. Euphorbia
Sta. IV, a. Upland, open woods
Sta. Ill, a. Silphium on black soil
Sta. II, a. Colony of S. laciniatum
Sta. IV, b. Ravine slope, open woods
Sta. IV, c. Dense climax forest cover

Fig. 2. Diagram of the relative evaporation in different prairie and forest
habitats, showing the great reduction in evaporation with the development of a closed
forest canopy of a climax forest; Charleston, Illinois.

74

August 19 to September 22) the facts are in harmony with similar
studies elsewhere covering a much longer period, so that there is valid
reason for confidence in them. The standard instrument was located,
as already mentioned, in an open, exposed cultivated garden, where the
intensity of evaporation was very high. The black soil prairie areas.
Stations II and III, a, have an average of 56.1 per cent. — a condition
much like that in the grassy-Euphorbia prairie at Loxa (Station II, a)
— or a little more than half that of the standard instrument. The dry
upland area of mixed prairie and young forest, on gray silt loam (Sta-
tion III, b), has an intensity of 80 per cent. This is in the region of
the most extensive grassy prairie about Charleston; the general ap-
pearance of the region is shown in Plate XIII. A surprising feature
of the table is the evaporation in the open-crowned upland oak-hickory
woods (Station IV, a). In this forest perhaps two thirds to three
fourths of the ground was shaded, and it was very well drained. The
evaporation here reached 54.2 per cent., being very near that of the
average of the black soil prairie (56.1 per cent.). I had anticipated
much less evaporation than on the prairie, a position more intermedi-
ate between the prairie and the lowland forest, or about 42 per cent,
(cf. Harvey, '14:95). The ravine slope (Station IV, b), although
somewhat open, has 31.5 per cent. — a very low rate of evaporation —
and is remarkably close to that of the densely crowned lowland for-
est (Station IV, c), at 26.9 per cent. The decline, however, in the
intensity of evaporation with the degree of completeness of the for-

Per cent, of standard

Sta. 11. Salt marsh outer margin

Sta. 3. Gravel slide, open

Sta. 1. Carnegie garden, standard

Sta. 9 and 10. Upper beach

Sta. 12. Salt marsh, inner margin

Sta. 2. Garden, high level

Sta. 4. Gravel slide, partly invaded

Sta. 5. Forest, open

Sta. 13. Fresh-water marsh

Sta. 6. Forest, typical mesophytic

Sta. 7. Forest, ravine type

Sta. 8. Forest swamp type

20

40

60

80

100

t20

■-

1 1 1 1 1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1

^IHBH^^HB

LX

■

f

Fig. 3. Diagram of the relative intensity of evaporation in the lowest stratum
of different kinds of habitats, Long Island, N. Y. (After Transeau.)

75

est crown, is strikingly shown in passing from the open upland
woods, at 54.2 per cent., to the ravine slope at 31.5 per cent., and on
to the lowland forest at 26.9 per cent.

A comparison of these results with those secured by Transeau
('08) on Long Island, is instructive. His standard instrument was
also in an open garden (Fig. 3), comparable with the Charleston
standard. A gravel slide, partly invaded by plants, had an evaporation
of 60 per cent., comparable with the open prairie at Charleston; the
open forest, 50 per cent., comparable with the upland open Bates
woods at 54.2 per cent. ; and the mesophytic forest, 33 per cent., com-
parable with the ravine and lowland places in the Bates woods at 31.5
and 26.9 per cent, respectively.

Association
Blowout (basin)
Blowout (slide)

Bunchgrass (Leptoloma consoc.)
Bunchgrass (Eragrostis trichodes con.
Standard
Beach

Quercus velutina woods
Quercus velutina
Willows (Acer part)
Willows (Salix part)
Mixed forest (margin)
Mixed forest (center)

Fig. 4. Relative intensity of evaporation in different kinds of habitats on sandy
soil, Havana, Illinois. (After Gleason and Gates.)

.CO

.40

.to

.80 1

.00 1

.so 1

J 40

1.56
1.27
1.18

) 1.04
1.00
0.93
0.66

^^^_

^^^^

^^^^

— 1 1

^__

^^™

^^^

^^^

^^"

"^^

^^B

^^"

""

"""

■

^^"

^^^

^^^

■

^^"

"J"

^™

^^^

^^^

^^^H

m

0.55

n^i^i

^^m

0.56

^^^^

^^_

^^^*

^"*

0.44

IBlHi

1

0.36

m^B

0.29

1^1

Another series of relative evaporation observations was made by
Gleason and Gates ('12) on sandy soils at Havana, Illinois. As their
methods were similar to those used at Charleston, useful comparisons
may again be made. The standard instrument was in an open area
comparable to the garden at Charleston. An examination of Figure 4,
summarizing the results of their study, shows that upon the grass-
covered sand prairie (bunch-grass) the evaporation was about no per
cent., that in open black oak (Q. velutina) woods (on sand) it was
about 60 per cent., and that in a denser hickory-black-oak-hackberry
mixed forest (somewhat open) it was about 31 per cent. There is thus
a close general correspondence between the conditions at Havana c^nd
Charleston, although the evaporation upon sand prairie appears to be
relatively much greater than upon the black-soil prairie.

Fuller ('11) and McNutt and Fuller ('12) have made comparative
studies in different kinds of forest in northern Illinois and in northern

76

Indiana. Their results are combined and summarized in Figure 5.
This diagram shows the relative evaporation near the surface of the
soil, the standard of comparison being the evaporation in a maple-
beech climax forest, where evaporation is relatively low. The aver-
age daily amount, in c.c, shows that there is a progressive increase in
evaporation as follows: 8.1 c.c. in a maple-beech forest, 9.35 c.c. in
the oak-hickory upland forest, 10.3 c.c. in an oak dune forest, 11.3 c.c.
in a pine dune forest, and an increase to 21.1 c.c, on the cottonwood
dunes. This expressed on a percentage basis is, in inverse order, re-
spectively 260 per cent, in the cottonwoods, 140 per cent, in the pines,
127 per cent, in the oak dunes, 115 per cent, in the oak-hickory for-
est, and 100 per cent, in the maple-beech forest.

20 ^0 60 eo 100 120 r40 160 160 200 22Q gdO 260 260

ndard

"

140
7%

SU

■ \?

115%

"

^00

Intensity of evaporation

Sta. A. Cottonwood dunes

Sta. B. Pine dune

Sta. C. Oak dune

Sta. D. Oak-hickory

Sta. E. Maple-beach forest

Pig. 5. Diagram showing the relative rate of evaporation in different kinds of
forest in northern Illinois and Indiana. [Data from Fuller ('11) and McNutt and
Fuller ('12).]

Shimek ('10, '11) has made valuable observations on the relative
rate of evaporation on the prairie of western Iowa. He used the open-
pan method in four representative habitats. His results show very
clearly that the rate of evaporation is much greater in exposed places
than where there is shelter from the sun and wind. I have put his
data in a form comparable with those which have just been discussed
(Fig. 6), and have made the cleared field area, Station 4, the standard
of comparison, as it more nearly approaches the standard used at
Charleston and by others. Station 3 is on a high bluff, exposed to the

Intensity of evaporation .

200

Sta. 3. Open, much exposed prairie
vegetation

Sta. 1. Open, exposed slope of bluflf,
prairie

Sta. 4. Open, cleared area, partly pro-
tected

Sta. 2. Bur-oak grove, protected

Fig. 6. Diagram of relative evaporation in prairie and forest habitats, in western
Iowa. (Data from Shimek.)

77

west and south winds, and, as might be expected, it has an excessive
evaporation — 184 per cent. Station i, also covered by prairie vegeta-
tion, and exposed to west and southwest winds but shehered from
winds from the south and southeast, also shows a very high evapora-
tion — 132 per cent. Station 4, which was made the standard, had been
cleared of forest, and was an open place protected by a ridge. Station
2 was apparently a dense grove composed of bur oak. basswood, elm,
and ash, with considerable undergrowth. Here the rate of evapora-
tion dropped considerably — to 36 per cent. The general character of
this forest calls to mind the denser oak forests on sand at Havana,
Illinois. An important feature of these observations is that they were
made far out upon the "prairie", bordering the plains, most other
studies on relative evaporation having been made much farther east.

In Ohio, Dachnowski ('11) and Dickey ('09) have recorded the
relative evaporation of the air, using a campus lawn as unity. In the
central grass-like area of a cranberry bog the evaporation was 69.2
per cent., and in the marginal maple-alder forest it was 51.2 per cent.

Harrington ('93:96-102), in summarizing European studies on
the relative evaporation (with a water-surface as standard) in the
open and in German forests shows that the "annual evaporation in the
woods is 44 per cent, of that in the fields." Compared with evapora-
tion in the open, that under deciduous trees is 41 per cent., and that
under conifers is 45 per cent. — a difference most marked in the sum-
mer. Ebermeyer's Austrian observations (I.e. .-99) show that the
"evaporation from a bare soil wet is about the same as that from a
water surface," both in the open and in the forest. A saturated soil
under forest litter gives an evaporation of only 13 per cent, of that
of a free-water surface in the open. Harrington (I.e.: 100) con-
cludes that "About seven-eighths of the evaporation from the forest
is cut off by the woods and litter together." Sherff ('13a, '13b) has
shown that in the Skokie Marsh, north of Chicago, the absolute
amount of evaporation near the soil was less at the center of a Phrag-
mites swamp than at its margin (Fig. 7), that a swamp meadow

Intensity of evaporation.

Sta. D. White oak-ash forest

Sta. B. Phragmites swamp, margin

Sta. C. Swamp meadow

Sta. A. Phragmites swamp, center

Fig. 7. Diagram of relative evaporation in Skokie Marsh area, near Chicago,
at 10 inches (25 cm.) above the soil. Eecalculated. (Adapted from Sherff.)

78

was in an intermediate position, and that in an adjacent white oak-ash
forest evaporation was about twice as much as in the swamp meadow.
Sherff used as standard the forest (D). This gave him for the center
of the swamp (A) 38 per cent., for the swamp meadow (C) 54 per
cent., and for the outer swamp margin (B) 105 per cent. In Figure
7, I have used his swamp meadow as 100 per cent., and by recalcula-
tion this gives the forest (D) 185 per cent., for the swamp margin (B)
105 per cent., and for the center of the swamp (A) 70 per cent. These
figures indicate a concentric arrangement of the conditions of evap-
oration about the swamp.

Intensity of eraporation

1907:

Sta. A. Above vegetation. 4
inclies above soil

Sta. B. Middle of vegetation.
2 inches above soil

10 20 30 40 50 60 70 80 90 100 MO

Sta. C. Lower vegetation,
above soil

feet, 6
2 feet,
5 inches

1908:

Sta. A. Above

vegetation. 5 feet, 6
inches above soil

Sta. B. Middle of vegetation. 2 feet,
2 inches above soil

Sta. C.

liower vegetation,
above soil

5 inches

i007.

»

■1 'x&y.

-

|32.

17.

■■i

lOO-/.

rr

r

HBPB M.7%

1

Fig. 8. Diagram showing the relative evaporation at different vertical levels in
a marsh in England, the evaporation in the lower layers of the vegetation being much
greater than in the upper strata or in the air above it. (Data from Yapp.)

Thus far, attention has been devoted solely to the horizontal differ-
ences in evaporation. There are also important vertical ones, vary-
ing above the surface of the substratum. Important observations on
this subject have been made, by a porous-cup method, in an open
grassy marsh in England, by Yapp ('09). The vegetation grew to a
height of two to five feet. From his data the accompanying diagrams
(Figs. 8, 8a) have been prepared. This shows that when the stand-
ard was made the rate of evaporation above the general level of the
vegetation, within the grass layer evaporation was reduced from about
one half (Sta. B, 1908, 56.2 per cent.) to one third (Sta. B, 1907,
32.8 per cent.) at 2 feet 2 inches above the soil; and that at 5 inches
above the soil it was reduced to between one fourteenth (Sta. C, 1907,
6.6) and one seventh (Sta. C, 1908, 14.7) of that above the vegeta-
tion. Yapp (1. c. : 298) concludes from his studies that "In general,
the results of the evaporation experiments show that the lower strata
of the vegetation possess an atmosphere which is continually very much

79

more humid than that of the upper strata, and farther, that the higher
and denser the vegetation the greater these differences are." This is
shown in Fig. 8a.

Intensity of evaporation.

Sta. A. 60 inches above ground, above
vegetation

Sta. B. 12 inches above ground among
vegetation

Sta. C. 3 inches above ground, among
vegetation

I

30

40

50 60

70

do go

71-:

50";

100'

Fig. 8a. Diagram showing the relative evaporation at different vertical levels in
a marsh in England, the evaporation in the lovFer layers of the vegetation being much
greater than in the upper strata or in the air above it. (Data from Yapp.)

In America only a few records have been made on vertical gra-
dients in evaporation, two of these in marsh areas, one in Ohio by
Dachnowski ('ii), and the other near Chicago by Sherff ('13a, '13b).
The Ohio observations, made upon a small island in a lake, in a cran-
berry-sphagnum bog, show that the rate of evaporation above the vege-
tation is much greater than among it, and that this diminishes as the
soil is approached, these results agreeing with those obtained by Yapp.
Sherff's observations were made in Skokie Marsh, north of Chicago,
and show that the relative evaporation also varies with different kinds
of swamp vegetation. From his data a diagram has been made (Fig.
9) in which the rate of evaporation in the upper part of the reeds

Intensity of evaporation

Phragmites
Sta. A. Within vegetation, 198 cm. (77
inches) above soil. Standard.

Sta. B. Within vegetation, 107 cm. (42
inches) above soil

Sta. C. Within vegetation, 25 cm. (10
inches) above soil

Sta. D. At soil surface

Typha

Sta. A. Within vegetation, 175 cm. (69
inches) above soil

Sta. B. Within vegetation, 107 cm. (42
inches) above soil

Sta. C. Within vegetation, 25 cm. (10
inches) above soil

Sta. D. At soil surface

Pig. 9. Diagram of relative evaporation at different vertical levels above the soil
within the vegetation of Skokie Marsh. (Adapted from Sherff.)

80

(Phragmites) at yy inches is taken as lOO per cent, or the standard.
Lower down, at 42 inches, the rate is 70 per cent., at 10 inches, 53 per
cent, and at the surface, 33 per cent. Among the cattails {Typha), in
the upper part of the vegetation, at 69 inches evaporation was 85 per
cent. ; at 42 inches it was 36 per cent. ; at 10 inches, 20 per cent. ; and
at the surface, 8.5 per cent. These results show that at successively
lower levels in the vegetation the rate of evaporation is greatly .re-
duced. They tend also to confirm the results of Yapp and Dachnow-
ski. It seems, then, fair to conclude that the rate of evaporation above
the swamp vegetation increases rapidly with downward progression,
and probably with upward progression also. A vegetable layer, com-
parable to the mulching of straw used by gardeners, thus acts as a pow-
erful conserver of moisture. There are great differences within a few
vertical feet in the open ; what is the condition within the forest ?

Intensity of evaporation.

Sta. A. Maple-beech forest. 6 feet (2m.)
above soil

Sta. B. Maple-beech forest. 10 inches

(25 cm.) above soil

Sta. C. Maple-beech forest. On slope of

ravine 30 feet deep (10 m. )
13.3 feet (4 m.) below general
surface.

20 ao 60 60 100 l?0 MO 160 180

;

100 5i

055 daL/sl

Standard

I

60%

""

Fig. 10. Diag:ram showing the relative evaporation in a beech-maple woods, six
feet above the soil (A), near the surface of the soil (B), and in a ravine (C).
[Adapted from Fuller ('12).]

The character of vertical differences in evaporation within the for-
est has not been given as much attention as the similar changes in the
open; but attention has already been called to the moisture-conserving
effect of a forest litter, the evaporating rate in one instance being only
13 per cent, when compared with that from a water surface in the open.
McNutt and Fuller ('12) have shown that grazing in an oak-hickory
forest changed the average daily rate of evaporation for 189 days
from 9.89 c.c, in the ungrazed forest, to 12.74 c.c, in the grazed for-
est, at Palos Park, 111. There are thus, within the forest, changes in
evaporation with differences both in the ground cover and in the litter
on the forest floor which correspond to the change in the vegetation in
open places.

Vertical differences in evaporation have been tested in a maple-
beech-forest in northern Indiana by Fuller ('12b), who used the po-
rous-cup method. His results have been summarized in Figure 10.
This diagram shows that the evaporation at six feet above the surface
is nearly twice as much as that at 10 inches above the surface, and

81

that in a ravine, 13.3 feet (4 m.) below, it was 80 per cent, of that 10
inches above the surface. The relative seasonal activity from May to
November is shown in Figure 11. This diagram shows that after the
leaves appear the highest evaporation takes place in July. This is
probably the critical season for some animals.

MAY

JUNE

JULY

AUGUST

SEPTEMBER

OCTOBER

-

—

a

\

20

\

/

\

\

^

s

/

\

\

\

\

/

\

/

\

-

15

\

\

,b

\

\

\

N

c

\

/

\

k,

10

\

\

\

/

\

N.

\

N

s

\

/

f

V

s

\

\

\

/

■^

/

/

\

\

^

—

\,

S

1,

\;

\

/

f

/

\

\

\

\

\

\

/

/

—

\

^

■—

\,

\

^

5

\ ^

f

/

^

\,

s,

■~-

\

/

>

s

-\

,

V

>5

Fig. 11. Diagram showing the average daily rate of evaporation in beech-maple
forest, six feet above soil (a), near the surface of soil (b), and in a ravine (c).
(From Fuller.)

In the forest, Libernau (Harrington, '93 : 34) found that the "rela-
tive humidity increases and decreases with the absolute humidity,
whereas it is known in general, and also at the Station in the open
country, that these two climatic elements are inverse. This is ac-
counted for by the fact that the forest is a source of atmospheric
aqueous vapor as well as of cooling." (L. c : 104: "The absolute
humidity decreases in the forest from the soil upwards. The rate of
decrease is usually the greatest under the trees and the least at the level
of the foliage. The rate above the trees is intermediate between the
other two. This rate is least in the late hours of the night, when it
may be zero. It increases with the increase of the temperature of the
air, becoming greatest in the midday hours, when, under exception-
ally favorable circumstances, it may make a difference of 10 per cent.

82

or even more. Occasionally, in high winds, the absolute humidity is
greater over the trees. Over the field station the daily progress of ab-
solute humidity was about the same as in the forest, but the maximum
difference was only about half as great. The absolute humidity in and
above the forest is greater than that over the open fields, and there is
some trace of an increase of this difference to the time of maximum."
A greater relative humidity has been found over evergreen trees
than over deciduous trees, which is slight (I.e.: 104), but the psy-
chrometer was close to the evergreens and farther above the decidu-
ous ones.

Intensity of evaporation .

10 20 30 40 SO 60 7

80 90 100

1^

wm

100;.

^

■■

\

1 31 X

r

■1

■ -. .

Neb./

(Julii 15- Sept IS, 62 d^Lts, Uncoln.

Sta. A. 20 rods (330 ft.) from wind-
break, 25 to 40 feet high.
Standard

Sta. B. 12 rods (198 ft.) from wind-
break

Sta. C. 3 rods (49.5 ft.) from -niud-
break

Fig. 12. Diagram showing relative retardation of evaporation by a windbreak,
Lincoln, Nebraska. [Adapted from Card ('97).]

The border of the Illinois forest and prairie was characterized by
tongues and isolated groves of forest and by glades. The forest had
the same kind of influence as windbreaks upon the leeward areas and
glades, and therefore the influence of windbreaks upon the evaporating
power of the air is of interest. Card ('97) made a valuable study of
this series of problems at Lincoln, Nebraska. The influence of wind-
breaks upon evaporation is summarized in Figure 12. This diagram
shows that leeward of a close windbreak ranging from 25 to 40 feet
in height, the rate of evaporation in terms of the standard (A), which
was 330 feet leeward, was 91 per cent, at a distance of 198 feet (B),
and 71 per cent, at 49.5 feet (C), thus showing, a marked reduction
with proximity to the windbreak. These observations covered 62 days.

Nearer to Illinois, similar though very limited observations were
made in central Wisconsin by King ('95) which agree with Card's
on the retardation of evaporation by windbreaks. His results are
shown graphically in Figure 13.

Recently Bates ('11) has made an elaborate study of the effects of
windbreaks upon light, soil, moisture, velocity of wind, evaporation,
humidity, and temperature. His results confirm those just given and
give additional facts which, however, with one exception, will not be
mentioned. The paper itself should be consulted. This investigation
by Bates shows that in proportion to the perfection of the windbreak

88

a quiet, stagnant air strip is formed to the leeward, and that this fa-
vors excessive heating during clear days and low temperatures on clear
nights. Years ago Harrington ('93:119) suggested this idea and
called attention to the close relation existing between the leeward con-
ditions of windbreaks and forest glades. The glade climate is more
rigorous, or extreme, than that upon plains (I.e.: 19, 84-88, 119).
Such a climate is thus a bit more "continental" during the spring, sum-

1

D 2

1 i

40

50

60 70

ao 90

lOO 110

100 •/.

1 lOlV.
1 101'/.

I

5%

1

7^\

^^

""

""

^

Intensity of evaporation

Distance from windbreak 12 inches high :

Sta. F. 500 feet leeward. Standard

Sta. E. 400 feet leeward

Sta. D. 300 feet leeward

Sta. C. 200 feet leeward

Sta. B. 100 feet leeward

Sta. A. 20 feet leeward

Fig. 13. Diagram showing the relative evaporation, May 31, at different dis-
tances leeward of a windbreak, Almond, Wis. [Adapted from King ('95).]

mer, and autumn. These glades are very hot in the early afternoon
and cool on clear nights, and the air is relatively stagnant ; as Harring-
ton says, it is "lee for winds from all directions." The center of a
dense forest may thus possess physical conditions quite different from
those of the glade forest margin or in the open. Beginning with the
relatively stable conditions within a forest toward its margin, the diur-
nal temperature variations are much more extreme (Harrington,
1. c. : 89) "to a distance of a score or so of rods where it reaches a max-
imum. The amplitude is greater in glades. Hence the extremes of
temperature are exaggerated just outside the forest." The annual soil
temperatures of a glade are intermediate between that of the forest and
the plain. The forest margin is thus seen to possess many of the char-
acteristics of the glade, for its climate is somewhat more extreme than
that in the open, far from the forest.

5. Temperature Relations in the Open and in Forests

The temperature relations in open and forested regions are often
very different. The density of the vegetable covering in the open and
in the forests varies much and may have considerable influence upon
animals. Yapp ('09) observed that the marsh vegetation in England

84

caused marked vertical differences in temperature in the vegetational
stratum. He summarizes these results as follows (p. 309) : "The
temperature results show that the highest layers of the vegetation pos-
sess a greater diurnal range of temperature than either the free air
above or the lower layers of the vegetation. Regularly, especially in
clear weather, both the higher day and the lowest night temperatures
were recorded in this position."

Dachnowski ('12: 292-297) studied the temperature conditions in
a cranberry bog substratum in central Ohio. He found that at a time
when ice formed from 8 to 15 inches thick on the adjacent lake, in the
bog it was only 3 to 5 inches thick, and there were small patches where
it did not form at all. At a depth of 3 inches in the peat the tempera-
ture ranged from 33° to '/'/° F. (.5 "-25.0° C). In the bordering
maple-alder zone, at 3 inches depth it ranged from 33° to 72° F. (.5''-
22.0° C. ). His observations indicate that the temperature relations
within the maple-alder zone are more stable than those in the open
central area.

Cox ('10) has also shown that the character of the vegetation in
Wisconsin cranberry bogs has much influence upon temperature rela-
tions in this habitat.

It seems very probable that similar conditions hold over prairie
vegetation, but I do not know of any observations on this point. We
are all familiar with the common practice of gardeners of using a mulch
of straw to retard temperature changes under it ; prairie vegetation
must have a similar influence. (Cf. Bouyoucos, '13: 160.)

The relative air temperatures within and without the forest show
a distinct tendency to reduce the maxima and minima, and to lozver
the mean annual temperature. Harrington ('93:53) concludes,
therefore, that ''the forest moderates (by reducing the extremes) and
cools (by reducing the maxima more than the minima) the tempera-
ture of the air within it. The moderating influence is decidedly greater
than the cooling effect." These effects are not uniform, but are much
more marked in the summer, and Harrington further says : "The cool-
ing effect tends to disappear in winter. The moderating effect is the
most important one and it is the most characteristic" (p. 56).

The temperature relations within the forest crown show that in
general the effects are similar to those found at an elevation of about
5 feet. The maxima are lowered, the minima are elevated, and there
is a cooling effect. The differences are most pronounced during the
summer, and the temperatures are intermediate in position between
those at the five-foot level and those in the open (I.e.: 66). At a
height of 24 feet, deciduous trees showed a marked summer cooling

85

effect, while evergreens showed much less, though they are much more
uniform for 9 months of the year. Again, he says: "In summer the
average gradient under trees is about +2°; that is, it grows warmer
as we ascend at the rate of two degrees per 100 feet (31 m.). Out-
side in the general average it grows colder by about a quarter of a de-
gree." This warmer air above the cooler in the forest favors its sta-
bility or relative stagnation, although as a wdiole the forest air is cool-
er and heavier than the surrounding air and tends to flow outward.
The forest thus tends to produce a miniature or incipient barometric
high. In conclusion Harrington (p. 72) states that "The surface of
the surface of the forest is, meteorologically, much like the surface of
the meadow or cornfield. The isothermal surface above it in sun-
shine is a surface of maximum temperature, as is the surface of a
meadow or cornfield. From this surface the temperature decreases in
both directions." In the case of a beech forest the warm diurnal layer
above the forest crown was only 6.5 feet thick (p. 34).

The conditions above the forest are thus representative of the at-
mospheric conditions above dense vegetation in general, and are in per-
fect harmony with Yapp's observations upon the temperature above a
marsh ('09: 309), quoted on a previous page, to the effect that tem-
perature changes are extreme here, and greater than in the free
air above or in the lower layers among the vegetation. The forest is
thus to be considered as a thick layer of vegetation in its influence upon
meteorological conditions. The conditions above the forest, there-
fore, exemplify a general law.

In general terms, the temperature of the soil below the zone of
seasonal influence is that of the mean annual temperature for a given
locality. The surface zone, however, varies with the season. Har-
rington ('93) has summarized the German observations on the rela-
tive soil temperatures in the open and in the forest. In the following
c|uotation the minus sign indicates a forest temperature less than a cor-
responding observation in the open. These temperatures were taken
about 5 feet above the soil. He says (p. 43) : "The average of the
seventeen stations (representing about two hundred years of observa-
tions) should give us good and significant results. It shows for the
surface — 2°. 59, for a depth of 6 inches (152 mm.) — 1°.87, and for
a depth of 4 feet (1.22 m.) — 2°. 02. The influence of the forest
on the soil, then, is a cooling one, on the average, and for central
Europe the cooling amounts to about two and a half degrees for the
surface. The cooling is due to several causes: The first is the shade;
the foliage, trunks, branches, and twigs cut off much of the sun's
heat, absorb and utilize it in vegetative processes, or in evaporation, or.
reflect it away into space. Thus the surface soil in the forest receives

■ 86

less heat than the surface of the fields. The same screen acts, how-
ever, in the reverse direction by preventing radiation to the sky, thus
retaining more of the heat than do the open fields. The balance of
these two processes, it seems from observation, is in favor of the first

and the average result is a cooling one The dififerences of

temperature at the depth of 6 inches (152 mm.) are more than half a
degree less than at the surface. In this is to be seen the specific effect
of the forest litter; it adds a covering to that possessed by the sur-
face, so that while the deeper layer is cooled as much by the protec-
tion from the sun's rays as is the surface, it is not cooled so much by
radiation of heat to the sky. Its temperature is, consequently, rela-
tively higher, and approximates somewhat more the field tempera-
tures."

"The forest soil is warmer than that of the open fields in winter,
but cooler in the other seasons, and the total cooling is much greater

than the warming one The forest, therefore, not only cools

the soil, but also moderates the extremes of temperature" (p. 46).

The character of the forest, whether evergreen or deciduous, in-
fluences the temperature conditions of the soil, as is seen by a com-
parison of these conditions in the forest and in the open. The two kinds
of forest are much alike in winter; during the spring the soil warms
up more rapidly under conifers. Temperature variations are slightly
greater under deciduous trees.

6. Soil Moisture and its Relation to Vegetation

The moisture in the soil is derived largely from precipitation, but
part of it, in some localities, comes directly from the adjacent deeper
soils or rocks, and thus only indirectly from precipitation. As Illinois
lies at the bottom of a large basin, there must be some subsurface flow
from the adjacent higher regions, but to what extent is not known.
McGee ('13a: 177) estimates that the general ground-water level —
the level at which the soil becomes saturated — has, since settlement, de-
clined 10.6 feet in Illinois. This decline is not limited to drained re-
gions but is a general condition. In addition to these changes of level
there are seasonal fluctuations. Sherff ('13a: 583) observed in Skokie
Marsh that the water-table was at or above the surface in May, then
declined until early September, and then rose rapidly to the surface by
the middle of October. The wet prairie at Charleston has undergone
just such changes as these ; the ground-water level has been lowered
and there are marked seasonal changes.

Harvey ('14) has recently shown that the soil of Bryngium-Sil-
phium prairie at Chicago contains a large amount of water during

87

April and until late in May ; that the moisture falls and is low during
July and August, with a mean of 24 per cent, of saturation for these
months ; but that in October the soil is again at or near the point of
saturation.

The blanket of humid air which accumulates under a cover of vesfe-
tation, retards evaporation and conserves soil moisture. The denser
the vegetation the more marked is its influence. The litter — the or-
ganic debris in an early stage of decomposition — on the forest floor
has the same tendency, and has even a greater water capacity than the
soil itself. On the other hand, a forest is a powerful desiccator; as
Zon ('13:71) has recently put it: "A soil with a living vegetative
cover loses moisture, both through direct evaporation and absorption
by its vegetation, much faster than bare, moist soil and still more than
a free water surface. The more developed the vegetative cover the
faster is the moisture extracted from the soil and given off into the air.
The forest in this respect is the greatest desiccator of water in the
ground." This drying effect is shown particularly near the surface
of the soil, where roots are abundant and where drouth is so marked
that it may prevent the growth of young plants here (cf. Zon and
Graves, '11 : 17-18).

Warming ('09:45) says: "It may be noted that, according to
Ototozky, the level of ground-water invariably sinks in the vicinity of
forest, and always lies higher in an adjoining steppe than in a forest ;
forest consumes water."

McNutt and Fuller ('12) have made a study of the amount of soil
moisture at 3 inches (7.5 cm.) and at 10 inches (25 cm.) below the
surface in an oak-hickory forest, at Palos Park, Illinois. They found
that the percentage of water to the dry weight of the soil at the 3-inch
level averaged 18.9 per cent, and at 10 inches was 12.5 per cent, of the
dry weight of the soil. The greater moisture near the surface is due
to the humus present in this layer. The grazed part of the forest
possessed less soil moisture, and shows the conserving effect of vege-
tation. (Cf. also Fuller '14.)

The artificial control of soil moisture is well shown by the effect of
windbreaks. Card ('97) studied the moisture content of the soil to
leeward of a windbreak and found that in general there is a "de-
crease in the per cent, of water as the distance from the windbreak
increases." As the physical conditions leeward of windbreaks are
similar in many respects to those in forest glades and forest margins,
it is very probable that the conditions of soil moisture also will be very
similar in these places.

88
7- Ventilation of Land Habitats

«

The preceding account of the temperature, humidity, and evapo-
rating conditions in various habitats forms a necessary basis for an un-
derstanding of the processes of ventilation or atmospheric change in
land habitats. The differences in pressure due to the different densi-
ties of cool and warm air and to the friction and retardation of mov-
ing air currents, determine to an important degree the composition
of the air in many habitats. In such an unstable medium as air,
changes take place very rapidly through diffusion, and through this
constant process of adjustment there is a tendency to level off all local
dift"erences. These are naturally best preserved where diffusion cur-
rents are least developed — in the most stagnant or stable atmospheric
conditions ; therefore any factor which retards an air current and pro-
duces eddies, or slow diffusion, will favor local differentiation of
the air.

We have seen that any vegetable cover retards air currents, so that
the air within the vegetation becomes different from the faster moving
air above it. The accumulation of humidity at different levels above
the soil within the vegetation, clearly shows this. The denser the vege-
tation the more completely are the lower strata shut off and, to a cor-
responding degree, stagnant and subject to the local conditions. Two
factors have an important influence upon these conditions : the charac-
ter of the cover itself, and the character of the substratum. If both
of these are mineral rather than organic, in general comparatively
little local influence is to be expected, although in some localities CO2
escapes from the earth and on account of its density may linger in de-
pressions and thus kill animals (Mearns '03). Generally, however,
the organic materials are of most importance both as a cover and as a
substratum, and are often the source of carbon dioxide. Living vege-
tation may also add oxygen to such stagnant air, but the main source of
it is the free air itself. The forest litter, on account of its imperfect
stage of decay, consumes oxygen and gives off carbon dioxide ; in the
humus below it, shut off even more from free access to air, the carbon
dioxide is relatively more abundant and the oxygen relatively less so
or absent; and in the deeper mineral soil the amount of carbon
dioxide is relatively less on account of the absence of organic debris,
and a small amount of oxygen is present.

The aeration of the soil is influenced to a large degree by its poros-
ity; the looser it is, the freer the circulation. Buckingham ('04) has
shown that "the' speed of diffusion of air and carbonic acid through
these soils was not greatly dependent upon texture and structure, but
was determined in the main by the porosity of the soil. . . . the

. 89

rate of diffusion was approximately proportional to the square of the
porosity .... the escape of carbonic acid from the soil and
its replacement by oxygen take place by diffusion, and are determined
by the conditions which affect diffusion, and are sensibly independent
of the variations of the outside barometric pressure."

In the upper, better ventilated, moist, neutral or alkaline layers of
vegetable debris decomposition is brought about mainly by the agency
of fungi ; but in the deeper, poorly ventilated acid layers, lacking oxy-
gen, bacteria are the active agents (cf. Transeau, '05, '06). The
higher the temperature the more rapid the circulation, and on this ac-
count ventilation in the open is relatively more rapid than in the cooler
woodlands. The black soil prairies are thus favorable to a higher tem-
perature and better ventilation. Dry soil, according to Hilgard
('06: 279) contains from 35 to 50 per cent, its volume of air, and in
moist or wet soils this space is replaced by water. Thus the condi-
tions which influence the amount of water present have a very im-
portant influence upon aeration. As water is drained from the soil, air
takes its place ; so drainage and the flow of water through the soil facil-
itate ventilation. The part of the soil containing air is thus above the
water-table ; and as this level fluctuates with the season and from year
to year the lower boundary of this stratum is migratory. Hilgard
states that cultivated garden soil contains much more air than uncul-
tivated forest soil. Warming ('09:43) says that the "production of
acid humus in the forest leads to an exclusion of the air." If lime is
present, such an acid condition can not arise.

While the source of oxygen in the soil is the air, the reverse is the
case with carbon dioxide. The surface layers of the soil, among
dense vegetation, constitute an area of concentration of carbon
dioxide. Because this is more soluble than other gases, it is found
in rain water, according to Geikie, in a proportion 30 to 40 times
greater than in the air. Rains thus assist in the concentration of
carbon dioxide in the soil. This concentration is well shown by the
following table by Baussungault and Lewy (Van Hise, '04:474).

Character of soil air

CO3 in

10,000 parts

by weight

1.

Sandy subsoil of forest

38

2.

Loamy subsoil of forest

124

3.

Surface soil of forest

130

4,

Surface soil of vineyard

146

5.

Pasture soil

270

6.

Eieh in hiunus

543

90

The amount of carbonic acid in the atmosphere is by weight about
4.5 parts in 10,000. The amount in the air is, as Van Hise says, "in-
significant in comparison with the amount in soils in regions of luxu-
riant vegetation. In such regions the carbon dioxide is from thirty to
more than one hundred times more abundant than in the atmosphere."
This carbonic acid in the presence of bases, sodium, potassium, cal-
cium, and magnesium compounds, forms carbonates c'nd bicarbonates.
This is the process of carbonation — one of the most important proc-
esses of change in surface soils.

In view of the dominance of CO2 in soils we may anticipate that
many of the animals living in them possess some of the characteristics
of the plants, bacteria, fungi, etc., which are active in such soils. The
anaerobic forms live without free oxygen ; others live only where oxy-
gen is present. The animals which thrive in the soil are likely to be
those which tolerate a large amount of COo and are able to use a rela-
tively small amount of oxygen, at least for considerable intervals, as
when the soil is wet during prolonged rains. This is a subject to
which reference will be made later.

