CHARLES

W&^

^!«';

^1^

r.

e

fe

%

w

is^

11
V" i

ANIMAL ECOLOGY

ANIMAL

PRENTICE- HALL, INC

ENGLEWOOD CLIFFS, N. J.

(^.

ECOLOGY

S. CHARLES KENDEIGH

Zoology Department
University of Illinois

ANIMAL ECOLOGY

S. Charles Kendeigh

©-1961 by
Prentice-Hall, Inc.
Englewood Cliffs, N. J.

All rights reserved.

No part of this book

may be reproduced in any form,

by mimeograph or any other means,

without permission in writing

from the publisher.

Library of Congress Catalog Card No.: 61-12332

Fourth printing June. 1965

PRINTED IN THE UNITED STATES OF AMERICA

03 7 14 - C

To my teachers of ecology

LYNDS JONES, OBERLIN COLLEGE

JOHN E. WEAVER, UNIVERSITY OF NEBRASKA

VICTOR E. SHELFORD, UNIVERSITY OF ILLINOIS

Preface

The science of ecology, born at the beginning
of the present century after a gestation period
of several hundreds of years, has now matured
into an honored and respected scholarly disci-
pline and field of research. This book is an efifort
to summarize the basic concepts and principles of
the subject and present the elementary factual in-
formation with which a person to be competent in
this field should be familiar, especially as these
things apply to animals.

After a Background section for orientation,
local communities and habitats are discussed in
some detail. It is my firm belief that one begin-
ning the study of ecology should first of all be-
come thoroughly acquainted with the places where
animals may be found in nature, what kinds of
organisms occur in different habitats, the abun-
dance and interrelations of organisms in these
habitats, the behavior and the life requirements of
the principal species, and the structure and succes-
sion of communities. The reader well founded in
this knowledge is ready to understand the ecologi-
cal processes and community dynamics which are
presented in the third section. In the fourth and
final section, the reader is introduced to the broad
field of geographic ecology, which, will give him
some knowledge of how animals are distributed
over the world, and why they occur where they do.

Physiological ecology, the study of the manner
in which organisms respond and adjust to envi-
ronmental factors, is dealt with sparingly. The
proper development of this subject takes one ex-
tensively into laboratory experimentation, which
is best left to the advanced student. Our emphasis
is kept on the study of the free-living organism in
its natural environment. Although the quantita-
tive aspect of ecology is emphasized, I do not be-

lieve tliat in an intnuluctory Ixxik it is desirable
to approach the subject from a statistical point-of-
view. Too few readers will have an adequate sta-
tistical background, and the introduction to the
subject matter of ecology should not be delayed
until an adequate background of statistics is ob-
tained — as necessary as that is to advanced work
in the field.

1 have not thought it desirable to devote a spe-
cial chapter to applied ecology or, more particu-
larlv, to wildlife management. The fundamental
principles of wildlife management are the same
as the fundamental principles of ecology, since
wildlife biology is the ecology of game species.
Throughout the book, however, I have tried to
show the relation of basic concepts to problems
in wildlife management. The special obligation of
the wildlife manager is to make practical use of
these principles for the promotion of wildlife pop-
ulations.

This book is designed for a course given at the
junior-senior-graduate level, to students who have
at least a year's background in zoology. I give
such a course during the autumn semester. If I
were to give it during the spring, I would probably
change the order of study of the four sections to
I-IV-III-II. Section IV would here substitute
for Section II in providing the student with some
knowledge of communities before undertaking
section III. This would permit field studies late
in the spring to be more closely correlated with the
discussion of local habitats and communities.

During the first semester at the University of
Illinois I have a half-day or a full-day trip every
Saturday until winter weather sets in, and there
are two half -day winter trips. Also included in
the field work is one weekend camping trip to
study communities not found locally. The stu-
dents get to see at first hand a large variety of
animals, and to measure population sizes by quan-
titative methods that may be crude but are never-
theless effective in stream riffles and pools, ponds
of different ages, bogs, lakes, grassland, deciduous
and coniferous forests, and serai stages as they
develop on rock, sand, pond, bog, floodplain. and
abandoned strip-mine areas. Some experimenta-
tion is also done in the field to analyze the manner
in which both aquatic and terrestrial species re-

spdtul to eiiviroiuneiUal factors. There is a small
amount of laboratory work for learning quantita-
tive methods of counting plankton, examining dif-
ferent kinds of respiratory systems in a(|uatic
organisms, searching and identifying micro-
organi.sms in the soil, experiments in choice of
habitats, and map-making. Methods for measur-
ing i)roductivity are discussed but actual practice
with these methods is left for an advanced class.

Knough material is given on plants in this
luiok. it is hoped, to bring out their essential place
in the ecosystem and to emphasize the bioecologi-
cal point-of-view. I believe it would be possible
for an instructor to use this textbook in a course
in general ecology by supplementing in lectures
tlie material and concepts jiresented in the book
with additional material on plants.

Some care has been taken with taxonomic
nomenclature. Common names are used through-
out the text as far as possible, with the scientific
nomenclature restricted to the index. Authorities
followed for most scientific names are the follow-
ing. Mammals — North America : Miller and Kel-
logg 1955 ; Eurasia : I-^llerman and Morrison-Scott
1951. Birds— A.O.U. Checklist 1957. Reptiles
and amphibians — Schmidt 1953. Fish — Bailey
1960. Invertebrates — as given by authors, not
standardized. Trees — Dayton et al. 1953. Grasses
— Hitchcock 1951 ; and other plants, Fernald
1950, Rydberg 1954. Common names of mam-
mals are mostly from Hall 1957; birds, A.O.U.
Checklist 1957; reptiles and amphibians, Conant
1958; and fish, Bailey 1960.

Pertinent references to literature are cited in
the text in connection with each major topic.
These references serve only as an introduction to
the very extensive literature in ecolog)-.

Finally, I wish to acknowledge the help of
many persons in the preparation of this text, par-
ticularly, Stanley A. Cain, Edward S. Deevey,
Ralph W. Dexter, Paul L. Errington, F. E. J.
Fry, Clarence J. Goodnight, F. T. Ide, Bostwick
Ketchum, Ernst Mayr, Howard T. Odum, Or-
lando Park, F"rank Pitelka, W. 1--. Ricker, Gordon
.\. Riley, M. D. F. Udvardy, and R. H. Whitta-
ker. In addition, several of my colleagues at the
University of Illinois read and commented on
early drafts of chapters. John Riina, of Prentice-

Preface

Hall, Inc., was very cooperative in having two or
three and occasionally four people read various
chapters, and two persons read the entire manu-
script. The final manuscript was expertly edited
by Oren Hunt, whose help was invaluable.

Illustrations come from several sources. I am
most grateful to Dr. Victor E. Shelford for sup-
plying many drawings originally published in his
Animal Communities of Temperate America
(1913); to the Illinois Natural History Survey
for original illustrations from several of their
publications; to the U.S. Forest Service who al-
lowed me to select what I wanted from their ex-

tensive file of photographs; to the Friez Instru-
ment Division ; to the University of Wisconsin
News Service, to a number of individuals for sup-
plying photographs or other illustrative material
for which acknowledgement is made in the legends
of the figures, and to Colleen Nelson, Katherine
Little, and Nan Brown for preparing special
drawings.

University of Illinois
Champaign, Illinois

Preface

Contents

BACKGROUND 1 Scope and History of Ecology, i

2 General Nature of Environmental Responses, 6

3 The Biotic Community, Its Structure and Dynamics, W

4 Measurement of Populations, 31

LOCAL HABITATS,

COMMUNITIES,

SUCCESSION

5 Streams, 42

6 Lakes, 59

7 Ponds, Marshes, Swamps, and Bogs, 79

8 Rock, Sand, and Clay, 96

9 Grassland, Forests, and Forest-Edges, 120

ECOLOGICAL 1 Dispersal, Migration, and Ecesis, 145

PROCESSES
AND COMMUNITY

DYNAMICS

1 1 Reactions, Soil Formation, and Chemical Cycles, 163

1 2 Cooperation ami Disoperation, 174

13 Food and Feeding Relationships, 187

1 4 Energy Exchanges, Productivity, ami Yield, 200

1 5 Reproductivity and Population Structure, 210

IX

1 6 Regulation of Population Size, 219

] 1 Irruptions. Catastrophes, and Cycles, 234

] 8 Niche Segregation, 245

1 9 Speciation, 257

GEOGRAPHIC 20 Distributional Units, 26ti

DISTRIBUTION

OF COMMUNITIES

2 1 Paleoecology, 280

12. Temperate Deciduous Forest Biome, 293

23 Coniferous Forest, Woodland, and Chaparral Biomes, 301

24 Tundra Biome, 315

25 Grassland Biome, 324

26 Desert Biome, 332
Zl Tropical Biomes, 340
28 Marine Biomes, 351

Bibliography, 373
Index, 404

X Contents

TIk- word t-coloi/y. (lerivetl from tin- (iri'ek
words oik'os moaning liahitation, and loijos meaning
discourse or study, implies a study of tlie habitations
of organisms.

luology was first described as a sejwrate field of
knowledge in 1866 by the German zoologist Krnst
Haeckel, who invented the word oekoloijic for "the
relation of the animal to its organic as well as its in-
organic environment, particularly its friendly or hos-
tile relations to those animals or plants with which it
comes in contact."