The air is the main source of oxygen, and from the air it diffuses
into the soil ; thus the process of ecjuilibration is constantly in progress.
Carbonic acid, also present in the air, is washed down by rain and
concentrated in the soil, where it is increased by the decay of organic
debris and by respiring animals to such an extent that it exists under
pressure and diffuses into the air, thus contributing to the air. In the
soil, then, the process of decarbotiisation is of great importance to
animal life, and must not be neglected. The optimum soil habitat is
therefore determined, to a very important degree, by the proper ratio
or balance between the amount of available oxygen and the amount of
carbon dioxide which can be endured without injury. The excessive
accumulation of carbon dioxide, an animal waste product, is compar-
able to the accumvilation of plant toxins which may increase in the
soil to such a degree as to inhibit plant growth. Such substances
must be removed from the soil, or changed in it to harmless com-
pounds, or plants and animals can not continue to live in certain
places. I have used the term ventilation to cover both the oxvgena-
tion and decarbonization of land habitats, and the same principles
are applicable to life in fresh-water habitats.

We have just seen how atmospheric ventilation favors the removal
of certain injurious waste products from the air and soil. In addition
to gaseous waste products there are also liquids and solid kinds which
may be equally harmful in a habitat. These are known to exist in con-
fined liquids, as in acjuaria (Colton, '08; Woodruff, '12), where they

91

interfere with the welfare of the animals present, and it is probable
that they also exist in soils. The older naturalists elaborated the idea
that if organisms were not such active agents in the destruction or
transformation of plant and animal bodies such remains would soon
encumber the earth. Thus organisms themselves are among the most
active agents in influencing directly and indirectly the ventilation of
animal habitats.

8. The Tree Trunk as a Habitat

A living tree trunk is composed of wood, sap (moisture), and
bark, all of which are relatively poor conductors of heat. When the
trunks are cooled, as in winter, they are slow in warming, not only
because of poor conduction but also because of the slow circulation of
sap, which is derived from the cool ground-water. As the season
progresses, the trunks warm up, this process being retarded in part by
the shade and the cool forest conditions ; and in the fall, radiation of
the heat accumulated also takes place slowly. The tree trunk therefore
changes its temperature slowly, as does the soil. The animals which
live within wood thus live in a relatively cool and stable environment.
In living trees the humidity is relatively high, as it may also be in
fallen, deca,ying logs. Relatively dry logs, before progress of decay,
on the other hand, form a relatively dry and uniform habitat. (Cf.
on the temperature of trees : Harrington, '93, pp. 72-75 ; Packard,
'90, p. 23 ; and Jones, Edson, and Morse, '03, pp. 97-100.)

p. Prairie and Forest Vegetation and Animal Life

The dependence of animals upon plants for food is one of the most
fundamental animal relations. It is a world-wide relation, but its
mode of operations varies greatly in different enviroimients. For ex-
ample, many years ago, Brooks gave us a graphic picture of the role
of marine vegetation in the economy of marine animals. In the sea
there are no forests or grasslands, and no corresponding animals as-
sociated with these conditions, as on land ; but in the sea great numbers
of minute plants float, and upon these feed an immense number of
small crustaceans and other small animals. These small creatures
occur in such large numbers that at times the sea is a sort of gruel
which sedentary and stationary kinds may appropriate by simply al-
lowing the sea to flow into their mouths. The food here circulates in
their environmental medium, as plant foods do in the soil and air. This
condition has made it possible for vast numbers of plant-like animals
to grow over the sea floor as plants do over rocks and plains. The
living meadows of animals thus furnish pasture for a host of preda-

92

ceous kinds ; and upon these still others prey, so that flesh-eating ani-
mals make up the most conspicuous classes of marine animals. Quite
otherwise are the conditions on land, where no air current carries food
to the hungry mouths of animals. Plants with roots in the soil and
stems in the air are able, however, to secure their food from the cir-
culating medium, but being themselves fixed, they are easy prey to
animals — both the sedentary kinds, which live in or upon the plant tis-
sues, and the active wandering kinds, which forage over large areas.
The predaceous animals, either by active mind or body, must secure
their food from the plant-feeding kinds. The great expanses of grass-
land and forest tend to be devastated by a vast army of animals which
far outnumber the predaceous kinds. The conditions of life, there-
fore, found upon grassland areas, like the prairie, and in the forest,
are to the farthest possible extent removed from those found in the
sea. This, then, is one of the most fundamental contrasts in the con-
ditions of existence encountered by animals.

These considerations naturally raise the question to what extent
and in what particular manner does land vegetation influence animal
life ? Does a change in the vegetation as great as that between the for-
est and the prairie have a marked influence upon animals? In the
Charleston region we have just such a difference in the vegetation.

Many years ago Bates pointed out repeatedly in his "The Natural-
ist on the River Amazons" that the animals of that densely forested
region were to a marked degree distinctly arboreal and "adapted" to a
forest life. In most densely forested regions like conditions probably
prevail, and to a corresponding degree open lands harbor animals
equally characteristic and as truly terrestrial in habits. The contrast
between the conditions of life in the open and in the forest is one of
the most fundamental environmental conditions upon land. The sig-
nificance of this contrast seems to have been realized only in part. The
prairies or grasslands are representative of only one kind of open ;
they are caused by many kinds of factors limiting the extension of
forests. Open places are formed by lakes, ponds, and swamps ; by the
avenues through forests formed by different kinds of streams, as
brooks, creeks, and rivers ; by the small amount of soil on rock sur-
faces; and by still other kinds of limiting influences, such as the sea,
severe climate, and altitude. Among almost all of the major taxo-
nomic groups of land animals is seen the independent origin and pres-
ervation of animals suited for life in the forest; this clearly points to
the extensive influence and antiquity of this environment. The same
is true of animals living in the open. But to assume that it is solely
the kinds of forest trees serving as food for animals, or the cor-
responding kinds of vegetation in the open, which determines whether

93

an animal lives in the open or in the forest, would be unwarranted in
the light of the preceding discussion of the effect of vegetation upon
air temperatures, winds, humidity, relative evaporating power of the
air, and corresponding changes in the soil. Animal life is most
abundant in a narrow vertical layer above the earth's surface, by far
the most of it is within a few inches or feet of the surface ; and above
the level of the forest-crown it diminishes with great rapidity. Be-
low the surface of the soil the same general law holds; most of the
ground animals are within the first few inches of soil, only a small
number extending a few feet below the surface, and those found at
greater depths being indeed very few. The rate of decline is many
times more rapid below the surface than it is above it. There is, then,
above and below the surface a rapid and progressive attentuation of
the favorable conditions for animals and plants, and the animals do
not establish thriving communities far from those physical conditions
which are also favorable to vegetation. Animals are dependent upon
plants for food, but both are dependent upon a certain complex of
physical conditions near the surface of the earth.

It is well to recall at this point how the influence of the climate and
the vegetation exemplify certain general laws which operate in all hab-
itats. The differentiation of habitats upon the earth is primarily due
to temperature and the specific heat relations of the earth, which re-
sult in the several media — gases, liquids, and solids. With a higher
temperature all would be gas, and with a lower one all would be solidi-
fied. The present intermediate conditions, therefore, permit the pres-
ent differentiation. These media are further differentiated by tem-
perature about as follows : Since the source of solar energy, heat, and
light, and the oxygen supply, are above the surface of the earth, the
vertical attenuation of these influences is one of the most striking
peculiarities of animal habitats, both in water (where the causes have
long been recognized) and upon land. Any covering of the earth,
even the surface layer of vegetation, soil and water, tends to shut off
heat, light, and oxygen. At the same time such a layer tends to shut
in those influences which originate primarily in or below it. Thus car-
bonic acid originating under the cover, by organic decay, breathing
animals, or bacteria, or washed in by the rain, tends to be shut in.
Furthermore, heat once reaching here, either in water or on land, tends
toward slow radiation. Thus we may look upon the surface layer as a
partition which is under pressure from both sides, and through which
constant interchange is in progress, as the process of dynamic equili-
bration operates.

94

This attenuation of intensities, above and below the surface, pro-
duces vertical layers of relatively equal strength or pressure. Thus
the attenuation of temperature in gases (air) and in liquids (water)
causes different densities in air and in water which modify to an im-
portant degree the physical and chemical conditions in these media.
This results in their stratification: when the heavier layers are below,
stability is the tendency ; and when the reverse order obtains, a
change takes place toward the stable condition. With stratification,
fiowage tends to occur within the strata, and to be horizontal rather
than vertical; additional pressure is therefore necessary to cause the
vertical currents or circulation under such conditions. This is why
carbonic acid accumulates in the soil and in small deep lakes abound-
ing in organic debris, this accumulation being largely due, in both
cases, to the slow rate of exchange caused by the stratification pro-
duced by differences in density. This same relative stagnation is a
primary ^factor in the vertical differences in the relative evaporating
power of the air within a vegetable layer of the prairie or the forest.
Though on the prairie the vegetational layer is generally but a few
inches or a few feet thick, in the forest it is about eighty feet, or
more, thick ; and the forest thus influences atmospheric conditions
solely as a thick layer of vegetation.

Differences, then, in the character, structure, or composition of the
surface of the substratum are of fundamental importance in under-
standing its relative influence upon animals. Primarily these differ-
ences are due to temperature, secondarily to temperature in combina-
tion with moisture ; and they result in the relative humidity and the
relative evaporating power of the air. The most important difference
in the surface layer in the Charleston region is that of prairie and for-
est, and therefore the main features of these habitats will now be sum-
marized. It should not be overlooked that conditions on the prairie are
likely to be quite representative of open places in general, though they
will probably be somewhat unrepresentative in the case of open places
having wet or extremely dry substrata. It is also true that the condi-
tions produced by the forest are comparable, in some degree, with
those due to the influence of an elevation.

95

Summary of Environmental Features of the Prairie and the Deciduous Forest
— Temperature, Humidity, and Evaporation — during the Growing Season

Prairie

Ahove the Vegetation

Sim, maximum heated stratum.

In

Cooler above and below this stratum.
Absolute humidity less than in or over

forest.

Forest

Above crown, in sun, maximum heated
stratum. A thin layer. Cooler above
and below this stratum.

Absolute humidity greater than in the
open.

Prairie

Among the Vegetation

Temperature lower and higher than in

the forest — more extreme.
Temperature lower toward the soil, and

warmer than in the forest.

Absolute humidity progressively increases

toward the soil.
Relative evaporation decreases toward the

soil; greater than in the forest.

Forest

Temperature moderated — not as low or
as high as on the prairie.

Temperature lower toward the soil, and
cooler than in the open.

Absolute humidity progressively increases

toward the soil.
Relative evaporation decreases toward the

soil; less than in the open.

In the Soil

Prairie

Temperature averaging warmer than
forest, warmer near surface in sum-
mer, and cooler in winter. Warmer in
sun and cooler at night than in forest.

Temperature progressively more stable
downward. Soil moisture increases
downward.

Forest

Temperature cooler on the average and
in summer, and warmer in winter, near
the surface, than in the open. Cooler
in sun and warmer at night than in
the open.

Temperature progressively more stable
downward. Soil moisture, below the
surface layer, increases downward.

*tai

The conditions on the prairie and in the forest may be graphically-
shown as in the following diagrams. Figure 14 showing the tempera-
ture relations, and Figure 15 showing the relative evaporating power
of the air.

96

I

i

a
o

>

,0

o

o

o

a

c3

•ii-i
O

OH

a

o

3

ID

o

s

97

+
O

o

a
o

m
«

O

U
ei

'^1

«

a
B

•iH
02
03

tH
• iH

CS

«H

O

u
o

»i

bo

.9

a
u
o
p<

>

<s

>

bo

■?

o

.a

(t

60
«

d

98

lo. Sources and Role of Water used by
Prairie and Forest Anim,als

The bodies of animals contain a very large proportion of water —
from 60 to 95 per cent. Growing animals in particular require water
in relatively large amounts. Practically all foods gain entrance into the
body in acpeous solutions, and are transported by water to all parts ;
and by the same means, the waste products, with the exception of the
excretion of carbonic acid, are removed. The methods by which aquatic
animals secure water are relatively simple, because they live in a
liquid medium; but the conditions upon land are quite different. Here
osmotic pressure does not operate as in water, and the air varies from
saturation to a very dry condition. This dryness tends to cause strong
evaporation from animals living in such a medium, and a proper bal-
ance between intake and water-loss is one of the most potent influences
in the life of land animals. In this relation lies the importance of the
sources of water available to them. These sources are as follows :
with the food, by drinking, from the atmosphere, and by metabolism.
The loss is by excretion and evaporation, the relative humidity and the
evaporating power of the air being, therefore, important considera-
tions. The loss of water is retarded in many ways. Some animals
possess a relatively impermeable skin, or a covering, as hair or feath-
ers, which retards air currents and evaporation through the skin, just
as a cover of vegetation retards soil evaporation. Other animals con-
serve their moisture by modes of behavior, being active mainly during
the cooler night, thus escaping the excessive evaporation of the heated
day; and still others live in burrows in the soil, where the humidity is
higher than in the air. Many animals can live only where the air is
humid. There is thus an almost endless series of conditions relating
animals to the supply and loss of water.

On account of the herbivorous food habits of so many animals a
large number secure much water with the juicy vegetation eaten, and
others from nectar or from the sap drawn or escaping from plants.
The predaceous animals secure a large amount of water from the fluids
of the animals they devour or the juices sucked from their bodies, as
in the case of certain Hemiptera and some parasites. In addition to
the fluids derived from plants and animals, many animals also drink
water, some in small amounts and others in large quantities. Innu-
merable observations have been made by naturalists on the drinking
habits of animals, but I know of no general discussion of this subject,
and particularly of none from the standpoint of the variation of their
behavior in this resoect in different environments. But the sources
of water mentioned are not the only ones available to animals, although

99

they are the most obvious, and familiar to us. An important addi-
tional source is that formed within the body of the animal by the proc-
esses of respiration and dehydration; this is metabolic water. The
relation of this source to others and to water-loss has recently been
summarized in an important paper by Babcock ('12:87, 88, 89-90,
91, 160, 161, 171-172, 174-175, 175-176, 181). The following quo-
tations from this paper will serve to give a concise statement of the
general principles involved in this important process. He says (pp.
87-88) : "There are, however, particular stages in the life history of
both plants and animals in which metabolic water is sufficient for all

purposes for considerable periods of time This is also true

in the case of hibernating animals that receive no water from external
sources for several months, although water is constantly lost through
respiration and the various excretions. In addition many varieties of
insects such as the clothes moths, the grain weevils, the dry wood bor-
ers, etc., are capable of subsisting, during all stages of development,
upon air-dried food materials containing less than ten per cent water ;
in these cases, nearly all of the water required is metabolic. . . . Many
organisms also, when deprived of free oxygen, are capable of main-
taining for a short time, certain of the respiratory functions, and de-
riving energy from food material and from tissues by breaking up the
molecular structure into new forms of a lower order. This is known
as intramolecular respiration, and like direct respiration, results in the
production of both water and carbon dioxide." (Pp. 89-90) : "The
substances oxidized by both plants and animals, to supply vital energy,
consist of carbohydrates, fats, and proteins. All of these substances
contain hydrogen, and their complete oxidation produces a quantity
of water equal to nine times the weight of hydrogen present in the orig-
inal substances. . . . Most of the the fats yield more than their weight
of water, while proteins, when completely oxidized, give from 60 to

65 per cent of water Animals, however, are unable to utilize

the final products of protein metabolism which are in most cases
poisonous and must be removed from the tissues by excretion in vari-
ous forms, the principal of which are urea, uric acid, and am-
monia The amount of metabolic water formed by oxidation

during any period is proportional to the rate of respiration

(Page 91 ) : "With parasitic plants, and with animals, which derive all
of their organic nutrients from chlorophyl producing plants, im-
bibed water is not so essential to life; with these the chief function
of imbibed water is to aid in the removal of waste products, the
metabolic water being in most cases sufficient for transferring nutri-
ents and for replacing the ordinary losses incurred by respiration
and evanoration." (Page 160): "Another and more im-

100

portant difference is the inability of animals to resynthesize the or-
ganic waste products of respiration into substances that may be

again utilized as nutrients This is especially the case with the

soluble products arising from protein metabolism. With most animals
these nitrogenous products are excreted in solution through the kid-
neys, chiefly as urea, but birds, reptiles, and all insects excrete most
of the nitrogenous waste matter as uric acid, or its ammonia salt, which
being practicall}^ insoluble in the body fluids, is voided in a solid con-
dition." (Page 6i) : "The need for water is much less for ani-
mals that excrete uric acid than for those that excrete urea, since
uric acid, being practically insoluble in the the body fluids, is not so
poisonous as urea and is voided solid with a minmum loss of water.
Many animals that excrete uric acid instead of urea never have access
to water and subsist in every stage of their development upon air dried
food which usually contains less than lo per cent water. The most
striking illustrations of this kind are found among insects such as the
clothes moths, the grain weevils, the dry wood borers, the bee moths,
etc. The larvse of these insects contain a high per cent of water, and
the mature forms, in spite of the development of wnngs which are rela-
tively dry, rarely contain less than 50 per cent of water." (Pp. 171-
172) : "Serpents and other reptiles that live in arid regions and rarely
if ever have access to water, except that contained in their food, are
said by Vauquelin to excrete all of the waste nitrogen as salts of uric
acid. The same is true of birds that live on desert islands where only
salt water is available. It is essential that animals of these types should
produce as much metabolic water as possible from the assimilated food,
and the waste of water through the excretions should be reduced to a
minimum. Since the food is largely protein both of these ends are at-
tained by the excretion of uric acid which, as already stated, contains
the least hydrogen of any nitrogenous substance excreted by animals so
that the maximum amount of metabolic water has been derived from
the food consumed." (Pp. 174-175) : "There are many animals that
are able to go long periods without having access to water except that
contained in their food, in which water usually amounts to less than
20 per cent of total weight, and the metabolic water derived from oxi-
dation of organic nutrients. A notable example of this is the prairie
dog which thrives in semi-arid regions. These small animals feed
upon the native herbage which for months at a time is as dry as hay.
It has been surmised that the burrows in which they live extend to
underground water courses, but this does not seem likely since in many
of .these regions wells must be sunk hundreds of feet before water is
reached. It is more probable that they depend chiefly upon metabolic
water. They feed mostly at night when the temperature is low and

101

during the hottest hours of day remain in their burrows where the air
is more nearly saturated with moisture and evaporation is relatively
small." (Pp. 175-176) : "An application of these principles would
undoubtedly serve to prolong life, when suitable water for drinking
is not available. In such cases the food should consist of carbohy-
drates and fats. Proteins should not be used The water re-
quired for preventing uremic poisoning under these conditions is small
and if the relative humidity of the surrounding air is high enough to
prevent rapid evaporation of water from the body, the metabolic water
arising from the oxidation of nutrients may be ample for the purpose."
(Page 181) : "Metabolic water derived from the oxidation of organic
nutrients would probably be sufficient for all animal needs were it not
for the elimination of poisonous substances resulting from protein de-
generation."

The preceding cjuotation brings out very clearly the harmful effects
of an accumulation of uric acid upon the animal. This is only a special
case illustrating a general law, for except water the main end products
of metabolism are acid. There is thus a constant tendency for acid to
accumulate, as Henderson ('13a: 158-159; see also '13b) has said:
"This tendency toward acidity of reaction and the accumulation of acid
in the body is one of the inevitable characteristics of metabolism; the
constant resistance of the organism one of the fundamental regulatory
processes. Now it comes about through the carbonate equilbrium that
the stronger acids, as soon as they are formed, and wherever they are
formed, normally find an ample supply of bicarbonates at their dis-
posal, and accordingly react as follows .... The free carbonic acid
then passes out through the lungs, and the salt is excreted in the urine."

Recently Shelford ('13b, see also '14a) has summarized the phys-
iological effects of water-loss by evaporation and other methods. It
is probable that the carbonic acid excretion is retarded by drying, and
that by this means irritability may be increased.

It is not simply loss of water, but loss beyond certain limits that
interferes with the life of animals. Thus loss is not an unmixed evil,
because, in addition to removing excretions, evaporation is an impor-
tant factor in the control of temperature within the bodies of animals.
Loss of water also tends to concentrate the body fluids, and when this
loss brings about a relatively dry condition, such tissues are in a con-
dition which is favorable for the endurance of relatively extreme low
or high temperature (Davenport, '97:256-258), and even dryness
(see references, Adams, '13 : 98-99). This is a reason why it is dif-
ficult to distinguish, in nature, between the effects of aridity and tem-
perature extremes, and hence arise the puzzling interpretations of con-

102

tinental climates. These extreme conditions are characteristic of many
habitats.

It is readily seen how the general principles just summarized apply
to the land animals of the prairie. Many of these are active during the
day, live in the bare exposed places, or near the level of the vegetation,
where evaporation is greatest and water-loss is correspondingly large,
and feed upon the dry haylike vegetation. Others remain among the
humid layers of the vegetation or in the moist soil, and feed upon
juicy plants and other moist food. Predaceous and parasitic animals,
deriving their moisture from their prey, occupy both the dry and humid
situations. These are representative cases, between which there are a
large number of intergradations.

In the forest, where evaporation is more retarded than in the open,
a large number of animals live in the forest crown, at the forest mar-
gin, in glades, and in wood, of all degrees of dryness, and eat food
varying similarly from juicy leaves to dry wood. On the other hand,
some live in moist logs, among damp humus, or in the soil, and feed
upon dripping fungi or soggy wood. Many of these animals possess
little resistance to drying.

The optimum for prairie and forest animals thus involves a
dynamic balance between the intake of water and its loss by evapora-
tion and excretion.

ANIMAL ASSOCIATIONS OF THE PRAIRIE
AND THE FOREST

I. Introduction

In an earlier chapter of this paper the habitats and animals found
at the different stations were discussed, and in the preceding section
the general characteristics of the physical and vegetational environ-
ment of the prairie and forest have been described ynd summarized.
We are now in a better position to consider the relations of the inverte-
brates, not only to their physical environment, but also to the vege-
tation, and, furthermore the relations which these animals bear to one
another. We wish also to consider both the prairie and the forest as
separate units, and to see how the animals are related to their physical
and biological environment. As previously stated, the special locali-
ties studied were described by stations both to give a precise and con-
crete idea of the prairie and its animals, as now existing in a limited
area, and also to preserve as much of the local color as the data would
permit. I wish now to reexamine these animals from another stand-
point, that of the animal association as a unit. The prairie as a whole

103

is not homogeneous from this point of view ; it is a mosaic composed
of a number of minor social communities. Each of these smaller units,
however, is fairly homogeneous throughout.

Our present knowledge of these minor associations is imperfect,
and for this reason they are arranged in an order approximating that
which we might reasonably expect to be produced if the initial stage
were made to begin with a poorly or imperfectly drained area and to
advance progressively with corresponding vegetational changes, toward
a more perfect condition of drainage. Upon the prairie a perfect series
would include every stage from lakes, ponds, and swamps to well-
drained dry prairie. But cultivation and drainage have obliterated so
much, that now only very imperfect remnants exist in the vicinity of
Charleston. Although the sequence followed therefore does not in-
clude all stages of the process it is approximately genetic.

There are three essential features in every animal association, or
community; certain physical conditions; certain kinds of vegetation,
which also modify the physical conditions; and representative kinds of
animals. Occasionally an effort is made to divorce these, to separate
organisms from their normal habitat, but such an effort is deceptive,
for no organism can live for any considerable period without a normal
environment.

I have not attempted to treat these associations with equal fullness.
In the sections devoted to the description of the stations it was possi-
ble in some cases, on account of the uniform character of a station, to
describe the animal association rather fully. In such instances the
detailed account is not repeated. In other cases I have elaborated the
community relations more fully here than elsewhere. The descriptions
of the stations and the associations, and the annotated lists, are in-
tended to be mutually supplementary.

II. The Prairie Associations

I. Szvainp Prairie Association

The swamp prairie community lives in a habitat characterized by
shallow water, which stands approximately throughout the growing
season of the vegetation. The soil is black, and rich in vegetable de-
bris. The characteristic plants are bulrush (Scirpus), flags (Iris),
swamp milkweed (Asclepias incarnata), beggar-ticks (Bid ens), and
young growths both of willow (Salix) and cottonwood (Popnliis del-
toides). The abundant growth of vegetation and the wet soil are con-
ditions favorable for the production and accumulation of organic de-
bris, which tends to fill the depressions and to supplement the inwash

104

from the surrounding slopes. At the same time, burrowing animals,
particularly the crawfish, also bury debris and work over the soil. In
the Charleston area this community was developed at Station I, d, and
in part at I, g.

The representative animals of this community are those living in
the water, such as the prairie crawfish, Canibariis gracilis (PL
XXXVI), the snail Galha iirnhilicata, and such insects as the nine-
spot dragon-fly, Lihellula piilchella (PI. XXXVIII, fig. 2), and the
giant moscjuito, PsoropJiora ciliata, whose immature stages are spent
in the water. In addition to these are other representative species
whose presence is, to an important degree, conditioned by the pres-
ence of certain kinds of vegetation — such species, for example, as
those which feed upon the dogbane (Apocymim) , the brilliantly col-
ored beetle Chrysoclms aiiratiis; upon milkweed, the milkweed bugs
Lygccits kalrnii and Oncopeltus fasciatus (PI. XL, figs, i and 3), and
the milkweed beetle Tetraopes; and, finally, the rather varied series of
flower visitors feeding upon pollen or nectar, such as the soldier-beetle
(CJiauliognatlius pcnnsylvamcus) , Eiiplioria scpulchralis, and several
species of butterflies, moths, bees and wasps, including the honey-bee,
bumblebees, and carpenter-bee (Xylocopa virginica), and the common
rusty digger-wasp (Chlorion ichneumoneiim) . Visiting the same flow-
ers, but of predaceous habit, were found the ambush spider (Mismncna
aleatoria) and the ambush bug (Phyniafa fasciata). Small insects
were preyed upon by the dragon-flies (Libellnla piilchella ), and the
dragon-flies in turn were entangled in the webs of the garden spider
(Argiope aurantia).

No animals were taken on the flags, but Needham ('00) has made
an important study of the population inhabitating flags at Lake Forest,
Illinois, and shows that it is an extensive one. He gives an excellent
example showing how the injury by one insect paves the way for a
train or succession of others. For example: the ortalid fly Chcrtopsis
ccnea Wied. (PI. XVIII, fig. i), bores into the stem of the buds and
causes them to decay (Cf. Forbes, '05, p. 164; Walton, Ent. News,
Vol. 19, p. 298. 1908). This condition affords a favorable habitat for
a pomace-fly (Drosophila phalerata Meig.*), an oscinid (Oscinis
coxendix Fitch, Plate XVIII, figures 3 and 4), a beetle, parasitic
Hymenoptera, and, after the decaying buds were overgrown by fungus
threads, the bibionid fly Scatopse pulicaria Loew. This paper by Need-
ham is one of the very few in which the population of a plant has been
studied as a biotic community. Forbes ('90, pp. 68-69; 02, p. 444)
has shown that snout-beetles {Sphenophorus ochreus Lee, Plate

*Mr. J. E. Malloch informs me that Z>. phalerata is not an American species.

105

XVIII, figures 5, 6, and 7) breed in root-bulbs of Scirpiis, and that
these beetles eat the leaves of Phragmites. Webster ('90, pp 52-55)
observed these beetles feeding on the leaves of Scirpus and the larvse
feeding on its roots. I have found great numbers of these beetles cast
up on the beach of Lake Michigan. Evidently they breed in the
swamps about the lake, fall into it when on the wing, and are washed
ashore.

2. The Cottonwood Community

Ordinarily we are accustomed to think of the prairie as treeless,
and yet one large tree was relatively abundant upon the original prairie
of Illinois, particularly upon wet prairie, or, when pools were present,
even upon the uplands. This was the cottonwood, Popiilus deltoides.
These trees were often important landmarks when isolated ; and today
the large trees or their stumps are important guides in determining the
former extent of the prairie. In the region studied there were no large
mature cottonwoods, although saplings were present, but north of
Charleston in the adjacent fields mature trees were found. They grow
normally at the margins of wet places, as about prairie ponds and
swamps, or along the small ill-defined moist sags and small prairie
brooks. This tree is usually solitary or in irregular scattered rows
when along streams, and does not, as a rule, form clumps or groves.
This relatively isolated habit may be a factor in the comparatively
small number of invertebrates which are associated with it, or at least
in the amount of serious injury which they do to these trees upon the
prairie. Many of the larger trees are mutilated, or even destroyed by
lightning (Cf. Plummer, '12), and such injury favors entrance of in-
sects on account of the rupturing of the thick bark.

The galls on the leaves and twigs of the trees often attract atten-
tion. A large irregular gall on the ends of the twigs becomes conspic-
uous in winter. This is formed by the vagabond gall-louse, Pemphigus
ocstlitndi Ckll. (PI. XIX, fig. i) {vagabundiis Walsh, Ent. News,
Vol. 17, p. 34. 1906). I have found these galls abundant upon the
prairie at Bloomington, 111. At this same locality I found a large
bullet-like gall at the junction of the petiole and the leaf — that of Pem-
phigus popidicanlis Fitch (PI. XIX, fig. 2), and at Urbana, 111., on
other large prairie cottonwoods, a somewhat similar gall, on the side
of the petioles, caused by P. populi-transversiis Riley (PI. XIX, fig. 3).
I have also taken large caterpillars of the genus Apatela on leaves of
cottonwood, and September 3, at Urbana, upon its cultivated form, the
Carolina poplar, A. poptdi Riley (PI. XX, fig. 6). These caterpillars
have bodies covered by yellow hair penciled with black. At dusk
swarms of May-beetles (Lachnostcrna) can be seen and heard feeding

106

among the leaves of the cottonwood and the CaroHna poplar. It is
noteworthy that I have made these observations at Urbana, Illinois,
upon cottonwoods growing upon what was originally prairie.

Forbes ('07a) has shown, as the result of extensive collections of
]May-beetles from trees, that they have a decided preference for Caro-
lina poplar (p. 456) and willow. This same paper also contains im-
portant observations on the nocturnal flights to and from the forest,
from the normal habitat of the grubs, and from the daytime abode of
the beetles in the open fields. Wolcott ('14) has recently emphasized
the point that the grubs live only in open places in proximity to wood-
land where the beetles can secure food. These observations show very
clearly that May-beetles are animals primarily of the prairie or forest
margin, and probably lived upon the original prairie, scattered, where
cottonwoods or willows grew. A glance at the map of the prairie and
forest (frontispiece) shows that the marginal area was very extensive,
and must have furnished an optimum habitat for these beetles. This is
a good illustration of the fact that the cottonwood exerted an influence
upon the prairie far beyond its shadow.

In some localities another beetle (Melasorna scripta Fabr. ) feeds
upon the leaves of the cottonwood, and may become a serious pest to
poplars and willows, but I have not seen this species abundant on iso-
lated mature trees upon the prairie. I have taken these beetles (July
2) under cottonwoods at Bloomington, 111. Packard ('90, pp. 426-
474) has published a list of the insects known to feed upon Popuhis.

Willows (Salix) are frequently associated with the cottonwoods
upon the prairie, but, in marked contrast with these, they generally
grow in colonies and are eaten by a great variety of insects. Packard
('90, pp. 557-600) lists 186 species of insects on them, and Chitten-
den ('04, p. 63) extends the number to 380 species. Of course in any
given locality the number of species found will be relatively small, and
the number is further limited by the environmental conditions —
whether the land is upland or low and flooded. The degree of prox-
imity of willows and cottonwood is likely to influence the relative
abundance of the insects feeding upon these trees, since a large number
of insects which feed upon willow also feed upon the cottonwood. Col-
onies of willow are thus likely to become sources of infestation for
the cottonwood; this relation, however, is a mutual one. Walsh ('64)
and Heindel ('05) have published very interesting studies of the com-
munity life of the insect galls on Illinois willows. Cockerell ('97, pp.
770-771) has listed the scale insects found upon willows and poplars.

107

J. Swamp-grass Association

The prairie swamp-grasses, slough grass (Spartina), and wild rye
(Blymiis) were growing in relatively pure stands or colonies in de-
pressions which were dry in the late summer. The prolonged wetness
of the habitat and the dominance of the few kinds of grasses are char-
acteristic features of the environment of this association. These con-
ditions were found at Station I, a and c, north of Charleston. As these
stations were rather homogeneous and have already been discussed
somewhat fully, only a summary will be given here.

On account of the grassy vegetation the abundance of Orthoptera
is not surprising. Representative species are Melanoplus dijferen-
tialis, M. femur-rubrum, Scudderia texensis, Orchelimuni vidgarc,
Xiphidiuni strictimi, CEcanfhiis nigricornis, and CB. qiiadripunctatus.
Other representative animals are Argiope aurantia and the swamp fly
Tetanocera plumosa. The list of species is probably very incomplete;
during the wet season there are undoubtedly a number of aquatics ;
furthermore, there are still other species which feed upon Spartina and
Blymus, particularly some Hemiptera, and stem-inhabiting Hymenop-
tera, and certain Diptera. Thus Webster ('03a, pp. 10-13, 26, 32, 38)
has recorded a number of chalcids of the genus Isosoma which live
in the stems of Blymus virginiciis and canadensis. In this same paper
he discusses their parasitic and predaceous enemies (pp. 22, 27, 33).
A fly also breeds in Blymus, the greater wheat stem-maggot, Mer-
oinysa amcricana Fitch (PI. XX, figs. 1-5), as recorded by Fletcher
(1. c, p. 48). This species is of economic importance, having spread
from grasses to the cultivated grains. It has been studied in Illinois
by Forbes ('84). He found a fly parasite of this species, and Webster
reports a mite preying on it. Webster (1. c, p. 53) reports another
fly, Oscinis carbonaria Loew, bred from Blymus by Fletcher.

In another paper Webster ('03b) has published a list of insects in-
habiting the stems of B. canadensis and virginicus. Osborn and Ball
('97b, pp. 619, 622; '97a) have discussed the life histories of certain
grass-feeding Jassidce which feed upon Blymus. Osborn ('92, p. 129)
records a plant-louse, Myzocallis, from Blymus canadensis in Iowa,
and a species of leaf-hopper has been recorded by Osborn and Ball
('97b, p. 615) from Spartina. On the same plant, Osborn and Sirrine
('94, p. 897) record a plant-louse on the roots. In a list of the plant-
lice of the world and their food plants Patch ('12) lists a few from
Spartina. This same list includes (pp. 191-206) many grasses and
the associated aphids, those on Blymus on page 196.

108

4- Low Prairie Association

The moist black soil prairie, a degree removed from the wet or
swamp condition, with ground water in the spring relatively near the
surface, is fairly well characterized by the rosin-weed (Silphiiim) , par-
ticularly S. terehintJiinacenm. Other plants likely to be associated with
vS'. terebintJiinaceiiin are SUpJiinm lacimatuni and S. integrifoUum,
Bryngium yuccifoliinn, Lcpachys pinnata, and, to a less degree, Lac-
tuca canadensis.

In the Charleston area this condition is represented by Station I, a,
north of the town, and Station III, a, and in part b, east of the town.
The proximity of ground water is shown at Station I, e, by the pres-
ence of crawfish burrows, probably those of Cambarus gracilis. At
Station III the proximity of water was also evident where v?. terebin-
thinaceum was most abundant in the railway ditches. Such perennial
plants are indicative of the physical conditions for a period of years,
and are thus a fairly reliable index of average conditions — much more
so than the annuals.

It is difficult to decide which kinds of animals are characteristic of
this kind of prairie. Provisionally I am inclined to consider the fol-
lowing as being so: Cambarus gracilis; Argiope aurantia; the grass-
hoppers Bncoptolopluis sordidus, Mclanoplus differcntialis, M. fcniur-
rubrum, Scuddcria texensis, and Xiphidiuin strictum; CEcantlius nigri-
cornis; Phyniata fasciata; and asilids. The presence of Lcpachys was
clearly an important factor in determining the presence of Melissodes
obliqita and Epcolus concolor. At Station III, b, east of Charleston,
Bpicanta pennsylvanica and Boinbus pennsylvaniciis, aiiriconnis, and
inipaficns were taken on the flowers of Silphiuni tcrebiniliinacenm.

Robertson ('94, pp. 463-464; '96b, pp 176-177) has published lists
of insect visitors to the flowers of Silphiuni and Lcpachys ('94, pp.
468-469), at Carlinville, 111. Recently Shelf ord ('13a, p. 298) has
published a long list of animals inhabiting Silphiuni prairie near Chi-
cago. Forbes ('90, p. 75) has reported the snout-beetle RhyncJiites
hirfits Fabr. as feeding upon Silphiuni intcgrifoliiivn.

In a colony of prairie vegetation at Seymour, 111., which included
much Silphiuni and Bryngium, the following insects v/ere taken Octo-
ber 7 from the ball-like flower clusters of Bryngium yuccifolium: the
bugs Lygarus kalmii, TJiyanta custator Fabr., Buschistus variolarius,
and Trichopcpla scmivittata Say (No. 539, C. C. A.), the last named
in large numbers, the nymphs in several sizes as well as the adults, a
fact which suggests that both may hibernate upon the prairie. Rob-
ertson ('89, pp. 455-456) has summarized his collections of insects
from Bryngiuni and on Biiphorbia corollata ('96a, pp. 74-75).

109

Upon remnants of prairie vegetation growing at Urbana, Illinois,
I have found several kinds of insects centered about a wild lettuce,
Lactuca canadensis. Upon the upper, tender parts of this plant, the
plant-louse Macrosiphum riidheckicc Fitch, thrives late in the fall, in
very large numbers. Some seasons nearly every plant is infested. The
lice become so abundant upon these tender parts that the entire stem
for a distance of a few inches is completely covered. They migrate
upward with the growth of the stem and keep on the fresh, tender
parts. Among the plant-lice, and running about on the stem of the
plant, attending ants abound; eggs, larvae, and adults of lace-wing flies
(Chrysopa) also abound; and several species of coccinellids, syrphid
larvae, and a variety of small parasitic Hymenoptera are present.

5. Upland Prairie Association

The w^ell-drained prairie, a degree removed from the permanently
moist prairie, is fairly well represented by the physical and biological
conditions in which Bnphorbia corollata, Apocynum mediuin, and
Lactuca canadensis, are the representative plants. The plant ecologist
would consider the conditions favorable to mesophytic plants. In the
Charleston region these conditions are approximated at Station II,
where drainage has doubtless changed the area from a somewhat
moist, to its present well-drained, condition.

Representative animals of this community are as follows : Argiopc
aitrantia, Misumena aleatoria, BncoptolopJius sordidus, Melanopliis
hivittatus, M. dijferentialis, Orchelininni vidgare, Xiphidimn strictum,
Biischistus variolarius, Phyniata fasciata, Chaidiognathiis pennsylvan-
icus, Bpicauta niarginata and B. pcnnsylranica, RhipipJwrus diniidia-
tns and R. linibatus, Anunalo, Bxoprosopa fasciata, Fromachns verte-
hratus, Bonihus pennsylvanicns, and Myzine sexcincta.

On dry prairie at Mayview, 111., September 26, I found the plant-
louse Aphis asclepiadis Fitch on the leaves and stems of the dogbane
(Apocynum) and the lice attended by the ant Formica fusca L. A
beetle, Languria mozardi Latr., whose larva is a stem-borer, inhabits
Lactuca canadensis. Its life history and habits have been discussed
by Folsom ('09, pp. 178-184).

6. The Solidago Community

A common community in the late summer and early fall is centered
about the goldenrod (Solidago). This plant was not abundant or in
blossom at any of the stations studied in detail, but it grew in small
widely scattered colonies or clumps. Observations were made in two

110

colonies, north of Charleston, both west of Station I, a, and I, g. The
collections made (Nos. 20, 26, 42, 43) are as follows:

Ambush Bug Phymata fasciata 20, 26

Stink-bug Buschistus variolariiis 26

Black Blister-beetle Epicauta pennsylvanica 26

Noctuid moth Spragiieia leo 20, 26

Conopid fly Physoccphala sagittaria 26

Empidid fly Enipis claiisa 43

Halictid bee Halictus fasciatus 26

Myzinid wasp Myzine sexcincta 20, 26

Ant . Formica fuse a suhsericea 20

It is important to know that these collections from Solidago were
made just as the flowers were beginning to blossom. Collections a few
weeks later would probably have given many more kinds. It should
be noted, too, that all these plants were far out upon the prairie and
far from woodlands — a factor which may influence to some extent
the kinds of visitors. As a rule the lists which have been published
state little or nothing at all as to the conditions in which the plants
were growing. If this factor is neglected, the presence of some vis-
itors remains puzzling. Thus on some goldenrods the locust beetle,
Cyllene rohinicc, is abundant; but this is conditioned in part by the
proximity of the yellow locust, which is absent on the Charleston
prairie.

Phymata was found copulating upon the flower, and with an em-
pidid fly, Bmpis claiisa (No. 43), in its grasp. Two kinds of galls
formed by insects were found on this plant : one formed by the fly
Cecidomyia solidaginis (No. 43), which forms a rosette of leaves;
and the other the spindle-like stem-gall, formed by a^ small caterpillar,
Gnorinwschcina gaUcrsoUdaginis (No. 7462 Hankinson). September
20 the moth Scepsis fidvieollis Hiibn. was found in goldenrod flowers
near Station I, a. Its larva feeds on grass. A large noctuid larva,
Ciienllia aster oides Guen., was found in a mass of flowers. As the day
was cloudy and cool. Scepsis was resting or sleeping on the flower
masses, as were also the black wasp CJdorion atratiitn Lep., and Pol-
istes — both the light form variatus Cress., and the darker one, paUipes
Lep. On October 23, 1893, I found the curculionid Centrinophus
helvimis Casey (det. H. F. Wickham) on goldenrod at Bloomington,
111.

Needham ('98, pp. 29-40) has given a good popular account of
the insects associated with goldenrod, and Riley ('93, pp. 85-87) has
published an extensive list and given a number of observations on their
food habits.

Ill

Pierce ('04, pp. 173-188) has published a long list of bees found
visiting Solidago in Nebraska. He also mentions the following beetles :
Cliauliognathus pennsylvaniciis, Nemognatha imniaculata and A^
sparsa, Zonitis hilincata, Bpicauta pennsylvanica, and Myodites soli-
daginis Pierce. Myodites is a rhipiphorid beetle which appears to lay-
its eggs upon Solidago. Here the larva develops, and from here, by
attaching itself to different flower visitors, it is carried to their nests.
The nesting sites are often populated by several kinds of insects, a
social community, and thus the larva is thought to be carried in close
proximity to the bee Bpinomia, upon which it is parasitic. This bee
does not visit Solidago, but frequents the stmflower (Hclianthus), and
thus is only infested at the nest (see also Canadian Entomologist, Vol.
XXIV, 1902, p. 394). This is a good example of the complex rela-
tions existing among the animals of the prairie. Robertson ('94, p.
455) found Myodites fasciatiis Say on Solidago at Carlinville, III, and
he also lists (1. c, pp. 454-458) many species of insects which he found
on different species of goldenrod. As Bpinomia is not known from
Illinois it is probable that some other bee is host for Myodites.

y. Dry Prairie Grass Association

The dry prairie grass association includes those animals which live
on the driest of the black soil prairie among the tall prairie grasses
Andropogon and Sporobolus. Upon the original prairie this was
probably a relatively stable habitat.

About Charleston these grassy habitats occupied only very small
areas north of the town, at Station I, g (in part), and Station HI, b
(in part).

Representative animals of this community are the following: Argi-
ope aiirantia, Brachynemiirus ahdoniinalis, Chrysopa oculata, Syrbula
admirabilis, Bncoptolophiis sordidiis, Melanoplns differentialis, M.
femur-riibrum, Scudderia texensis, Orchelimiim vidgare, Conocepha-
lus, CBcantJiiis nigricornis and CH. 4-punctatus, Bnschistus variolarius,
Sinea diadcina, Phymata fasciata, Chaidiognathus pennsylvanicus,
Tetraopes tctraophthalmns, Rhipiphorus dimidiatus, Bxoprosopa fas-
ciata, Promachus vertebratits, Bombus pennsylvanicus, auricomiis, im-
patiens, fraternns, and separatus, Melissodes bimacidata, and Myzine
sexcincta.

Probably a number of insects breed in the roots and stems of An-
dropogon and Sporobolus, but none were secured.

Although Blymus has contributed many insect pests to cultivated
grains, it seems that Andropogon has not, if we except the chinch-bug
(Blissus leiicopterus Say). This insect was not related to Andropo-

112

gon as Isosoma is to Blyiniis, but this and other prairie grasses which
grow in bunches or stools evidently formed the optimum hibernating
quarters of these pests when they lived upon the original prairie
(Fitch, '56, p. 283; Marlatt, '94a; Schwarz, '05) and upon the sea-
shore. Osborn and Ball ('97a and '97b) have listed several grass-
feeding Jassidcc from Andropogon and Sporobolus. Osborn and Sir-
rine ('94, p. 897) found a plant-louse on the roots of Andropogon,
and Patch ('12, p. 191) lists Schizoneura corni Fabr. on A. furcatus.

8. A Milkweed Community

Bordering the gravelly ballast along the rails north of Charleston at
Station I (PI. II, fig. 2) may be seen a large-leaved plant, the common
milkweed (Asclepias syriaca). This plant flourishes along the track
in many places, and wherever it was found there tended to appear a
small but very well-defined animal community. To determine the com-
position of this social community, a few collections were made at vari-
ous points within Station I. That this milkweed is the hub of this
microcosm is clearly shown by the fact that no similar association was
found grouped around any other plant in the area, not even about the
other milkweeds, A. sullivantii, or A. incarnata. The collections are
numbered as follows : Nos. 27-30, 33, 34, and 154.

The terminal young and tender leaves of the plant are often densely
covered with the plant-louse Aphis asclepiadis Fitch (Nos. 28, 29),
and these lice are attended by the workers of the ant Formica fitsca
suhscricea Say (Nos. 30, 154). On another plant no plant-lice are
recorded, but upon it were found their common enemy, the nine-
spotted ladybird, Coccinclla p-notata; two species of ants (Formica
pallid e-fiiha schaufttssi incerta, and Myrmica rubra scahrinodis sabu-
leti) ; besides, running about on the leaves, the pretty, metallic, long-
legged flies Psilopus siplio (No. 27). They run with a singular rapid
glide, stop suddenly for a moment, and then continue their rapid pace.
Certain flies of this family are said to be predaceous, but I have never
seen Psilopus capture any small animal. On the same plant just men-
tioned a small bug, Harmostes reflexulus, was also taken ; and in the
flowers of this plant were hundreds of a small dark-colored empidid
fly, Bmpis claiisa (No. 27). Two other animals were found on this
plant; Zonitis bilineata Say (No. 33), and a jumping spider (attid),
which had in its jaws what appeared to be the remains of the beetle
Diabrotica 12-punctata (No. 34). Contrary to my usual experience,
these plants did not abound with milkweed beetles (Tetraopes) or with
the common milkweed bugs (Lygcciis kahnii and Oncopeltus fascia-
tus), which are usually numerous. The proximity of the fragrant

113

blossom of Asclepias incarnata may explain this paucity at this time
and place. The milkweed butterfly, Anosia plexippiis, is of course a
member of this community.

W. Hamilton Gibson ('oo, pp. 227-237) has discussed, in a very
interesting manner, the relations of this plant to its insect pollinators,
and calls attention to the variety of insects which are entrapped and
killed by its flowers. He also points out that the dogbane (Apocy-
mim) has a similar habit.

Robertson, our leading American authority on the relations of
flowers and insects, has published extensive lists of the flower visitors,
not only of A. syriaca (cormiti) but of other Illinois milkweeds
(Bot. Gaz., Vol. XI, pp. 262-269; Vol. XII, pp. 207-216, 244-250;
and Trans. St. Louis Acad. Sci., Vol. V, No. 3, pp. 569-577).

III. Relation of Prairie Animals to their Environment

The relation of prairie animals to the major features of their phys-
ical and biotic environment presents several facts of unusual interest.
On account of the relatively heavy precipitation during June, the slight
topographic relief of the region, and its imperfect drainage, unusually
large areas of the original black soil prairie are wet or swampy. Cer-
tain animals are able to tide over this early, unfavorable wet-summer
period because they are not fully roused from their winter inactivity ;
others, in their immature stages of development, require less food than
later; still others survive by migration to the drier uplands. At the
same time, other animals, preferring moist or wet habitats, flourish,
and then decline in numbers as the season advances. Toward August,
on account of the eastward migration of the continental peninsula of
aridity and intense evaporation, those animals whose activity is re-
tarded by the earlier wet season find the conditions progressively more
favorable, and thrive and grow accordingly. This is the acme of the
season for dry-prairie animals, and great numbers of slowly maturing
composite plants now make the landscape yellow with their flowers.
The Orthoptcra are now mature, and when flushed, or, when not
flushed, by their sounds, are noticeable. That these conditions cause
these animals to thrive, is only too evident during exceptionally dry
seasons, when the ordinary August drouth begins in July and extends
into September.

In the conditions just indicated, the imperfect drainage, the wet
season followed by the dry, we are touching closely upon the real causes
of the prairie. Yet to me it seems fruitless to search for the cause of
the Illinois prairie ; the causes are probably multiple. In the midst of
the Great Plains, the "short grass country" the causes of grass-land

114

may be relatively few, because the dominating conditions are so thor-
oughly established and extreme. But near the eastern margin of this
dominance, upon the prairies — the "long grass country" — the number
of limiting factors increases greatly, and even a relatively trivial local
influence is able to overcome the slight momentum which this domi-
nance possesses. In Illinois, then, the causes of the prairie biota, men-
tioning only the larger groups of influences, seem to be as follows : a,
a sandy character of the soil, resulting in sand prairie ; b, loam and
good drainage, resulting in black soil prairie ; c, very imperfect drain-
age, resulting in wet prairie. A shallow soil underlaid by rock might
also produce prairie, but I have not seen any large area of this kind in
Illinois.

We have, then, in the wetness and the dryness of the prairie two
of the important controlling influences upon the prairie associations.
On the prairie aquatic animals may thrive, particularly those which
develop early and mature rapidly, and possess some power to resist
or tide over the dry season, either as adults of non-aquatic habits by
estivation, or in some resistant immature stage. We can see how
aquatic animals, in this manner, are capable of enduring these extreme
conditions and remain numerous upon the prairie. Where crawfish
holes are abundant, many small aquatic animals are able to utilize them
and thus escape drying. Crawfish holes should be examined during
dry seasons with this idea in mind. On the other hand, the prairie
is inhabited by many animals which can not endure much moisture, and
live best in conditions of moderate or extreme dryness. These are the
kinds which find their optimum during the driest part of the season,
and in very dry years. When there is an abundance of moisture, some
of these, for example the chinch-bug, are particularly susceptible to
disease. The maximum development of this arid type as seen on the
Illinois sand prairie has been studied by Hart ('07) ; more recently by
one of my students, Vestal ('13b, '14) ; and about Chicago and north-
ern Indiana by Shelf ord ('13a). An examination of the lists of sand
invertebrates given by Hart (1. c, pp. 230-257) and Vestal ('13b,
pp. 14-60), in comparison with those for the black soil prairie at
Charleston, will show many differences, not only in kinds but also in
their relative abundance. Some allowance must also be made for the
fact that the animals of the black soil prairie are not as fully pre-
served as those of the sand areas.

I. The Black Soil Prairie Community

The soil population of both sand and black soil prairie has never
received thorough study, although observations from the sand areas

115

have been recorded by Hart, and his observations ampHfied by Vestal.
In the black soil area many observations have been made by Forbes
('94) on the life histories and habits of certain species of economic
importance, particularly those injuring corn and grasses in the soil. In
his studies are included many insects, such as elaterid larvae, aphids,
ants, and white-grubs. The physical conditions of life here yet await
careful investigation.

A very large number of the animals living on and above the sur-
face of the soil spend a part of their lives within it. Thus among the
Orthoptera, the acridiids lay their eggs in the soil — this is probablv
true of most of the beetles ; and even the parasitic animals often spend
most of their life in the soil with their hosts. This is true also of the
wasps and a great number of hibernating animals, and of a large num-
ber of grass-inhabiting, and other, Lepidoptera. Such characteristic
flies as the asilids and bombyliids spend much of their life in the soil,
as do many other flies, at least during their pupal period. It is very
probable that upon the original prairie a large number of noctuid and
crambid moths and tipulid and elaterid larvae inhabitated the prairie
sod, and with them, of course, were associated their enemies — preda-
ceous beetles, and parasitic flies and Hynicnoptcra. For an account of
grass-feeding crambids Felt ('94) and Fernald ('96) should be con-
sulted.

The stage of development, structure, and behavior of soil-inhabit-
ing animals are often quite different from those living above the sur-
face. Some kinds, as pupae or adults, have spines or setae, which enable
them to wriggle in the soil, as, for example, do the pupal asilids or the
adults of Myzine and Tipkia. Locomotion in such a dense medium
is attended by many difficulties, and it is not surprising that animals
living here have peculiarities of structure and behavior, and that a
large number are relatively sedentary.

In the discussion of the ventilation of habitats, attention was called
to the fact that soil-inhabiting animals probably possessed considera-
ble resistance to an abundance of COo and to a lack of oxygen. We
are all familiar with the abundance of earthworms, Liimhriciis and its
allies, crawling upon the surface and entrapped upon our walks and
pavements after prolonged rains. In these cases the saturation of the
soil has driven out the air. Apparently the earthworms are relatively
less resistant to the lack of oxygen than many other soil animals, for
they come to the surface in a much more marked degree. Since earth-
worms live in burrows, have an easy route to the surface, and are pos-
sessed of good powers of locomotion, they contrast strikingly with
many other sedentary soil animals. Bunge ('88, p. 566) found that
earthworms were able to survive one dav in an oxygen-free liquid.

116

Cameron ('13, p. 190) speaks of the resistance to drowning of elaterid
larvae as follows: "I myself have kept specimens of the larvae of
Agriotes lincatus, our commonest wireworm, in water for as long as
six days without their being drowned, but those which were thus
treated for a period of seven or eight days did not generally recover
from the deleterious effects of immersion. Leather- jackets and sur-
face caterpillars submitted to the same treatment succumbed in a much
shorter time, one to two days for the caterpillars, depending on their
state of development — much shorter time than this for very young
forms — and from one to three days in the case of leather- jackets, the
latter being in all cases fully mature."

Dr. R. D. Glasgow informs me that it is probable that the soil-
inhabiting white-grubs, Lachnostcrna, may be able to close their spira-
cles when the soil is saturated and thus resist drowning, as in the case
of the European Mclolontha (Cf. Henneguy, '04, p. 105 ; Packard, '98,
p. 442). With this closure of the spiracles there is probably corre-
lated a powerto resist a lack of oxygen and an excess of COo. In any
case, this is a subject worthy of experimental investigation. Cam-
eron ('13, pp. 197-199) has called attention to the marked resistance
to a lack of oxygen found in muscid (dipterous) larvae; they endure
submersion for long periods and recover rapidly. He says (1. c, p.
198) : "A faculty of resistance and power of adaptability to adverse
circumstances is of peculiar advantage to the insect inhabitants of the
soil, which, owing to the varying climates and atmospheric conditions,
are often subjected to the most severe extremes of heat and cold, w^et
and drouth. The more sluggish maggots of Diptcra have a greater
plasticity than the active larvae of predaceous Coleoptera. On consid-
ering these two orders by themselves, amongst Diptera the larvae of
Muscida have a greater power of resistance generally than the larvae
of Nematocerous and Brachypterous families, whilst among Coleop-
tera the grubs of RhyuclwpJwra are not so easily affected as those of
Carabidcc and StaphyUuidce and other active families. This is just
what we might expect, seeing that nature, which has deprived Dipter-
ous maggots and Weevil grubs of legs that they might readily escape
danger, has compensated them to some extent by endowing them with
a greater power of resistance to adverse conditions."

Upon the black soil prairie the snout-beetles Sphenophorus
abounded in the roots of swamp plants, where they were particularly
liable to submersion with varying rainfall. It is, however, possible that
this resistance may be entirely independent of the footless condition.

The optimum soil conditions for insects have thus been summa-
rized by Cameron ('13, p. 198) as follows: "Soils that are of a light
and open texture are, as we have already seen, the ones most fre-
quented by soil insects, all other conditions, such as those of food, being

117

equal A porous subsoil is also conducive to the well-being of

insect life, in that the rain can quickly penetrate, and, as it passes
through, air is drawn into the more superficial layers in order to take
its place. Hence a reason why soil insects are only rarely found in the
deeper subsoil; for the increased amount of moisture, together with the
decrease in aeration, is decidedly detrimental to their activities."

The density, moisture, solutions, and ventilation of the soil, its
fresh and decaying vegetation, make conditions possible both for a
population consisting of vegetable feeders and, preying largely upon
them, a series of predaceous and parasitic associates.

It is desirable that the prairie ground fauna should be made the ob-
ject of special investigation, particularly from the standpoint of soil
solutions, moisture content, ventilation, humus content, and the in-
fluence of the living vegetation. For this reason several papers are
here mentioned which will be valuable in such a study. Diem ('03)
has made an elaborate quantitative study of the ground fauna of the
Alps. He studied a variety of conditions, including pasture, meadows,
and coniferous forest soils. He describes his methods of study and
gives many references to the literature. Other papers which should be
studied in this connection are by Dendy ('95), Cameron ('13), Motter
('98), and particularly those by Holdhaus ('10, '11 a, 'lib). Banta's
('07) paper on cave animals will also prove valuable because of the
close relation of cave animals to those living in the smaller openings in
ordinary soil.

Near the soil surface, among the stools of grass and on the ground,
vegetable litter is most abundant, and humidity is high, evaporation
slow, and the temperature lower and also more equable than higher
up. It is in this layer that a vast number of animals hibernate, and
in it also many, active at night, are hidden during the day. In tliis
layer live the animals which feed largely on organic debris. Buml)le-
bees often build their nests at this level, or in depressions in the ground.
Some of our species of Bomhiis may nest deep in the soil and ventilate
the nest by vibrating their wings, as do certain European species ( Sla-
den, '12, pp. 47-49). This is a very interesting response to a subter-
ranean life and merits investigation.

2. The Prairie Vegetation Community

Above the surface of the soil, among the vegetation, quite another
environment exists. This varies greatly not only with the character of
the substratum but also with the character and density of the prairie
vegetation. The fertility of the black soil, and the rapidity with which
it is occupied by vegetation, makes areas of bare soil of short duration.

118

The prevailing condition is therefore one of dense vegetation. I know
of no detailed study of the amount of life which develops in this layer
of prairie vegetation. For this reason certain observations made in
meadows and pastures are of interest. McAtee ('07) examined a
grassy meadow and the surface of the soil for bird food, and a corre-
sponding area of four square feet of a forest floor. He concluded that
the population in a meadow is much more dense than that in a forest.
This conclusion, however, is not valid, as Banks ('07) has pointed out,
because the two areas are not strictly comparable ecologically. In the
meadow life is concentrated near the surface; in the forest it is
largely in the trees and not on the forest floor. Clearly the ecologically
comparable areas of the open and the forest are their subsurface soils,
the surface soil and the layer of vegetation, and the space above the
vegetational layer. As previously pointed out in this paper, the forest
should be looked upon as a very thick layer of vegetation. Another
estimate of the population of pasture vegetation has been made by
Osborn ('90, pp. 20-23). This is a rough estimate, but it shows that
there were about one million Jassidcu present per acre. He further
estimated that that the amount of vegetation per acre eaten by insects
amounted to about one half of that eaten by a cow. This example aids
one in understanding how it was possible for the insects of the origi-
nal prairie to influence the amount of food available for the buffalo,
particularly during dry seasons when there was limited grass growth,
and when grasshoppers throve in large numbers. In this layer of vege-
tation, in addition to the general feeders, eating almost any kind of
vegetation, there is a rather extensive population which has a restricted
diet, feeding upon a single food plant, or on only a few species. There
are a number of cases where, though an insect has several food plants,
all, or nearly all, belong to the same plant association, and often have
much the same geographic range. A good example of this among
prairie animals is the case of the plant-louse MacrosipJunn riidheckicc
Fitch, which lives on a variety of prairie plants ; as Vcrnonia, Solidago,
Bidcns, Ambrosia, Cirsimn, SUphinni, and Lactiica (Cf. Hunter, '01,
p. 116). The beetle Chrysochus and the bugs Lygccus kalniii and On-
copeltiis fasciatits are often found on Asclepias and Apocynum; Aphis
asclcpiadis lives on Asclepias and on Euphorbia. Though pollen- and
nectar-feeding insects often forage over many kinds of plants, some
of them have clearly defined preferences, almost amounting to limita-
tion to a single food plant. Thus the bee Melissodes obliqiia seeks
pollen largely from Lepachys pinnata, and the Pennsylvania soldier-
beetle, though very abundant on flowers, is not numerous in corn
fields even when pollen is excessively abundant.

119

Many kinds of insects are recorded as "sleeping" among rank
growths of vegetation and on flowers. In such places on cloudy or cool
days, late in the evening or in the early morning, insects are found at
rest and in a sluggish or torpid condition. The cause of this behavior
is not known. They may be "sleeping," or they may only have been
trapped there by a lowering of the temperature, as at sundown, when
their activity slowed down and they came to a rest on the last flower
visited. In this connection it should be recalled that it is near the gen-
eral level of the surface of the vegetation that the most extreme tem-
peratures are found, — the most warmth in the sun and the greatest
coolness at night. This is the main zone also of flowers visited by in-
sects.

In this same layer of vegetation is found the usual grouping of
vegetable feeders, scavengers, predators, and parasites. As the nectar-
drinkers visit the flowers, certain predators spring upon them, just as
the large members of the cat family seize their prey at the margins of
streams and lakes when the herbivores come to drink. Other preda-
ceous insects such as the wasps, robber-flies and dragon-flies, live
active lives and seek their prey on the wing.

Above the general surface of the prairie vegetation no inverte-
brates live permanently, unless the parasites, external and internal,
of the swifts and swallows can be so considered. Winged forms fre-
quent this region during flights in which they find food and mates.
Spiders, by their cottony "balloons," utilize the winds and are thus
transported. All of these are transients, and not permanent inhabi-
tants of the open area.

->

Interrelations ivithin the Prairie Association

In concluding this discussion of the conditions of life on the prairie,
we may profitably consider some parts of the network of interrelations
which bind together the animals and the environment. As the kinds of
animals and the number of factors involved are so numerous, only a
few selected animals will be considered. In this choice I have not lim-
ited myself solely to the kinds taken at Charleston, but have utilized
common and well known prairie animals. As representatives of the
soil-inhabiting forms the white-grubs and May-beetles (Lachnos-
terna) and the corn-field ant (Lasiiis niger americanus) have been
chosen; as representatives of those which live above the surface and
mainly among the vegetation the differential grasshopper and Boni-
biis have been chosen ; and as representatives of the active predators
and parasites, Promachns, Chlorion, Tiphia, and the parasitic fungi
Bmpiisa and Cordyceps. Statement of the available supply of water

120

and oxygen, the temperature, etc., is omitted for simplicity, not
because these matters are unimportant. Some of the main features
of these interrelations are summarized in the following diagram. Fig-
ure 1 6. This shows that the white-grubs living in the soil and devour-
ing the roots of plants are preyed upon in turn by an aggressive fun-
gus (Cordyccps) and by a wasp (Tiphia) — an external parasite; and
that Tiphia is parasitized in turn by Bxoprosopa and by the larva of
the small beetle Rhipiphorus. The adult May-beetles feed upon the
leaves of trees, and although many show a decided preference for trees
living in the open, as the cottonwood and willows, others feed largely
upon forest trees. Thus the prairie animals exert a direct influence
upon the forest community as well as upon the prairie. The differ-
ential grasshopper feeds upon the vegetation, and jumps or flies into
the webs of Argiope, where it may be killed even if it should not be
eaten. The eggs which this grasshopper lays in the soil are devoured
by the larvae of Chaiiliognathus and Bpicauta, and the adults are killed
by the fungus Bmpnsa, or mutilated by the mite Tromhidium — an ex-
ternal parasite (PI. XXI, figs, i and 2). The rusty digger-wasp,
CJdorion ichiieuiiwucmn, feeds upon the nectar and pollen of flowers,
and provisions its burrows in the ground for its larva with grasshop-
pers (Orchelimum) ; this larva, again, is probably devoured by the
small parasitic fly Metopia. The larvae of the soldier-beetle Chaidiog-
natJuis are predaceous, and eat other larvae ; thus they influence many
species; the adults frequent flowers as pollen-feeders. Although
Bpicauta devours eggs of grasshoppers during its larval stage it feeds
upon vegetation in the adult stage. The larvae of Bomhiis live upon
nectar and pollen supplied them by the female or worker, and the adult
is also a nectar- and pollen-feeder, Bomhus thus being solely sustained
by vegetation. They are preyed upon by a host of predaceous enemies,
as Phyniata and Proniaclius; and parasites, including the flies Fron-
tina, Brachycoma, probably Conops, and the false bumblebee (Psithy-
rus); their nests, moreover, form a habitation for a great variety of
insects, mites, and other animals too numerous to be put in the dia-
gram. These bees, then, on account of their large size, their large col-
onies, and the large amount of concentrated food which they amass at
the nest, combine to make themselves attractive to a great number of
animals, and become the hub of a busy microcosm, an extensive com-
munity of mutually interrelated kinds.

The root-louse of grass, Schizoneura panicola Thos. (Forbes, '94,
pp. 85-93), through the attention of several kinds of ants, Lasins niger
americamts Emery, L. flavus De G., L. inter jectiis Mayr, and Formica
schauftissi Mayr, is cared for from the egg to the adult stage; these
ants keep the plant-lice on fresh roots from which they suck their food.

121

.•^

si;

c

t^

t-

5

"^

?

0)

o

•1^

^

■or

o

t-

-o

^-

/

^

/-r

(o

//

//

■S

,i/

■*-^ «

'r

^-^ /

<b '-

Ij^/

""-^

1

1

1

1

% I

:s 1

^

&l^

y^

^'/^

-C 1

!^ '

C- 1

1

X
^

col y

^ /

I

^ c

/

■♦- 1 —

-«jL

'^ 1

7~-

c •

/

"Or."

/

.O \

/

/

~^ \ i

/

^ y

s \ /

P^ rt

122

In return the ants secure honeydew and wax from the lice. A closely
related aphid, Schizoncura corni Fabr. lives from '"September until
June on the dogwood (Cornus), and from June until September on the
roots of certain grasses" (Forbes, 1. c, p. 89). This insect, upon the
original prairie, was probably an inhabitant of the forest margin, or
lived near moist places where dogwoods abounded. (This point should
be determined at some favorable locality. ) In such a complex, inter-
woven community as that of the prairie it is immaterial where one
takes up the thread of relations, for if followed carefully without
interruption it will lead one about, from one animal to another suc-
cessively, until the intimate life of every animal and plant in the com-
munity has been reached, and influenced to some extent. Thus the
animals living in the soil, at the surface, and among the vegetation are
bound together, not only by their changes of habitats, as when a sub-
terranean maggot matures and becomes a flower fly, but also by their
movements, as when an active wasp or grasshopper burrows in the
soil, so that there is a complex interpenetration of relations which ex-
tends to all depths, to all horizontal relations, and binds together the
entire social community.

In this discussion only the invertebrates have been considered, but
this phase of the subject should not be concluded without emphasiz-
ing the fact that a.11 the organisms of a region form a single biotic com-
munity, each member of which is related to all the others and to the
physical environment.

ly. The Forest Associations

/. Introduction

In a study of forest animals their relation to the physical and vege-
tational environment must be kept constantly in mind, in order that
their progressive changes may be clearly understood. If the woodland
animals and associations are considered broadly, it is possible to study
the progressive transformation of the habitats and associations by
agencies which erode the land and thus develop the drainage, and to
combine with this a study of the successive changes in the vegetation
(including vegetable products). In the Charleston region this trans-
formation includes the progressive invasion of the prairie by the for-
est. From this standpoint it is also possible to arrange the forest as-
sociations in a genetic series.

There is little doubt that this entire region was once treeless or
prairie, that in time the forest invaded it, mainly or almost exclusively
along the streams. Even at the time of settlement the forests had not
spread far from the larger streams ; but by normal forest extension and

123

drainage development the prairie was encroached upon and restricted.
The trees farthest from the streams, speaking in general terms, may be
looked upon as the pioneer guard of the extending forest. Such trees
are oaks and hickories of various kinds, which are hardy and able to
live on wet, acid, or very dry soils, as, for example, the shingle oak
(Qucrcns imhricaria) and post oak (Q. michaiLrii=minor) . In the
Charleston area all such forest remnants are so closely pastured that
they were not studied ; therefore our series is incomplete. The upland
forest in the Bates woods (Station IV, a) may be considered some-
what representative of a second stage in forest development. This,
however, is not a primeval condition, but one which has been modified
by man ; for example, the mature trees have been removed. It is, how-
ever, clearly an oak-hickory forest.

A third stage in forest development is found upon the bottom,
nearer the river, the most favorable habitat for tree growth in the re-
gion, where the red oak (Qiierciis rubra) and hard maple (Acer
sacchariim) form a dense, humid shady forest — a climax mesophytic
forest. With these changes in the vegetation there have been corre-
sponding changes in the physical environment. The relatively open
oak-hickory forests are dry, both in the air and in the ground ; they
are well lighted ; they are warmer and cooler relatively ; and they have
soil which contains less litter and humus. Fallen trees and stumps
decay more slowly on account of the dry environment. As the open
woods become closed by the development of a dense forest crown,
these conditions are changed in important ways : the woods become
progressively darker, more stable in temperature, more humid in air
and soil ; the litter and humus increase ; and all wood decays more
rapidly both on account of the moisture, fungi, etc., and the activity of
animals. The earlier stages in forest development result in the com-
bination of glade and grove — islands of open, and islands of trees — but
with the extension of the forest by its encroachment upon the glades
the forest crown becomes complete and continuous, and a climax for-
est has become established. These relations show what kind of factors
must be considered in striving to group forest habitats in a develop-
mental series.

The forest associations are here considered in the same sequence
as that given in the description of the forest stations, and for this
reason the discussion will be brief, being mainly intended to give a
uniform treatment to all the animal communities studied^ about
Charleston. A more general discussion of the ecological relations of
our common forest invertebrates follows.

124

2. Dry Upland (Querciis and Carya) Forest Association

The upland oak-hickory forest community is upon high well-
drained land. It is bordered by a ravine and a valley, so that the pre-
cipitation drains away rapidly. The soil, in contrast with that of the
black soil prairie, is a gray loam, containing little organic debris.
Through clearing, the woods have become relatively open, so that the
sunny spots are rather numerous. The characteristic vegetation con-
sists of oaks and hickories, such as white oak (Qiicrciis alba), black
oak (0. vchitina), shag-bark hickory (Carya ovata), pignut (C. gla-
bra) ; and of rose, raspberry, sassafras (Sassafras variifolinm) , sumac
(RJiiis glabra), young trees, horsemint (Monarda) everlasting (An-
tennaria) and tick-trefoil (Desmodiiim) . The conditions are those
of Station IV, a, the upland Bates woods, and the open ravine slopes,
lY,h.

Representative animals of this community, including numerous
ground-inhabiting Orthoptera — many of the acridiids being short-
winged forms — are Dicliromorpha viridis, Chloealtis conspersa,
SpJiaragcnion bolli, Mclanoplus atlanis, amplcctens, obovatipennis,
and sciiddcri, Scuddcria furcata, Microcentruni laurifoUiim, Orcheli-
muni cuticulare, Xipliidiitni neinorale, Nenwbius fasciatits and macn-
latus, Apitluis agitator, Cicindcla nnipunctata, Calosoiua scruta-
tor, Chrysochus auratus (on dogbane in an open area), Myrmeleoni-
dcr, and SphcurophtJialma. Several species of butterflies were seen on
the wing in the sunny openings. A number of cecidomyid and cyni-
pid galls on oaks and hickories are more characteristic of the upland
forest than of the lowland forest on account of paucity or absence
of white and black oaks and hickories upon the bottoms. Other
upland plants determine in a similar manner the presence of other
animals.

As a forest develops, upon what has previously been a treeless
tract, and as wood therefore becomes an available animal habitat, a
very complex factor is added to the environment. Not only is a log
food for certain animals, but also, if it lies upon the ground, it affords
conditions favorable for still others. It tends to conserve moisture
under it, and as it decays and disintegrates, fungi grow upon and in it;
hence other food is produced for animals which are not eaters of wood.
As decay progresses, furthermore, the log itself readily absorbs and re-
tains moisture, thus giving to some animals within it a habitat with
atmospheric conditions of relatively high humidity, in which land mol-
lusks, diplopods, etc., thrive. Such conditions furnish an important
factor in the extensive range of certain animals throughout several
kinds of forest ; for though the kinds of trees may change, nevertheless

125

when once the log habitat is developed certain animals are able to per-
sist. Nor is the log the only factor of this character in the forest;
the moist soil, abounding in vegetable debris, has a similar influence ;
and besides, when once a dense canopy is developed the retarded evap-
oration and the shade, with the accompanying reduction in heat rays,
have a marked influence. The presence of logs and vegetable
debris upon the forest floor determines to a very important degree
the presence of the land mollusks, diplopods, Vermes, Galerita janus,
and Meracantha contracta; it determines, upon the slopes (Station
IV, h,), the presence of Ischnoptera, MclanoHis, Passalns cornutiis,
and Scolecocampa liburna.; and it probably determines, too, many of
the ants on the upland and on the forest slopes. Among the forest
shrubby growth and tree trunks Bpeira verrucosa and Acrosoma
riigosa (and probably spinea) spread their webs and appear to thrive
only in deep shady woods. A large number of butterflies and moths
feed upon the foliage of forest trees, being thus distinctly arboreal, as
are also Cicada (nymph, subterranean), Diapheromera, Calosoma
scrutator (predaceous), Tremex cohmiba (and its parasite Thalessa
lunator), and CyrtopJiyllus perspicillafus. Gcotriipcs splendidns is a
ground scavenger. The presence of Ammopliila abhreviata is due to
the presence of numerous caterpillars on the foliage.

J. Artificial Glade Community in Lozvland Forest

In the dense humid lowland forest of the Bates woods (Station
IV, c) a small open area has been formed by cutting; an artificial
glade, as contrasted with a natural open forest. This may be consid-
ered an experimental glade. Although it is on the river bottom and
completely surrounded by a dense forest community, it is clearly not
related to that community, but rather to the open upland forest, and
for this reason is here interposed between the discussion of the upland
and lowland associations.