Ecology has been variously defined by other in-
vestigators, as "scientific natural history," "the study
of biotic communities," or "the science of community
populations" ; probably the most comprehensive defi-
nition is the simple one most often given : a study oj
animals and /slants in their relations to each other
and to their environment.

OBJECTIVE.S

Ecology is a distinct science because it is a
body of knowledge not similarly organized in any
other division of biology : because it uses a special set
of techniques and procedures ; and because it has a
unique point-of-view. The essence of this science is a
comprehensive understanding of the import of these
phenomena :

1. The local and geographic distribution of
organisms :

2. Regional variations in the abundance of or-
ganisms ;

3. Temporal changes in the occurrence, abun-
dance, and activities or organisms ;

4. The interrelations between organisms in popu-
lations and communities :

5. The structural adaptations and functional
BctchsrOimd: adjustments of organisms to their physical

environment :

6. The behavior of organisms under natural con-

The Scope and 7. The evolutionary development of all these in-

terrelations : and
History of EcoloSy S. The biological productivity of nature and how

this may best serve mankind.

METHODS

To the achievement of these objectives tlie
following methods or points of attack are funda-
mental :

Observation in detail of how organisms live un-
der natural conditions.

1

PROMINENT LEADERS IN THE DEVELOPMENT OF ECOLOGY
IN AMERICA

C.C. Adams, animal ecologlst {courtesy Dorothy Kehaya).

F.E. Clements, plant ecologist

Aide Leopold, wildlife manager.

2 Background

Concentration of studies not on the rare but on
the most abundant and influential organisms in
the community.

Measurement and evaluation of physical fac-
tors in the actual microhabitat occupied by or-
ganisms.

Correlation of findings of experimental studies
of organisms in the laboratory with observations
of those organisms in the field.
Use of quantitative — not just qualitative — tech-
niques in field studies as well as laboratory
studies.

A study of organisms in the field may bring to
light problems which will be most expediently worked
out in the laboratory ; but field and laboratory investi-
gations must be integrated. The investigator must
often study the morphology of dead organisms in the
laboratory, and there perform experiments on living
animals and plants held under carefully controlled
experimental conditions. But unless such studies are
perspective to the normal life of an organism, as it is
lived in natural conditions, they are not ecology.

The use of exact quantitative techniques is, of
course, a general characteristic of all science. But
special difficulties arise when such techniques are ap-
plied to free-living organisms in natural conditions.
For example, size of animal populations has, in the
past, often been described in such vague terms as
"rare," 'common," or "abundant." These are subjec-
tive terms, based largely on an impression gained by
the observer of the apparent conspicuousness of the
species. As James Fisher, an English naturalist,
wrote in 1939, a species has usually been indicated
as "rare" when actual numbers expressible in one's
and two's could be recorded ; "common" when the
observer began to lose count ; and "abundant" when
he became bewildered. One of the chief problems of
the ecologist is to develop methods by which to meas-
ure the absolute size of populations and the produc-
tive capacities of different habitats so that the activi-
ties of widely varying types of species may be
compared. For setting up experiments and organiz-
ing and analyzing studies under natural conditions,
it is becoming more and more essential that the ecolo-
gist become familiar with and employ good statistical
procedures (Williams 1954).

As a contribution to human knowledge and under-
standing, ecology is in the fortunate position of being
concerned with the most complicated systems of or-
ganisation, apart from human societies, with which
we have to deal. For this very reason it provides a
constant challenge to the imagination as well as to
experimental ingenuity. It is more difficult to analyse
and isolate the relevant factors in a living community
than in a simpler system, but the gain in significant
understanding of the material world and in compre-

Iictuiin;/ till' beauty oj its on/anicatiou is perhaps
better in proportion ( Mac lady en 1957: 246).

RKLATION TO OTHER SCIENCES

Ecology is one of tlie three main divisions
of biology : the other two being morphology and phys-
iology. The emphasis in morphology is on under-
standing the structure of organisms ; in physiology,
on how they function : and in ecology, on their ad-
justments to the environment. These divisions over-
lap broadly. To appreciate fully the structure of an
organ, one needs to know how it functions, and the
way it functions is clearly related to environmental
conditions. The morphologist is concerned with prob-
lems of anatomy, histology, cytologj', embryology,
evolution, and genetics : the physiologist, with inter-
preting functions in terms of chemistry, physics, and
mathematics : and the ecologist, with distribution, be-
havior, populations, and communities. The evolution
of adaptation and of species is of mutual interest to
the ecologist and to the geneticist ; bioclimatology is
a connecting link between ecology and physiology.
All areas, in the final analysis, are simply different
approaches to an understanding of the meaning of
life.

SUBDIVISIONS OF ECOLOGY

Ecology may be studied with particular ref-
erence to animals or to plants, hence animal ecology
and plant ecology. Animal ecology, however, cannot
be adequately understood except against a consider-
able background of plant ecology. When animals and
plants are given equal emphasis, the term bioecology
is often used. Courses in plant ecology usually dis-
miss animals as but one of many factors in the en-
vironment. Synecology is the study of communities,
and autccology the study of species.

In this book we shall survey the fundamentals and
basic facts of animal ecology. We will study com-
nmnity ecology, the local distribution of animals in
various habitats, the recognition of community units,
and succession ; ecological dynamics, the processes of
dispersal, ecesis, reaction, coaction, productivity, com-
petition, speciation, and regulation of abundance ; and
geographic ecology, geographic distribution, palaeo-
ecology. and biomes. We will also be interested
throughout the text with how species and individ-
uals respond and adjust to the physical factors of
their environment, but a full study of physiological
ecology must be left to another time and place.

When special consideration of their ecology is
given to one or another taxonomic group, we speak

E.A. Birge. limnologlst.

Henry C. Cowles, plant ecologist (courtesy R.J. Pool).

Victor E. Shelford. aninnal ecologist.

The scope and history of ecology

of mammalian ecology, avian ecology, insect ecology,
parasitology, and so on. When emphasis is placed on
habitat, we speak of oceanography, the study of ma-
rine ecology; limnology, the study of fresh- water
ecology; terrestrial ecology, and so on. Animal eth-
ology is the interpretation of animal behavior under
natural conditions : often, detailed life history studies
of particular species are amassed. Sociology is really
the ecology and ethology of Mankind.

Ecological concepts, which may be grouped to-
gether as applied ecology, have many practical appli-
cations ; notably wildlife management, range man-
agement, forestry, conservation, insect control,
epidemiology, animal husbandry, even agriculture.

This preview of ecology indicates the great
breadth and unique character of the subject material
which justifies the view of ecology as one of the three
basic divisions of general biological philosophy.

HISTORY

That certain species of plants and animals
ordinarily occur together and are characteristic of
certain habitats has doubtless been common knowl-
edge since intelligent man first evolved. This knowl-
edge was essential to him for procuring food, avoid-
ing enemies, and finding shelter. However, it was not
until the fourth century bc, that Theophrastus, a
friend and associate of Aristotle, first described in-
terrelations between organisms and between organ-
isms and their environment. He has, therefore, been
called the first ecologist (Ramaley 1940).

The modern concept that plants and animals occur
in closely integrated communities began with the
studies of August Grisebach, a German botanist, in
1838; K. Mobius, a Danish investigator of oyster
banks, in 1877; Stephen A. Forbes, an American,
who described the lake community as a microcosm in
1887; and J. E. B. Warming, a Danish botanist, who
emphasized the unity of plant communities in 1895
(see Kendeigh 1954 for further details and literature
citations). C. C. Adams recognized and described
many animal communities in his ecological surveys of
northern Michigan and of Isle Royale in Lake Su-
perior, published in 1906 and 1909. V. E. Shelford
presented a classic study of animal communities in
temperate America in 1913, and Charles Elton pub-
lished an outstanding analysis of community dy-
namics in 1927. Although an appreciation of the fact
that the whole community is one biotic unit, rather
than one unit of plants and another of animals,
may be discerned in the writings of some early in-
vestigators (eg., J. G. Cooper in 1859), the fact has
been brought to modern emphasis in the work of
F. E. Clements and V. E. Shelford, especially in their
Bio-ecology published in 1939.

Succession of plant species after burns and in
bogs has been known in a general way since about
1685 ; and European ecologists have studied succes-
sion since the late nineteenth century. The present-
day interest in succession, however, especially in
North America, dates from the plant studies of
Henry C. Cowles in 1899 on the sand dunes at the
south end of Lake Michigan, and the work of Fred-
eric E. Clements, 1916. C. C. Adams and V. E. Shel-
ford, in the citations noted were among the first to
apply the concept to animals.

Geographic ecology, in the modern sense, dates
from the generalizations on the world-wide distribu-
tion of animals made by the French naturalist,
Georges L. L. Buffon (lived 1707-1788), and the
explorations of the German botanist, Alexander von
Humboldt (lived 1769-1859). There was lively inter-
est and many important contributions in this general
field during the nineteenth century ; notably, the life-
zone concept of C. Hart Merriam (1890-1898) needs
special mention. During the present century the con-
cept of biotic provinces is identified with L. R. Dice
(1943) and the biome concept with F. E. Clements
and V. E. Shelford (1939). The broad survey of
ecological animal geography made by R. Hesse in
1924 exerted considerable effect and this treatise was
later translated into English and revised by W. C.
Alleeand Karl P. Schmidt (1951).

The study of population dynamics, so important
in modern ecology, dates back at least to Malthus, who
pointed out in 1798 the limitation to population
growth exerted by available food. Darwin, in 1859,
recognized the importance of competition and preda-
tion in developing his theory of evolution. Pearl,

1925, analyzed mathematically the characteristics of
population growth, and Lotka, 1925, and Volterra,

1926, developed theoretical mathematical equations
to show the manner in which populations of different
species interact. These studies led to the classic ex-
periments of Gause, 1935, with interacting popula-
tions of predators and prey. Nicholson's publication
in 1933 stimulated much thinking concerning the fac-
tors that stabilize populations at particular levels.
Andrewartha and Birch, 1954, emphasized the im-
portance of climate and other factors on determining
the size of populations.

The measurement and analysis of energy use by
organisms for existence and growth is now of very
great interest in ecology. Attention to biological pro-
ductivity began in the 1930's in connection with prac-
tical pond-fish culturing in Europe and the limnologi-
cal studies of Thienemann in Europe and of Birge
and Juday at the University of Wisconsin, but the
modern crystallization of the subject came with the
fresh-water and marine investigations of Lindeman,
Hutchinson, and Riley at Yale University (Ivlev
1945) and of Howard and Eugene Odum. An early

Background

study of energy relations within terrestrial communi-
ties is that of Stanchinsky (1931 ).

Physiological ecology had its historical beginnings
in the correlation of biological phenomena with vari-
ations in temperature stimulated by Galileo's inven-
tion of a hermetically sealed thermometer about 1612
AD. The French naturalist Reaumur summed the
mean daily teni[>eratures for Ajjril, May, and June in
1734 and again in 1735, and correlated the earlier
maturing of fruit and grain during the first year with
the greater accumulation of heat. A discovery of
parallel significance was of oxygen in 1774 by the
English clerg\'man. Priestley, and the finding by
Lavoisier, a Frenchman, in 1777 that it was an essen-
tial part of air. Claude Bernard, another French
physiologist, enunciated the principle of homeostasis
in 1876. This concept originally referred to regula-
tory mechanisms which maintained the "internal en-
vironment" of the body constant in the face of chang-
ing external conditions. Later, the concept came to
be applied also to maintenance of community inter-
relations. \'an't HofT, a Dutch scientist, contributed
to physiological ecology in 1884 in describing how
the speed of chemical reactions increased two- or
three-fold with each rise of 10°C. K. G. Semper and
Charles B. Davenport clearly established physiologi-
cal ecology in bringing together pertinent information
in 1881 and 1897-1899 respectively. More recent
summaries of knowledge and methods in this general
field have been made by V. E. Shelford in Laboratory
and Field Ecology (1929) and by Samuel Brody in
Biocncrgctics and Growth (1945).

The development of animal behavior or ethology
may be traced back through the natural history of
ancient times. More recently the 13 volumes of
Thierleben. prepared by A. E. Brehm during the
period 1911 to 1918, are noteworthy. H. S. Jennings,
1906, and Jacques Loeb. 1918, made valuable contri-
butions to the understanding of the behavior of in-
vertebrates. Precise modern techniques and concepts
as applied to vertebrates began to take form about
1920 with the development of banding and marking
of individual animals by S. Prentiss Baldwin (1919)
and the recognition of territories in the nesting of
birds by H. E. Howard (1920). The formulation of
the concept of releasers as controlling instinctive be-
havior by Wallace Craig (1908), K. Lorenz (1935).
and X. Tinbergen (1951) has produced a profound
effect on present-day thinking.

In regard to other divisions of ecology, the crys-

tallization of studies in oceanography may be credited
to i-:dward Forbes 1843, Maury 1855, Alexander
Agassiz 1888, Petersen and his colleagues 1911, and
Murray and Hjort 1912; limnology to Forel 1869,
Hirge 18^)3, Juday 1896, Ward and Whipple 1918,
Thienemann 191.^1935, and Xaumann 1918-1932;
and wildlife management to Aldo Leoi)old 1933.

Ecology, then, is of comparatively recent develop-
ment as a distinct science, but its roots extend well
back into the past. Doubtless the most comprehensive
treatment of the subject in all its aspects is Principles
oj Animal Ecoloijy by Alice, I-Jiierson, Park, Park
and Schmidt, published in 1949 (for citations of his-
toric interest in this chapter, see this reference).
.Since ecology is a young science, it should be empha-
sized that its concci)ts and techni(|ues have not be-
come standardized and that there is opportunity and
stimulus here for many new investigators.

The Ecological Society of America was founded
in 1915. and in 1960 had a membership of over two
thousand. The British Ideological Society, organized
in 1913, has a membership of about one thousand.
The society in America has given birth to several off-
spring during its 45 years of existence : The Wildlife
Society, Society of Limnologists and Oceanog-
raphers. The Nature Conservancy, and a Section on
Animal Behavior and Sociobiology. Several of these
organizations have their own journals. The Ecologi-
cal Society of America publishes two periodicals:
Ecology for short papers and Ecological Monographs
for long ones. The British Ecological Society also
publishes two : Journal oj Ecology for plant papers
and Journal oj Animal Ecology for animal papers.
Oikos began publication in 1949 to represent ecolo-
gists in Denmark, Finland. Iceland, Norway, and
Sweden. Announcement was made in 1960 of the or-
ganization of the International Society for Tropical
Ecology to include India and adjacent countries.
There will be a Bulletin. The Ecological Society of
Australia was organized in 1960. The New Zealand
Ecological Society came into existence in 1952 and
regularly ])ublishes Proceedings of its annual con-
ferences. The Japanese Journal oj Ecology, begun in
1954, is the official publication of the Ecological So-
ciety of Japan. Most of its articles are in Japanese,
but they have summaries in a European language.
Finally, many papers of interest to ecologists appear
in zoological journals of various sorts that do not
carry the word ecology in their titles.

The scope and history of ecology

Background:

The General Nature

of Environmental

Responses

Ecology, by definition, deals with the interrela-
tions of organisms with each other and with their
environment. These interrelations become estab-
lished as organisms respond in various ways to con-
tacts with one another and with the ever-changing
environment.