The glade was about 25 feet in diameter ; only on the north side,
where the sun had the best access, had brush (sassafras) made much
progress in closing the borders of this open area. It was therefore in
direct communication with the dense surrounding lowland forest.
Such a small glade permitted direct sunlight on the ground only dur-
ing the middle hours of the day, and it was during this time that ani-
mal life was most active. On account of the dense shade of the sur-
rounding forest there was little undergrowth, but in parts of the glade
there was a dense growth which covered the ground. It was com-
posed of grasses, large masses or colonies of Bupatorium caicstimim
in flower, Actinomeris altermfolia, with wood nettle (Laportea cana-

126

densis), and clearweed (Pilea purnila) surviving as relics of the low-
land forest vegetation.

Representative animals of this community are the following: Mi-
smnena aleatoria, Lycosa scutulata, Bpeira dorniciliorum, Aulacises
irrorata, Jalysits spinosus, Dichromorpha znridis, Melanoplus amplec-
tetts, gracilis, and sciidderi, Anihlycorypha rotimdifolia, Conoceph-
alus nehrascensis, OrcheUniuin ciiticulare and glaberrimum, Xiphidiuni
nemorale, Nenwbius fasciatus, Acanthocerus galeator, Autographa
precationis, Bpargyreus tityrns (larva on sassafras), Derotnyia dis-
color, Milesia ornata, and, apparently as wanderers from the forest,
Calopteron reticidatum, Thalessa lunator, and Pelecimis polytiirator.

4. Hinnid Loivland (Hard Maple and Red Oak)
Forest Association

This lowland forest community is upon a well-drained but moist
slope of the valley of the Embarras River. The soil is damp, and con-
tains a large. amount of vegetable debris. The forest canopy is com-
plete, and the forest is relatively dark. Representative trees are the
hard maple (Acer saccharum) , red oak (Qnerciis rubra), and the elm
(Uhnus americana) ; the herbaceous plants are nettle (Laportea cana-
densis) and the clearweed (Pilea pumila).

Representative animals are the various forest mollusks, Bpeira tri-
vittata, Acrosonia spinea and rugosa, Acarus serotincc, Bittacns stig-
inatcrus (and probably strigosus and apicalis), Asaphes vieninonius,
Calopteron tenninale, probably Thalessa lunator, Pelecimis polytura-
tor, and Tapinoina sessile and other ants. Boletotlierus hifurcus is
dependent upon the shelf-fungus Polyporus, which grows most abun-
dantly on decaying stumps and logs in moist woods. The species of
Bittacns are as representative of shady, moist woods as are the nettle
Laportea and the clearweed (Pilea). Such an insect as Bittacns might
live in the park-like groves of an open forest, but its optimum habitat
is in the dense climax forest. Perhaps the most striking contrast be-
tween the open and closed shady forest is due to the absence of nu-
merous Ortlwptera which are generally abundant in open grassy places.
That these forms are able to thrive on the bottoms when the proper
conditions are present is seen by their abundance in the glade in the
lowland forest. In the uplands also, Papilio and Polygonia frequent
the open spaces, but in the shady lowland forest the slow, low-flying
Bnodia and Cissia are the characteristic butterflies seen on wing.

127

[5- Animal Association of a Temporary Stream]

The prairie animal communities were arranged in an order to aid
in looking upon the prairie habitats as so many different degrees or
stages in the progress of drainage development, this being a dominant
physical environmental factor upon the prairie. Similarly, the forest
communities are easily arranged in a developmental sequence depend-
ent upon the combined influence of the progress of erosion and drain-
age and the advance of forest upon the prairie. Thus the prairie and
forest are given an orderly sequence, and the only remaining important
habitat, in the region examined, is that of the stream series.

Very little time was devoted to the study of the stream animals,
and mention of it is made here mainly because of this opportunity to
show the harmony and continuity of treatment which it is possible to
give to all the habitats and communities of a limited forested region.

This small temporary stream formed the southern boundary of the
area which was studied in the Bates woods. It formed Station IV, e,
and is an early stage in stream development. To understand just what
this means it is necessary to consider the processes which have been in
operation and which have reached the present stage of stream develop-
ment. This stream flows in a steep-sided ravine cut in the unconsoli-
dated glacial deposits which form the sides of the Embarras valley, a
ravine between 75 and 100 feet deep when it enters the valley, which
narrows rapidly, turns to the northwest, and soon ascends to the sur-
face of the upland oak-hickory forest. The upper parts and head end
of the ravine are dry, except during rains and soon -after ; but the lower
part may retain water in the basins for a number of days after rains.

The same conditions which we now find at the head of this ravine
once existed at the edge of the valley. That is, at one time there was
no ravine in this region. As the rainfall from the uplands flowed over
the edge of the valley it started a small gully; this, once formed, be-
came the trail for waters of other rains, each shower tending to cut
the ravine deeper and wider and to advance it into the upland. This
process has continued until now the head of the ravine has cut back
about one half of a mile. The unconsolidated debris is not composed
of homogeneous materials, and has therefore been washed away more
rapidly at some places than at others. In this manner pools have
formed where less resistant materials were, and between these pools,
over more resistant gravel or stone, miniature cascades or rapids have
been formed, the tendency thus being towards an alternation of pools
and cascades. In these pools Mr. T. L. Hankinson took a number of
vertebrates, and upon the surface of the pools were many water-stri-
ders, Gerris remigis. From the burrows along the margin of the

128

stream Mr. Hankinson secured Cambariis diogenes. Thus by the
growth of this ravine a new community is developing at this place —
that of a temporary stream.

In time such a stream will cut down to ground-water level, the
pools will become permanent, and a constant current will be main-
tained between the pools, and a permanent stream will become estab-
lished. The manner in which this ravine and stream grow, at the
expense of the upland forest, is an indication of how the upland for-
est may be changed and by degrees become converted into a lowland
forest and even into an aquatic habitat.

V. Relation of the Deciduous Forest Invertebrates to their

Environment

We have seen that the forest should be looked upon as a thick
layer of vegetation in its efifect upon the physical conditions which in-
fluence animal life. This thick layer is of relatively slow growth, and
in its early stage it is composed of shrubs and young trees. But "as
the vertical extent of the forest increases and the forest crown mi-
grates upward, the intervening trunk, bark and branch habitat ....
enlarges and the leaf-eating inhabitants of the forest crown rise up-
ward. This crown fauna retains or rather continues some of the char-
acteristics found at the marginal zone, with which it retains direct con-
tinuity" (Adams, '09, p. 162). In addition to this vertical upward mi-
gration of the forest crown, the forest also tends to spread laterally,
by arms or peninsulas of forest, which expand upon the open, or by
the excentric growth of groves, which in time fuse and form a contin-
uous forest. The original forest margin and adjacent prairie was
characterized by "groves", as they were commonly called by the early
settlers, and also by more or less open woods or "oak openings," which
are the homologs of the open oak forests yet found on the Illinois
sand areas. This interdigitation of forest and prairie produced penin-
sulas of forest extending into the prairie, peninsulas of prairie ex-
tending into the forest, islands of prairie surrounded by forest, and
islands of forest surrounded by prairie. Where the forest was advanc-
ing, the open places or glades are to be considered as prairie relics;
and when the prairie was for any reason encroaching on the forest the
forest is to be considered the relic. The glade and the grove are thus
comparable communities, and are to be considered as relics or pio-
neers according to the direction of advance of the local association.
The development of adequate drainage and all that is associated with
this process, the character of the soil, the extension or retreat of the
forest, the changes in composition of the forest, and the kinds of

129

animals composing the communities are the dominating influences in
the woodland environments. In the Charleston area the soils are loam,
and therefore sand need not be considered. The forests are of two
main types, the oak-hickory of the uplands and the red oak-maple
of the lowland. At present the forests are declining; in fact, the low-
land Bates forest has been converted into a corn field since these
studies were made.

The kinds of animals present in the woods are strikingly different
from those of the prairie, as is seen almost at a glance, and as is quite
clear by a comparison of the annotated lists of the prairie and forest
animals. Prolonged study will probably serve to enhance this differ-
ence. A small number are found both in the forest and upon the
prairie, but this is the marked exception. Furthermore, the open oak-
hickory woods, and the glade-like clearing which furnishes an open
habitat within the woods, contained a vast majority of the animals
found common to the prairie and the forest. These animals are to be
looked upon as pioneers (or relics) of the prairie, and are not to be
confused with the dense forest inhabitants. On a previous page atten-
tion was called to the vast importance of the marked discontinuity
which exists between the kinds of animals living in the open and in the
forest. This distinction is so marked as to merit comparison with the
contrast existing between land and fresh-water animals. Possibly
on land it ranks second only to this in its fundamental character. When
the same kind of animal lives both in the open and in the forest, it
often behaves differently in the two situations. It is significant that it
rec^uired more than a generation for the southern woodland human
pioneers of Illinois to change their behavior sufficiently for life on the
prairie. Undoubtedly there are many examples of just such changes
in behavior.

I. Forest Soil Community

The animals of our woodland soils have not been specially investi-
gated. Many observations on the life histories of soil invertebrates
have been recorded, but not as much is known of them as of prairie
soil animals because of the smaller numbers which attack cultivated
crops. Undoubtedly the native underground inhabitants of raspber-
ries, currants, blackberries, and other wild shrubs have continued to
thrive on the cultivated kinds (see Webster '93 for a paper on rasp-
berry and blackberry insects), and the same is true of the crab-apples
and the haws. Few subterranean animals, however, inhabiting these
shrubs and trees of the forest have been studied in detail, with the
notable exception of the periodical cicada. It is very probal^le that a
number of animals which lived in the prairie soil continue to do so in

130

the forest glades; and many ground-inhabiting Orthoptera in the for-
est oviposit in the soil as do their congeners on the prairie. On Isle
Royale, Michigan, I found that the carabid beetles which lived in the
openings were likely to extend into the coniferous forest in the humus
layer, which corresponds to this habitat in the open, and this is prob-
ably true to some degree in Illinois forests.

In the denser forest, in marked contrast with the prairie, there is
generally a large amount of litter on the forest floor. The prairie soils
are dark, but the surface contains a relatively smaller amount of or-
ganic materials comparable to forest litter. In the forest, however,
though the sub-surface soil is relatively light in color, the surface con-
tains much fresh and partially decayed organic debris.

McAtee ('07) has made a careful count of all the invertebrates
found upon an area of four square feet of the forest floor, at or near
the surface. This is the only quantitative study made of our forest
soil animals known to me. His observations were made during the
hibernating season.

Representative plant-feeding ground animals are the two cicadas
linnci and septendecim, which suck sap from the roots of trees. Their
underground enemies seem to be largely mites. The arboreal habit of
the adults subjects them to many enemies. The periodical cicada, as
the result of subterranean life, in the moist soil, displays little resist-
ance to drying, and when exposed to the air soon shrivels, as shown
by Marlatt ('07, p. 123). When conditions in the soil are unfavorable
(1. c, p. 96) as the period of emergence approaches, some individuals
respond by building a mud tube, similar to the crawfish chimneys,
which are closed with a plug of mud. That saturated ground seems
to be an unfavorable condition at this stage suggests that resistance
to the lack of oxygen decreases as the insect matures. Most of the
nymphs of this species live within less than two feet of the surface,
though some rather inconclusive observations indicate that the
nymphs have a wonderful resistance to submergence, as is shown
by the following quotation from Marlatt ('07, p. 125) : "A
curious feature in connection with the underground life of
this insect is the apparent ability to survive without injury in
soil which may have been flooded for a considerable period. Doctor
Smith records a case of this kind where a gentleman in Louisiana in
January, 181 8, built a milldam, thus overflowing some land. In March
of the following year the water was drawn off and 'in removing a hard
bed of pipe clay that had been covered with water all of this time some
6 feet deep the locusts were found in a fine healthy state, ready to make
their appearance above ground, that being the regular year of their
appearance.' Another case almost exactly similar is reported by Mr.

131

Barlow. In this instance the building of a dam resulted in the sub-
merging of the ground about an oak tree during several months of
every summer, ultimately resulting in the death of the tree. This went
on for several years, until the dam was washed away in a freshet, when
digging beneath the tree led to the discovery of the cicada larvae in
apparently healthy condition from 12 to 18 inches below the natural
surface of the ground. In both of these instances the ground may
have been nearly impervious, so that the water did not reach the insects
nor entirely kill all of the root growth in the submerged soil."

The roots of plants, and particularly those of trees, penetrate rather
deeply into the soil, but finally die, leaving a large amount of organic
substance in the soil. As the large roots decay, animals are able
through the tunnels made to penetrate rather deeply and to find organic
food, in the shape of wood and fungi. Hotter ('98, p. 225) performed
an interesting experiment which shows that wood buried three feet
below the surface and dug up after two or three months contained
spiders, mites, Thysanura, psocids, a beetle, and flies. Although this
wood was buried in a cemetery, it is not unlikely that woodland soils
commonly have such a fauna. Davenport ('03, pp. 22-23) has tabu-
lated the habitats of many Collenibola and shows that many species live
in damp soil, in sand, under bark, under stones, in caves, etc. — condi-
tions corresponding to the soil habitat. These insects are very sensi-
tive to moisture, and some are able to resist submergence in sea water
from twelve to sixteen hours per day. Davenport says (page 17) :
"During all but about six to eight hours of the day these air-breathers
are below the surface of the sand, during which time they must take in
relatively little oxygen." During certain seasons, when the soil is sat-
urated, such resistance must be of great value to its possessor. I know
of no extensive observations or experiments on the resistance of these
soil animals to carbonic acid, to the lack of oxygen, or to various com-
binations of these conditions.

That the soil conditions in glades and forests are different has
already been pointed out. We have below a good example of the
response of a forest animal to an artificial glade or clearing. A num-
ber of observations have also been made on the hastened rate of emer-
gence of the periodical cicada where the soil has been abnormally
warm, as in a hothouse (Schwarz, '90a, p. 230), or where the ground
has been warmed by flues (Marlatt, '07, p. 90), or where a forest has
been burned, and possibly the heat from the fire in combination with
its greater absorption of heat after the fire, has caused the cicadas to
emerge (Marlatt, '07, p. 94). In a forest glade, made by clearing.
Schwarz ('90a) found the cicadas emerging when none were found
in the surrounding woods. Concerning this discovery he remarks:

132

"Now, a clearing made in the midst of a dense forest forms a natural
hothouse, the soil receiving much more warmth on such places than in
the shady woods. We should thus not wonder to see the Cicada ap-
pear earlier on such cleared spaces than in the woods.'' There is there-
fore reason to expect the season to be more advanced in glades than
in the surrounding woods.

The peculiar fossorial fore legs of the cicada nymphs are marked
structural features associated with the subterranean habitat. Very
naturally, too, cleaning reactions are correlated with such a burrower,
whose legs become begrimed with the soil.

Near the surface of the soil the variety of animal life is greatly in-
creased. Not only forms which inhabit the soil regularly are present,
but many live here for short periods as adults or during some imma-
ture stage. It is not possible to draw a sharp line between the soil
community, the humus layer community, and the community of the
decayed and solid wood for these reasons : the slightly decomposed
organic debris on the surface is progressively renewed by leaves,
stems, branches, and animal remains, and is transformed below into
the humus layer; this also grades upward by all degrees, through
decaying wood into solid wood, and on to the living trees. The acid-
ity of leaves during the early stages of decay and their alkalinity at
an advanced stage is a fact of great importance, as has been shown
by Coville ('14). This suggests the paucity or absence of animals
in dense matted layers of decaying leaves.

In considering the animals that live on or near the surface of
the soil in Bates woods, certain species seem more characteristic of
rather bare mineral soils, others are more representative of open
oak-hickorv woods, and still others are representative of much
humus. The acridiid locusts found in these woods, such as Cliloealtis
conspcrsa and Mclanoplus amplcctens, are woodland rather than
prairie in their haunts, and are commonly found near the bare soil and
oviposit in it. Here live the woodland cricket Apithus, the tiger-beetle
Cicindda iinipunctata, the scavenger Gcotrupes splendidits, the mutillid
ant SphccroplithaUua, the wasp Psannuochares ccthiops and Lycosa;
and Arninophila ahhrcviata buries its eggs here in the soil.

Among the loose litter harvest spiders (Liohunmn) were found
running about, although they are not confined to these conditions, for,
like Calosoma scrutator, they climb trees. The crickets Nenwhiiis
found here seem to avoid bare soil. The larva of the beetle Mera-
cantha contracta was found among decaying leaves.

The animals living in the humus layer of the soil, and in the much
advanced stages of decayed wood, are not wholly identical, because in
the humus layer roots of living plants and fungi are so often available

133

for food. On the other hand, many of the inhabitant:^ of decayed logs,
as snails and slugs, use the log as a retreat and sally forth at night and
during moist weather to devour vegetation. Rotten wood also con-
tains many fungi affording fresh, living plant tissue.

Representative animals of the forest litter, especially of its humus
layer, appear to be certain millipeds, as Callipiis and Clcidogona. Cook
('i lb, p. 451 ) has said of them : "Nearly all the members of the group
have essentially the same habits and live in clearly similar environ-
ments. They pass their lives buried in the humus layer of the soil or
among the dead leaves or other decaying vegetable matter that fur-
nishes them food." Elsewhere he says ('iic, p. 625) : "In nature at
large the millipeds have a share in the beneficial work of reducing dead
plant material to humus. Prussic acid and other corrosive secretions
may aid in the precipitation of colloidal substances m the humus, in
addition to the protection that they give by rendering the millipeds dis-
tasteful to birds and other animals that otherwise might feed upon
them. The precipitation of colloids enables the millipeds to keep their
bodies clean and protects them against the clogging of their spiracles."
Diem ('03, pp. 383-386) gives a good summary of the habitats and
foods of certain European diplopods. I am inclined to consider the
layer of litter as the habitat of the immature panorpid Bittacns, of
which three species were found in the Bates woods. The adults fly
about among the low vegetation much after the manner of the Tipuli-
dcc, with which they are easily confused when on the wing. It is prob-
able that the larva of Panorpa confusa West, has habits similar to
those of Bittacus. I have taken the adult of this species but once — at
Bloomington, Illinois, August 23, 1892, in dense damp woods. The
larvae of Panorpa are predaceous, and this is probably true of Bit-
tacus. The ant Stigmatomma paUipes is another representative of
this community (cf. Wheeler, '05, p. 373), as are probably also a
number of tipulid larvae.

The animals of the humus layer appear to live much more active
lives than those deeper in the soil. This activity in itself allows them
a chance to secure the necessary supply of oxygen, which tends to be
deficient among the decaying vegetation; at the same time, moreover,
their movements must aid in the ventilation of the soil. It is of inter-
est to observe that millipeds abound in a habitat relatively deficient in
oxvgen, abounding in carbonic acid, and are producers of prussic acid
(HCN), whose physiologic effect is to inhibit oxidation and nutrition.
Roth (Diem, '03, p. 385) submerged some diplopods in water from six
to eight hours and thev survived. (For the marked resistance of ge-
ophiloids, see Ent. News, 24:121.) In nature they must often
meet with such conditions in the soil. One of the most abundant kinds

1B4

of myriapods in the debris on the forest floor is Spirobolus marghiatus
Say, taken in Urbana, 111., in the Brownfield woods October 15, 18,
and May 23 (many specimens), and in the Cottonwood forest October
8 and 13; at White Heath, III, May 26; at Riverside, III, August 23;
at Tonica, 111., in September; and at Bloomington, 111. This is the
common large brown diplopod, our largest myriapod. Another large
and abundant species is Fontaria virginiensis Dru. This is largely
brown dorsally, with marginal triangular yellow spots, yellow below.
A chilopod, Bothropolys multidentatus Newp., was taken in the
Brownfield woods October 18 ; and in woods at Monticello, 111., in June
(M. Waddell), with Otocryptops sexspinosiis Say. In the Brownfield
woods it was taken October 15 and 18; and here also Polydesmiis scr-
ratus Say was taken May 23. CaUipiis lactariiis Say was taken in the
Cottonwood forest previously mentioned, October 8 from decayed logs,
and in the Brownfield woods October 15, associated with Scytonotus
granulatus Say and the chilopod LitJiobius voracior Chamb. (No. 491,
C. C. A.). These predaceous kinds must be considered important
members of the humus and rotten-log communities, and are somewhat
comparaljle to the predaceous clerid beetles upon the living tree trunks
in their influence upon the community. They are, however, very sensi-
tive to moisture and live in a humid atmosphere among damp debris.
Shelford ('13b) has shown that Fontaria corriigata Wood is very
sensitve to moisture. Myriapoda are infested by a number of gre-
garine parasites (Ellis, '13, pp. 287-288).

The following statement loy Coville ('14, p. 337) is of much in-
terest : "The importance of myriapods, however, as contributing to
the formation of leafmold has not been adequately recognized. In
the canyon of the Potomac River, above W^ashington, on the steeper
talus slopes, especially those facing northward, the formation of alka-
line leafmold is in active progress. . . . Here during all the
warm weather the fallen leaves of the mixed hardwood forest are
occupied by an army of myriapods, the largest and most abundant
being a species known as Spirobolus marginaUis. . . . On one
occasion a thousand were picked up by Mr. H. S. Barber on an area
10 by 100 feet, without disturbing the leaves. On another occasion
an area 4 by 20 feet yielded 320 of these myriapods, the leaf litter in
this area being carefully searched. Everywhere are evidences of the
activity of these animals in the deposits of ground-up leaves and rot-
ten wood. Careful measurements of the work of the animals in cap-
tivity show that the excrement of the adults amounts to about half
a cubic centimeter each per day. It is estimated on the basis of the
moist weight of the material that these animals are contributing each

135

year to the formation of leafmold at the rate of more than 2 tons
per acre."

The burrows of earthworms aid in the ventilation of the soil and in
carrying down into it vegetable debris, as Darwin long ago observed.
In the blackened decayed leaves at Urbana, III, on November. 18, I
found enchytraeid worms abundant, and in the adjacent soil, below a
decayed log, a Diplocardia (No. 547, C. C. A.).

In the Brownfield woods at Urbana, among the dead leaves and in
logs during the cool season hibernating females of the white-faced hor-
net, Vespa macidata Linn. (PI. XXI, fig. 3) are often found. Females
were taken from among leaves or in decayed wood October 8, and 12
(in rotten wood), October 15 (No. 491, C. C. A.), and November 9.
The Bloomington records of hibernating females are April 23 and
October 18. In such situations two ichneumons have been taken in the
Brownfield woods : Hoplismemis nwridus Say on November 14, and
Ichneumon cincticornis Cress., November 9; also the two ground-
beetles Anisodactyhis interstitialis Say and Lehia grandis Hentz (PI.
XXI, fig. 4) on October 18; and CcuthophUus sp., Lehia grandis, Ga-
lerita jamis, the larva of M eracantlia contracta, and the large black
predaceous bug Melanolestes picipes H. S. (PI. XXII, figs, i and 2)
October 12, under bark and under logs. Melanolestes was also found
in the Cottonwood forest November 14, with the "slender-necked bug,"
Myodocha scrripcs Oliv. (PI. XXII, fig. 3). These examples show
how during the hibernating season many animals are to be expected
here which at other seasons live in other habitats. Vespa is arboreal,
as shown by the large nests seen in these woods.

Baker ('11, p. 149) has listed many mollusks found under fallen
logs and under bark in the forest of southern Michigan. As various
scavengers thrive in this zone, eating not only the vegetable debris, but
also the animals which die in it or fall upon it, the digestive peculiari-
ties of these animals are in part a response to the conditions of this
habitat. The animal carcasses which fall to the ground are compar-
able to the similar slowly falling remains which tend to accumulate
upon the bottom of bodies of standing water. The student of this
community will find of interest Dendy's ('95) paper on animals in
the soil, under stones and bark.

2. The Forest Fungus Community

Many fungi grow up through the humus layer and are food for a
great number of animals. Still other fungi grow only on and in wood. I
will not now attempt to emphasize this difference. The fleshy fungi are
very short-lived at the surface, and soon decay or are devoured by
various animals. A large number, if not most, of our land MoUusca

186

devour them. On a stump in the upland Bates woods Zonitoides
arborea, Pyramidula perspectiva, and Philomycns carolinensis were
found upon a felt-hke growth of fungi ; it is to be remembered, too,
that with the other snails lives the snail Circinaria which preys upon
them. At the time the Bates woods was examined, it was rather dry,
so that fungi were not abundant. No millipeds were found on fungi,
but Cook ('lib, p. 625) states that "The mouth parts of millipeds are
not adapted for biting or chewing, but are equipped with minute scrap-
ers and combs for collecting soft, decaying materials. Dead or dying
tissues are preferred. The only living plants that are regularly eaten
by millipeds are the fleshy fungi. Some of the native millipeds in the
vicinity of Washington, District of Columbia, feed to a considerable
extent upon the local species of Amanita, Russula, and Lactarius.
Damage is sometimes done to other plants when millipeds gain access
to wounded surfaces of roots or cuttings." A horned fungus beetle,
Boletothcrus bifurcus, living on Polyporus on stumps, was found in the
Bates woods.

At Urbana, 111., in a dense maple-bass wood forest (Brownfield)
November 14 I took a very large number of the small mycetophagid
beetle Triphylhis hiimeralis Kby. (No. 545, C. C. A.) on a shelf-
fungus, Polyporus tomentosus Fries, growing on a much decaved log.
On the under side of this same kind of fungus numerous tipulid flies
were found, some individuals evidently ovipositing. These were deter-
mined by Mr. J. R. Malloch as belonging to the genus Trichocera.
These are flies which thrive in the far north, as in Greenland. One
species, brumalis Fitch (Lintner's Second Report, p. 243) is found
common in forests in the winter season, and even when the tempera-
ture is below freezing they are on wing. Such northern forms are
likely to be active in winter or vernal farther south. On another shelf-
fungus, Dccdalia sp. taken at Urbana, 111., I found numerous speci-
mens of Arrhenoplita bicornis Oliv. (PI. XXIII, fig. 2). This is a
small greenish tenebrionid in which the males have two large horns on
the head. I have the following woodland fungus-beetles taken at
Bloomington, Illinois: Bndomychidcc — Aphorista vittata Fabr., April
14 (A. B. Wolcott) ; Brotylidcr — Tritoma- thoracica Say, June 23 (on
fungi) and July 26; T. biguttata Say (Sept. 21 ), Mcgalodacne fasciata
Fabr., March 7 (A. B. Wolcott) ; Nitidididcc — Phenolia grossa Fabr.
(July 26), Pallodes pallidus Beauv., July 2 (on gilled fungus) ; Myce-
topJiagidcc — Mycetophagus bipushdatus Mels. (April 27), M. puncta-
tus Say April 18, and June 23 (on fungi) ; Tcncbriotiida: — Platydema
mficorne Sturm. March 13 and June 23 (on fungi), Diaperis macnlata
Oliv. {hydni Fabr.) (PI. XXIII, fig. i) July 26; Melandryidcc —
Bustro pints bicolor Say, June 23 (on fungi), and B. tomentosus Say,

137

June 23 (on fungi). In the Brownfield woods at Urbana, III, Penthe
ohliquata Fabr. and P. pimelia Fabr. were taken under logs October
15 (No. 491, C. C. A.). Ulke ('02, p. 53) says. "Penthe, on fungi
growing on logs and stumps." Cratoparis lunatus Fabr. (Anthribidcc)
was taken April 5 and 23, Bloomington, III, and August at Havana,
Illinois. Figures of some of these fungus-beetles are given in Felt's
report ('06, pp. 494-498).

The general animal population of fungi is so extensive, including
mites, sow-bugs, myriapods, and mollusks, in addition to insects, that
no attempt will be made to summarize it here. The student of Illinois
fungus animals will find Moffat's paper ('09) on the Hymenomycetes
of the Chicago region very helpful. (Cf. von Schrenk and Spauld-
ing, '09.) A few references to zoological papers will aid the student
who wishes to give more attention to this interesting and increasingly
important economic subject, and a short list follows.

Busck ('02). Mushroom pests.

Hubbard ('92). Insects in Poly poms z'ohatiis Peck; and ('97)
on the ambrosia beetles.

Johannsen ('io-'i2). Mycetophilidae.

Malloch ('12). Phoridse in fungi.

Popenoe ('12). Mushroom pests.

Patch ('12). Aphids on fungi, page 179.

Ulke ('02). Notes on food habits of fungus-beetles, of which
there are many families, including SilpJiida:, Staphylinidcc,
Eitidoinychidcc, Brotylidcv, Mycetopliagidcr, Nitidididcc, Scar-
ahccidce, Tenehrionidce, Melandryidce, Scolytidcc, etc.

Jager ('74, I, pp. 245-246) and Moller ('67, pp. 59-60) have given
short lists of the German fungus insects.

The subject of fungus insects can not be dismissed without special
mention of the ambrosia beetles of the family Scolytidce. These small
beetles have been studied by Hubbard ('97), who showed that they
rear fungi in their tunnels in wood, these fungi furnishing nourish-
ment to the larvae and beetles. Each beetle seems to grow its own kind
of fungus. They belong to the following genera: Platypus, Xyleb-
orus, Corthylns, Monarthrmn, Xyloteres, and Gnathotrichus. The
beetles of the genus Corthylns live in a variety of hardwood trees,
including maple, sassafras, dogwood, etc., and attack living trees. The
ambrosia beetles are thus dependent upon fungi growing in tlie trees.
They furnish a very striking example of a mutually dependent asso-
ciational relationship. Hopkins ('99, '93a, '93b) has published much
valuable data on the life history, habitats, and enemies of these beetles.
A study of them as a biotic community would be very interesting and

188

valuable, since such a good foundation has already been built by Hub-
bard and Hopkins.

J. The Forest Undergroivth Community

Above the soil, in the layer of herbaceous and shrubby vegeta-
tion in the Bates woods, lives a considerably different assemblage of
animals from that in the soil. Running about over this vegetation, or
resting on it, are found the harvest-spiders, and in webs spread between
trees and shrubs are found Epeira insularis and verrucosa, and Acro-
soma spinea and riigosa.

In the Cottonwood forest at Urbana, cutting has made rather open
spaces so that there is considerable undergrowth, including much spice
bush (Benzoin) ; among these bushes two spiders thrive, Bpeira in-
sularis Hentz and B. douiicilioritm Hentz. The leaf -footed bug, Lep-
toglossus oppositiis Say (PI. XXH, fig. 4) also abounded on these
plants. Insularis is also in the Brownfield woods. The jumping
spider Phidippus audax Hentz, and Acrosonia rngosa were also taken
in the Cottonwood forest. In a dense shady flood-plain forest at Mun-
cie, Illinois, Acrosonia rngosa and Bpeira verrucosa and lahyrinthica
were taken August 3. The harvest-spiders Liobununi are largely
animal scavengers, but the true spiders are of course strictly pre-
daceous. The location of the spider-webs, near the ground, attests
the flight of insects upon which they depend for food. The numerous
snails feed to a large degree upon the herbaceous plants of this lower
layer, as do plant-feeding Hemiptera and the grass-eating Lcpidoptera,
including the woodland butterflies Bnodia and Cissia, other Lepidop-
tera, and Bveres, Autographa, Polygonia, and, possibly the katydid
Amblycoryph-a. In the shrub layer Bpeira domiciliorum, folded
among leaves, is a characteristic animal. It seems to thrive best in
more open woods than those in which Acrosonia abounds. Nettles
(Laportea) and clearweed (Pilea) were not searched for animals,
but were undoubtedly inhabited by a number of kinds. The same is
true of the shrubs. Young trees in this layer appear less liable to
attack by gall-producing insects than larger trees are.

The following insects feed upon woodland shrubs, and were taken
at Bloomington: Ceramhycidcc — Liopus alpha Say, June 18 (bred
from sumac by Felt, '06, p. 482), and taken by me on elm during
June; Liopus fascicidaris Harr. {xanthoxyli Shimer), June, re-
corded as from prickly ash, Zanthoxylum (Packard, '90, p. 659) ; and
Molor chits biniaculatus Say, copulating April 17, reported from dog-
wood, redbud, twigs of maple and hickory, (1. c, '90, p. 293, 424).
The curculionid Conotrachelus senicnlus Lee, was taken October 10,
1891, from the inside of a very ripe papaw at Bloomington; another

139

specimen was captured during August at Havana, 111. Felt ('06, p.
582) records seniciilus as from hickory and butternut. Attclabus
rhois Boh. was taken July 4, on hazelnut, at Bloomington. It is re-
corded from sumac, dogwood, alder, and oak.

For lists of Coccidcc living on woodland (and other) shrubs see
Cockerell ('97).

4. The Forest Crozvn Comynunity

Instead of next turning to the animals of decayed wood on the
forest floor, I wish to begin at the other end of a series, with the ani-
mals of the living tree, and then to follow an order which passes pro-
gressively through enfeebled, dying, fermenting, seasoned, and solid
wood to all stages of its decay. The decay of a fallen trunk commonly
begins with the sap-wood, thus loosening the bark, and extends in-
ward until the whole becomes soft or is changed to brown powdered
wood, which gradually changes to humus. This is a series of progres-
sive humification, and, speaking in general terms, follows the course
through which all forests tend to pass ; although fire, flood, and ani-
mals, including man, divert much wood from such a fate.

To investigate such a series fully is far beyond the scope of the
Charleston studies, and yet our material, supplemented to some de-
gree, may serve at least to outline one. The difficulties of studying
the animals of the forest crown are serious, and so far as known to
me no comprehensive work on this community has been done in this
country. Many members of it have been studied individually, but
the animals have not been studied as a community. About the
woodland insects a vast fund of facts has been accumulated in the
study of the economic problems of shade, fruit, and forest trees;
furthermore, investigations have shown that among the invertebrates
insects have a controlling or dominating influence in the forest. But
the relations of the other forest invertebrates to the forest crown have
received very little attenion from our students.

The animals of the forest crown, and particularly those of the
foliage, are more exposed to changes of temperature, moisture, wind,
and evaporation than those below the crown and protected by it. With-
in the crown there are, in fact, an upper, exposed part, and the lower,
protected part. Many of the animals of the forest crown live rela-
tively free from the influence of the substratum, as other animals in
the open water are similarly free from the influence of the bottom.
Others divide their time, part of it being spent in or on the earth, and
a part of it in the trees. Conditions of poor ventilation, darkness,
density of medium, relative stability, excess of moisture, and cor-
responding conditions in the soil, are here replaced by conditions of

140

good ventilation, intense light, and changing and a relatively dry
medium. The problems involved in these conditions vary accordingly.

The relative scarcity of molkisks and myriapods in trees is m
marked contrast with their abundance in habitats in proximity to
the soil. In the Bates woods the cherry-leaf gall-mite, Acariis, is
arboreal, but spiders are almost entirely absent. The w^alking-stick,
Diapheromera, is arboreal in part, but its eggs fall to the earth and
hatch there. The Severins ('lo) ha^'e shown that the emergence of
walking-sticks from the eggs is influenced to a very marked degree bv
moisture, dryness being distinctly injurious and moisture favorable.
The molting of the young animals seems similarly dependent upon
moisture, and may be prevented by keeping them in a "well-aerated
breeding-cage" (Severins, 'iic). This is another clear case of a
forest animal sensitive to moisture. To the fact that there is greater
moisture near the soil are therefore related the egg-laying habits and
the development of the immature insect, a development in marked
contrast with that of the strictly arboreal katydids. Of the katy-
dids, Microcentnmi and Cyrtophyllns are distinctly arboreal through-
out life, as the eggs are attached to the twigs, and they are relatively
independent of the ground. Curiously the Bates woods specimen of
CyrtophyUus was taken among low sprouts. Aniblycoryplia, how-
ever, lives near the ground. The cicadas are distinctly arboreal dur-
ing the imago stage. The larvae of Papilio tiirnits and cresphontes,
Bpargyrcus tityrus, Cressouia juglaudis (and parasite), Telca, ath-
eroma, Basilona, Halisidota, Datana, Nadata, Hetcrocauipa, Eus-
troina, YpsolopJius, and the slug caterpillar are all arboreal. Many
of these pupate on the branches or among the leaves and do not de-
scend to the earth. The sphingid Cressonia, however, pupates in the
soil. There is a marked tendency for the Lcpidoptcra to Ije com-
pletely arboreal. Even noctuid caterpillars such as Pcridroma saucia
and its allies, which live during the day on the ground, climb trees
at night (Packard, '90, p. 173; Slingerland, '95). Many of the
gall-flies are limited to certain kinds of trees and are arboreal, as, for
example, the several species of Cccidomyia found in the Bates woods.
The same is true of certain cynipid gall-makers, such as Holcaspis,
Ainpliiholips, and Andricus. It will be seen that the above-listed
kinds are largely defoliators and leaf-gall producers. Auunopliila is
a predator. Tragus and the small hymenopters (cocoons) on Cres-
sonia are parasitic.