The term environment describes, in an unspecific
way, the sum total of physical and biotic conditions
influencing the responses of organisms. More spe-
cifically, the sum of those portions of the hydrosphere,
lithosphere, and atmosphere into which life pene-
trates is the biosphere. There are no characteristic or
permanent inhabitants of the atmosphere, although
the air is traversed by many kinds of animals and
plant propagules. Of the hydrosphere, there are two
major biocycles, the marine and fresh-water ; of the
lithosphere there is one, land (Hesse et al. 1931).

A habitat is a specific set of physical conditions
(e.g., space, substratum, climate) that surrounds a
single species, a group of species, or a large com-
munity (Clements and Shelford 1939). The ultimate
division of the biosphere is the niicrohabitat, the most
intimately local and immediate set of conditions sur-
rounding an organism ; the burrow of a rodent, for
instance, or a decaying log. Other individuals or
species are considered as part of the community to
which the organism belongs and not part of its habitat.
The term biotope defines a topographic unit charac-
terized by both uniform physical conditions and uni-
form plant and animal life.

In order for organisms to exist they must respond
or adjust to the conditions of their environment.
The first living organisms probably evolved in the sea
and must have possessed very generalized adjust-
ments to this relatively uniform and favorable habitat.
However, these early organisms had inherent in them
the potential for expansion, as they later spread into
other and more rigorous habitats, particularly fresh-
water and land. As evolution proceeded, organisms
became more and more limited in the range of their
ability to respond as they became specialized in their
adjustments to particular habitats. This led to the
great diversification of species that we see at the
present time, with each species restricted to its par-
ticular niicrohabitat and place in the community.

Organisms respond to differences or changes in
their environment in four principal ways : morpho-
logical adaptations, physiological adjustments, be-
havior patterns, and community relations. Chapters
2 and 3 are a resume of these responses, the general
fundamentals of which must be understood before the
subtle relations of an organism to its environment
that are the substance of ecology can be appreciated.
Probably the most important of distinctions be-
tween organisms in a consideration of their morpho-
logical responses to the environment is whether
they are sessile or motile (Shelford 1914). Most

plants arc, of course, sessile ; most animals, motile.
There are, however, some motile ])lants among uni-
cellular forms and male gametes, and there are many
sessile or slow-moving animals in a(|uatic habitats.
Sessile organisms respond to variations of the en-
vironment primarily by changes in form ; motile ani-
mals. ])riniarily by changes in behavior.

MOKIMIOI.OGICAL .AD.APT.^TIONS

('hangcs in form and siriirliiro

is a similar moriihological res|)onse to a similar en-
vironment (Schmaulhausen 1949). If the growth-
form ])ersists through many generations and appears
to he an adajjtation. even though not inherited, it is
often called an ccad. If and when the growth-form
becomes inherited as the result of evolutionary proc-
esses, it then becomes an fiolyf<c. Lije-form is a gen-
era! term referring to the sha|)e or appearance of an
organism irrespective of how formed (Daubenmire
1947 ). The prevalence of particular life-forms among
the im])ortant organisms helps to separate and char-
acterize biotic communities, as we will see repeatedly
in later discussions.

Consider a sessile organism, the tree. It is essen-
tial to the tree that its foliage be exposed to sunlight.
As it grows within a forest, it is usually tall and
slender, and little branched except at the top. where
the cap of foliage reaches into the full sunlight. Grow-
ing on the forest's edge, the tree is shorter, and
branching and foliage are dense both at the cap and
on that side exposed to full sunlight. The tree which
grows solitary in an open place is short, but branch-
ing and foliage are dense and uniformly distributed,
often starting close to the ground. In similar manner,
the variations in form assumed by sessile colonial
animals, such as sponges and corals, reflect vicissi-
tudes imposed by habitat (Wells 1954).

Morphological variations induced by peculiarities
of habitat do occur in motile animals : thickening of
the shells of clams subjected to strenuous wave ac-
tion ; variation in number of vertebrae, scales, and fin
rays among fish subjected to different temperatures
at critical periods in their growth (Taning 1952) ;
changes in the number of facets in the bar-eye of the
fly Drosophila as a correlative of temperature varia-
tions during a short critical period in larval growth
(Krafka 1920) ; the many variations in form and size
of internal parasites, depending on crowding and
other environmental conditions (Baer 1931) ; pointed
tails in certain flatworms crawling over a substratum
during growth, contrasted with rounded tails occur-
ring in the same species when individuals are experi-
mentally prevented from crawling over the sub-
stratum (Child 1903).

That individuals of the same species are so much
alike attests the great extent to which the course and
outcome of morphological development is genetically
determined. But that there are variations between
individuals, and between groups of individuals, of the
same species shows that morphological development
is also responsive to environmental influences. Modi-
fications induced by the environment emerge as the
individual develops and are not specifically inherited
by the succeeding generation. These modifications
are called growth-forms. If the generation following
is similar in growth-form to the parent generation, it

Life-forms of plants

The life-form of a plant is characterized by its
vegetative form, its length of life, the arrangement
and character of its leaves, whether its stem is her-
baceous or woody, its manner of growth, and its
means of overwintering. Life-form categories some-
times agree with large taxonomic units, such as ferns
and mosses, but on the other hand some taxonomic
groups contain s])ecies exhibiting a variety of life-
forms and some life-forms include species only re-
motely related taxonomically.

There have been many systems proposed for the
classification and terminology of the life-forms of

FIG. 2 1 Form assumed by the coral Modreporo as If develops
In (a) deep water; (b) barrier pools; (c) rough water (from
Wood-Jones 1912).

The general nature of responses 7

plants, but for the animal ecologist the following
simplified classification, based on Pound and Clem-
ents (1900), is sufficient for general purposes:

1. Annuals: Passing the winter or dry season in
seed or spore form alone, no propagation or
accumulation of aerial shoots ; living one year.

2. Biennials : Passing one unfavorable season in
the seed or spore form and the next in a vege-
tative stage ; no accumulation of aerial shoots :
living two years or parts of two years.

3. Herbaceous perennials: Passing each unfa-
vorable season in both seed or spore and vege-
tative form : no accumulation of aerial shoots ;
living several to many years.

a. Broad-leaved herbs : mostly terrestrial

b. Sod grasses : a continuous turf

c. Bunch grasses : scattered clumps

d. Succulents : some broad-stemmed cacti

e. Water plants:

( 1 ) Submerged : vegetative body entirely
underwater.

(2) Floating: leaves floating on water sur-
face ; water lilies, duckweed.

(3) Emerging: leaves extending above
water surface : cattails, sedges, rushes.

f. Ferns

g. Mosses

h. Liverworts
i. Lichens
j. Fungi
k. Algae

4. Woody perennials: Passing the unfavorable
season as aerial shoots or masses, often as seeds
also ; living many years as a rule.

a. Lianas: vines

b. Succulents : some tree or barrel cacti

c. Bushes: much -branched, low growth, sev-
eral stems

d. Shrubs: single stem and tree-like but
smaller

e. Trees:

(1) Deciduous: shedding leaves during
unfavorable season.

(2) Evergreen: leaves shed irregularly
and tree never completely bare.

(a) Needle-leaved: narrow, usually-
elongated leaves.

(b) Broad-leaved: leaves much as on
deciduous trees but shed irregu-
larly.

Life-forms of animals

Systems to classify the life-forms of animals
have been little developed (Remane 1943, 1952).

The major life-forms of animals more often agree
with their taxonomy than do plants, but some life-
forms include representatives from several different
taxonomic groups. There can be recognized encrust-
ing forms such as the fresh-water bryozoa Pliiina-
tclla and some sponges : coral forms, including grass,
leaf, or shrub forms ; radiate forms, such as coelente-
rates and echinoderms generally ; bivalve forms ; snail
forms : slug forms ; worm forms ; crustacean forms ;
insect forms : fish, snake, bird, and four-footed forms.
Each of these major types may be subdivided ; for
instance, the four-footed form of mammals (Osburn
ct al. 1903) :

Aquatic (swimming) : seal, whale, walrus
Fossorial (burrowing) : mole, shrew, pocket

gopher
Cursorial (running) : deer, antelope, zebra
Saltatorial (leaping) : rabbit, kangaroo, jumping

mouse
Arboreal (climbing) : squirrel, opossum, monkey
Aerial (flying) : bat

Adaptations

Plants and animals of specific life-forms are
adaptations to live in particular habitats and to be-
have in particular ways (Klaauw 1948). The life-
forms listed for mammals are largely adaptations to
particular strata (water, subterranean, ground, tree,
air) within a community rather than to the habitat
as a whole ; for instance, the subterranean adapta-
tions of mammals living in the Arctic tundra are
similar to the subterranean adaptations of mammals
in the tropics. In communities lacking one or more
strata (for instance, the tree stratum in grassland),
animals specifically adapted to the missing strata are
also absent. In communities in which all strata are
present, a catholic variety of life-forms occurs.