Among the animals which live for a considerable part of their
lives in or on the soil and a part in the trees, are the two cicadas,
Calosoma, Cressonia, Aniuwphila, and certain ants, although no special
observations were made to learn to what degree the ants patrolled the

141

trees. The relatively large number of caterpillars present suggests
that in this woods they were attended by a large number of parasitic
flies and parasitic Hymenoptera in addition to predaceous insects.
The twig-pruners, Elaphidion, are referred to here because they be-
long to the crown commuity for at least a part of their lives. For
a summary of our knowledge of these beetles reference should be
made to Chittenden ('98 and '10,) and to Forbes ('11, pp. 50-53),
who gives a summary of their injury to oaks and hickories in Illi-
nois. The oak pruner, Blaphidion villosum Fabr. (PI. XXIII, figs.
3 and 4) was taken by me at Bloomington July 3. It is injurious to
hickory, maple, and other trees. The normal duration of the life
cycle appears to be one year, but in dry wood this period may be pro-
longed to four or more years (Hamilton, '87; Chittenden, '10, p.
5) — another example of the prolongation of life in dry wood. Mr.
W. P. Flint informs me that Oncideres cingulatiis Say is a common
Illinois beetle, which girdles hickory branches, and that in the dead
fallen branch its larva develops. It is reported from hickory and
basswood by Hopkins ('93b: 198.)

Additional defoliators of trees taken at Bloomington include
Macrobasis unicolor Kby. (PI. XXIV, fig. 2), taken June 27 on the
Kentucky coffee-tree, Gymnocladns. Other specimens were taken
June 4 and 12. Hamilton (Can. Ent., Vol. 21, p. 103) also records
this as defoliating locust. The larvse of the curculionid Conotrache-
liis elegans Say, taken September 5, is recorded as feeding on the
leaves of hickory. The imbricated snout-beetle, Bpicccrus imhrica-
tus Say (PI. XXIII, fig. i), was taken June 4, and, copulating,
June 27, at Bloomington. It has been recorded feeding upon the
leaves of wild cherry, plum, gooseberry, etc.

The nut-weevils may be properly considered as members of the
crown population. Of these Balanimis nasicus Say was taken August
I (on papaw) at Bloomington, and during September at Chicago.
This is recorded as from acorns, hazelnuts, and hickory-nuts. Bal-
animis nniformis Lee. was taken August 20, 21, and September 21
at Bloomington. This, too, is recorded as from acorns, as also is
B. carycu Horn, taken August 27. Miss Murtfeldt ('94) has ob-
served B. reniformis ovipositing in acorns and has described the
process. This weevil is associated and in competition with the acorn
codling-caterpillar, McHssopus latiferreana Walsm. These two in-
sects pave the way for a small caterpillar of the genus Gclcchia, and
for a second caterpillar, the larva of the acorn moth, Blastohasis
glandulella Riley, which feeds on the refuse within the acorn, and is
thus a scavenger. The debris of the predecessors is an essential for
the one that follows. Hamilton ('90) has given a good account of

142

the habits of Balaniiius (Cf. Chittenden, '08). On a previous page
mention is made of the habit of the May-beetles (LacJinostcrna) de-
foHating oaks.

The invertebrate animals of the forest crown are largely insects,
and for this reason some of the treatises on forest insects, and on cer-
tain families of Lepidoptcra, make excellent manuals for this as-
semblage. Thus Packard's "Forest Insects" ('90) and his mono-
graphs on the arboreal bombycine moths ('95; '05; '14) are very
essential. In his "Forest Insects" the various kinds of insects are
grouped according to the kind of tree and the part of the tree which
they inhabit, and thus one can readily find what is given concerning
those living upon or in the foliage, buds, fruits, twigs, etc. A some-
what similar arrangement is found in Felt's "Insects Affecting Park
and Woodland Trees" ('05, '06). The crown community varies with
the kind of trees composing it, as many kinds feed upon a relatively
small number of food plants, on allied kinds of plants, or on those of
members of the same plant association. The herbivorous species are
influenced in variety and abundance by the kind of vegetation ; their
predaceous and parasitic associates, however, are only indirectly in-
fluenced in this manner.

5. The Tree-Trunk Community

In an earlier section attention was called to the equable conditions
in tree trunks, and to the available moisture in the food of wood-
eating insects. The outer growing part of the tree contains the great-
est amount of water, insoluble starch, soluble sugar, and other food
materials ; the heart- wood, on the other hand, is dead and contains
only a small amount of water (see Roth, '95, p. 29). In view of these
relations it is but natural that the outer parts of living trunks should
be subject to attack by more animals than are the drier and less nour-
ishing inner parts. We should expect that young animals would thrive
best in the layers of the outer, moister wood, not only on account of
the softer wood being less difficult to chew, but also on account of
its greater nutriment and the larger supply of water in these layers.
The inner parts are thus protected not only by the outer layers, but
also by the general inability of many animals to digest dry wood.
Many of the insects which live in wood, particularl}^ in dry wood, re-
quire several 3^ears to attain maturity. This gradual rate of develop-
ment seems to be due in part to the slowness with which metabolic
water is produced by the growing larvae. There are many cases re-
corded in which developing larvae have apparently been delayed in
maturing for many years b}^ living indoors and in dry wood. Weis-
mann ('91. p. 48) has published an interesting case of Bitprestis splen-

143

dens which emerged from a desk which had been in use for thirty
years. He suggests that such prolonged Hves are a kind of starvation
sleep analogous to winter sleep. McNeil ('86) records that the beetle
Bburia quadrigeminata (PI. XXVIII, fig. 5) emerged from a door-
step in a house which had been built nineteen or twenty years, and
Packard ('90, pp. 687-688) records the emergence of the wood-
boring beetle Monohamimis confusor Kby., which came out of a piece
of pine furniture which had been in use "for fully fifteen years."
Felt ('05, p. 267) states that instances are recorded of Chioji cinctus
(PI. XXVIII, fig. 2) emerging from wood several years after the
furniture had been manufactured. The prolongation of the life cycle
of BlapJiidion villosum (PI. XXIII, figs. 3 and 4) in dry wood is
another case bearing upon this point. Other similar cases are known
which show that larval life is greatly prolonged in dry wood, or that
the adult in such conditions lives for many years. In such cases it is
not known just when the adult transformed.

Animals which live in living bark and living wood are in some
cases, with regard to moisture and to air, subject to peculiar conditions
brought about by the sap of the tree. In the case of hardwoods the
sap is watery, and in conifers the pitch or turpentine is gummy and
easily mires feeble insects, or suffocates them. Why is it that in hard-
woods, such as maple and box elder, all wood- and bark-boring in-
sects are not flooded in their burrows and drowned by the flow of sap
in the spring? I do not know how many factors are involved in this
problem. The gummy exudation on peach and cherry trees is evi-
dence of the influence of insects upon the flow of sap. Where sap
flows from trees many insects, particularly flies and Lepidoptera, are
attracted to and feed upon this fluid. Felt and Joutel ('04, p. 17)
state that the grubs of some members of the beetle genus Saperda
feed upon the sap, but they do not give the evidence for their opinion.

In the coniferous trees the flow of pitch has a marked influence
upon the bark-inhabiting scolytids. Hopkins ('99, pp. 404) says of
the pair of Dendroctomis frontalis, which work together to establish
the brood, that "In this operation in healthy living bark filled with
turpentine, it is necessary for one of the beetles to move back and
forth in the burrow continually in order to keep it open and push back
and dispose of the borings and inflowing turpentine. . . . From
the time they penetrate the outer layer of living bark there must nec-
essarily be an incessant struggle with the sticky, resinous mass which
is constantly flowing into the burrow and threatening to overcome
them." The larva of another bark-beetle, D. terebrans, is able to live
in this sap. Thus Hopkins (1. c, p. 418) says: "This social brood
chamber is often extended down towards or even into the bark of the

144

roots in such a manner as to hold the turpentine flowing into it. Thus
the larvae are often completely submerged in the viscid substance,
v^'hich does not appear to interfere w^ith their progress." There are
thus marked differences in these beetles in their response to sap. As
a result of utilization of the knowledge of this difference, the larvse
sensitive to an excess of sap may be killed in trees by diverting a large
amount of it into the infested bark. This plan was proposed and
practised by Robert on conifers as quoted by Packard ('90, pp. 29-
30) ; and by Hopkins ('99, p. 391) for the elm. By "cutting narrow
strips of bark from the trunks of infested elms, the Scolytids were
either killed or driven out by the increased vigor of the tree and the
greater flow of sap which it is well known will result from this treat-
ment."

The trunk of a tree is of such a substantial nature that it can not
be destroyed at once by animals. Such durability furnishes an oppor-
tunity to see how one kind of insect prepares the way for attack by
others, as is shown by the following examples. The elm borer, Sa-
pcrda tridcntata Oliv. (PI. XXIV, figs. 3 and 4), invades weakened
trees, and it is followed (Felt, '05, p. 70) by the weevils Magdalis
armicollis Say (PI. XXV, figs, i and 2) and M. barbita Say, Neo-
clyUts erythrocephulns Fabr., and, as a parasite of Sapcrda, Mclano-
hracon simplex Cress., and Bracon agrili Ashm., which is a parasite
of Neoclytus (1. c, p. 73). Four other insects have been found as-
sociated with Magdalis barbita (1. c, p. 74). Xylotrcchus colonus
Fabr. (PI. XXVIII, fig. 6) appears to be able to kill hickory, and
from such wood many insects have been reared by Felt ('05, p. 261).
Felt and Joutel ('04, p. 18) state that in hickories dying from injury
bv Scaly fits quadrispinosus Say (PI. XXV, fig. 3) the beetle Saperda
discoidca Fabr. follows, living under the bark.

The absence of woodland Ccranibycidar, Scolytidce, and Bupresti-
d(u in the Charleston collections eliminates the most important and
largest variety of insect inhabitants of tree trunks.* In addition to
the beetles which invade trunks, the large boring caterpillar, Prio-
noxvstus robinicT (cf. Packard '90, p. 53), and the horntail larva,
Treniex cohnnba, are able to do much injury. The caterpillar can

*I visited the Bates woods on June 8, 1914, and found a number of insects
in a recently cleared part of the upland area about a stump of a black oak (Q.
vclutina). Running about in the sun on the top of the stump was CJiri/sohothris fem-
orata Fabr., near the stump was a cerambj'cid, Stenosphenus notatus Oliv., and on
the bark, shaded by a vigorous growth of suckers, were the cockroach Isclinoptera
imvqualis Sauss.-Zehnt., the tenebrionid beetle Meracantha contracta Beauv., and the
otiorhynchid Fandeletejns hilaris Hbst. About the base of the stump was a large
funnel-shaped spider-web beneath which and in its meshes were remains of the fol-
lowing insects: Chion cinctns (cerambycid), Meracantha contracta, Chrysoiotkris
femorata (several specimens), an AgriUis, Passalus cornutus, and Laehnosterna.

145

kill living trees, but the horntail generally follows injury of some
kind. Within the tree trunk there is not the safety from enemies
which one might anticipate. A large number of wood-inhabiting im-
mature insects are footless, and have relatively small powers of loco-
motion. Their burrows are relatively small, so that when an enemy
once gains admission it can easily secure the owner. Tree trunks
infested with horntails often have a large number of females of Tha-
lessa on them. I have caught them literally by the handful in such
places. Many other parasitic Hynicnoptcra are easily taken upon
trees infested by boring larvae if watched carefully during the warm
parts of the day. Schwarz ('82) has called attention to a number
of beetles which live in the burrows of wood-boring insects. These
burrows may be invaded, not only while yet inhabited by their mak-
ers, but also after their abandonment. To find an insect in a burrow
is therefore not proof that the insect made it. A predaceous larva
which is reported to destroy bark- and wood-boring larvre is Alans
oculatus L. (PI. XVI, figs, i, 2, and 4). I have taken this larva in
the woods at White Heath, 111., May 26, and the beetle at Savanna,
111., May 30. The beetle was taken at Bloomington, 111., March 23
(A. B. Wolcott) ; in its hibernating cell in a rotten log in the Cot-
tonwood forest October 8 (No. 489, C. C. A.) ; and — an immature
larva — in the Brownfield woods May 23, Urbana, 111. Both the larva
and the beetle hibernate in logs. Hopkins ('04, p. 42) says of the
larva : "As a larva [it] preys upon numerous species of bark and
wood-boring insects in deciduous trees." Currie ('05, p. 102) says:
"The larvae prey upon and do much toward preventing the increase of
several of the destructive flat-headed borers (Buprcstidcc) in decidu-
ous trees." Snyder ('10, p. 8) reports the larva of Alans sp. "espe-
cially injurious" to decayed poles, and Lugger ('99, p. 130) states
that they live largely upon insects found in decayed wood. Evi-
dently the food habits of these larvae need investigation. Probably
other predaceous elaterid and trogositid larva; live in our trees.
Other predaceous beetles on trees are the following, taken at Bloom-
ington : Chariessa pilosa Forst., July 3, Clems qnadrigiittatns Oliv.
(PI. XXVI, fig. 3) lune 15, and Cv'matodcra haltcata Lee, July and
August 17. Hopkins ('93b, p. 187) reports that Chariessa pilosa
(PI. XXVI, fig. 6) is found under bark of walnut, and was taken
in a dead grape-vine, and reports also that it is predaceous. Felt
('06, p. 504) figures this species and reports it on trees infested
with borers.

The locust borer, Cyllene rohinicr Forst. (PI. XXVII, figs, i and
2), is a common insect "in manv localities, and the beetle is frequently
taken upon Solidago in the fall. The beetle was taken at Bloommg-

146

ton September 14 and October 2 (A. B. Wolcott), and on the prairie
at St. Joseph, 111., on flowers, September 26 (No. 310, C. C. A.).
Hopkins ('06, p. 8) has shown that although the larvae begin devel-
opment only in living wood, they are able to complete it in dry dead
wood, but in this case such conditions hasten development.

The apple borer Saperda Candida Fabr. (PI. XXVI, fig. 4) was
taken in the woods at Bloomington July 4. In the original forests it
probably bored in the wild crab apples and the haws (Cratcrgus). S.
tridentata Oliv., the elm borer, was also taken at Bloomington. This
is a serious pest to elms, and paves the way for Magdalis and Neocly-
his. Mr, W. P. Flint informs me that Saperda vestita Say is com-
mon throughout the state in the live bark of linden, and that Sinoxy-
lon hasilare Say lives mainly in weakened trees and in living wood.
He also tells me that Goes debilis Lee, G. tigrina DeG., and G. pid-
vendentns Hald., live in a variety of living trees.

The flat-headed apple-tree borer, Clirysobothris femorata Fabr.
(PI. XXVI, fig. 5), is known to attack the bark of enfeebled trees
and logs and stumps of oak, hickory, maple, basswood, and apple
(Hopkins, '93b, p. 183). The beetles were taken June 13, 25, 30,
and August 11, at Bloomington. Leptostylus acidiferus Say (PI.
XXVIII, fig. i) was taken April 17 in the same locality. Hopkins
('93b, p. 196) reports this insect infesting dying and dead maple- and
apple-trees. The larvae mine in the inner bark. Beutenmiiller ('96,
p. 79) states that it breeds under the bark of oak. The curculionid
Cryptorhynchiis parochus Hbst., is reported by Hopkins ('04, p. 34)
to mine as a larva in "the inner bark and sap wood of weakened and
recently dead walnut." It is also reported from butternut. Thirteen
specimens of this species were taken at Bloomington April 17. The
larvae of Romaleum atomarimn Dru. live in stumps and logs of re-
centlv dead oak (Hopkins, '04, p. 36), and are reported also from
harkberrv. The beetles were taken July 25 and August 8 at Bloom-
ington. Romaleum riifidum Hald. was taken at Charleston June 17.
This is reported from oak. The larvae of CJiion cinctus Dru. (PI.
XXVIII, fig. 2) are reported by Hopkins ('04, p. 36) to "mine the
inner bark and bore into the wood of trunk and branches of dying
and recently dead hickory, chestnut, oak, etc." This beetle was taken
at Urbana, and at Bloomington July 12. The larvae probably con-
tinue to live in the seasoned wood, as the beetles are recorded as
emerging from dry wood some years after furniture or lumber was
manufactured.

Certain species of insects live mainly in dead, though solid and
seasoned, wood, before decay causes any important changes; some
begin work in the living wood and continue in the dead wood; and

147

others begin in dead wood and continue there after it begins to de-
cay. Among the beetles which Hve in dead wood the hickory borer,
Cyllene carycu Gahan {pictus Auct.) is representative. This beetle
closely resembles the locust borer, but it appears in the spring and
early summer, rather than in the fall as does rohinics. I have taken
caryce at Bloomington April 20, 30, May 20, and June 20, and at
Urbana May 16. The larvae bore in dead branches and small trees of
hickory and mulberry, according to Hopkins ('93b, p. 194). Xylo-
trcchus colomis Fabr. (PI. XXVIII, fig. 6) lives in the bark and
wood of dying and dead timber of oak, hickory, elm, and ash (Hop-
kins, '93, p. 194). My Bloomington records of it are May 9, June
14, 25, and July i and 20. Bburia qiiadrigeminata Say (PI. XXVIII,
fig. 5) is a borer in ash and honey-locust (Packard, '90, pp. 541-
542), and has been taken on beech and elm (Hopkins, '93b, p. 193)
and in hickory. Bloomington records of it are July, August i (on
papaw), and August 28. Elsewhere mention has been made of its
long life in dry wood. Blaphidion miicronatiim Fabr. has been re-
corded by Chittenden ('98, p. 42) as emerging from dry wood as fol-
lows : "There is a divisional note on its having bred February 8,
1889, from a piece of dogwood (Cormis) which had been stored in
a carpenter shop some years to be used for hammer handles. The
larvae had worked principally under the bark where they produced
large and irregular channels, entering, when nearly full grown, the
solid wood, in which they transformed." It also lives in healthy liv-
ing wood. The larvae of Arhopalus fiilnnnaus Fabr. is reported to
live in the inner bark and sap-wood of oak. This was taken during
May at Bloomington, and Dicerca hirida Linn., a hickory borer, was
taken at Chicago August 8 and at Bloomington June 13.

The oak pruner, Blaphidion villosimi Fabr. girdles hickory
branches, which fall to the ground. From seasoned wood thus
formed Hamilton ('87) reared from branches one half to one inch
in diameter, the following beetles : "ClytantJms riiricola and alho-
fasciatus, Neoclytiis luscus and crythrocephaliis, Stenospheniis no-
tatns, etc." Such seasoned wood is particularly liable to attack,
according to Hopkins ('09, p. 66), by beetles of the family Lyctidcc
(cf. Kraus and Hopkins, '11). In such wood, too. white ants
(Termes) and carpenter-ants (Camponohis) will make extensive ex-
cavations. The northern brenthid. Bupsalis miniita Dru.. (PI.
XXVIII, figs. 3 and 4) occasionally lives in living weakened trees,
but is generally in dead wood. Hopkins ('93b, p. 207) records it as
from oak, elm', and beech, and Packard ('90, p. 69) as from white
oak. I have taken it at Bloomington June 15. 25 (copulating), and
July 2. Neoclytus liiscus Fabr., a hickory and ash borer, was taken

148

there October 15. The larvae of Neoclytus erythrocephaliis Fabr.
(PI. XXVIII, fig. 4) are associated in dead ehns with Magdalis
(Packard, '90, p. 228; Felt, '05, p. 70), and appear to follow injury
by Saperda tridcntata. In hickory, Neoclytus has been found as-
sociated with Xylotrechiis colonus Fabr., Chrysohothris femorata
Kahr., Catogenus rufus Fabr., and Tremex and Thalessa (Felt, 05,
p. 261). The cucujid Catogenus rufus was taken at Springfield July
20 by A. B. Wolcott. Liopus rariegatus Hald., taken at Blooming-
ton June 1 1 and July 22, is reported under the bark and from several
kinds of trees. The ceramb3^cid Smodicum ciicujiforme Say is also
reported from under bark, and was taken July 6 at Bloomington.
Calloidcs nohilis Say, reported from oak stumps and hickory, was
taken at Chicago in June. From oak also Purpuricenus lunncralis
Fabr. is reported. This was taken at Chicago, and Jvme 9 at Bloom-
ington. The rare lymexylid, Lymexylon sericeutn Harr., "a borer
in old oak wood," was taken at Bloomington July 2. The larva of
the flat-headed borer Dicer ca divaricata Say bores in the dead and
rotten wood of maple, cherry, etc. The beetles were taken May 9
and June 3 at Bloomington. Other wood borers whose records for
Bloomington should be given, are as follows : Leptura proxima Say,
a maple borer, June 13; Dorcaschema ivildii Uhler, an Osage-orange
and mulberry borer, June 19; Criocephalus ohsoletus Rand, July 14;
and Obcrea tripunctata Swed., whose larvae breed in twigs of cotton-
wood and blackberry, June 13 (Blatchley, '10, p. 1092).

6. The Decaying JVood Conununity

Thoroughly dry wood, or that submerged in water and thus shut
away from the air, remains sound for an indefinite period. In the
decay of wood, a certain amount of moisture, air, a favorable tem-
perature, fungi, and insects, are the main agents and conditions.
The fungi growing on wood remove the starch, sugar, and other
food materials, or they may dissolve the wood itself. This process
of course changes the character of the wood so that animals able to
derive sustenance from the solid wood now find it unsuitable for
their purpose ; and still other kinds, on the other hand, unable to
eat the solid wood, are now able to feed upon the softened product.

The rate of decay of trees varies greatly. The yellow locust (Ro-
hiuia) red cedar {Juniperus), mulberry {Morus), and hardy catalpa
(Catalpa) are very resistant. This catalpa is reported by von Schrenk
('02, p. 50) to serve as a railway tie for eighteen years and remain
sound ; as fence posts it has served from twenty-three to thirty-eight
years. Large stumps of white oak and walnut are also very durable.

149

On the other hand, cottonwood (Popidus), basswood {Tilia), and
silver maple (Acer saccharinum) decay rather rapidly. I have found
little definite information on the rate of decay of our trees. The
most definite information I have found concerning the durabil-
ity of wood in contact with the soil is in a study of fence posts
by Crumley ('lo). He shows that heartwood is particularly lial)le
to decay (1. c, pp. 613-614). He gives (pp. 634-635) the following
scale of durability, beginning with the most durable; Osage orange,
yellow locust, red cedar (woodland grown), mulberry, white cedar,
catalpa, chestnut, oak, and black ash. The following have poor dura-
bility: honey-locust, sassafras, black walnut (young trees; old trees
are durable), butternut, and elm. Red cedar growing "in the open
is about the same as oak in durability." These observations aid in
giving some idea of the relative rate of decay of logs and stumps in
contact with the soil. In the West, Knapp ('12, p. 7) has shown
that the upper part of the bole of fire-killed Douglas fir "deteriorates
more rapidly than the lower part because of the larger proportion
of sapwood. . . . Down tiuiber is less subject to insect attacks than
standing timber but decays more rapidly." Hopkins ('09, p. 128)
publishes a photograph of an Engelmann spruce forest, at an eleva-
tion of 10,000 feet on Pike's Peak, which was killed about 1853-56,
about fifty 5^ears previously; there were, however, still preserved on
the trunks, the markings of the beetles which killed the trees. The
rate of decay in warm moist regions is relatively more rapid than
that in cool and dry regions.

As wood decays it loses the characteristics which distinguish the
living and solid trees. For this reason we anticipate that animals
showing a preference for different kinds of trees are more charac-
teristic of the living and sound wood, and decline in numbers as
disintegration progresses, being replaced by the kinds which live in
and upon decaying wood. There is thus with the decay of wood a
progressive increase in the kinds of animals characteristic of humus.
This is true in general terms, for certain animals even show a pref-
erence for certain kinds of decayed wood, while others are general
feeders upon almost any kind of such wood. Hamilton ('85, p. 48)
has observed that "Cucujus clavipcs feeds on locust, maple, sycamore,
wild cherry, hickory, white oak, elm ; Clinidunn sculptile on spruce,
hemlock, tamarack, black oak, hickory, chestnut, ash, gum, poplar,
birch; Synchroa punctata on all species of oak, hickory, apple, cherry,
mulberry, Osage orange, chestnut ; Dendroidcs canadensis on nearly
everything."

The decay of wood begins when moisture and fungi are able
to gain entrance, as at some point of injury — an insect burrow, a

150

broken branch, a fire scar near the soil, etc. — and spreads from such
source. The time of year, and the method by which a tree is killed
will often have an important influence upon the kind of invasion by
animals. A tree which is killed and remains standing is not so liable
to rapid decay as one which lies upon the ground and becomes moist.
It is readily seen that there are a vast number of causes which oper-
ate to produce all degrees of decaying wood. A fallen hardwood
trunk and its stump are liable to begin decaying at the sap-wood
layer, just under the bark. The bark loosens; and moisture, fungi,
and animals mutually hasten each others' activities, and the processes
of disintegration. Lender such bark, in the Bates woods, were found
the following: Cjueens of the carpenter-ants {Camponotus) estab-
lishing their colonies ; the flat-bodied larvae of Pyrochroa; the large
Carolina slug {Philomyciis) ; the beetle Passaliis cormitiis; white
ants (Termes flavipcs) ; the rotten-log caterpillar {Scolccocauipa
lihurna) ; the snails Zonitoidcs arborca and Pyrainidiila perspectiva;
Polydcsmns, Galerita jouus, and a Mclanotus larva. These are fairly
representative kinds of animals of the log community at this stage
of development. It will be noted that the ants, the white ants, and
Galerita are predaceous, but that the remainder are probably sus-
tained largely by rotten wood, herbaceous plants, and fungi. With
the progressive radiate (when beginning within) or convergent (when
beginning without) growth of decay this animal community migrates
into the log or stump as its favorable habitat increases in area and
thickness. When this process has made considerable advance and
the log has become soft, the animals which began at the surface or
within are able to penetrate the entire log. This may be considered
an intermediate stage in the transformation of the log to humus.
This biotic community, composed of fungi and animals, commonly
begins its work at the surface (most frequentlv, in the case of fallen
trees, on the under side where the log touches the ground) and moves
progressively inward, transforming the log as it goes. In its wake
there follows a later stage of the transformation — the dark-colored
humus layer, composed of decayed wood, the dead bodies of animals,
and their excrement. The large number of years involved in such
a transformation makes it possible for many kinds of animals to find
this sort of habitat, — just as old artificial ponds are more fully stocked
with animals than newlv excavated ones. Slowly developing ani-
mals are thus able to live here, the conditions prevailing being at the
other extreme from those suited to a life in the ephemeral fungi.

As a fallen or standing trunk dries out. particularly upon the up-
per surface, if fallen, the bark often curls, cracks, and loosens from
the wood. In such a situation in the Cottonwood forest at Urbana,

151

the spider CoriracJme versicolor Keys, was taken by me March 23.
At times such places are relatively dry, and in them I have frequently
found, in large numbers, the tenebrionid beetle Nyctohates pcnmyl-
vanica DeG. This species was taken at Bloomington March 9 and
June 15. A similar-appearing relative, with enlarged femora, Mcri-
niis Icuz'is Oliv., was taken June 15 and July 29. When Nyctobates
is placed in a corked vial it proceeds to chew the cork (which is about
of the firmness of the bark and wood in which it lives) and makes a
fine sawdust. Nyctohates was taken by me November 18 under loose
dry bark of the sugar maple {Acer sacchanim) in the Cottonwood
forest (No. 549, C.C.A.). The March and November records of
this species clearly indicate that the beetle hibernates in the wood.
Scotobates calcaratiis Fabr. and Xylopimis saperdioides Oliv. are
other tenebrionids which live under bark. I have taken Scotobates
at Bloomington June 29 and July 2, and Xylopimis June 29, July 2
and 26. The cucujid beetle Brontes dubitts Fabr. was taken at Bloom-
ington March 9, April 5, July 25, 26, and September 21, and Cucujus
clavipes Fabr. (PI. XXVIII, fig. 8), whose larvae Smith reports to
be predatory, was taken March 6. Hopkins ('93b, p. 177) reports
both of these beetles from the bark of dead deciduous trees. Town-
send ('86, p. 66) reports both under the bark of decayed basswood,
and Packard ('90, p. 223) records clavipes from under oak bark.
Another common beetle, a spondylid, Parandra bninnea Fabr. (PI.
XXIX, figs. I, 2, and 5), I have taken from decayed wood at Bloom-
ington. The larvae, pupse, and beetles were found in rotten wood
July 23. and the beetles also on July 25, 26, and August 6. Hart
('it, p. 68) calls this the heart-wood borer on account of its methods
of boring in this part of several kinds of deciduous trees. It burrows
largely in rotten, and, also, according to Mr. W. P. Flint, in sound,
walnut heart-wood. In recent years Snyder ('11, p. 4) reports much
injury by it to telephone poles. He says : "The insect attacks poles
that are perfectly sound, but will work where the wood is decayed ;
it will not, however, work in wood that is 'sobby' (wet rot), or in
very 'doty' (punky) wood." As this injury is near the ground, the
invasion is probably begun in rotten wood by the young larva and ex-
tended later into the firm wood. This same author ('10, pp. 7-8)
lists several other insects associated in poles with Parandra. Clearly
this beetle is an inhabitant of wood in the early stages of decay. It
apparently does not kill trees, nor remain to the last in the log with
Passalus, but occupies an intermediate position. This is a repre-
sentative of a class of insects whose ecology has been rather slighted
in the past because of the economic conditions which permitted the
neglect of insects which were not supposed to be of much importance.

152

But with increased economic efficiency this class of insects which
hasten the decay of wood will receive more attention. Mr. W. P.
Flint informs me that in the southern part of Illinois the white ants
(Termes flavipes) and the ant Cremastogaster lineolata are very
active in decaying wood. Other inhabitants of damp rotten wood,
logs, and roots, are the larvse of the large scarabaeid Xyloryctes saty-
riis Fabr. I have taken them at Urbana, 111., October i, 12, and 15
in the Brownfield woods, and in the Cottonwood forest October 8.
Smith ('10, p. 321) reports the larva feeding in the roots of ash, and
Walsh (Proc. Boston Soc. Nat. Hist., Vol. 9, p. 287. 1863), from
the roots of grass. Osmoderma scabra Beauv. (PI. XXIX. fig. 4)
was taken at Bloomington, 111., July 26, and O. eremicola Knoch
(PI. XXIX, fig. 3) in June at Bloomington, and at Springfield, 111.,
in July by A. B. Wolcott. The larvae of both these species are known
to live in decaying wood ; the adults are found under the bark, and
according to Packard ('90, p. 283) in heart-wood. Prionus imbri-
cornis L. (PI. XXIX, fig. 6) lives under bark and in decaying wood.
One individual was taken at Bloomington July 22. Orthosoma
brunneuin Forst., another species with larval habits similar to Prionus,
was taken at the same place during July. It lives in a great variety
of decaying wood. The larvse of the common rose flower-beetle,
Trichius piger Fabr. (PI. XXIX, fig. 7), taken by me June 16, 18,
19, 22, 25, and July 7, and at Savanna, 111., May 30, live, accord-
ing to Smith ('10, p. 322), in "old oak stumps." The larvae of
Lucanus dama Thunb. (PI. XXXI, figs, i and 2) live in decaying
wood. The beetle was taken June 30, in July, and August i under
wood. The beetles of Dorcus parallcliis Say were taken May 12,
July 25, and August 6. Ceruchus piceiis Web. was taken April 5,
and one taken July 25 was covered with white fungus threads. The
larva of Dorcus and Ceruchus feed mainly or solely in rotten wood.
On Plate XXX the larva of Meracantha contracta is seen in its bur-
row in decayed wood. These insects from decayed wood are among
the most common of woodland insects.

In concluding this part on insects of rotten wood the following
papers should be mentioned, which will be of assistance to one pur-
suing this subject: Townsend ('86), on beetles in decaying bass-
wood; Packard ('90, pp. 222-223), o^i insects of decaying oak, (1. c,
pp. 283-284) in decaying elm, (p. 424) in decaying maple, and (p.
612) in hackberry; Felt ('06, pp. 484-494) on insects in decaying
wood and bark of deciduous trees; and Shelford ('13a, pp. 245-247)
on insects of decaying beech. Dury (Ent. News, Vol. 19, pp. 388-
389, 1908) states that he took over three hundred species of beetles

153

from a much decayed log; unfortunately, however, he does not pub-
lish the list.

Some of the animals which invade the log in its earliest stages
of decay continue to hold possession throughout the transformation.
Thus Passaliis arrives early, as soon as the bark begins to loosen,
and remains to a late stage in the process — when the log or stump
can easily be kicked to pieces. The rotten log caterpillar Scolccocampa
has a somewhat similar history in the log. When a log reaches such
a condition that it looks like brown meal, and is nearly level with the
surface of the ground, it may during the summer become so dry that
it affords a favorable haunt for myrmeleonid larvae ; probably the
ant-lion of Myrinclcon iininaculatus DeG., a woodland species.

In the foregoing manner the tree trunk decays and naturally sinks
lower and lower, the woody fibers disappear, the debris becomes
darker in color, the autumn leaves, twigs, and other litter of the
forest gradually add layer to layer, and finally the remains of the
log become blended with the humus of the forest floor. Thus is com-
pleted one of the most important cycles of transformation to be found
in the forest habitat. The following diagram. Figure 17, has been
prepared to show the general train or succession of insects correspond-
ing to these changes in the conditions in trees.

It will simplify this discussion of changes in the animal associa-
tions, caused either by changes in the character of the forest trees or
by changes in the woodland vegetable products, to state concisely the
main general factors involved in these changes. To explain zoological
facts it is often necessary to utilize the products of the allied sciences,
and the student may even be forced to make some investigations for
himself in these fields, because these sciences may not have especially
treated his specific problems. All relations become of zoological sig-
nificance, however, when they bear upon a zoological problem. The
major group of causes or processes which operate in such a way as to
initiate changes in forests may be grouped provisionally as follows.

1. Geological and physiographic processes: crustal movements
of the earth, as earthquakes ; the wearing down or erosion of the
land, as the mowing down of forests by landslides.

2. Climatic processes : wind storms, tornadoes, ice and sleet
storms, etc., which injure trees and destroy forests; lightning and
fires, — in brief, any climatic factor which is able to injure or kill
trees.

3. The processes of competition and succession of forest vege-
tation; based upon plant activities, as when an oak-hickory forest
is followed by a red oak-hard maple forest, or when fungi kill trees.
These causes are largely botanical problems.

154

155

4- Destruction of trees by animals: the processes of defolia-
tion, borings in branches, bark, trunk, or roots, and the girdling of
trees. Fires started by man, depending on the degree of destruction,
cause new cycles of succession. Both beavers and man build dams,
flood areas, and thus kill trees.

5. Combinations of physical and organic processes; the flood-
ing of river bottoms by driftwood rafts which become converted into
dams and thus submerge large areas.

Since it is most usual for these causes to act, not singly but in
various combinations, and since they also vary greatlv in their degrees
of influence, their operation is extremely complex. The drowning of
the forests along the Mississippi River through the sinking of the
land by the New Madrid earthquake, is a good example, showing how
a large tract of forest may be killed and much dead and decaved wood
formed, as has been shown by Fuller ('12) — (Plate XXXH). Tarr
and Martin ('12) have shown how destructive to forests the earth-
quakes are in Alaska. The influence of the New Madrid earthquake
upon animal life has not been investigated, but it is not too late even
today, after more than one hundred years, to make important studies
on this subject. On the other hand, the processes of erosion operate
more continuously than the periodic earthquakes, and tend to degrade
the land, lower the water-level, and to change the habitats in swamp
and other forests.

The results of climatic influences are seen in the amount of injury
done by sleet, which, weighing down the branches, breaks many of
them and leaves the fractured stubs as favorable points for attack
by fungi and insects. Webb ('09) has shown that when a tornado
passed through Mississippi and Louisiana the felled pine forests were
from one to three miles wide. Practically all of this timber became
infested with the larvse of Monohaimuus titillator Fabr. After a
severe frost in Florida the dead wood of the orange-trees became in-
fested bv wood-boring larva?, which spread from this wood to the
enfeebled living wood, as Hubbard (Howard, '95) observed. Light-
ning (Plummer, '12) kills and maims many trees, producing dead
wood, and through fires started in the same manner much more dam-
age is done. Hopkins ('09) considers that much of the iniurv at-
tributed to fire is primarily due to insects which made the dead and
dry fuel for the destructive fire work.

That competition among trees weakens some of them is well
known. This weakening makes them more susceptible to attack by
fungi and insects. In a forest where the shade-enduring trees can
shade out all competitors, the shrubs and trees which are intolerant
show just such a lack of resistance. As an example of this process

156

the following case may be cited : INIr. W. P. Flint informs me that
he has observed that shaded, suppressed white oaks in southern Illi-
nois are much more heavily infested bv the bark-louse Aspidiotiis
obscunis Comstock, and by the beetle Phymatodes varius Say than
are the vigorous trees.

Trees may be injured and killed by animals in many ways, as by
defoliating them, boring in the twigs, trunk, or roots, and by the de-
struction of the bark and sap-wood of the trunk. Of injuries caused
by insects the work of defoliators of hardwoods is one of the most
conspicuous kinds. Repeated defoliation of elms by the elm leaf-
beetle Galerucella luteola Miill. will, according to Felt ('05, p. 61),
so weaken a tree that Tremex coluniba finds suitable food in its dis-
eased and dying substance. With Tremex present its parasite Thalessa
also arrives. The maple borer, Plagionotiis speciosiis Say, may also
weaken a tree and pave the way for Tremex and Thalessa. A study
of the after effects of the prominent defoliators of shade and forest
trees, such as the fall w^eb-worm {Hyphantria cunea), the white-
marked tussock-moth {Hemerocanipa leucosfigiua, Plate XXXI, figs.
3, 4 and 5), the bag-worm {Tliyridopteryx epheniercrformis), the
larch saw-fly {Neiuatiis erichsonii), the gypsy moth {Portlietria
dispar), and the brown-tailed moth {Buproctis chrysorrhoca) , would
doubtless throw much light upon the details of successions caused by
insects. I have not been able to learn that this subject has been studied
carefully in this country. Such injuries are clearly not limited to
hardwoods, for many similar observations have been made in conif-
erous forests. Hewitt ('12, p. 20) has listed some of the beetles
which follow the defoliation of larches by the larch saw-fly. Hop-
kins ('OT, pp. 26-27) found that the spruces of New England were
being killed by the bark-beetle Dendroctomis piceaperda Hopkins;
that following the damage done came other beetles, such as Polyg-
ra pints rufipennis Kby., which attacks the weakened tops of the
trees, following the attack of its predecessors on the trunk or base;
and that also, following Dendroctomis, came Tetropiiim cinnamop-
ieriiin Kby., which mines in the dead trees. The yellow pines of the
West are killed by the bark-beetle Dendroctomis ponderosa, and this
is followed by many kinds of insects which live on the decaying bark
and wood, as Hopkins ('02, pp. 10—16) has shown. He also states
('OQ, p. 68) that in the Appalachian Mountains Dendroctomis fronta-
lis Zimm. killed a large part of the trees in an area "aggregating over
75,000 square miles." Such examples of multiple attack show the
complexity of the causes influencing forest life. When the great
amount of influence which insects are able to exert and do exert upon
forests is considered, the question is raised as to what may be their

157

influence in determining the kind of trees that compose what the
plant ecologists (Covvles and others) consider the cHmax forest of
eastern North America — the maple-beech forest. It has long been
known (Packard, '90, p. 515) that the beech has remarkably few in-
sect enemies, perhaps about fifty species being recorded. Its associate,
the hard maple (Acer sacchariim) , has many more, and the oaks and
hickories, which are largely absent from the climax forest and char-
acterize the changing stages preceding the climax, are preyed upon
by more insects than any other of our trees, their number possibly
equaling the sum total of all the other forest-tree insects.

A good example of the combined influence of physical and organic
factors is seen in the huge rafts of driftwood which have accumulated
in the Red River of Louisiana and Arkansas (Veatch, '06) — (Pis.
XXX and XXXIV) — on such an extensive scale that hundreds of
acres of the bottoms were flooded and the forests killed, producing
vast quantities of dead and decaying wood. With the opening of
the drainage canal, connecting Lake Michigan with the Illinois River,
the bottoms w^ere so flooded that willows, maples, cottonwoods, etc.,
on the lowest ground were killed along the river for many miles, and
presented a view similar to that shown on Plate XXXV. In this
manner vast quantities of dead and decaying wood have been made
available as food and habitat for wood-inhabiting invertebrates.

7. Interrelations within the Forest Association

The dependence of the animal upon the physical and organic en-
vironment is primarily a phase of the problem of maintenance. In
the forest these relations are so intricate, and involve the lives of so
many kinds of animals, that a forest, like the prairie, must be looked
upon as a mosaic composed of a vast number of smaller animal, or
biotic communities, each one not only interrelated at many angles
within itself, but similarly connected with the other communities of
the forest. Walsh ('64, pp. 549-550) has given us a graphic ac-
count, not of the forest as a whole but of one of its smallest units —
those which he found clustered about the galls of willow trees, the
willow leaf-gall community. He says :

"Nothing gives us a better idea of the prodigious exuberance of
Insect Life, and of the manner in which one insect is often dependent
upon another for its very existence, than to count up the species which
haunt, either habitually or occasionally, one of these Willow-galls,
and live either upon the substance of the gall itself or upon the bodies
of other insects that live upon the substance of the gall. In the single
gall S [alicis]. brassicoides n. sp. there dwell the Cecidomyia which

158

is the maker of the gall — four inquilinous Cecidomyia — an inquilinous
saw-fly (Hymenoptera) — five distinct species of Microlepidoptera,
some feeding on the external leaves of the gall, and some burrowing
into the heart of the cabbage, but scarcely ever penetrating into the
central cell, so as to destroy the larva that provides them with food
and lodging — two or three Coleoptera — a Psocus (Pseiidoneiiroptera)
— a Heteropterous insect found in several other willow-galls — an
Aphis which is also found on the leaves of the willow, but pecu-
liarly affects this gall — and preying on the Aphides the larva of
a Chrysopa (Neuroptera) and the larva of a Syrphide (Diptera) —
besides four or five species of Chalcididae, one Braconide Ichneumon
(Hymenoptera) and one Tachinide (Diptera), which prey on the
Cecidomyia and the Microlepidoptera — making altogether about two
dozen distinct species and representing every one of the eight Or-
ders. ... If this one little gall and the insect that produces it
were swept out of existence, how the whole world of insects would
be convulsed as by an earthquake ! How many species would be com-
pelled to resort for food to other sources, thereby grievously disar-
ranging the due balance of Insect Life! How many others would
probably perish from off the face of the earth, or be greatly reduced
in numbers ! Yet to the eye of the common observer this gall is noth-
ing but an unmeaning mass of leaves, of the origin and history of
which he knows nothing and cares nothing!"

With this conception of a community in mind it is only necessary
to refer to the following diagram (Fig. i8) to see how immaterial
it is as to where one begins to take up this thread of interrelations,
for sooner or later every animal and plant in the association will have
to be passed in review and its influence recognized as a response to
its conditions of life.

ECOLOGICALLY ANNOTATED LIST

I. Prairie Invertebrates

An exhaustive study of the animal ecology of a region or an as-
sociation must be based upon a thorough investigation of the ecolog-
ical relations of the individual animals composing it. An ideal an-
notated list in an ecological paper should, therefore, include for each
species a complete account of its life history, its behavior, its physi-
ology, and the structural features which would in any way contribute
to an understanding of the response of the animal to its organic and
inorganic environment. At present we have no such knowledge of
the animals of any locality or of any complex association of animals.

159

>!3

\ pooM pdReodQ

pOOM 0UlF)eO9Q
pOOM pe9Q

' SdAGQ-j

to

"«3

CD

Q

0)

■Or

c

c
o

o
o

a
o

at

a

V
CS

V

«

o
bo

o

O

" a

•a

. o

bx) o

■g (U

|5
_ >-i

S ts

as fe

bt2
Q.9

o

d S

fn OS
v

160

In a preliminary study, like the present one, it is desirable to record
rather fully the observations made in the region studied, because
we have so few descriptions of the conditions of life on our prairies.
An effort has been made to give for each species the date of obser-
■vation or collection, the locality or "station" where found, observa-
tions on habits and life history, and the field numbers of the speci-
mens secured. These numbers illustrate how observations may be
accumulated, upon a large number of individuals, without the ob-
server's being familiar with them, or even knowing" their scientific
names.

It is really surprising how little is recorded about some of the
commonest animals of the prairie and forest in zoological literature.
Other animals, particularly those of economic importance, are
treated rather fully, but generally with little relation to their natural
eiivironment. In this list it has been considered desirable not to
give an extended account of each kind of animal, but to refer to
some of the most important literature concerning it, so that one may
gain some general idea of the ecological potentialities of each kind of
animal.

MOLLUSCA

4

Physid^
Physa gyrina Say.

Three half-grown young and an adult shell were taken among
swamp milkweed, Asclepias incaniata (Sta. I, ^), Aug. ii (No. 19).
All show distinct varices ; the last one formed on the adult shell is
very distinct. These scars mark a period of rest or slow growth
which was probably due to hibernation or the drying-up of the
swamp. Physa, as a rule, can not endure such extreme desiccation as
can Lyuuicca, and to that degree is indicative of a more permanent
water supply. Our specimens were all dead, but some of them so
recently that fly maggots came from them.

Lymn^id^

Galha uinhiUcata (C. B. Adams).

A single specimen of this small snail was taken among swamp
milkweeds (Sta. I, d) Aug. 11 (No. 18). Mr. F. C. Baker, who de-
termined the specimen, writes me that this is the first record of this
species for Illinois. Baker remarks ('ir, p. 240) that this species is
"abundant in still water in sheltered borders of rivers, in small
brooks, ditches, and streams, and in shallow overflows. Clings to
dead leaves or other submerged debris, or crawls over the muddy

161

bottom of its habitat, in shallow water. Associated with Galba
ohrussa, Aplexa hypnorum, and the small planorbes (Baker). In
ditches and brooks in pastures (True). Common in damp places
and in ditches along roads where water collects only in rainy weather
(Nylander)."

Our specimen was taken where the water was very shallow (only
a few inches deep) and overgrown with vegetation. This species ap-
pears to be a strictly shallow-water marginal form, and has consider-
able power of enduring desiccation.

CRUSTACEA

ASTACIDiE

Caiiiharus gracilis Bundy. Burrowing Prairie Crawfish. (PI. XXXVI.)
The prairie crawfish was abundant at Sta. I, d, on the wet parts
of the prairie. T. L. Hankinson dug some specimens from their
holes, which proved to be of this species. Specimens were captured
Apr. 23, 191 1, and Aug. 9, 1910 (No. 7442).

Crawfish burrows were observed to traverse the dense vellow
clay with which the railway embankment had been built over a
swampy place at Sta. I, d. Burrows were also observed at Sta. I, e,
among the colony of SUphium terebinthinaceum and Lepachys pin-
tiata, and also at Station I, g.

I have found the characteristic claw of this species on wet prairie
along the railway track at Mayview, 111. At this time, September 26,
19 1 2, burrows with fresh earth were numerous, far from anv stream.
(No. 482, C. C. A.)

Combanis diogcncs Girard. Diogenes Crawfish.

Crawfish of this species were taken by T. L. Hankinson at Sta.
I, d (No. 8047A). The presence of this chintney builder at this sta-
tion suggests that the numerous chimneys shown in Figure 2, Plate
IIIB are in part the work of this species though they are in part also
the work of gracilis.

AEACHNIDA

PhAIvANGIIDA

Phalangiid^

Liobnnum. politum Weed. Polished Harvest-spider. (Pl. XXXVII,
fig. 3.)
Two small phalangiids, both probably of this species, were found
under moist wood upon the prairie (Sta. I, ^) Aug. 8. Concerning

162

these specimens, Mr. Nathan Banks writes me that they are "young,
not fully colored, but probably Liobununi politum Weed."

Weed ('91) reports that this rather rare species occurs in fields
and forests, and is seldom found about buildings. He has found it
among river driftwood, and says ('92a, p. 267) : "It sometimes oc-
curs under boards in fields, and is often swept from grass and low
herbage." When disturbed it emits, as do others of its family, a
liquid with a pungent odor. Weed ('91) has made some observa-
tions on its breeding habits. He notes that in confinement it ate
plant-lice.

L. forniosiini Wood was taken by me upon the lodged drift-
wood of a small brook on the border of a forest at White Heath,
111., May 4, 191 1. (No. 505, CCA.) This species, according to
Weed ('89, p. 92), hibernates as an adult.

Arandida
Epeirid^

Argiopc auraiitia Lucas {^riparia Hentz). Common Garden Spider.
(PI. XXXVII, figs. I and 2.)
This is very abundant, and the most conspicuous spider on the
prairie. Found among the prairie grasses (Sta. I,**^) Aug. 8 and 12
(Nos. 6 and 39) ; in its web among goldenrod, Solidago (Sta. I),
Aug. 12 (No. 26); among the swamp grasses (Sta, I, a) Aug. 28
(No. 179) ; and among Ely nuts (Sta. I, c) Aug. 24 (No. 153) ;
from sweepings made in the colony of Lepachvs pinnata (Sta. I, e)
Aug. 12 (No. 40) ; and on the Loxa prairie (Sta. II) Aug. 13 (No.
49), Aug. 27 (No. 178), and Aug. 28 (No. 179); in an open
area in the upland Bates woods (Sta. IV, a) Aug. 17 (No. 93);
and in an open glade in the lowland forest (Sta. IV, c) Aug. 22
(No. 143). In its webs in the swamp-milkweed colony (Sta. I, ^)
Aug. 9 the large dragon-fly LihcUula pulehcUa Drury was found en-
trapped ; a grasshopper, Mclanoplus differcntialis Thomas, was also
found entrapped (Sta. I, a) Aug. 28 (No. 179) ; and a large butter-
flv, Papilio polyxenes Fabr., was discovered (Sta. I, d) Aug. 12 (No.

45)-

The openness of an area rather than its prairie character appears

to determine the habitat of this spider. This is evidenced by its

presence in open spaces within the forest. It flourishes in gardens

for similar reasons. Years ago I found this species very abundant

in the late summer and fall at Bloomington, 111., in an asparagus bed,

after the plants had been allowed to grow up and form a rank mass

163

of vegetation. This species has received considerable study.
McCook ('90) and Porter ('06) record many observations on this
species. Howard ('92b) has discussed its hymenopterous parasites
and those of some other spiders.

No specimens of Argiopc traiisz'crsa Emerton, the transversely
black-and-yellow^-banded relative of aiirantia, were observed at
Charleston, although they are fairly abundant in colonies of prairie
vegetation near Urbana, e. g. at Mayview, 111., Sept. 26, and on Nov.
26, 191 1. I have seen this species only among colonies of prairie
vegetation along railway rights-of-way.

Thomisid.e

Misumciia alcatoria Hentz. Ambush Spider.

This crab-like flower spider was abundant upon flowers : on the
mountain mint, Pycnanthemiim flexiwsum (Sta. l,g), Aug. 8 (No.
6) ; on the mint, (Sta. I) with a giant bee-fly, Exoprosopa fasciata
Macq., Aug. 12 (No. 31); on the Loxa prairie (Sta. II) with the
same kind of fly, Aug. 13 (No. 47) ; on the prairie (Sta. I, g) on the
flower of the swamp milkweed, Asdcpias incarnata, Aug. 24 (No.
157) with a male bumblebee, Bonibus separatus Cress.; on Andropo-
gon (Sta. I, g) with a large immature female of ConocepJialiis, Aug.
24 (No. 1159) ; on the Loxa prairie (Sta. II) on flowers of Bryn-
giuni yiicci folium, Aug. 27 (No. 178) ; in the colony of Blymiis
(Sta. I, a) Aug. 28 (No. 179) ; and in the open glade of the low-
land Bates woods (Sta. IV, c) on the flowers of Bnpatoriuui avlcs-
tinum, with a very large syrphid fly, Milesia ornata Fabr. {=virgin-
iensis Drury), Aug. 26 (No. 184). These insects captured by the
spiders vary from about five to ten times the size of their captor. There
is considerable variation of color in this series of spiders.

It would be well worth while for some one to make a special
study of this spider, and give us an account of its methods of cap-
turing food and finding fresh flowers, with a full account of its life
history. McCook ('90, Vol. 2, pp. 367-369) gives some informa-
tion alDout the habits of an allied species of spider, but the account is
meager. Some observations on the breeding habits of this species
have been made by Montgomery ('09, p. 562); and Pearse ('11)
has recently published the results of an interesting study of the rela-
tion between the color of these spiders and the color of the flowers
they frequent. He conclud'^s thnt nitb' i^di tliis spider may change
its color slowly (from yellow to white), it does not do so witli
rapidity or in such a way as to match its surroundings, and, further,
that it does not seek an environment or a flower colored like itself.

164

He finds, however, that on zvhite flowers, ivJiite spiders occur gen-
erally, that on yellow flowers, yellow spiders occur, and also that
upon flowers of colors other than white and yellow, such as purple,
pink, and blue (p. 93), white spiders predominate.

Attid^
Pliidippiis sp.

This jumping spider was taken Aug. 12 (No. 34) on the common
milkweed, Asclepias syriaca, along the railway tracks (near Sta.
T, a), and when captured had in its jaws fragments of what seemed
to be Diabrotica 12-piinctata Oliv. ; but as the fragments were lost
during the process of capture, this determination was not made
certain.

ACARINA

Trombidiid^

Trouihidiiiui sp. Harvest-mites. Chiggers. (PI. XXI, figs, i and 2.)

These are the immature six-legged stage of a mite or mites which
when mature have eight legs. The young are parasitic on insects
(Banks, Proc. U. S. X"at. Mus., \''ol. 28, pp. 31-32, 1904) ; the
adults prey upon plant-lice and caterpillars ; one species also eats
locusts' eggs.

These mites were very abundant on the prairie north of Charles-
ton (Sta. I), and became such a pest that relief had to be sought
in a liberal application of flowers of sulphur to our legs and arms,
as is recommended by Chittenden ('06).

INSECTA
Odonata

LlBELLULID.^

Syuipctrnui ruhicunduluiii Say. Red-tailed Dragon-fly.

This dragon-fly was taken in the prairie grass zone (Sta. \, g)
Aug. 8 (No. 4.) It is one of our commonest kinds. The nymphs
live in small bodies of standing water. The adults forage for small
insects in open places, along hedge rows, and in open forest glades.
For the habitats of dragon-fly nymplis. reference should be made
to Needham (Bull. 68, N. Y. State ^lus., p. 275. 1903). William-
son ('00, pp. 235-236) has observed robber-flies carrying this species,
and has found this and other species of dragon-flies in the webs of
the spider Argiope.

165

Lihelhtla pitlchella Drury. Nine-spot Dragon-fly. (PI. XXXVIII,
fig. 2.)
Individuals were abundant in both colonies of swamp milkweeds
(Sta. I, d and g) and several were seen entrapped in webs of Argiopc
aiirantia (Sta. I, d) Aug. 9. This is one of the most abundant of
our large dragon-flies. It frequents small bodies of water and slug-
gish pond-like streams. Williamson has taken it also in the webs of
Argiopc. This large powerful insect is able to do considerable dam-
age to a spider-web and then make its escape. Among the milk-
weeds (Sta. I, d) an individual was seen by T. L. Hankinson to
escape from a web. This dragon-fly, like most of its kind, captures
small insects on wing; one kind, however, is reported to have dug a
cricket out of the ground (Psyche, Vol. V, p. 364. 1890).

Neuroptdra
Myrmeleonid^

Brachynemurns ahdominalis Say. Adult Ant-lion.

A single specimen was taken along the railway track north of
Charleston (near Sta. 1, g) Aug. 12 (No. 36). This is a species
which frequents dry habitats. The larva is unknown, but is prob-
ably predaceous — as other ant-lion larvae are and as the adult is sup-
posed to be.

Two adult females were taken July 19 and 20, 1907, at Cincin-
nati, Ohio, in my room, to which they were attracted by the electric
light. Another female was taken Aug. 8, 1901, at Gate City, Vir-
ginia (near Big Moccasin Gap). Determined by R. P. Currie.

Chrysopid^

Chrysopa ocidata Say. Lacewing. (PI. XXXVIII, fig. i.)

A single specimen of this insect was taken among prairie grasses
(Sta. I, ^) Aug. 12 (No. 44). The larvae feed upon plant-lice, and
the adults are also considered predaceous. Howard (Proc. Ent.
Soc, Wash., Vol. 2, pp. 123-125. 1893) has given a list of their
numerous hymenopterous parasites. Mr. T. L. Hankinson captured
one also (Sta. I) July 3, 191 1 (No. 7665). Fitch ('56) published
many observations on the members of this genus; and Marlatt ('94^1)
has written on the life history of this species.

166
Orthoptera

ACRIDIID/E

Syrhula admirahilis Uhler.

One specimen of this grasshopper was found in the tall prairie
grasses blue-stem Andropogon and Paniciim (Sta. I,^') Aug. 8 (No.
3). Morse ('04, p. 29) says this species frequents "open country"
and is "common in upland fields amiid Andropogon and other coarse
grasses."

Bncoptolophus sordidns Burm. Sordid Grasshopper. (PI. XXXIX,

^^- ^-^
One nymph of this species was taken in the prairie-grass colony

north of Charleston (Station 1, g) Aug. 12 (No. 44) ; another (No.

158) on Aug. 24 in the colony of Lcpachys pinnata (Sta. I, e) ; and

an adult (No. 48) Aug. 13 at Loxa (Sta. II, a) from the flowers of

Silphhmi in tegrifoliiini.

This is a species characteristic of dry open places, where the

vegetation is low. The peculiar snapping sound made by the male

when on wing is cjuite characteristic. (Cf. Hancock, '11, pp. 372-

373-)

Dissosteira Carolina Linn. Carolina Grasshopper. (PI. XXXIX,

A very reddish specimen of this species was taken in a cleared
bottom forest at River Mew Park, about three miles southeast of
Charleston, Aug. 19 (No. 95). Many specimens were observed in
the pasture above the "Rocks," on the Embarras River about three
miles east of Charleston. These individuals exhibited to a marked
degree the hovering, undulating flight which is so characteristic of
this species during the hot days of summer and early autumn. Town-
send (Proc. Ent. Soc. Wash., \'ol. i, pp. 266-267. 1890) has made
interesting observations on this habit, and finds that it is mostly the
males which participate in this courting ceremony, as he considers it.
There appears to be more or less of a gathering of individuals when
one of the locusts performs. There were perhaps half a dozen per-
forming in the colony observed at the "Rocks." Townsend (Can.
Ent., Vo\. 16, pp. 167-168. 1884) has considered this flight as re-
lated to breeding. Some one might study this subject with profit,
and determine its meaning. Poulton's paper "On the Courtship of
certain Acridiidse" (Trans. Ent. Soc. London, 1896, Pt. II, pp. 233-
252) might prove helpful in this connection.

This species seems to have been influenced by man to a marked
degree. Its original habitat appears to have been natural bare spots.

167

such as sandy beaches, banks of streams, sand-bars, and burned
areas. In a humid forested area such places are usually in isolated
patches, or in more or less continuous strips as along shores ; but
since the activities of man produce large cleared areas and bare
spots, such as roads, railways, and gardens, the favorable area of
habitat for this species has been vastly increased. Consult Han-
cock ('ii, pp. 340-347) for observations on the habits of this
species.

Schistocerca alutacea Harr. Leather-colored Grasshopper. (PI.
XXXIX, fig. 3.)
One specimen of this large grasshopper was taken cast of
Charleston, on the prairie which grades into the forest (Sta. Ill, a)
Aug. 15 (No. 59). Morse ('04, p. 39) and Hart ('06, p. 79) rec-
ognize that this species lives among a rank growth of vegetation
and brush. In general the local conditions are open or transitional,
and may be compared to those of a shrubby forest margin, and not
to those of the distant open prairie or to conditions within the for-
est. (Cf. Hancock, '11, pp. 366-370.)

Mclanoplus hiznttatiis Say. Two-striped Grasshopper. (PI. XL.

%• 3-)
This grasshopper was taken from flowers of the rattlesnake-
master, Bryugiiini yuccifoliitui, on the prairie at Loxa (Sta. II),
Aug. 13 (No. 55). It is a little surprising that it was so rare this
season on the prairie areas examined, as it is usually a common
species. Hancock ('11, pp. 356-359) has discussed this grasshopper.

Mclanoplus dijferentialis Thomas. Differential Grasshopper. (PI.
XXXIX, fig. 5, and PI. XL, fig. i.)

This species was generally common in open areas, especially on
the prairie, but was also found in open places in the forest. It was
very abundant in the colonies of swamp prairie grasses, Spartlna
and Hlymns (Sta. I, a), Aug. 28 (No. 179); in the upland prairie
grasses, as Andropogon and Panicuin (Sta. l,g), Aug. 12 (No. 39) ;
and in colonies of tepachys (Sta. I, c) Aug. 12 (No. 40); also at
Loxa on Silphimn integrifoluiin (Sta. II, a) Aug. 13 (No. 48).

This must be considered as one of the most common and char-
acteristic of prairie animals. Notwithstanding the destruction of
the original prairie,, its habitat has been perpetuated, particularly
upon waste and neglected areas, such as fence rows, roadsides, rail-
way rights-of-way, and vacant city lots.

168

Melanophis femur-riibrmn DeG. Red-legged Grasshopper. (PI.
XXXIX, fig. 2.)
This species also is one of the most common and generally dis-
tributed insects upon open areas. It was found among the prairie
grasses Andropogon and Sporoholus (Sta. I. g) Aug. 8 and I2 (Nos.
3 and 39) ; in the Lepachys colony (Sta. I, e) Aug. 12 (No. 40) ;
and in Blyinus and Spartina (Sta. I, a and c) Aug. 24 and 28
(Nos. 153, 179, and 180). As Hart ('06, p. 81) has remarked, it
is cornmon in cultivated areas. Cultivation appears to be distinctly
favorable to it ; differ entialis, on the other hand, seems to thrive best
in waste places.

LOCUSTID^

Scudderia texcnsis Sauss.-Pict. Texan Katydid.

This is the common and characteristic katydid of the prairie
areas. It was found (Sta. I, g) among the tall swamp milkweeds
Aug. 8 (No. 2) ; in the tall blue-stem Andropogon and in Panicum
Aug. 12 (No. 44) ; in the Lepachys colony (Sta. I, e) Aug. 12 (No.
40) ; and among the swamp prairie grasses Spartina and Blynuis
(Sta. I, a and c) Aug. 28 (Nos. 179 and 180). Consult Hancock,
'11, pp. 330-331, for the life history of this species.

Conocephohis sp., nymph.

A large female nymph was secured on blue-stem Andropogon
(Sta. I, g) Aug. 24 (No. 159), having been captured by a crab-
spider, Misumena aleatoria Hentz.

Orchelinmm vidgare Harr. Common Meadow Grasshopper. (PI.
XL, figs. 2 and 4.)
This grasshopper was taken east of Charleston on the flowers of
broad-leaved rosin-weed, Silphium terehintJnnaceum (Sta. Ill), Aug.
26 (No. 175) ; on the Loxa prairie (Sta. II) Aug. 27; on the flow-
ers of rattlesnake-master, Eryngimn yuccifoliuni (No. 178) ; and
on the prairie north of Charleston from the colonv of wild rye,
Blymus (Sta. I, a), Aug. 28 (No. 179). A scjueaking individual
(No. 180) captured here confirmed observations made in other
places — particularlv in the tall prairie grasses Andropogon and
Sporoholus (Sta. ^,g), where the first specimen (No. 3) was taken
Aug. 8. Nymphs, very probably of this species, were also in the
prairie grasses Andropogon and Sporoholus (Sta. 1, g) Aug. 8 (No.
3) ; and Aug. 28 (Nos. 179 and 180) in the swamp grasses Blyinus
and Spartina (Sta. I, a, c). This species is preeminently a tall-grass
frequenter, whose penetrating seeing during the sunny hours serves
to locate grass plots and low, rank weedy growths.

169

Blatchley ('03, p. 384) has observed the species feeding on small
moths, and once saw an individual on goldenrod eating a soldier-
beetle, ChaiiliognatJms pcnnsylvanicus DeG. Forbes ('05, p. 144)
reports that its food consists mainly of plant-lice, and leaves of grass,
fungus spores, and pollen. It is thus evident that it eats both animal
and vegetable food.

Xiphidhim attenuatiim Scudd. Lance-tailed Grasshopper. (PI. XL,

fig. /•)
On the prairie at Loxa (Sta. II), on flowers of the arrow-leaved

rosin-weed, Silphium integrifolium, a single individual of this species

was found Aug. 13 (No. 48).

According to Blatchley ('03, pp. 380-381) it frequents the coarse

vegetation bordering wet places. He also states that the eggs are

placed between the stems and leaves of "tall rank grasses."

Xiphidinm strictiim Scudd. Dorsal-striped Grasshopper. (PI. XL,

This prairie species was taken on prairie clover, Petalostemiim
(Sta. I, &), Aug. II (No. 21) ; in sweepings among the cone-flower,
Lcpachys piniiata (Sta. i,e), Aug. 20 (No. 40); on the mountain
mint Pycuanthcmiim flexiwsum (Sta. I) Aug. 12 (No. 35); on P.
flexiiosum or P. pilosiim (Sta. II) Aug. 13 (No. 57) ; among the
swamp grasses Blyinus and Spartina (Sta. I, a and c) Aug. 28 (Nos.
179, 180) ; on the Loxa prairie on Silphium integrifolium (Sta. II)
Aug. 13 (No. 48) ; and on purple prairie clover, Petalostciumu pur-
pur ewn (Sta. II), Aug. 13 (No. 50).

Forbes ('05, p. 147) gives its food as plant-lice, fungi, pollen
and, larp^ely, other vegetable tissues. He also states that it frequents
the "drier slopes in woods and weedy grounds" (p. 148).

Gryllid.^

Q^canthus nigricornis Walk. Black-horned Meadow Cricket. (PI.
XL, fig. 5, PI. XLI, figs. I and 2.)

This prairie cricket was taken in sweepings from the cone-flower
(Lcpachys piiinata) colony (Sta. 1, e) Aug. 12 (No. 40); on the
transitional prairie east of Charleston (Sta. Ill, &) Aug. 15 (No.
62) ; and from the swamp cord-grass, Spartina (Sta. I, a), Aug. 28
(No. 179).

Blatchley ('03, p. 451) says: "In August and September, nearly
every stalk of goldenrod and wild sunflower along roadsides, in open
fields or in fence corners, will have from one to a half dozen of these
insects upon its flowers or branches. It is also especially abundant

170

upon the tall weeds and bushes along the borders of lakes and ponds,
and in sloughs and damp ravines."

Blatchley (1. c, p. 452) made some incomplete observations on
the peculiar courting habits of this species, a subject which has been
elaborated by Hancock ('05). Hancock also describes the method
of oviposition. The female first gnaws the plant stem ; then bores a
hole and deposits an egg ; and next, again gnaws the stem. The eggs
are laid in stems of blackberry, goldenrod, and horseweed (Leptilon).

Houghton (Ent. News, Vol. 15, pp. 57-61. 1904) has published
interesting observations on the carnivorous habits of nymphs of
CE. niveiis DeG. Cf. Parrott and Fulton, '14.

Ashmead (Insect Life, Vol. 7, 241. 1894) reports that CE. nigri-
cornis ( fasciatns) is preyed upon by the wasp Chlorion harrisi Fernald
{Isodontia philadclphica St. Farg.).

Qicanthus quadripunctatus Beut. Four-spotted White Cricket.

This prairie species was found among the tall prairie grasses
blue-stem Andropogon and Panicimi (Sta. I, ^) Aug. 8 (No. 3);
and among the colony of cord grass, Spartina (Sta. I, a), Aug. 28
(No. 179).

Blatchley ('03, p. 453) reports it on "shrubbery and weeds in
fence-rows and gardens; and along roadsides." This indicates how
a prairie species adjusts itself to the conditions produced by man.
Parrott (Journ. Econom. Ent., Vol. 4, pp. 216-218. 191 1) gives
figures of the eggs of this species and describes its method of ovipo-
sition in raspberry stems.

Hemiptera

ClCADIDJE

Cicada dorsata Say. Prairie Cicada.

Although this species was not taken at Charleston, a single speci-
men (No. 185) was captured at Vera, Fayette county, 111., Septem-
ber I, on a giant stool of blue-stem Andropogon. Osborn (Proc.
Iowa Acad. Sci., Vol. 3, p. 194. 1896) reported one specimen from
Iowa; Woodworth, (Psyche, Vol. 5, p. 68. 1888) says: "On the
prairies, Illinois to Texas"; and IMacGillivray (Can. Ent., Vol. 33, p.
81. 1901) adds Missouri, Colorado, and New Mexico.

Membracid.e
Campylenchia ciirvata Fabr.

This bug was taken in sweepings made in the colony of cone-
fiower, Lcpachys pinnata (Sta. l,e), Aug. 12 (No. 40).

171

Jassid^
Flatynietopms frontalis Van D.

This leaf-hopper was taken in sweepings in the cone-flower col-
ony (Sta. l,e) Aug. 12 (No. 40).

Aphidid^

Microparsus variabilis Patch.

This plant-louse infests the leaves of the Canadian tick-trefoil,
Desnwdiiini canadense, and causes the leaves to curl. Quite a colony
of these plants found infested (near Sta. I, /) Aug. 24, were stunted
and deformed by these plant-lice (No. 160). Consult Patch (Ent.
News, Vol. 20, pp. 337-341. 1909) for a description of the insect
and a plate showing the injury which it causes; also Williams (Univ.
Studies, Univ. Neb., Vol. 10, p. 76, 1910) and Davis (ibid., Vol. 11,
p. 28. 1912).

Aphis asclepiadis Fitch. Milkweed Plant-louse.

Plant-lice of this species were abundant upon the younger ter-
minal leaves of the common milkweed, Asclcpias syriaca, along the
railway track north of Charleston (Sta. I) Aug. 12 (Nos. 28, 29,
and 154). Associated with them were workers of the ants Formica
fusca Linn. var. subsericea Say (Nos. 28, 29, and 154) and For-
mica fusca Linn. (No. 28). On a milkweed plant which lacked the
plant-lice were found associated another ant, Formica pallid e-fiilva
Latr., subsp. schouftissi Mayr, var. incerta Emery, and the metallic-
colored fly Psilopus sipJio Say.

At LTrbana, 111., a very abundant plant-louse on wild lettuce,
Lactiica canadensis, is Macrosiphuni rndbeckicc Fitch (det. by J. J.
Davis). The upper, tender branches of these plants are in the fall
covered with vast numbers of these lice, both wingless and winged.
That this species feeds upon a number of other prairie plants is a
point of much interest because of their distinctly prairie character.
It is reported from Vernonia, Solidago. Bidens, Ambrosia, Cirsimn,
Silphinui, and Cacalia (Thomas, Eighth Rep. State Ent. 111., p. 190.

[879).

Pentatomid>33

Biiscliistus variolariiis Beauv. (PI. XLI, fig. 3.)

This common plant-sucking bug was taken on flowers of the
swamp milkweed, Asclepias incarnata (Sta. I, d), Aug. 9 (No. 12) ;
from the blue-stem Andropogon colony (Sta. l,g), where a large
lobber-fly, Promachus vcrtcbratiis, was taken astride a grass stem
with one of these bugs in its grasp Aug. 12 (No. 39) ; at Station

172

I by T. L. Hankinson, July 3, 191 1 (No. 7665) ; on the Loxa prairie
(Sta. II), with insects from flowers of the purple prairie clover,
Petalostemum purpureum, Aug. 13 (No. 50) ; and on flowers of the
mountain mint Pycnanthemum pilosiim or P. flextiosum (Sta. II),
Aug. 13 (No. 52). Consult Forbes ('05, pp. 195, 261) for a sum-
mary of its life history, and references to literature. It feeds upon
a great variety of plants (Olsen. in Journ. N. Y. Ent. Soc, Vol. 20,
p. 53. 1912) and on soft-bodied insects.

Stirctrus ouchorago Fabr. (PI. XLI, fig. 5.)

This highly colored bug was taken, Aug. 23 (No. 146), not
upon the prairie proper but at the margin of the Bates woods (near
Sta. IV, a), where the clearing had been so complete that only
sprouts and young trees occurred, associated with many plants which
frequent open, sunny places, such as ironweed (Vernonia) and
Pycnanthcuium pilosiim.

This bug sometimes feeds upon the larv?e of the imported as-
paragus beetle, Crioceris asparagi (Chittenden, Circ. No. 102, Bur.
Ent., U. S. Dept. Agr., p. 6. 1908). This circular contains figures
of the nymph and adult. Olsen reports it as feeding upon cater-
pillars and beetle larvee and on the plants Asclepias and Rhus (Jour.
N. Y. Ent. Soc. \"ol. 20, pp. 55, 56. 1912).

Thyreocorid.e

Thyrcocoris puUcariiis Germ. Flea Negro-bug. (PI. XLII, fig. 2.)

This negro-bug was taken on the flowers of goldenrod, Solidago
(near Sta. I, a), Aug. 12 (No. 26). Forbes and Hart ('00, p. 100)
state that this insect abounds on Bidens, a plant which grew in great
abundance near the goldenrod referred to. Taken (Sta. I) by T. L.
Hankinson July 3, 191 1 (No. 7665).

Lyg^id.5:
Ligyrocoris sylvestris Linn.

This insect was taken while sweeping vegetation in the cone-
flower (Lepachys) colony (Sta. 1, e) Aug. 12 (No. 40).

Lygcrus kalniii Stal. Small Milkweed Bug. (PI. XLII, fig. i.)

This is one of the commonest insects found upon milkweeds of
the prairie. Specimens were taken on the flowers of the swamp
milkweed, Asclepias incarnata (Sta. l,g), Aug. 8 (No. i) ; on flow-
ers of the mountain mint, PyciiautJicniuuu flexiwsum (Sta. I, g) ,
Aug. 8 (No. 6); and on swamp milkweeds (Sta. 1, d) Aug. 9
(No. 12).