In addition to adaptations to stratum and habitat,
there occur ecologically significant adaptations for
food-getting and metabolism, protection, and repro-
duction. The variety of teeth found in mammals and
lizards, the variation in shape and size of bills of
birds, the different mouth parts of insects, the siphons
of clams, the suckers of leeches, the water canal sys-
tems of sponges are but a few special anatomical fea-
tures especially designed for food-getting. Associated
with food-getting is a great diversity in structural
adaptations for the digestion of the food, for respira-
tion, for circulating food materials and gases through
the body, for excreting wastes, for support and move-
ment, and for nervous and hormonal regulation. All
these internal organs and structures are necessary to
the animal for utilizing the energy resources of the
environment.

8 Background

-Ml animals are subject to predatioii or coinpeti-
tioii and must liave means of protecting themselves
or offsetting losses in the struggle for existence. Such
adaptations take a variety of form such as body-
armor, concealing coloration, attack weapons, or be-
havior ])atterns of escajie. High rates of mortality are
oH'set by high rates of reproduction or, in some lower
organisms, by considerable power of regenerating
wiiole organisms from fragmented parts.

The manner in which reproduction occurs and tlie
special structures concerned with rejiroduction vary
with each type of animal and often with each indi-
vidual species. These adaptations are universal and
too numerous even to attempt to classify at this point
but are certainly obvious to all. Tlie primary objec-
tives in the life of each species are to maintain the
e.xistence of the individual and to reproduce its own
kind, and all adaptations to live in favorable habitats
are designed toward these ends.

Maliiral sclrrtion

To be heritable, a variation must have been
caused either by mutations in the genes or chromo-
somes of the individual or by new combination of
genes. Mutations are produced at random, and
mostly independent of natural environmental condi-
tions, although there is some experimental evidence
that they may be induced or increased in frequency
by cosmic rays, ultraviolet rays, heat, and certain
chemicals. There is no reason to believe, however,
that environmental factors can ordinarily influence
the kind of mutations that occur.

Heritable variations in the structure of organisms,
and in their physiology and behavior as well, may be
favorable, unfavorable, or of neutral value to the ex-
istence of the species. Variations that decrease the
efificiency of a species in its struggle for existence
against competitors and unfavorable environmental
conditions usually disappear, but variations that in-
crease this efificiency give those individuals that pos-
sess them a better chance for survival and for giving
birth to similar offspring. Thus, there is natural se-
lection of the fitter individuals, and a gradual im-
provement in the relations between the species and
its environment. It is in this way that adaptations
are established. A better understanding of the eco-
logical relations between different species and be-
tween species and the environment will contribute to
a better understanding of the process of evolution.
At the same time a thorough understanding of the
processes of evolution is necessary to understand how
organisms become adapted to live in particular habi-
tats (Simpson 1953).

A close study of differences between individuals
shows that within many species convergent evolution

occurs under similar environmental conditions.
Many of these variations an- genetic and apparently
due to natural selection. The i)est established corre-
lations are the following, aithougii even they are sub-
ject to fre(|uent exceptions (Mayr 1942. Uobzhansky
1*)51 ) :

I^ergmann's rule: geographic races of a species
possessing smaller body-size are found in the warmer
parts of the range, races of large body-size in the
cooler parts. This appears true for cold-blooded as
well as warm-blooded animals (Ray 19()0).

.\i.i.i;.\"s kii.e: tails, ears, bills, and other ex-
tremities of animals are relatively shorter in the
cooler parts of a species' range than in the warmer
l>arts.

Gloger's rule: in warm-blooded species, black
pigments increase in warm and humid habitats, reds
and yellow-browns prevail in arid climates, and pig-
ments become generally reduced in cold regions.

Races of birds in the cooler parts of a species'
range lay more eggs per clutch than races in the
warmer parts of the range. Likewise the number of
young per litter of mammals averages higher in cooler
climates.

The stomachs, intestines, and caeca of birds that
live on a mixed diet are relatively smaller in the
tropical- than in the temperate-zone races of a species.

The wings of birds that live in a cold climate or
in high mountains are relatively longer than those of
close relatives that live in lowlands or in a warm cli-
mate.

Races of birds in cool climates are more often and
more strongly migratory than races in warm climates.

Races of mammals in warm climates have less
under-fur and shorter contour hairs.

Fish of cool waters tend to have a larger number
of vertebrae than those living in warm waters. In-
crease in salinity tends to induce the same result as
low temperature.

Fish that inhabit swift waters tend to be larger
and more streamlined than inhabitants of sluggish
or stagnant waters.

Cyprinid fishes, isolated in desert springs, tend to
lose their pelvic fins.

Land snails reach their greatest size in the area
of optimum climate within the range of the species.

The relative weight of snail shells is highest in
the forms exposed to the highest radiation of the sun
or to the greatest aridity.

Land snails tend to have smooth, glassy, brown
shells in cold climates, and to have white or strongly
sculptured shells in hot dry climates.

It would appear at first glance that several of
these rules have a physiological basis: for instance,

The general nature of responses 9

large body size and short appendages give less sur-
face area per volume of body and thus minimize heat
loss from the body, in cold climates. It is doubtful,
however, that the smaller surface area thus attained
gives enough reduction of heat loss to be significant
in warm-blooded animals and would not apply to
cold-blooded ones. Rather the ability of warm-
blooded animals to live in cold climates depends more
importantly on the insulation of the body surface, its
exposure, its vascularization, and its ability to tolerate
a cold tissue temperature (Scholander 1955, Irving
1957). The older explanations of selective value of
many of these rules are therefore doubtful.

PHYSIOLOGICAL ADJUSTMENTS

Nature of adjustments

Probably the first response of any organism to
a change in the environment is physiological. A
physiological response must certainly precede any
change in form or structure which requires growth.
Even a change in behavior must follow a change in
some receptor or sense organ followed by nervous
function ; a fall in air temperature, for instance,
brings a drop in the metabolic rate of cold-blooded
organisms but a rise in the rate of warm-blooded or-
ganisms. Cold may stimulate nerve endings in the
skin of birds or mammals and produce shivering and
a search for protective cover. Transference from the
dark to light may immediately initiate photosynthesis
in resting chloroplastids within a plant cell, or a
change in turgescence on opposite sides of a sessile
zooid may result in a turning movement, an orienta-
tion to or away from the light source. Physiological
responses are thus internal responses to factors of the
environment. Often they are difficult to detect.

Types of response

Environmental factors influence organisms
physiologically in various ways (Fry 1947). These
effects may be classified as follows :

Lethal : causing death ; for instance, extreme heat
or cold, lack of moisture, and so forth.

Masking : modifying the effect of some other fac-
tor. Low relative humidity increases the rate
of evaporation of moisture from body sur-
faces so that warm-blooded animals are able
to survive at otherwise intolerably high air
temperatures.

Directive: producing an orienting response in
relation to some environmental response so

that the organism gets itself into favorable
conditions.

Controlling: influencing the rate at which some
process functions, but not entering the reac-
tion. Temperature, pressure, and viscosity,
for instance, affect secretion, locomotion, and
metabolism.

Deficient : curtailing an activity because some es-
sential ingredient, such as a salt, oxygen, or
the like is absent or at unfavorably low con-
centration.

The same environmental factor may produce dif-
ferent effects at different times and under different
conditions. Temperature may be lethal, if extreme ;
masking, as when cold reduces the demand of cold-
blooded organisms for food ; directive, by inducing a
search for more favorable locations ; or controlling,
as a modifier of the rate of metabolism. Often the
distinction between controlling and deficient factors
is not made, or they are considered as together con-
stituting limiting factors.

Threshold and rate

Every environmental factor varies through a
wider range of intensivity than any single organism
could tolerate. Characteristically, there is for each in-
dividual organism a lower and an upper limit in the
range of an environmental factor between which it
functions efficiently. For any one factor, different or-
ganisms find optimal conditions for existence at dif-
ferent points along the range : hence their segregation
into different habitats.