173

This is another common insect about which very little is known.
Its food plants and life history are worthy of study. I have taken
this species from Mar. 20 (adult, 1894) to Nov. 4 (adult, 1893) 'it
Bloomington, 111.; at Havana, 111., during August; and at Chicago
June 8 (1902). That it probably hibernates in the adult stage is
shown by the fact that I captured an adult as early as Mar. 22 at
Urbana, 111. This bug, like the squash-bug (Anasa), may have
an active migratory period in the fall, and only those individuals
survive the winter w4iich happen to be in favorable places w^hen the
cold weather sets in. I have captured this bug in the dense Brown-
field woods (Urbana), where it was crawling on a log Oct. 12 (No.
312, C.C.A.). Hart ('07, p. 237) records it from Asclepias cornuti
(--=A. syriaca) at Havana in the sand area, and also from Teheran,
Illinois.

Oncopeltiis fasciatus Dall. Large Milkweed Bug. (PI. XLII, fig. 3.)
This large red plant-bug I took but once — on flowers of the
swamp milkweed, Asclepias incarnata (Sta. l,g), Aug. 8 (No. i);
T. L. Hankinson, however, captured another specimen (Sta. I) July
3 1911 (No. 7665).

I have found it in years past abundant on prairie colonies of
milkweed at Bloomington, 111., from June into September, and at
Havana and Chicago during August. On Sept. 26, at Mayview, 111.,
along the railway among prairie plants this plant-bug was found on
dogbane (Apocyniim). A pale yellow color may replace the red.

CORElTfM

Harmostcs rcflcxiilns Say.

This bug w-as found in flowers of Asclepias syriaca along the
railway track (Sta. I) Aug. 12 (No. 27).

Eeduviid.e

Sinea diadcma Fabr. Rapacious Soldier-bug. (PI. XLI, fig. 4.)

One specimen of this bug was taken from the flowers of the
mountain mint, Pycnanthenmm flexuosum, in the prairie grass col-
ony (Sta. l.g). Aug. 8 (No. 6). I took it at St. Joseph, 111., in a
colony of prairie vegetation along the railway track Sept. 26, 191 t
(No.495^C.C.A.).

This bug preys upon caterpillars and many other insects. The
little we know of its life history has been recorded by Ashmead ('95,
Insect Life, Vol. 7, p. 321); its predaceous habits, however, have
attracted considerable attention from economic entomologists. For

174

numerous references to this phase see Caudell, Jour. N. Y. Ent. Soc,
1901, \'ol. g, p. 3. The young feed upon plant-hce.

Phymatid^

Phvinata fasciata Grav (icolffi Stal). Ambush or Stinging Bug.
(PI. XLII, fig. 4') "

This is one of the most abundant and characteristic of prairie
insects. It was taken from the flowers of the swamp milkweed.
Asclepias iucarnata (Sta. I, g), Aug. 8 (No. i); among the same
flowers, at Station \,d, Aug. 9; on goldenrod, Solidago (near Sta.
1,0-), Aug. II (No. 20); and again on goldenrod (Station I)
Aug. 12 (No. 43), in copula, and with an empidid fly in its clasp;
on flower of mountain mint, Pycnanthcmuni flcxitosmn (Sta. I),
Aug. II (No. 24); from goldenrod (Sta. I) Aug. 12 (No. 26);
in sweepings from the colony of LepacJiys pinnata (Sta. I, e) Aug.
12 (No. 40) ; from the flowers of the mountain mint, P. flexuosiiui,
on the Loxa prairie (Sta. II) Aug. 13, with a large beefly, Exo pro-
se pa fasciata, in its clutches (No. 57) ; on the following flowers
(Sta. II) Aug. 13 — rosinweed, Silphiuin integrifolium (No. 48),
mountain mint PycnanthcnuDU pilosimi and P. flexiiosum (No. 52),
Culver's-root, Veronica zirginica (No. 54), and rattlesnake-master,
Bryngiuni yuccifolinin (No. 55) ; in the partly cleared area north of
Bates woods (Sta. IV) in flowers of the mountain mint P. pilosuni
Aug. 23 (No. 146) ; and on the Loxa prairie, at telegraph pole No.
12323 (Sta. II), on the flowers of rattlesnake-master Aug. 27 (No.

178).

At May view, 111., in a colony of prairie vegetation, one speci-
men was taken by Miss Ruth Glasgow with the butterfly Pontia pro-
todicc Sept. 26, 1912; a second had captured a dusky plant-bug,
AdclpJiocoris rapidus Say. At the same time and place Miss Grace
Glasgow took from a flower another bug with the bee-fly Sparnopo-
liiis fidinis Wied. This fly is parasitic on white-grubs, LacJinosterna
(Forbes. '08. p. 161). Among prairie vegetation at St. Joseph, 111.,
Sept. 26, 191 1, I took from a flower an ambush bug with a large
cutworm moth, Fcltia suhgotJiica Haw. (No. 302, C.C.A.). (PI.
XLIII, figs. I and 2.)

Packard {^"J^, p. 211) records that Phyniata fasciata had been ob-
served feeding upon plant-lice on linden trees in Boston, and Walsh
(Amer. Ent., Vol. i, p. 141. 1869) states that it feeds habitually
upon bees and wasps, and shows skill in avoiding their sting. Cook
(Bee-keeper's Guide, ninth ed., pp. 323-324, 1883) reports that it
destroys plant-lice, caterpillars, beetles, butterflies, moths, bees, and

175

wasps. The ambush bug and the ambush spider {Misuuicna alea-
toria Hentz) are in active competition upon flowers for much the
same kind of food.

MlRID^

Adclphocoris rapidus Say. Dusky Leaf-bug. (PI. XLIT, figs. 5
and 6.)
This leaf-bug was taken from the flowers of the rattlesnake-
master, Eryngium yiiccifoliiim (Sta. 11, a), Aug. 13 (No. 55). It
was taken in a colony of prairie vegetation at Mayview, 111., Sept. 26,
1912, by Miss Ruth Glasgow, who found it captured by Phymata
fasciata. It feeds upon a large variety of plants.

Lygiis pratefisis Linn. Tarnished Plant-bug. (PI. XLIII, figs. 3
and 4. )
This common plant-bug was taken, copulating, from the fl.owers
of the swamp milkweed, Asdcpias incarnata (Sta. I, d), Aug. 9 (No.
12). It is a common fruit and garden pest. Consult Forbes ('05,
pp. 119, 263) for figures of this species and references to its life his-
tory and habits, and Crosby and Fernald ('14) for a very full account
of this species.

coleoptera

Carabid/e
LcptotracJicliis dorsalis Fabr.

This ground-beetle was taken in the Spartiiia colony on the
prairie north of Charleston (Sta. I, a) Aug. 28 (No. 179). It is
supposed to be predaceous. Its life history is not known to the
writer. Blatchley ('10, p. 138) records it as from "low herbs in
open woods", and Webster ('03b, p. 22) states that the larva of this
beetle destroys the larvae of Isosoma grande Riley in wheat fields.

Although no special effort was made to secure members of this
family of beetles from the prairie, where they must abound, it is sur-
prising that some members of the genus Harpalus were not so
abundant as to demand attention. More attention to the ground
fauna and less to that found on vegetation would doubtless have
given other results. Generally in this family the food habits are
predaceous, but there are exceptions, and these include kinds which
frequent open places. On September 25, 1900, the writer found
specimens of Harpalus caliginosus Fabr. feeding on the flowers or
seeds of ragweed, Ambrosia, which grew in a neglected field along
Holston River near Rogersville, Tenn., and at Rockford, Tenn., on
Sept. 25, 1901, similar observations were made upon Harpalus penn-
sylvaniciis DeG. Many years ago Webster ('80, p. 164) made simi-
lar observations on this species, and also found it eating wheat, timo-

176

thy seeds, the prairie grass Panicum crusgalli Linn., and even a small
beetle, Ips 4-giittatns Fabr. He also observed H. caliginosus feed-
ing upon seeds of ragweed. Ambrosia artemisiifolia. (See Forbes —
'80, pp. 156-157 and '83a, pp. 45-46 — for further observations upon
the food habits of the beetles of this genus.) Clarkson (Can. Ent.,
Vol. 17, p. 107, 1885) observed caliginosus feeding upon ragweed
on Long Island; and Hamilton (Can. Ent., Vol. 20, p. 62, 1888) re-
cords similar observations for this beetle and for pennsylvaniciis.
Both species are reported to injure strawberries. Coquillett (Insect
Life, Vol. 7, p. 228, 1894) observed caliginosus feeding upon a
grasshopper.

COCCINELLID^

Hippodamia parenthesis Say. .Parenthetical Ladybird.

This insect was taken only by T. L. Hankinson (Sta. I) July 3,
191 1 (No. 7665).

Coccinclla novemnotata Hbst. Nine-spotted Ladybird. (PI. XLIV,

%• 2).
This insect was taken on the common milkweed, Asclepias syri-
aca, (Sta. I) Aug. 12 (No. 27). This species is another example of
one of the commonest insects to which so little attention has been
given that we really have no full account of its life history and ecol-
ogy. Many scattered observations have been made, but none are ex-
tensive. Forbes examined the stomach contents of five specimens
and found that they had eaten plant-lice, fungus spores, and a few
lichen spores ('80, pp. 157-159, and '83a, pp. 53-54)-

Lampyrid^e

Clwuliognatluis pennsylvaniciis DeG. Soldier-beetle. (PI. XLIII,
figs. 5 and 6.)

This is one of the most abundant beetles found on flowers in late
summer and fall, particularly upon goldenrods (Solidago), and
other composites. The first specimens were taken in a cleared area,
with much sprout growth and open patches, where the mountain mint
Pycnanthemum pilosiun abounded, (near Sta. IV, a), Aug. 23 (No.
146). On the following day they were first found on the prairie —
copulating as usual — on the flowers of the swamp milkweed, Asclepias
incarnata (Sta. I, d), Aug. 24 (No. 156.)

They were taken from the flowers of the broad-leaved rosin-
weed, SilpJiiiim, terebinthinaceum, on the prairie east of Charleston
(Sta. Ill, b) Aug. 26 (No. 175), and on the Loxa prairie (Sta. II,

177

Pole No. 12323) on the flowers of the rattlesnalve-master, Bryngium
yuccifoliimi, Aug. 27 (No. 178).

According to Riley (Second Rep. U. S. Ent. Comm., p. 261.
1880) the eggs of this species are deposited on the ground in irregu-
lar bunches. He quotes Hubbard, who says that the larvae huddled
together when ready to moult, and that afterwards they became very
active. The insect passes the winter as a nearly mature larva, and
matures about August. The larvae are known to eat beetle larvae and
caterpillars; the adults feed upon nectar and pollen.

SCARAB.EID^

Liiiphoria scpiilchralis Fabr. Black Flower-beetle. (PI. XLIV,

fig- 4-) .

Only two specimens of this beetle were taken : one on the flowers
of the swamp milkweed, Asclepias incarnata (Sta. I, d), Aug. 24
(No. 156) ; the other from the flowers of Pycnanthemiun pilosum in
the cleared area bordering the upland Bates woods (Sta. IV^ a) Aug.
23 (No. 146). Blatchley ('10, p. 997) reports it at sap, on various
flowers, and especially on goldenrod ; and Webster has found it eat-
ing into kernels of corn (Insect Life, Vol. 3, p. 159).

B. inda (PI. XLIV, fig. 3) has been observed by Wheeler ('loa,
p. 384) to fly to an ants' nest and bury itself; he suggests that it may
live in such nests. Schwarz ('90b, p. 245) considers the inda larvae
abundant at Washington in nests of Formica integra. For the life his-
tory of this beetle see Chittenden (Bull. 19, N. S., Bur. Ent., U. S.
Dept. Agr., pp. 67-74. 1899).

Pelidnota punctata Linn. Spotted Grape Beetle. (PI. XLIII, fig. 5.)

Only one specimen of this beetle was taken. It was found upon
a prairie containing some forest relics, on a grape leaf (Sta. Ill, b)
Aug. 15 (No. 58). This insect is a forest or forest-margin insect; as is
indicated by the fact that the larva feeds upon the decaying roots
and stumps of oak and hickory. The adult devours leaves of the
grape and of the Virginia creeper (Cf. Riley, Third Rep. Insects
Mo., p. 78).

Cerambycid/e

Tetraopes tetraophthalmns Forst. Four-eyed Milkweed Beetle.

This is one of the commonest insects in the prairie parts of Illi-
nois. Nevertheless, though almost every schoolboy who ever made
a collection of insects has it in his collection, very little is known of
its habits or life history.

178

At Charleston it was taken Aug. 8 on flowers of the swamp milk-
weed, Asclcpias incarnata, at Sta. \, g (No. i) and at Sta. I, d (No.
12) ; on the flowers of the mountain mint Pycuantheuuim virgiiii-
'annrn (Sta. I) Aug. 12 (No. 35); and T. L. Hankinson took the
beetle (Sta. I) July 3, 191 1 (No. 7665).

Robertson (Trans. St. Louis Acad. Sci. A^ol. 5, p. 572. 1891)
states that this beetle and Epicauta vittata Fabr. gnaw the flowers of
the swamp milkweed; and in the same volume (p. 574) reports that
the rose-breasted grosbeak (Hahia ludoviciana) cleared these beetles
from A. syriaca in his yard. Beutenmiiller (Jour. N. Y. Ent. Soc,
\'ol. 4, p. 81. 1896) says that the larva bores into the roots and lower
parts of the stems of Asclcpias, and suggests that the other species
have similar habits.

Tetraopes femoratus Lee. (?) Milkweed Beetle.

A peculiar individual (No. i) was taken Aug. 8 on the swamp
milkweed Asclcpias incarnata (Sta. L d). Mr. C. A. Hart, who de-
termined the specimen, remarks that it "is very remarkable — thorax
of fenwratiis, antennae and pattern nearest to 4-ophtlialmus.''

Chrysomelid.e

CryptoccpJialus venustiis Fabr.

This leaf-beetle was taken from the flowers of prairie clover,
Petalostcmiim (Sta. L &). Aug. 11 (No. 21). Blatchley ('10, p.
1 123) states that it is found on the flowers of Erigcron in timothy
fields, on iron weed, and on wild sweet potato. Chittenden ('92,
p. 263) has observed the var. simplex Hald. on ragweed, Amhrosia
irifida, "dodging around the stem after the manner of a scjuirrel or
lizard on a tree-trunk The insect is a polyphagous leaf-eater."

ChrysocJuts aiirafus Fabr. Dogbane Beetle.

Only two specimens of this usually common metallic-green beetle
were seen and secured. One (No. 14) was taken Aug. 9 on the
dogbane or Indian hemp. Apocynuin incdiuni, growing among the
swamp milkweeds. Asclcpias incarnata (Sta. I, d) ; and the other on
dogbane in the upland part of Bates woods (Sta. IV, a), Aug. 20,
1910 (No. 103). Later, July 3. 191 1, T. L. Hankinson (Sta. I) also
secured this beetle (No. 7665). The food plant was abundant, but
the beetles appeared to be exceptionally rare. This is another widely
recognized but really little known insect. It is also found on the
leaves of milkweeds. Zabriskie (Jour. N. Y. Ent. Soc, \^ol. 3. p.
192. 1895) describes the egg-capsules of this species, which he found
early in July on fence posts, near plants of the spreading dogbane.

179

Apocynum androsccmifoliurn, and especially upon the under surface
of the leaves of this plant. A single ^gg is deposited within a conical
black mass, which is probably the excrement of the beetle. To this
note Beutenmiiller adds that "the larvae, after hatching drop to the
ground and live on the roots of the plant."

With so much of a clue, the complete life history of this species
ought to be worked out without much difficulty. Forbes once re-
ported this species injuring potato (Lintner, Fourth Report on the
Injurious and other Insects of the State of New York, p. 142).

Nodonota convexa Say.

This small leaf-beetle was taken in sweepings of vegetation in a
colony of the cone-flower, Lcpachys pinnata (Sta. 1,(7), Aug. t2
(No. 40). Blatchley ('10, p. 1149) states that it occurs in low-
places on ragweed. Ambrosia trifida. This cone-flower colony was
on rather low land containing crawfish holes.

Trirliabda tomentosa Linn.

This insect was taken at Station I by T. L. Hankinson July 3,
191 1 (No. 7665). It is common on Solidago. Schwarz (Am. Nat..
Vol. 17, p. 1289. 1883) reports it as a defoliator of prickly ash
(Zantlioxylmn).

Diabrotica 12-punctata Oliv. Southern Corn Root-worm. (PI.
XLV, fig. 3).
This common corn pest was taken in sweepings of the vegetation
in a colony of Lcpachys pinnata (Sta. I, c) Aug. 12 (No. 40), and
T. L. Hankinson captured it (Sta. I) July 3, 191 1 (No. 7665). A
few feet away w^as a large corn field. It was also taken on the
flowers of Brynginui vuccifoUuni on the prairie at Loxa (Sta. II)
Aug. 13 (No. 55). Here also a field of corn stood only a few feet
away.

Diabrotica longicornis Say. Western Corn Root-worm. (PI. XLV,
fig. I.)
This beetle was found upon the flower-masses of the mountain
mint Pycnanthcrmim pilosum, growing in a forest clearing (near
Sta. IV, a) Aug. 23 (No. 146). It feeds upon the silk and pollen
of corn, and probably on the corresponding parts of other plants.

Diabrotica atripcnnis Sav.

One specimen of this beetle was taken on the flowers of the
swamp milkweed, Asclcpias iiicaniata (Sta. l,d), Aug. 8 (No. i).
Very little appears to be recorded on this species except that it feeds
upon the pollen and silk of corn, the pollen of composites, and the
blossoms of beans (Forbes, '05, p. 189).

180

Meloidje

Zonitis hilineata Say. Two-lined Blister-beetle. (PI. XLIV, fig. i.)

This beetle was taken on the apical leaves of the common milk-
weed, Asclepias syriaca (Sta. I), Aiig. 12 (No. 33). Blatchley
('10. p. 1356) records it as from the flowers of the wild rose.

Bpicaitta Ji'ittata Fabr. Old-fashioned Potato Beetle or Striped
Blister-beetle. (PI. XLV, fig. 5.)

Several specimens were taken by T. L. Hankinson at Station I
July 3, 191 1 (No. 7665).

Bpicauta marginata Lee. Margined Blister-beetle. (PI. XLV, fig. 2.)

This beetle was taken at Station I by T. L. Hankinson only — July
3, 191 1 (No. 7665) ; it was taken also from the leaves of the rosin-
weed, Silphium integrifolimn, on the Loxa prairie (Sta. II) Aug. 13
(No. 48) ; from an open ravine in Bates woods (Sta. IV, h) Aug. 22
(No. 124) ; and in the lowland glade (Sta. IV, c) Aug. 22 (No. 143).
For accounts of the common Illinois species of blister-beetles see
Forbes and Hart ('00, pp. 487-490, and Forbes, '05, pp. 111-114).

Bpicauta pcnnsylvanica DeG. Black Blister-beetle.

This beetle was collected from flowers of goldenrod, Solidago
Sta. I, a), Aug. 12 (No. 26); on the Loxa prairie (Sta. II) from
flowers of the rosin-weed, Silphimn intcgri folium, Aug. 13 (No.
48) ; on flowers of Silphium tcrehinthinaceiim (Sta. Ill a), Aug. 20
(No. 119) ; in the cleared margin of Bates woods (near Sta. IV, a),
on flowers of Pycnanthemnm pilosum Aug. 23 (No. 146) ; again on
goldenrod, Solidago (near Sta. I, fl), Aug. 24 (No. 152) ; and from
the Loxa prairie on flowers of rattlesnake-master, Bryngium yucci-
folimn, (Sta. II. a) Aug. 27 (No. 178).

The larvae of this and some other species of blister-beetles prey
upon locusts' eggs. (Cf. Riley, First Rep. \J. S. Ent. Comm., p. 293.
1878.) The beetle lays its own eggs in the vicinity of the locusts'
eggs.

Rhipiphorid.;e

Rliipiphorus dimidiatus Fabr.

Five specimens of this mordellid-looking little beetle were taken
on flowers of the mountain mint Pycnanthemum flexuosum (Sta.
I, ^) Aug. 8 (No. 6) ; and three specimens on flowers of the moun-
tain mints P. flexuosum and P. pilosum on the Loxa prairie (Sta.
II) Aug. 13 (No. 52). Blatchley ('10, p. 1366) reports it as from
the flowers of P. linifoliuin Pursh.

181

These small beetles are black except the basal two-thirds of the
elytra, which are pale yellow. The larvae are parasitic on wasps, as
has been shown by Chapman for the European species paradoxus
(Ann. Mag. Nat. Hist., Ser. 4, Vol. 5, p. 191, and Vol. 6, p. 314.
1870). The larvae undergo a very peculiar metamorphosis wliich is
related to their parasitic habit. It is desirable that the life histories
of the American species should be studied.

Ashmead (Psyche, Vol. 7, p. yy. 1894) reared this beetle from
the cells of the wasp Bumenes fratcrna Say. Riley (Sixth Rep. Ins.
Mo., p. 125. 1874) states that he bred Rhipiphorus pectinatus Fabr..
var. ventralis Fabr., from the cocoons of the wasp (Tiphia) which
preys upon the grubs of Lachfio sterna. Melander and Brues ('03,
p. 26) found another member of the same family of beetles, Myo-
dites fasciatiis Say, on wing over nests of Halictus. Pierce ('04)
has made a valuable study of the ecology of Myodites solidaginis,
giving particular attention to its host, a bee (Bpinomia triangulifera
Vachal). Pierce (I.e., p. 185) states that the tiger-beetle Cicindela
puncUilata Fabr. is an active enemy of Bpinomia and Myodites. I
have found this a very abundant beetle in open sunny places on bare
ground, as, for example, along a footpath through a timothy meadow
at Bloomington, 111. Such situations are the favorite haunts of many
burrowing Hymenoptera.

Rhipiphorus linihatus Fabr.

A single specimen was taken on the flower of the rattlesnake-
master, Brynginm ynccifolium, on the Loxa prairie (Sta. II, a) Aug.
27 (No. 178). This species is yellow, with black elytra, and a large
black spot on the dorsum of the prothorax. Blatchley ('10, p. 1367)
reports it from various composites. Robertson (Trans. St. Louis
Acad. Sci., Vol. 6, pp, 106, 107. 1892) reports this beetle from Car-
linville. 111., on the flowers of several species of Pycnanthemum, and
(idem. Vol. 5, p. 571) he also records it from milkweeds (Asclepias).

Rhynchitid^

Rhynchites crneus Boh.

This snout-beetle was taken on the prairie west of Loxa from
flowers of the rosin-weed, Silphiuiii iutcgrifolium (Sta. II), Aug.
13 (No. 48). It has been taken from other flowers (Pierce, '07, p.

250-

Galandrid^

Sphenophorus venatus Say {placidus Say). (PI. XLV, fig. 4.)

This "bill-bug" was taken from the colony of tall blue-stem An-
dropogon and foxtail, Panicum (Sta. l,g), Aug. 12 (No. 39).

182

Forbes ('03 — 22d Rep. State Ent. 111. — p. 8) gives a summary of what
is known of this species. It is a corn pest, has been found widely
dispersed in Illinois, and hibernates as an adult beetle. A tachinid
fly has been bred from the larva of .S'. robiistiis Horn. (Coquillett,
'97, p. 18.)

CURCULIONID.i:

Ccntrinus peniccllus Hbst.

This snout-beetle was taken on the flowers of goldenrod, Soli-
dago (near Sta. I, a), Aug. 12 (No. 26); another specimen was
taken from Sullivant's milkweed, Asclepias siillivantii (Sta. I), Aug.
12 (No. 41). Forbes and Hart ('00, p. 493) state that it has been
taken in the "latter part of July and August." It injures beet leaves,
but its early life history is not known.

Ccntrinus scitfcUiini-album Say.

This beetle was taken at Station I, July 3, 191 1. by T. L. Hank-
inson (No. 7665). It has been taken from a number of flowers in
which it fed upon pollen (Pierce, '07, p. 284). The larva of Cen-
trimis picumniis Hbst. has been found injuring Setaria (Webster, in
Insect Life, \^ol. I, p. 374. 1889).

LEriDOPTERA
Papiugnid^e

Papilio polyxcncs Fabr. Celery Butterfly.

This common butterfly was taken on wing along the railway
track near the swamp milkweed (Asclepias incaruata) colony (Sta.
I, d) Aug. 9 (No. 15), and from a web of the common garden
spider Argiopc aurantia, among these milkweeds (No. 45). Chitten-
den (Bull. 82, Bur. Ent. U. S. Dept. Agr., pp. 20-24. iQOQ) gives
a brief account of this common species which feeds upon umbellifers.
It was very abundant on parsley in the J. I. Bates garden (near
Sta. R', a)'Aug. 26 (No. 174).

PlERID^

Pontia rap(c Linn. Cabbage Butterfly. (PI. XLVI, fig. i.)

A mutilated specimen of this butterfly, which had been captured
bv a robber-flv, was secured bv E. N. Transeau (Sta. HI, h, Aug. 15 ;
No. 61).

Bnrynuts philodicc Godart.

This butterfly was taken on the flowers of Pycnanthcnuim pilo-
sitni in a cleared area bordering the Bates woods (near Sta. IV, a)

1S3

Aug. 23 (No. 146) ; and on flowers of the swamp milkweed, A. in-
carnata (Sta. I, d), Aug. 9 (No. 12).

Nymphalid^

Argynnis idalia Drury. Idalia Butterfly.

This species was taken from the flowers of the swamp milkweed,
A. incarnata (Sta. l,d), Aug. 12 (No. 37).

Anosia plcxippus L. Milkweed Butterfly. (PI. XLVI, fig. 3.)

This common butterfly was abundant upon the prairie at Sta-
tion I. It was observed copulating on willows at Sta. I, d, Aug. 9.
and when on wing was able to carry its mate, whose wings were
folded. It was observed on flowers of the thistle Cirsiiini discolor
at Station I (No. 155).

Lyc/t:nid.e

CJirysoplianiis tlwe Boisd. & Lee. Thoe Butterfly.

This butterfly was taken on flowers of the rattlesnake-master,
Bryngiuni yuccifoUiim, on the Loxa prairie (Sta. II) Aug. 13

(No. 55). '

The caterpillar feeds upon smartweeds (Polygonum) and dock
(Riimcx) , and also upon prickly ash, Zanthoxylum.

Sphingid.^

Heniaris diffinis Boisd. Honeysuckle Sphinx.

This hawk-moth was taken upon flowers of the swamp milkweed.
A. incarnata (Sta. I, rf), Aug. 12 (No. 32), and by T. L. Hankin-
son July 3, 191 1, at Station I (No. 7655). This moth flies during
bright daylight. The caterpillar lives on bush honeysuckle, snow-
berry, and feverwort.

Arctiid.e

Amrnalo eglenensis Clem, or tencra Hiibn.

This caterpillar was taken on dogbane, Apocyniim medium, on
the Loxa prairie (Sta. II) Aug. 13 (No. 53).

Bglcncnsis is reported to feed upon Asclcpias tuhcrosa and
Apocynum.

NOCTUID/T]

FJiodophora gaurcu Sm. and Abb.

This interesting larva was not taken at Charleston, but on the
prairie near Vera^ Fayette county. 111., on Gaura biennis Sept. i
(No. 186). This specimen was determined by W. T. M. Forbes. It

184

is of interest that this larva, which is recorded from the "Southern
and Southwestern States" and Colorado, was found on the prairie
of Illinois. It is another example illustrating the southwestern and
western affinities and origin of many elements in the prairie fauna.
Mr. C. A. Hart informs me that he took the moth at a light Sept. lo
and 17, 1909, at Urbana, and that it was taken at Pekin, 111., in August.

Spragiieia leo Guen.

This little moth was taken once on the flowers of Solidago (near'
Sta. I, a) Aug. II (No. 20) ; again, in a similar situation, Aug. 12
(No. 26) ; and a third time in the cleared area near the Bates woods
on the flowers of Pycnanthcuiiun pilosmn (Sta. IV, a) Aug. 23
(No. 146, two specimens).

Gelechhd^

Gnorimoschema gallccsolidaginis Kilty. (Caterpillar Gall) (PI. XLVI,

fig- 4-)

This common gall was taken by T. L. Hankinson on Solidago at
Sta. I, Aug. 8, 1910 (No. 7462).

Cf. Riley (First Rep. Ins. Mo., pp. 173-175. 1869) and Busck
(Proc. U. S. Nat. Mus., Vol. 25, pp. 824-825. 1903).

DiPTERA
Cecidomyiid^

Cecidomyia solidaginis Loew. (Goldenrod Bunch Gall.) (PL
XLVI, fig. 5.)

This gall was taken on Solidago Aug. 12 at Sta. I (No. 42),
and by T. L. Hankinson at Sta. I, on Aug. 8, 1910 (No. 7462).
This gall forms a rosette or terminal bunch of leaves on Solidago.

Cecidomyia sp.

A willow cone-gall was found Sept. 13 by T. L. Hankinson on
willows at Sta. I. (Cf. Heindel, '05.)

CULICID^

Psorophora ciliata Fabr. Giant Mosquito or Gallinipper.

This is our largest species of mosquito. It was taken among the
swamp milkweeds, Asclepias incarnata (Sta. l,d), Aug. 10 (No.
73); and in the prairie grass colony (Sta. I, ^) Aug. 12 (No. 44).
Both of these places were near moist or wet areas. Individuals were
not abundant, although the species is particularly adapted to living
where the moisture is variable. jMorgan and Dupree (Bull. 40, Div.

185

Ent., U. S. Dept. Agr., p. 91. 1903) have concluded that all the eggs
do not hatch with the first rain after their deposition, but that hatch-
ing is completed with the alternation of wet and dry weather.

Mycetophilid^

Bugnoriste occidentalis Coq.

A single specimen of this small fly was taken on the flowers of
Solidago (Sta. I) Aug. 12 (No. 26). The specimen was determined
by J. R. Malloch. It had been previously recorded from goldenrod
flowers by Aldrich ('05, p. 148).

Sciara sp.

These small flies were taken from the flowers of the mountain
mint, Pycnanthemum fle.vuostmi (Sta. l,g), Aug. 8 (No. 6).

BOMBYLIIDiE

Bxoprosopa fasciata Macq. Giant Bee-fly.

This was one of the most abundant and characteristic insects of
the prairies and cleared areas, and belongs in the same class as the
red milkweed beetle (Tctraopes) and the milkweed bug. Lygcctis kal-
juii. It was taken from flower masses of the mountain mint Pycnan-
themum flexuositm (Sta. I, g) Aug. 8 (No. 6); on the flowers of
Verbena stricta Vent, (near Sta. I, a) Aug. 11 (No. 23) ; again from
P. flexiiosnm (Sta. I) Aug. 11 (No. 24); and on the flowers of
Liafris scariosa (Sta. 11, a) Aug. 27 (No. 176). Two specimens
had been captured by the flower spider Misnmcna alcatoria Hcntz :
one on flowers of the rosin-weed, Silphiuni integrifolinm (Sta. II),
Aug. 13 (No. 47), the other on flowers of the mountain mint Pycnan-
themum flexuosum (Sta. I) Aug. 12 (No. 31) ; and a third was cap-
tured by the ambush bug, Phymata fasciata Gray, on the flowers of
the mountain mint (Station II) Aug. 13 (No. 57).

This was a very common species on the prairie patches at Bloom-
ington. 111., July 26 to Aug. 23, and in pastures abounding in Verbena
at Kappa, 111, and Havana, III, in August. Graenicher ('10, pp. 94-
95) has listed "several species of flowers from which this fly has been
taken. It is probable that it preys upon some wasps, since a related
species, B. fascipcnnis Say, has been bred from the cocoons of the
white-grub wasp, Tiphia (Forbes, '08, p. 160).

Systa:chns vulgaris Loew.

In the cleared area bordering the Bates woods, on flowers of the
mountain mint Pycnanthemum pilosum (near Sta. IV, a), a specimen

186

of this bee-fly was taken Aug. 23 (No. 146). Graenicher ('10, p. 93)
has Hsted a variety of plants visited by this fly.

The habits of this species appear not to be known, but the larvse
of an alHed species, S. oreas O. S., preys upon the eggs of grasshoppers
(Riley, Second Rep. U. S. Ent. Comm., pp. 262-268. 1880). Shel-
ford ('13c) has found that Spogostylum anale Say is a parasite on the
larva of Cicindela. A related fly, Sparnopolius fuli'us, is parasitic
on the grubs of Laclinostcrna (Forbes, '08, p. 161). Holmes ('13)
has shown the relation of light to the hovering flight of Bomhylins.

Mydaid^

Mydas clavahis Drury. Giant fly.

A single specimen of this giant fly was taken on flowers of the
swamp milkweed, Asdepias incarnata (Station \,d), Ax\g. 9 (No.
12). I have taken this species at Chicago during July, and at Bloom-
ington, 111., on June 29.

Harris (Insects Injurious to Vegetation, p. 607. 1869) describes
briefly the larva and pupa; and Washburn (Tenth Ann. Rep. State
Ent. Minn., PI. II, fig. 15. 1905) gives a colored figure of the species.

The larvae of this family live in decaying wood and prey upon
insects, and the adults are also predaceous (Hubbard '85, p. 175).

Howard (Insect Book, p. 136) states that the larva of Mydas
fnlvipcs Walsh "lives in decaying sycamore trees and is probably
predatory on other insects living in such locations." He also states
that the adults are predaceous.

ASILID^

Deromyia sp.

This robber-flv was taken on the Loxa prairie (Sta. II) Aug. 13
(No. 51).

The larvae of some members of this familv feed upon rhubarb
roots (Harris, Ins. Inj. to Vegetation, p. 605. 1869), and others, as
Brax hastardi, are known to prey upon the eggs of grasshoppers
(Riley, First Rep. U. S. Ent. Comm., pp. 303-304, 317. 1878).
Adults of several species of robber-flies feed upon grasshoppers ;
others kill bees (Riley, Sec. Rep. Ins. Mo., pp. 1 21-124. 1870).

Proinachus vcrtehratiis Sav. \"ertebrated Robber-flv. (PI. XL\^I,
fig. 6.)
This is an abundant fly upon the prairie. A specimen was taken
on the Loxa prairie (Sta. II) Aug. 13 (No. 56) ; and on the prairie
east of Charleston (Sta. Ill, h) Aug. 15 (No. 62). Here a robber-
fl}' was seen with a cabbage butterfly, Pontia rapce (No. 61) ; since the

187

fly escaped, however, the species is not known. Another was found
astride a grass stem (Sta. I, g) with the stink-bug Buschistiis variola-
riiis grasped in its legs Aug. 12 (No. 39). Aug. 12, among the prairie
grasses (Sta. I, (/), a pair of these flies was taken copulating (No. 44).
Walsh (Am. Ent., Vol. I, pp. 140-141. 1869) states that Asihts preys
upon Polistes and Bonibiis, which it grasps by the head-end, to keep
out of the reach of the sting, from the bodies of which it sucks the
juices. It handles a harmless grasshopper very differently.

I have observed a large species of robber-fly at Havana, 111., which
hung suspended from grass while devouring its prey; and Aldrich
(Proc. Ent. Soc. Wash., Vol. 2, p. 147. 1893) observed a robber-fly
suspended by its fore feet, apparently asleep, holding a large beetle.
Cook (Bee-keepers' Guide, ninth ed., pp. 317-321. 1883) has seen a
species of robber-fly capture a tiger-beetle, Cicindela; many of these
flies furthermore prey upon the honey-bee. The introduction of this
bee into the prairie associciation must have had considerable influence
upon flower-frequenting insects, and especially upon the predaceous
kinds.

The capture of the cabbage butterfly by an asilid is another obser-
vation which Cook has recorded for Proctacanthns viilberti Macq.
(Asihts niissouricusis Riley). He says (1. c. p. 318) : "It has been ob-
served to kill cabbage butterflies by scores." Wallis (Can. Ent., Vol.
45' P- 135- 1913) observed this fly capturing Cicindela. Punnett
(Spolia Zeylanica, Vol. 7, pp. 13-15. 19 10) has recently shown that in
Ceylon robber-flies are important enemies of large butterflies. Procta-
canthus milbcrti has been observed to prey upon locusts (Riley, First
U. S. Ent. Comm.. p. 317. 1878). For an elaborate account of the
food and feeding habits of this family see Poulton, ('07).

As very little is known of the breeding habits of the American
species, the observations of Hubbard on the oviposition of Mallophora
orcina Wied. (Second Rep. U. S. Ent. Comm., p. 262. 1880) are of
interest. He saw a female of this Florida species bury its abdomen in
the ground, where it deposited five or six eggs at a depth of half to
two thirds of an inch. The eggs hatched in a week. Bra.v lateralis
Maccj. has been recorded as predaceous upon May-beetle larvse (Titus,
in Bull. 54. Bur. Ent., U. S. Dept. Agr., pp. 15-16). Titus gives fig-
ures of the larva and pupa.

D0LICHOPODID.E

Psilopus sipho Say. Metallic Milkweed Fly. (PI. XLVI, fig. 2.)

This pretty metallic-colored fly, observed by almost every field
student or collector, is one of our commonest insects. It runs rapidly

188

over the upper surface of the leaves of the common milkweed, Ascle-
pias syrioca^ and is so nimble that it requires a little care to catch it. A
large number of the flies were secured from the common milkweed
along the railway track (Sta. I) Aug. 12 (No. 27), and also on the
milkweeds infested with the plant-louse Aphis asclepiadis Fitch. Al-
though some species of Dolichopodidcc are said to be predaceous, I
have never seen this species attack any insect.