The threshold is the minimum quantity of any
factor that produces a perceptible effect on the or-
ganism. It may be the lowest temperature at which
an animal remains active, the least amount of mois-
ture in the soil that permits growth of a plant, the
minimum intensity of light at which a photoreceptor
is stimulated, and so forth. Above the threshold, the
rate of a function increases more or less rapidly as
the quantity of heat, moisture, light, or other environ-
mental factor is augmented, until a maximum rate is
attained. Above the maximum, there is usually a de-
cline in the rate of a process either because of some
deleterious effect produced, the interference of some
other factor, or exhaustion. The curve of decline at
high temperatures is usually steeper than the curve
of acceleration at low temperatures.

Law of toleration

For each species there is a range in an environ-
mental factor within which the species functions at
or near an optimum. There are extremes, both maxi-
mum and minimum, towards which the functions of

10 Background

a species are curtailed, tlieii iiiliil)ited. In stiiiie organ-
isms, such as tisli, the u/^f^cr limit oj tolerance is
readied before activity is reduced to zero. At low
temperatures, the loiver limit of tolerance may he
readied wliile the animal is still cajjahle. potentially,
of considerable activity, and death is the result of
other factors. On the other hand, some organisms
may survive in an inactive or dormant state under
environmental conditions that do not permit activity,
only to become functional again when critical factors
rise above the threshold. Before the limits of tolera-
tion are reached there are zones of increasing physio-
logical stress.

The species as a whole is limited in its activities
more by conditions that produce physiological dis-
comforts or stresses than it is hy the limits of toler-
ation themselves. Death verges on the limits of tol-
eration, and the existence of the species would be
seriously jeopardized if it were frequently e.xposed to
these extreme conditions. In retreat before condi-
tions of stress there is a margin of safety, and the
species adjusts its activities so that limits of tolera-
tion are avoided. There is variation in hardiness of
individuals within a species, so that some hardy indi-
viduals find existence possible under conditions that
disrupt other individuals. The population level of a
species becomes reduced therefore before the limits
of its range are actually reached. It is desirable to
test by acclimation and breeding experiments whether
these differences in physiological adaptiveness be-
tween individuals or populations are genetic or
phenotypic (Prosser 1955).

Species vary in their limits of tolerance to tiie
same factor. The Atlantic salmon, for instance,

THRESHgLD

Limits of tolerance for some organisms

TEMPERATURE

High

FIG. 2-2 Infsractlon betwean environment end cold-blooded
organisms: organism activity as a function of environmental
temperature (modified from Fry 1947).

spends most of its adult life in the sea, but goes an-
nually into fresh-water streams to breed. Most other
marine fishes are killed quickly when placed in fresh-
water, as are fresh-water fish when placed in salt
water. The following terms are used to indicate the
relative extent to which organisms can tolerate vari-
ations in environmental factors. The prefix steno-
means that the species, population, or individual has
a narrow range of tolerance and the prefix eury- in-
dicates that it has a wide range : thus stcnohaline

relation to distribution and population level — often a normal curve (modified from Shelford I" I )•

^Lower limit of tolerance Upper limit of tolerance -

High

GRADIENT -

•High

The general nature of responses

11

and euryhaline in respect to salinity, stenohydric and
euryhydric in respect to water, stenothermal and
eitrythermal in respect to temperature, stenophagic
and euryphagic in respect to food, stenoecious and
euryoecious in respect to niche or habitat, and so on.

Law of the minimum

An organism is seldom, if ever, exposed solely
to the effect of a single factor in its environment. On
the contrary, an organism is subjected to the simul-
taneous action of all factors in its immediate sur-
roundings. However, some factors exert more influ-
ence than do others, and the attempt to evaluate their
relative roles has led to the development of the law
of the minimum.

The first elaboration of this law was made by the
German biochemist, Justus von Liebig, in 1840, who
stated :

// one of the participating nutritive constituents
of the soil or atmosphere be deficient or wanting or
lacking in assimilability , either the plant does not
grow or its organs develop only imperfectly. The de-
ficient or lacking constituent makes those that are
present inactive or lessens their activity. If the de-
ficient or lacking constituent he added to the soil or
if occurring in insoluble form it be made soluble, then
the other nutrients become active (Browne 1942).

Blackman (1905) developed the more compre-
hensive concept of limiting factors when he listed five
factors involved in controlling the rate of photosyn-
thesis : amount of COo available, amount of HoO
available, intensity of solar radiation, amount of
chlorophyll present, and temperature of the chloro-
plast. Any one of these factors will control the rate
of the process if the factor is present in least favor-
able amount, or may actually stop it when insufficient,

320

16 20 24 28 32 36 40

TEMPERATURE, °C

FIG. 2-4 The relation between maximum respiration rate, tem-
perature, and oxygen tension (mm Hg as shown by values in
the graph) in young goldfish acclimated to each temperature
before measurements were taken (Fry 1947).

even though all other factors occur in abundance.
The same principle applies to animal functions.

Since the rate of a process may be controlled by
too great an amount of a substance, such as heat, as
well as by too small an amount, and since the pres-
ence or abundance of an organism may be limited by
a variety of environmental factors, biotic as well as
chemical and physical, and since the limiting effect
may be due to two or more interacting factors rather
than a single isolated one (Shelford 1952), the laiv
of the minimum may be restated in broad ecological
terms, as follows : the functioning of an organism is
controlled or limited by that essential environmental
factor or combination of factors present in the least
favorable amount. The factors may not be continu-
ously effective but only at some critical period during
the year or perhaps only during some critical year in
a climatic cycle (Taylor 1934).

BEHAVIOR RESPONSES

Orientation

Behavior responses to changes in environ-
mental factors can usually be detected immediately
as turning or locomotor activities on the part of the
organisms (Fraenkel and Gunn 1940). These move-
ments tend to take the organism away from points of
danger and into more favorable locations, or to per-
form sotne task essential to existence, or to reproduc-
tion. If the movement involves curvature or a turn-
ing movement either toward or away from the source
of stimulus, the movement is called a tropism. Motile
organisms frequently respond by actual locomotion
toward or away from the stimulus rather than mere
turning, and such guided or directed locomotor move-
ments are called taxes. When the movements of the
animal are random in direction, and there is no im-
mediate orientation to the source of stimulus, but the
frequency of turning or speed of the movements is
dependent on the intensity of stimulation, such re-
sponses are termed kineses. As the result of kineses
an animal may arrive by chance in a favorable en-
vironment, by which the intensity of the stimulus is
reduced or entirely eliminated. To identify the stim-
ulus to which the organism is responding, the fol-
lowing prefixes are employed : thermo-, tempera-
ture ; photo-, light: geo-. gravity; hydro-, moisture;
chemo-, chemicals ; thigmo-. contact ; baro-, pressure ;
rlico. current; and galvano-. electricity.

Jacques Loeb, during the period 1888-1918, vig-
orously maintained that all tropisms and taxes of
organisms were mechanical, automatic, and explain-
able in simple concepts of physics and chemistry.

. . . the overzvhelming majority of organisms
have a bilaterallv s\nnmetrical structure. . . . Nor-

Background

))/(j//v ///(• froccs.u's indiiciiKj locoiiioliou arc equal in
both lialvrs of the central ncn-oiis system, and the
tension of the syniinctrical ntiisclcs hein;/ equal, the
animal moves in as straight a line as the imperfec-
tions of its locomotor apparatus permit. If, however,
the velocity of chemical reactions in one side of the
body, c.j/.. in one eye of the insect, is increased, the
physiological symmetry of both sides of the brain and
as a consequence the equality of tension of the sym-
metrical muscles no longer e.vist. The muscles con-
nected ti'lth the more strongly illuminated eye are
thrown into a stronger tension, and if new impulses
for locomotion originate in the central nen^ous sys-
tem, they will no longer produce an equal response
in the symmetrical muscles, but a stronger one in the
muscles turning the head and body of the animal to
the source of light. The animal «•('// thus be com-
pelled to change the direction of its motion and to
turn to the source of light. . . (Loel) 1918).

The idea that all instinctive activities of organ-
isms were forced and invariable responses to en-
vironmental factors met many objections. H. S. Jen-
nings (1906) pointed out that many Protozoa are
asymmetrical in body structure and hence could not
lend support to the tonus theory. Furthermore, the
movements and responses of many organisms to en-
vironmental stimuli were not stereotyped, but random
in nature : of a trial and error sort. Although much
of Loeb's theory has been disproven experimentally
and appears untenable on the basis of observations
of animal activities under natural conditions, it crys-
tallized the need for objective analysis and interpre-
tation of animal behavior, and the avoidance of teleo-
logical and anthropomorphic explanations. The study
of orienting responses of organisms is of utmost
ecological significance since it is largely by means of
such responses that organisms find their proper and
favorable habitats.

Prejerendum

The behavior responses of animals and their
orientation in respect to most environmental factors
can be tested experimentally, and results thus ob-
tained correlated with the animal's behavior under
natural conditions. There is a variety of procedures
and equipment suitable to these purposes (Shelford
1929, Warden, Jenkins, and Warner, I, 1935) and
there is distinct value in verifying field observations
with e.xperimental analyses.

When the number of favorable responses at each
unit intensity of an environmental factor is plotted
against the entire range of that environmental factor,
the usual result is a normal or Gaussian curve. The
maximum number of responses normally occurs near
the center of the range, with a progressive reduction

in nniui)cr toward each cxtn-me. \u extension in
eacii direction from the i)eak of the rcs|)onscs to in-
clude ."^O, 2^, or some smaller percentage of tiie total
responses is called the preferendum for that animal
or griiuii of animals.

IniKilr hihniiiir

.Much of the behavior of organisms is deter-
mined by heredity and is characteristic of the species
in its ])roper environment. This behavior may be evi-
dent at birth or it may not develoj) until the nervous
.system, including both the receptor and eflfector mech-
anisms, is fully matured. Such innate behavior is of
various degrees of complexity. A reflex is a quick,
automatic response of a single organ or organ system
to a simjile stimulus: for instance, the knee jerk in
man. Tropisms, taxes, and kineses may involve a
series of reflexes and represent a higher level of in-
tegration. An instinct, or inherited behavior pattern,
is a complex fixed behavior that is activated, more or
less automatically, when the animal is presented with
the proper stimulus (Thorpe 1951).

The anatomical basis for these various grades of
iiehavior lies in the structure of the nervous system
and especially, in higher types of animals, in the in-
terarrangement of neurones and synapses with each
other and in the neural pathways that become estab-
lished. Behavior patterns become elaborated through
evolution, are as subject to mutation as any struc-
tural part of the body, and are a means whereby ani-
mals respond advantageously to the various factors
in their normal environment.

Stimuli

Before an action will take place the nervous
mechanism must be released by the reception of a
stimulus. Stimuli may be either external or internal
to the organism. Protoplasm is sensitive to any kind
of stimulation, provided it is intense enough. In
higher organisms, however, specialized tissues have
become particularly sensitive to one kind of stimulus,
and these tissues, or sense organs, are called recep-
tors. There are several forms of receptors : photo-
receptors, phono-receptors. mechano-receptors,
chemo-receptors, thermo-receptors. and stato-recep-
tors. Not all types of receptors are present in all
organisms, and the structure and effectiveness of
those present varies from one kind of animal to an-
other. The efficiency of the receptor mechanisms is
important, as they largely determine the environ-
mental factors to which the animal will respond and
the degree of sensitivity involved.

Stimuli may be internal, and derive either from

The general nature of responses

External factors-

- Motivation

/

releaser mood

\ /

Neural mechanism

Behavior

FIG. 2-5 Factors involved in the activation of an instinct.

hormones or as kinesthesia involving changes in the
tension of muscles and tendons or changes in shape
or form of muscle fibers. Motivation is established
when there is an accumulation of internal stimuli po-
tentials as the result of hormone action, kinesthetics,
or changes of metabolism. A combination of motiva-
tion with proper external conditions and stimuli sets
up a drive, such as the hunger drive, or reproduc-
tive drive (Richter 1927).

Once a major drive is initiated, satisfaction of it
requires a series of events and stimuli at different
levels of integration, so that a hierarchy of drives, ac-
tions, and stimuli is established. The significance of
this hierarchy is that a major activity in the life cycle
of an animal does not take place until the organism
is in a proper physiological state, which depends,
often in large part, on the environment, and then one
action leads to another until consummation is com-
pleted. In the male stickleback, for instance, the re-
productive drive is not initiated until hormone stimuli
are released as the result of gonad enlargement and
response to lengthening daily photoperiods. Once the
reproductive stimulus is given, the first secondary

FIRST LEVEL SECOND LEVEL THIRD LEVEL

drive is the establishment of nesting territories by
fighting among male fishes. Then the nest is built.
Only after this is completed is the male ready to re-
ceive the female.

Even though an animal may have potential ca-
pacities in its sense organs with which to respond to
the whole environment, a particular action is trig-
gered by stimuli from only a very small part of the
environment. This is a fundamental characteristic of
innate behavior, and the discovery of these critical
sign stimuli or releasers is necessary for an apprecia-
tion of the interrelation of animals in a community
and how they respond to their environment (Lorenz
1935, Tinbergen 1951).

The complete enactment of mating behavior in
the stickleback proceeds step by step in an orderly
manner, each action a releaser for the ne.xt. If any
one step is changed, or is interrupted, the behavior
subsequent in the sequence does not take place. Re-
leasers are of a variety of sorts in different species,
but commonly involve particular colors or color
patterns, call-notes or songs, shapes, chemicals, or
contacts, as well as associated acts, positions, or move-
ments on the part of another animal. If these trig-
gers are not presented, the behavior does not become
expressed even though a specific nervous mechanism
is present. The analysis of behavior through obser-
vation and experimentation with the objective of un-
derstanding how an animal acts under natural con-
ditions constitutes the science of ethology, an essential
branch of ecology. Ethology differs from psychology
in that it is concerned with understanding not only
the causality of behavior but also the survival value
of behavior patterns under natural conditions, and
the evolution of these patterns. Psychology is con-
cerned more with analyzing the nervous mechanisms
that are involved.

Reproductive [
drive

• Fighting

Chasing

Biting

Threatening

Digging

Testing of materials

Boring

Gluing

Zigzag dance
Leading female to nest
Showing entrance
Quivering
Fertilizing eggs

Fanning
Rescuing eggs

FIG. 2-6 The hierarchy of drives an<
sticltleback (after Tinbergen 1951).

Learning

All behavior is not, of course, automatic and
inherited. Much of it represents the adjustment of
fixed patterns to changes in and conditions of the
animal's surroundings (Thorpe 1956). Learning may
be defined as the adaptive change in individual be-
havior as a result of experience.

The simplest form of learning is habituation, that
is, learning not to respond to stimuli which tend to
be without significance in the life of the organism.
Young animals, for instance, have an innate tendency
to respond to a wide variety of danger stimuli, such
as any sudden movement or noise. However, when
such stimuli are presented repeatedly without asso-
ciation with further effects, the young animal learns
to disregard them. There is some evidence, on the
other hand, that instinctive recognition of a special-

1 4 Background

ized predator of a species shows little or no habitna-

Conditionintj is a form of learning and consists of
the establishment of a connection between a normal
reward or punishment and a new stimulus, that is,
one that hitherto has had no meaning to the animal.

Imprintiny is especially well shown in waterfowl
and gallinaceous birds. Grey-lag geese reared from
the egg in isolation react to their human keepers, or
to the first relatively large moving object that they
see. as they would the parents by following. This
imprinting of the parent companion is confined to a
very definite and usually very brief period following
shortly on emergence from the egg. Once thoroughly
established, the behavior is very stable, if not totally
irreversible. Furthermore, this imprinting of a hu-
man being as a substitute for its own species will call
forth, a year or more later, sexual reactions to man
in the mature bird. There is no innate recognition by
birds of parent, species, se.x, or home locality, but
there is evidence that these are learned through as-
sociation and contact during the course of develop-
ment. Imprinting doubtlessly also occurs in other
animals than birds.

Imitation is another form of learning. An indi-
vidual in a flock or herd may start to feed or run
when it observes other individuals feeding or run-
ning. A young animal learns much that is traditional
of the species by imitating its parents. Vocal imita-
tion is conspicuous in the elaborate songs of some
birds.