The peculiar breeding habits of some of the members of this fam-
ilv have been described by Aldrich (Am. Nat., V^ol. 28, p. 35-37.

1894).

Syephid^

Syrphus americanus Wied. (PI. XLVII, figs. 3, 4, and 5.)

This fly was taken along the railway track (Sta. I) Aug. 9 (No.
11). Its hum when on wing sounded much like that of the small yel-
low-jacket, Vcspa. Metcalf ('13, p. 55) found it feeding on aphids
infesting Phraguiites.

Certain syrphid larvae prey upon plant-lice, and the adults are
abundant on flowers, especially unbellifers, feeding on their nectar. For
good accounts of both larvae and adults consult Williston (Bull. 31,
U. S. Nat. Mus., pp. 269-272. 1886) and Metcalf ('13).

Mesogramma polituin Say. Corn Syrphid. (PI. XLVII, figs, i and 2.)
This syrphid was found in great numbers on the Loxa prairie (Sta.
II) Aug. 27 (No. 177).

The larvae are pollen feeders, as has been shown by an examination
of the contents of the alimentary canal (cf. Riley and Howard, Insect
Life, Vol. I, p. 6). Also consult Forbes ('05, p. 162), who figures the
species. Upon the original prairie the species probably fed on the pol-
len of various grasses or other plants.

Allograpta ohliqua Say. (PI. XLVII, figs. 6 and 7.)

This insect was taken on the Loxa prairie (Sta. II) in company
with great numbers oi Mesogramma polituin Say, Aug. 2^ (No. 177).
For figures of the larva, pupa, and adult see Washburn (Tenth Ann.
Rep. State Ent. IMinn., p. loi. 1905) and Metcalf ('13, p. 58). It
feeds upon aphids.

COXOPID.^

Physocephala sagittaria Say.

This insect was taken on the flowers of goldenrod, SoUdago (Sta.
I), Aug. 12 (No. 26). Also taken on a small-flowered aster at Ur-
bana. 111., Oct. 8. The larvae of this family are parasitic on other
insects. There is a figure of an allied species on Plate XLVIII, fig-
ure I.

189

Tachinid.^

Cistogaster iminaciilata Macq.

A single specimen of this fly was taken on the flower of rattlesnake-
master, Br yngium yucci folium (Sta. II) Aug. 13 (No. 55).

The larva is parasitic on lepidopterous larvae (Townsend, Psyche,
Vol. 6, p. 466. 193) ; and has been bred from the army-worm, Leiicania
unipuncta Haw. Two undetermined species of tachinids were taken
by T. L. Hankinson (Sta. I) July 3, 191 1 (No. 7665).

Trichopoda ruficauda V. d. W.

A single specimen of this fly was taken along the railway track
(Sta. I) Aug. 12 (No. 38).

An allied species, T. pemiipes Fabr., has been bred from the
squash-bug (Cook, Rep. Mich. State Board Agr., pp. 1 51-152. 1889),
and another, phunipes Fabr., has been bred from a grasshopper, Dis-
sostcira vcmista Stal (Coquillett, '97, p. 21).

SCIOMYZID^

Tctanoccra plumosa Loew. (PI. XLVIII, fig. 2.)

Taken in a colony of Spartina (Sta. I, a) Aug. 28 (No. 179).
This species is figured by Washburn (Tenth Ann. Rep. State Ent.
Minn., p. 121. 1905). The larvae of this family are aquatic. Need-
ham (Bull. 47, N. Y. State Mus., pp. 580-581, 592, PI. 14. 1901)
describes and figures T. pictipes Loew. (Cf. Shelford, '13a.)

Trypetid^e
Buaresta cequalis Loew.

This insect was taken in sweepings among a colony of the cone-
flower, Lepachys pinnata (Sta. I, e), Aug. 12 (No. 40). Marlatt (Ent.
News, Vol. I, p. 168) records the rearing of this fly from the seed-pod
of the cocklebur (Xanthium).

EmPIDID/E

Bnipis claiisa Coq.

A specimen of this fly was taken from a pair of copulating ambush
bugs, Phymata fasciata, on the flowers of Solidago (Sta. I) Aug. 12
(No. 43), and great numbers, so many that they darkened the flowers
on which they rested, were seen upon Asclepias syriaca (Sta. I) Aug.
12 (No. 27). The specimen was determined by J. R. Malloch.

McAtee (Ent. News, Vol. 20, pp. 359-361. 1909) gives an account
of the habits of Bmpididcc, and Schwarz (Proc. Ent. Soc. Wash., \'ol.
20, pp. 146-147. 1893) states that one kind captures small flies, and

190

suspended by its foreleg, eats its prey. This position when eating is a
curious habit, independently acquired by several predaceous insects, as
Biffacus. Vcspa, and certain Asilidcc.

Mr. Malloch has called my attention to British observations made
upon the peculiar habits of these flies. Thus Howlett ('07) has shown
that the male supplies the female with an insect for food during copu-
lation. These observations have been confirmed by Hamm ('08).
Poulton ('07) discusses the food habits of these flies in much detail.

Hymexoptera

Cynipid^

RJwditcs ncbitlosiis Bassett. (Rose Gall.)

This gall was taken on a wild rose, Rosa, in the mixed forest and
prairie colony east of Charleston (Sta. Ill, h) Aug. 15 (No. 60).

Braconid;e

An undetermined species was taken from the flowers of Pycnau-
thenuini pilosuui in the cleared area with sprout growth bordering the
Bates woods (near Sta. IV, a) Aug. 23 (No. 146).

FORMICID.^

Myriuica rubra Linn., subsp. scabrinodis Nyl., var. sabitlcti Meinert.

This ant was found upon the prairie on flowers of the common
milkweed, Asclepias syriaca (Sta. I), Aug. 12 (No. 27). It was asso-
ciated with Formica fusca subscricca Say and Formica pallid c-fiilva
schaufussi inccrta Emery.

Wheeler ('05. pp. 374, 384) regards this as one of the heath ants,
which "inhabit rather poor, sandy or gravelly soil exposed to the sun

and covered with a sparse growth of weeds or grasses It

nests in sandy or gravelly sunny places such as open pastures, road-
sides, etc." These requirements are admirably met by the conditions
along the gravelly and sandy road-bed of the railway where the milk-
weeds flourish.

Formica fusca Linn., var. subscricca Say.

This ant was found on flowers of the goldenrod, Solidago (near
Sta. I, c), Aug. II (No. 20); on leaves of the common milkweed
(Asclepias syriaca) infested with the plant-louse Aphis asclcpiadis
Fitch (Sta. I) Aug. 12 (No. 30) and again Aug. 24 (No. 154) ; and
in the upland Bates woods (Sta. IV, a) Aug. 26 (No. 163).

According to Wheeler ('loa, p. 458) this ant is enslaved by For-
mica sanguinea Latr. and the following subspecies : aserva Forel, ruhi-

191

cimda Emery, suhmida Emery, snhintegra Emery, and puherula
Emery. Wheeler has seen Formica sauguinca "plunder a suhscricea
nest nearly every day for a week or a fortnight." In raiding a nest
the ants carry off the larvae and pupse to their own nests, to serve as
slaves when matured.

Wheeler (1. c., p. 374) states that suhscricea mav live in a great
variety of situations — an unusual trait, but indicated in our collect-
ing by its presence in both forest and prairie.

Formica pallidc-fiik'a Latr., subsp. scJuuifussi Mayr, var. inccrta
Emery.

This common reddish ant was taken on the prairie from flowers
of the common milkweed, Asclcpias syriaca (Sta. I), Aug. 12 (No.
27) ; and on the Loxa prairie from flowers of the mountain mint
Pycnanthemiim pilosum or P. flcxuosum (Sta. II) Aug. 13 (No. 52).

This ant was associated on the milkweeds with Myrmica rubra
Linn., subsp. scabrinodis Nyl., var. sabuleti Meinert, and Formica
fuse a subs eric ea Emery.

Wheeler ('05, pp. 373, 374) lists this species as frequenting glades,
"open sunny woods, clearings, or borders of woods," and further adds
that the glade and field faunas are not separated by a sharp line, for
"Formica schaufussi, for example, seems to occur indifferently in
either station." That open patches in woods or glades often contain
ants which also frequent open places, is thus in harmony with a gen-
eral rule for this association, not only in the case of animals but also
of plants, so that it applies to the entire biota of such situations.

Wheeler ('loa. p. 393) lists a small wingless cricket, Myrmccophila
pergandei, as living with Formica pallide-fulra. These lick the sur-
faces of the ants, and seem to feed upon the products of the dry bath.

Wheeler says ('05, p. 400) that the food of schaufussi appears to
be "largely of the excrement of Aphides and the carcasses of insects."

Wheeler ( '04, pp. 347-348 ) states that the nests are usually found
under a stone, and that Formica difficilis Emery var. consocians
Wheeler is a temporary parasite upon incerta, but "only during the
incipient stages of colony formation" (p. 358). This is a temporary
parasitism of one colony upon another, during which the parasite mul-
tiplies and becomes strong enough, at the expense of its host, to estab-
lish a new independent colony. This is what Wheeler calls a "tem-
porary social parasite, a true cuckoo ant, which sponges on another
species only so long as necessary in order to gain a successful start
in life." Schwarz ('90b, p. 247) records several species of beetles as
living with schaufussi. Not only does this species suffer from tempo-
rary ant-parasites, but it mav be enslaved by some form of Amazon-

192

ant, as Polyergiis lucidus (Wheeler, 'loa, p. 482; Tanquary, '11,
p. 302).

MUTILLID/i:

Sph(rropJithalma sp. Velvet Ant.

This wasp was taken on the bare footpath at the margin of the
Bates upland woods (near Sta. IV, a) Aug. 23 (No. 151). It is prob-
ably parasitic in the nests of bees.

MyZINID/E

My:::ine scxcmcta Fabr.

This black-and-yellow-banded wasp was very abundant on flowers.
It was taken Aug. 8 (Sta. I, g) on flowers of Asclcpias incarnata (No.
i) and irom Pycnantheuiiun flexitosuui (No. 6) ; from the flowers of
goldenrod, Solidago (near Sta. I, a), Aug. 11 and 12 (Nos. 20 and
26).; by T. L. Hankinson (Sta. I) July 3, 191 1 (No. 7665) ; on flow-
ers of Pycnanthcimim (Sta. II) Aug. 13 (No. 52) ; and from the
flowers of Brynginm yuccifolmm (Sta. II) Aug. 13 (No. 55); and
from the cleared area bordering Bates woods (Sta. IV, a) Aug. 23
(No. 146).

Packard (Guide to the Study of Insects, 8th ed. p. 177. 1883)
states that this wasp flies "low over hot sandy places." This is one
of the species found by Banks* (Jour. N. Y. Ent. Soc, Vol. 10, p. 210,
1902) to sleep in grass, and by Brues (idem, Vol. 11, p. 229. 1903)
resting during the day and night upon plants.

ScoLnD.^
Scolia bicincta Fabr.

This hirsute black wasp, with two yellow transverse dorsal bands
on the abdomen, is represented in our series by four specimens. Three
of these were taken on flowers of Pycnantheuuim pilosinii from the
clearing bordering the upland portion of the Bates woods (near Sta.
IV, a) Aug. 23 (No. 146) ; the others, from an open space in the up-
land forest (Sta. IV, a) Aug. 26 (No. 163). I have also taken this
species at Bloomington, 111., Aug. 23, 1892, and Aug. 25, 1896.

Packard (Guide to the Study of Insects, 8th ed., p. 176. 1883)
states that in Europe Scolia bicincta burrows sixteen inches in sand
banks, and that it probably stores its nest with grasshoppers. Riley
(First Rep. U. S. Ent. Comm., p. 319. 1878) states that species of
Scolia are known to have the habit of stinging grasshoppers and
digging nests, provisioning these with grasshoppers, on which they
lay eggs as does the wasp CJilorion cyancnm Dahlb. (C. ccrrulciim
Drury). (Cf. with Kohl, Ann. des K. K. naturhist. Hof museums, Bd.

193

5, pp. 1 21-122. 1890.) Forbes ('08, pp. 157-160) has found that
Tiphia is parasitic upon the grub of the May-beetles (Lachnosterna).
The wasp crawls into the ground in search of the larva, stings it, and
lays its eggs upon it. It is not unlikely that Scolia has similar habits.
The sleeping habits of hiQincta and some other Hymenoptera have
been described by Banks (Journ. N. Y. Ent. Soc, vol. 10, pp. 127-130.
1902), Brues (idem. Vol. 11, pp. 228-230. 1903), and Bradley (Ann.
Ent. Soc. Amer., Vol. i, pp. 127-130).

Scolia tricincta Fabr.

One specimen was taken — in the clearing bordering the Bates
woods on flowers of Pycnanthemum pilosimi (Sta. IV, a) Aug. 23
(No. 146).

EUMENID^

Odyneriis vagus Sauss. Potter Mud-wasp.

An oval mud nest, about 18 mm. long and ip mm. in diameter, was
found on a stem of dogbane, Apocynnm mcdiiini (Sta. I), Aug. 12
(No. 46). The nest was placed in a vial; and later, a single wasp of
the above species came from an opening which was made at the point
where the mud cell was formerly attached to the plant.

This is a predatory wasp, which stores its nest with caterpillars
(Peckhams, in "Wasps, Social and Solitary," pp. 94-95. 1905).

Vespid^

Polistcs — probably variatus Cress.

A small nest was observed in a grassy area near Station l,e, but
was not secured. The adults feed the young with caterpillars and nec-
tar. See Enteman (Pop. Sci. Monthly, Vol. 61, pp. 339-351. 1902)
for an excellent account of the habits and life history of these social
wasps.

That these wasps will build their nests in an open area is of inter-
est, because the nests are so commonly found under eaves and on the
under side of roofs — situations which were originally lacking on the
prairie.

As Walsh stated, the social wasps do not store up food, because
"they feed their larvae personally from day to day."

PSAMMOCHARID^

Priocncnwidcs iinifasciatus Say {Priocncuns). Spider Wasp.

This wasp was taken in the cleared area bordering the Bates woods,
on flowers of Pycnanthemum pilosmn (near Sta. IV, a) Aug. 23 (No.
146).

194

A specimen was taken Aug. 21 at Bloomington, 111. The yellow
wings and antennae, and yellow subapical wing spot on the smoky
wings make this a conspicuous species. The family name Pompilida;
was formerly used for these wasps.

Sphecid/e
Ammophila nigricans Dahlb.

A single specimen was taken from the flowers of Pycnanthemum
flexnosuni (Sta. I) Aug. 11 (No. 24).

This is a very common Illinois species. I have taken it at Bloom-
ington from June 22 to September 9, at Havana during August, and
at Chicago, August 19 and 28. A specimen taken August 2 at Bloom-
ington, 111., was digging in the ground when captured.

Chlorion ichnenmoncuin Linn. (Sphex ichneumonea Fabr.). Rusty
Digger-wasp. (PI. L, fig. i.)

This insect, abundant on flowers of the swamp milkweed, Ascle-
pias incarnata, August 8, was taken on them at Sta. I, g, Aug. 8 (No.
i) and at Sta. l,d, Aug. 9 (No. 12); and on the mountain mint
PycnantJicnimn fle.vuositni (Sta. I) Aug. 8 (No. 6). It was also
taken by T. L. Hankinson July 3, 191 1 (No. 7665).

This is a very common insect on flowers in central Illinois. I have
found it abundant at Chicago during August; at Bloomington, 111.,
from June 24 to Oct. i ; at IMayview on Sept. 26 in a colony of prairie
vegetation.

Packard (Guide to the Study of Insects, pp. 167-168. 1870) tells
how these wasps dig holes four to six inches deep in gravel walks, and
after capturing long-horned grasshoppers, Orchclirnum vidgare or
O. gracile, and stinging and paralyzing them, proceed to bury them.
The tgg is deposited on the locust before the soil is scraped in. (Cf.
Walsh, Am. Ent., Vol. i, p. 126. 1869). For an excellent account of
the habits of this species constdt the Peckhams, "Instincts and Habits
of the Solitary Wasps" (1898). See Fernald ('06) for the recent
synonymy.

Chlorion pcuusylvanicmn Linn. Pennsylvania Digger-wasp.

This wasp was taken on the flowers of Bryngium yuccifolium
(Sta. II) Aug. 13 (No. 55). On Aug. 8, 1893, I captured a specimen
at Chicago. (Cf. Fernald, '06, p. 405.)

Chlorion horrisi H. T. Fernald {Isodontia philadclpJiica Auct.). Har-
ris's Digger-wasp.

One specimen of this wasp was taken on flowers of the mountain
mint Pycnantheniuni flexuosuni (Sta. I) Aug. 11 (No. 24).

195

I have also taken this species at Bloomington, 111., Aug. 21 and
Sept. 7 and 11.

This wasp has been known in North Carolina to build its nests
in the funnel-like bases of the leaves of the pitcher-plant Sarracenia
flaz'a (Jones, Ent. News, Vol. 15, p. 17 and PI. III. 1904), and
provisions its nest with QHcanthus. Ashmead (Insect Life, Vol. 7, p.
241. 1894) states that it "preys upon the cricket OEcantJnis fasciatus
Fitch."

Chloriou atratuvi Lepeletier {Priononyx atrata St. Farg. and Sphex
hninneipes Cress.). Black Digger-wasp.

This species was taken from the flowers of Bryngium yticcifolium
(Sta. II) Aug. 13 (No. 55). I have also taken it at Havana, 111.,
during August, and at Bloomington, 111., on September 3, 5, and 12.

In a colony of prairie vegetation near St. Joseph, 111., when out
with a class on an ecological excursion, Sept. 26, 191 1, I made some
interesting observations on this wasp. Along the Big Four railway
track between Mayview and St. Joseph, 111., fresh sand and gravel
had very recently been placed upon the road-bed. In this fresh sand
we observed a large black wasp, CJilorion atratuin, digging. The wasp
was about two thirds of her length in the hole when first observed,
and when captured later she was more than her length in the hole.
She would scratch out the sand so that it fell near the mouth of the
hole, and then come out and, standing over the pile, she would scrape
it far out of the way by rapid movements of her legs. Every now and
then she would come out of the hole with gravel in her jaws; several
of such samples were preserved. As the sand was loose the gravel
was of course not firmly imbedded. Of the small stones carried out
five of the largest range from one fourth to one half an inch in diam-
eter. In bulk each of these is larger than the thorax of the wasp.
Four small flies were seen to hover about the hole ; some which alighted
on small stones near by were captured by a member of the party and
proved to be small tachinids (No. 309, C.C.A.), which Mr. J. R.
Malloch determined to be Metopia Icucocepliala Rossi. (Cf. Cociuillett,
'97, p. 127.) Mr. Malloch also called my attention to recorded obser-
vations on other tachinid flies which inhabit the burrows of Hymenop-
tera in Great Britain, and are parasitic in habit (Malloch, 09). Hamm
('09b) has described how one of these flies, Setulia grisea Mg., follows
the females of Cerceris as she provisions her burrow with weevils.
Thev were observed to enter and to come out of the burrow. Me-
lander and Brues ('03, pp. 9, 20) state that M. leiicoccphala infests
the bee Halictus Ijy choosing "the moment when the incoming bee
pauses at her threshold quickly and quietly to oviposit on her pollen
mass and thus infect her ofi^spring." This fly has been reported to be

196

viviparous. Cf. Aldrich ('05, p. 476). The Peckhams ('98, p. 37) ob-
served a small fly at the burrows of CJilorion ichnciiinoncum. Brues
(Jour. N. Y. Ent. Soc, Vol. 11, p. 228. 1903) has observed this species
near Chicago sleeping in sweet clover. ( See also Bradley, in Ann. Ent.
Soc. Amer., Vol. i. pp. 127-130. 1908.)

For the habits of this species see the Peckhams, "Instincts and
Habits of the Solitary Wasps," pp. 171-173. This species provisions its
nest with the Carolina locust, Dissosteira Carolina. Coquillett (Insect
Life. Vol. 7, p. 228, 1894) says that this species shows a preference for
Melanopliis femur-rubrum DeG. in provisioning its nest.

Stizid^

Stilus hrevipennis Walsh. Digger-wasp.

A single specimen of this large wasp was taken on flowers of
PycuantJicmum flcxuosuin (Sta. I) Aug. 12 (No. 35) ; another was
taken by T. L. Hankinson (Sta. I) July 3. 191 1 (No. 7665).

Walsh (Am. Ent., Vol. i, p. 162. 1869) found this species on flow-
ers of the wild parsnip at Rock Island, 111. An allied wasp, Sphcciiis
speciosiis Drury, preys upon the cicada or dog-day harvest-fly, Cicada
pniiiiosa, on which it lays its egg and upon which its larva feeds. Con-
sult Riley (Insect Life, Vol. 4, pp. 248-252. 1891) for an excellent
account of this wasp. As Walsh infers, hrevipennis and speciosns prob-
ably have similar habits. A tachinid fly, Scnotainia trilineaia V. d.
W., has been bred from the nest of speciosns (Coquillett, '97, p. 20).

HaLICTID/E

Halictus obsciirus Rob.

A single specimen was taken — on the Loxa prairie from the flow-
ers of Bryngium ynccifolium (Sta. II) Aug. 13 (No. 55).

Halictus fasciatns Nyl.

This bee was taken Aug. 13 on the Loxa prairie (Sta. II) from
the flowers of SilpJmini integrifoliuni (No. 48) and from those of
Pycnanthemum flexiiosmn or P. pilosiini (No. 52) ; and on goldenrod,
Soli dago (Sta. I), Aug. 12 (No. 26).

Halictus virescens Fabr.

A single male of this small bee, with metallic green head and
thorax, was taken on flowers of verbena (Sta. I) Aug. 11 (No. 23).

NOMADID^

Bpeolus concolor Rob.

This species was taken on the heads of the cone-flower, Lepachys
pinnata (Sta. I, e), Aug. 8 (No. 8) ; very abundantly from flowers of

197

the mountain rnmt Pycuanthernuni' flcxuosmn (Sta. I, ^) Aug. 8 (No.
6) ; from flowers of Silphium intcgrifolimn (Sta. II) Aug. 13 (No.
48) ; and from flowers of Pycnanthcmnm flcxuosimi or pilosum (Sta.
II) Aug. 13 (No. 52).

It is said to be "parasitic on the species of Colletes," but Robertson
('99' PP- 35' 37) does not accept this view, and Ashmead (Psyche, Vol.
7, pp. 41-42. 1894) states that Bpeolus donatus Smith makes a nest
in the ground and provisions it with a honey-paste. He describes the
burrows, tgg, and larva.

Robertson has published keys to the Carlinville (111.) species of
Bpeolus (Can. Ent, Vol. 35, pp. 284-288. 1903).

EUCERID.E

Melissodcs aurigenia Cress.

A single female of this species was taken from flowers of ver-
bena ( near Sta. \,h) Aug. 11 (No. 23).

The homing behavior of this genus of bees has been studied by
Turner (Biol. Bull, Vol. 15, 247-258. 1908). He concludes that
memory is utilized.

Melissodes bimaculata St. Farg.

This bee was taken from the heads of the cone-flower, Lepachys
pinnata (Sta. I, e), Aug. 8 (No. 8) ; abundantly from flowers of the
mountain mint PycnantJicimiin fiexuosuin (Sta. I, ^) Aug. 8
(No. 6) ; on the Loxa prairie on flowers of the rosin-weed, Silphium
integrifolinm (Sta. II), Aug. 13 (No. 48; and on the cleared margin
of the Bates woods on flowers of the mountain mint, P. pilosum (Sta.
IV, a), Aug. 22 (No. 146).

Some observations on the "sleeping habits" of this bee and of other
Hymenoptcra have been made by Banks ( Journ. N. Y. Ent. Soc, Vol.
10, pp. 209-214. 1902). Graenicher ('05, p. 164) has recorded ob-
servations on the habits of M. trinodis Rob. and also on its bee para-
site Triepeolus. Ashmead (Psyche, Vol. 7, p. 25. 1894) found the
burrows of bimactdata eight inches deep in the soil.

Melissodes desponsa Smith.

This bee was taken on the cleared margin of the Bates woods on
flowers of the mountain mint Pycnanthemum pilosum (near Sta.
IV, a) Aug. 22 (No. 146).

Melissodcs ohliqua Say.

This bee was found abundant upon flowers of the cone-flower, Le-
pachys pinnata (Sta. I, e), Aug. 8 (No. 8) ; it was taken from flowers
of the white mint, Pycnanthemum flexuosum (Sta. I), Aug. 11 (No.

198

24) ; and a female was taken from the flowers of Silphiurn integri-
folium (Sta. I) Aug. 13 (No. 48). According to Robertson (Trans.
Acad. Sci. St. Louis, Vol. 6, p. 468. 1894) this bee is the most abun-
dant bee visitor to the cone-flower, and it also shows a marked prefer-
ence for this plant.

Megachilid^

Megachile mendica Cress. Leaf-cutting Bee.

A single specimen was taken on flowers of the swamp milkweed,
Asclcpias incarnata (Sta. l,g), Aug. 8 (No. i).

The habits of our leaf-cutting bees have received little attention,
although the circular areas which they cut from rose leaves are a fa-
miliar sight. Putnam (Proc. Essex Inst., Vol. 4, pp. 105-107. 1864)
describes the nests of Megachile centuncularis Linn., and Packard, one
of its hymenopterous parasites (idem, pp. 133-137).

Megachile hrevis Say. Short Leaf-cutting Bee.

A single female was taken by T. L. Hankinson (Sta. I) July 3,
191 1 (No. 7665). This species is known to use plum leaves for its
nest. Its habits have been briefly described by Reed (Sec. Rep. Ent.
Soc. Ont., pp. 24-26. 1872; Can. Ent., Vol. 3, pp. 210-21 1. 1871).
The nest is formed of a leaf which is wrapped about the disks cut from
the leaves, and is not in the ground or in cavities in wood as is the case
with many species. Packard (Jour. N. Y. Ent. Soc, Vol. 5, p. 109-
III. 1897) describes and gives figures of the immature stages of what
is possibly M. centuncularis Linn. See also Packard ('73), Ashmead
('92), and Howard ('92a).

Some of the species of this genus are parasitized by bees of the
genera Stelis and Caiioxys as has been shown by Graenicher ('05) ;
some also are parasitized by certain flies (Howard, in Proc. Ent. Soc.
Wash., Vol. 2, p. 248. 1893).

Xylocopid^

Xylocopa z'irginicaDrnry. Carpenter-bee. (PI. XLIX.)

Only four specimens of this bee were taken, and these were found
on flowers of the sw^imp milkweed, Asclcpias incarnata (Sta. I, (/),
Aug. 8 (No. i) and 24, (No. 156).

The carpenter-bee has much the appearance of a large bumblebee.
The female cuts tunnels in wood to make a nest for the young Pack-
ard has described the larva (Journ. N. Y. Ent. Soc, Vol. 5, p. 113.
1897). The same author records observations by Angus on the boring
habits of this species (Our Common Insects, pp. 21-24. ^^73)- He
found the larva of a bee-fly, Anthrax sinuosa Wied., parasitic on the

199

larva of the carpenter-bee. Felt, ('05, PI. 39, and '06, p. 484) has
given figures of the nest and has briefly described it. The burrows are
made in the seasoned lumber of houses, in telegraph poles, and in simi-
lar situations. On the prairie at Charleston, fence posts, telegraph
poles, and railway ties constitute the supply of wood available for nest-
ing purposes. It thus appears probable that this bee was not particu-
larly abundant on the original prairie, far from the forests or cotton-
woods, for such nesting habits imply a supply of wood for the bur-
rows. The larva is said to feed upon pollen, on which the eggs are
placed.

BOMBID^

Bontbus pennsylvanicus DeG. Pennsylvania Bumblebee.

This species was taken on the Loxa prairie from flowers of the
purple prairie clover, Petalostemnm piirpiireum (Sta. II), Aug. 13
(No. 50) ; on flowers of the mountain mint, Pyciiautliomtin pilosum
or P. flcxuosinii (Sta. II) Aug. 13 (No. 52) ; on flowers of the rattle-
snake-master, Bryngiuin yuccifoUuui (Sta. II), Aug. 13 (No. 553);
in an open glade in the lowland forest (Sta. IV, c) Aug. 22 (No.
143) ; on flowers of the thistle Cirsmm discolor (near Sta. I, d) Aug.
24 (No. 155) ; from the flowers of the broad-leaved rosin-weed, Sil-
phiiim terehinthinaceum (Sta. Ill, &), Aug. 26 (No. 175); and on
the prairie west of Loxa on the flowers of the blazing star, Liatris
scariosa (Sta. II), Aug. 27 (No. 176).

Banks (Jour. N. Y. Ent. Soc, Vol. 10, p. 212. 1902) has recorded
this species as sleeping on flowers.

The following papers on the habits and life history of the bumble-
bees will aid in the study of these neglected insects :

Coville, Notes on Bumble-Bees. Proc. Ent. Soc. Wash., Vol. i, pp.
197-202. (1890) — Putnam, Notes on the Habits of some Species of
Bumble Bees. Proc. Essex Inst., Vol. 4, pp. 98-104. (1864) — Packard
The Humble Bees of New England and their Parasites ; with notices
of a new species of Anthophorabia, and a new genus of Proctotrupidse.
Proc. Essex Inst., Vol. 4, pp. 107-140. (1865) — Marlatt, An Inge-
nious Method of Collecting Bombus and Apathus. Proc. Ent. Soc.
Wash., Vol. I, p. 216. (1890) — Howard, The Insect Book, (1904),
pp. 12-16; and Sladen, The Humble-Bee (1912). Marlatt describes
the use of a jug of w^ater in collecting bees from the nest. (This has
long been the common method of destroying these bees used by coun-
try boys and farmers of central Illinois.)

A very important systematic paper, which also contains much on
the life history and habits of the American Botnbidce has recently
been published by Franklin ('13).

200

A tachinid fly, Brachycoina davidsoni Coq. (Coquillett, '97, p.
10) has been bred from a larva of Bomhiis fcrvidus Fabr. The larva
of the syrphid fly Volucella lives as a scavenger in Bombus nests (Cf.
Metcalf, '13, p. 68). The conopid flies Physocephala and Conops are
parasitic on Bombus. A nematode parasite, SpJiccndaria bonibi, in-
fests hibernating queens. It has been found in B. pcnjisylvanicus, fer-
vidus, and consiinUis (Cf. Stiles '95).

Bombus auricomus Rob.

Two males of this species were taken from flowers of the large-
leaved rosin-weed, Silphiiun terebinthmaceum, on the prairie area
east of Charleston (Sta. Ill, b), Aug. 26 (No. 175). This bumble-
bee was also taken bv T. L. Hankinson (Sta. I) Julv 3, 191 1 (No.
7665). (Cf. Franklin, '13, Pt. I, p. 413.)

Bombus impaticns Cress. Impatient Bumblebee.

A sino-le female was taken from the flowers of the broad-leaved
rosin- weed, Silpliiuui terebintJiinaceum, east of Charleston (Sta.
Ill, b), Aug. 26 (No. 175).

Bombus f rat emus Smith.

Two females of this species were taken on flowers of the swamp
milkweed, Asclcpias iucaniata: one of them (No. i) at Station I, g,
Aug. 8; and the other (No. 12) at Station I, d, Aug. 9.

Bombus scparatus Cress.

This species was collected from the swamp milkweed, Asclcpias
incarnata, as follows : Station I, g, Aug. 8 (No. i) ; Station I, d, Aug.
9 (No. 12); Station l,d, Aug. 24 (No. 157) — the latter had been
captured by the flower spider Mismnena alcatoria Hentz ; and one
male from flowers of the horse mint, Monarda (Sta. I), Aug. 11
(No. 22).

Psithyrus variabilis Cress. False Bumblebee.

A single female was taken from the flowers of the horse mint,
Monarda (Sta. I), Aug. 11 (No. 22) ; and a male was taken on the
prairie west of Loxa from flowers of the blazing star, Liatris scar-
iosa (Sta. II), Aug. 27 (No. 176). These bees are parasitic in the
nests of Bombus. For an excellent account of the habits of the Brit-
ish species, Sladen ('12, pp. 59-72) should be consulted.

Apid^

Apis mellifera Linn. Honey-bee.

Workers of this species were extremely abundant on flowers of
the milkweed Asclcpias incarnata (Sta. I, and Sta. l,d,g) Aug. 8

201

(No. i). Milkweed flowers play a double role as food and enemy.
Robertson (Trans. St. Louis Acad. Sci., Vol. 5, p. 573) states that
honey-bees are frequently found hanging dead from the flowers of
the common milkweed, A. syriaca, and Gibson (Harper's Mag., Vol.
95' PP- 519-520. 1897) has found many of them entrapped by this
milkweed. Bees are not the only insects captured by this insect trap,
for Gibson found gnats, crane-flies, bugs, wasps, beetles, and small
butterflies hanging from the flowers. He also found that the dogbane
Apocymim thus captures moths.

n. Forest Invertebrates

MOLLUSCA
Helicid^

Polygyra albolabris Say. (PI. LI, figs. 2 and 3.)

A single adult dead shell (No. 91) of this woodland species was
found in the upland forest (Sta. IV, a). It is our largest species of
snail.

The natural history of our land-snails has received little attention,
but is worthy of careful study. The best account of the life history
and habits of this species is by Simpson ('01).

Polygyra claiisa Say.

A single dead immature shell was taken under a small decayed limb
on the ravine slope (Sta. IV, h) Aug. 26 (No. 164), associated with
many individuals of Pyramidiila pcrspcctiva, and one individual each
of Vitrea indcntata and V. rhoadsi.

Shimek ('01, p. 200) groups this species with those which frequent
"higher, more deeply shaded (often mossy and rocky) banks and
slopes, sometimes in deep woods."

CmciNARnD^

Circinaria concava Say. Predaceous Snail.

A large dead shell (No. 71) and several living specimens were
found in a decayed stump in the upland forest (Sta. IV, a). A young
individual (No. 113), diameter 6 mm., was taken Aug. 20 among the
vegetable debris washed from a ravine and deposited as a low fan in
the lowland forest (Sta. IV, c). With it were associated Vitrea in-
dcntata, and some kind of large snail eggs (No. 114). This is a car-
nivorous species.

202

ZONITID^

Vitrea indentata Say.

One specimen (No. 113) was taken Aug. 20, among a mass of
drifted rotten wood and dead leaves deposited at the mouth of a* ra-
vine in the lowland forest (Sta. IV, c), in company with a young speci-
men of the carnivorous Circinaria concava; and another (No. 140),
on Aug. 22, under leaves at the base of a ravine slope (Sta. IV, h),
in woods so dense that there was very little herbaceous vegetation, but
a thick ground cover of leaves and vegetable mold. The interesting
ant Stigniatonuna pallipes, Myrinica rubra scabrinodis scheiicki, and
the larva of Mcracantlia contracta were found here. Specimens were
also taken Aug. 26 (No. 164) under a small decayed limb on the
ravine slope (Sta. IV, ^) in company with Vitrea rhoadsi, Polygyra
ciansa, and Pyrainidula perspectiva.

Vitrea rhoadsi Pilsbry.

This snail was taken under a small damp decayed limb on a
wooded ravine slope (Sta. IV, h) in company with V. indentata,
Pyrainidula perspectiva, and Polygyra clausa (No. 64). Mr. F. C.
Baker informs me that this species has not previously been recorded
from Illinois.

Zonitoides arborea Say.

This snail was taken on a fungus wjiich was growing on a de-
cayed stump in the upland forest (Sta. IV, a) Aug. 17 (No. 71),
in company with the mollusks Pyramidula perspectiva, Circinaria
concava, and Philouiycus carolinensis, the ant Aphcenogaster fulva,
and the white ant Termes fiavipes. Also taken from a moist rotting
stump, on the slope of the valley (Sta. IV, h), Aug. 17 (No. 84),
in company with the snail P. perspectiz^a, the slug P. carolinensis,
newly established colonies of the ant Caniponotus herculeanus penn-
sylvanicus, and the beetle Passalus cornutus.

This snail appears to be mainly a species of the woodland, where
it occurs under decaying wood and vegetable debris.

Motter ('98, p. 219) records this species from an old grave. This
suggests a subterranean habit. (Cf. Baker, '11, p. 155.)

PhILOMYCID/E

Philoniycus carolinensis Bosc. Carolina Slug.

Several young specimens of this slug (No. 71), about 5 mm. long
when contracted in alcohol, were found (Sta. IV, a) Aug. 17 in the
upland forest on a well rotted stump overgrown in part by a felt-like
fungous growth. The finding of these young slugs and the finding

203

elsewhere in the forest of eggs, possibly of this species (Nos. 86
and 114), is of special interest. On the forested ravine slope (Sta.
IV, b) in another decaying stump, in which the bark was loosened
and the sap-wood quite decayed, soft, large examples of this slug
were found in abundance Aug. 17 (No. 89). They were associated
with newly established colonies of the carpenter-ant Camponotus
hcrciilcanus pennsylvanicus, and the horned Passalus, Passali