Trial and error learning involves trial responses to
a variety of stimuli with gradual elimination of all
responses and stimuli except the relevant ones. A
chick pecks at random at all sorts of objects until it
accidentally strikes one which is edible, whereafter
the chick has a greater tendency to peck at objects
that have a similar appearance. Repetition of the
same act usually leads to the formation of a habit.
Habits often appear stereotyped but differ from in-
stincts in that they have to be learned and are not
inherited.

Insight learning involves an apprehension of rela-
tions and the sudden adoption of an appropriate re-
sponse without previous trial and error behavior.
The mason wasp of India builds a cluster of clay
cells. After depositing an egg in each cell, the female
fills it with caterpillars and seals it with a lid. Even-
tually the whole cluster is covered with a layer of
clay. While a wasp was away hunting for its prey,
an experimenter made a large hole in the side of a
cell. On its return, the wasp put in a caterpillar
which fell out through the hole. A second caterpillar
stuck in the hole with a large part hanging out
through it. When the cell was completely provi-
sioned, the wasp appeared to notice the hole for the
first time and carefully examined it. With great and

FIG. 2-7 Courtship «nd maflng behavior of the three-spine
sticitlebecit (after Tinbergen 1951).

The general nature of responses 15

prolonged effort she managed to stuff the caterpillar
back into place. She then collected a pellet of clay
and mended the hole. Such behavior as this involves
an apprehension of relations and a sudden adaptive
response not preceded by trial and error. Insight
learning may be manifested in various ways as
through homing ability, detouring around obstacles,
tool-using, discrimination of forms and patterns, and
so forth.

Ecological life histories

Developmental life histories trace the origin and
growth of structures and functions of an animal from
the egg stage until maturity is reached. Such studies
are largely embryological in nature. Behavior life
histories attempt to analyze the activities of animals
in terms of innate and learned behavior, and the
neural mechanisms involved. In order to do this, it
is often necessary to trace the origin of each activity
to the manner in which it first makes its appearance
in the young animal. Ecological life histories, on the
other hand, are concerned with the activities of a
species throughout its life cycle, and in relation to
its adjustments to natural conditions. Ecological life
histories usually proceed with, first, analysis of the
behavior adjustments needed for the survival of the
mature animal ; then of its reproductive behavior ;
and, lastly, of the development of behavior and physio-
logical adjustments of the young animal. In general,
the proper procedure is. . . to discover and estab-
lish correlations betiveen the behavior of the organism
and the conditions in its environment, and then to
test the significance of the correlations by appropriate
experiments in nature or in the laboratory. The point
should be emphasized that you start with nature, that
is, with the organism in its environment. Also it
should be noted that morphology and physiology of
the organism are entirely subsidiary matters, al-
though most important to the person interested in
knowing how the organism behaves as it does. . .
(Huntsman 1948). The behavior of a species in re-
lation to its environment is called its mores (Shelford
1913).

The following are important items that should be
included in a complete ecological life history of a
species :

1. Phylogenetic and geological history.

2. Geographic and habitat distribution with an

analysis of adjustments to the physical en-
vironment and of biotic interrelations within
the community.

3. Variations in population, through time and
in space.

4. Changes in seasonal activities and physiologi-

cal states: breeding, migration, hibernation.

5. Food, enemies.

6. Parasites, diseases.

7. Reproductive potential, mortality, rate of pop-
ulation turnover.

8. Requirements for reproduction : home range,
territory, nest-site, nesting materials, etc.

9. Breeding behavior : mating, nesting, etc.

10. Development of offspring: rate, stages, gen-
erations per year, etc.

Useful outlines, methods, and bibliographies for
ecological life history studies of different kinds of
animals and plants have been published in the scien-
tific periodical Ecology since October, 1949.

Ecological niche

The ecological niche is the particular position in
a community and habitat occupied by an animal as
the result of its peculiar structural adaptations, its
physiological adjustments, and the special behavior
patterns that have evolved to make best use of these
potentialities. Important factors in the niches occu-
pied by white-footed mice and deer mice are de-
scribed in Table 2-1. Both mice are equipped with
large eyes for nocturnal vision, large external ears
for hearing, long vibrissae on the face for aid in run-
ning through dark underground burrows, and pro-
tective coloration. P. I. novcboracensis has a longer
tail than P. m. bairdii. which appears to be an adap-
tation for climbing. It is possible that these two
species are segregated into different niches because
bairdii is more tolerant of extreme temperatures and
low moisture conditions, and hence is more prevalent
than novcboracensis in the exposed grassland habitat,
but is unable to displace novcboracensis within the
forest because of the latter's tree climbing ability.

Every species has its own peculiar niche. No two
species can permanently occupy exactly the same
niche in the same locality. The living together of
many species in the same community is possible only
because their various niche requirements are differ-
ent. The analysis of the critical factors in these niche
requirements is often very dif^cult but is one of the
main objectives of ecology.

COMMUNITY INTERRELATIONS

The fact that species with similar tolerances
and requirements aggregate into similar environ-
ments to form communities is a response of special
interest. No organism occurs alone. Each must find
its place in the community and establish relations
with other members of it. The manner in which the
response of species to each other is affected is shown
in the structure and composition of the community

Background

TABLE 2-1 Comparison of nichas of fhe

Factor

white-footed and deer

both species of the genus P«iomyicu>.

P. leucopus noveboracensis

P. maniculatus bairdii

Vegetation, substratum or
space occupied

Microclimate

Food
Enemies

stratum where food found
Reproductive site; nesting
materials

Dlel activity
Seasonal activity

Deciduous forest; subterranean,

terrestrial, arboreal; home

range 0.12 hectare
Shade, rich humus, moderate

moisture, medium temperature
Seeds, nuts. Insects
Owls, foxes, weasels, shrews
Surface of ground
Nests of leaves in burrows,

logs, stumps, or tree

cavities
Nocturnal
Active throughout year

Sparse grassland; subterra-
nean and terrestrial only;
home range 0.24 hectare

Sunlit habitat, low moisture,
temperature often extreme

Seeds, grass, insects

Owls, foxes, weasels, shrews

Surface of ground

Nests of dried grass in bur-
rows, crannies, or clumps of
grass

Nocturnal

Active throughout year

and in its internal dynamics, succession, and distri-
bution. The analysis of the community responses and
interrelations of organisms is a major objective of
tiiis book.

SUMMARY

The environment, or specifically the habi-
tat, of an organism consists of the physical conditions
that surround it. In order to live in a particular habi-
tat, an organism must be morphologically adapted to
it. This may be accomplished to a certain extent dur-
ing growth, especially in sessile forms, but depends

mainly on long evolutionary processes of variation
and natural selection. I"!<ach organism must also be
physiologically adjusted to the various factors of its
environment. Species vary in their limits of toler-
ance, and those factors in their surroundings that are
most immediately unfavorable limit their habitat dis-
tribution. In order for an organism to take advan-
tage of its morphological and physiological adjust-
ments, it must have the proper behavior responses.
These inherited and acquired action patterns involve
selective orientation in response to environmental
stimuli. Occurrence of different species in the same
habitat necessitates the establishment of compatible
community interrelations.

The general nature of responses 1 7

The concept of the biotic community is basic to
an understanding of ecology. We will here be con-
cerned only with laying a foundation of general prin-
ciples. Details will come in later chapters, but for
proper orientation we must know something about
how the community is organized, how it functions,
and how it may be recognized.

INTERNAL STRUCTURE
AND PROCESSES

Communitv and ecosystem

Background:

The Biotic Community —

Structure

and Dynamics

A community, or biocenose, is an aggregate of
organisms which form a distinct ecological unit. Such
a unit may be defined in terms of flora, of fauna, or
both. Community units may be very large, like the
continent-wide coniferous forest, or very small, like
the community of invertebrates and fungi in a de-
caying log. The extent of a community is limited
only by the requirement of a more or less uniform
species composition.

A different community occurs in each diflferent
habitat and environmental unit of larger size, and in
fact the composition and character of the community
is an excellent indicator of the type of environment
that is present. Since plant communities and animal
communities occur together in the same habitat and
have many interrelations, the one can scarcely be con-
sidered independently of the other. Together they
make up the biotic community, and the biotic com-
munity along with its habitat is termed an ecosystem
(Tansley 1935). The ecosystem is the best unit for
the study of the circulation of matter and flow of
energy between organisms and their environment.

Comnumities may be distinguished as major or
minor. Major cotnmunities are those which, together
with their habitats, form more or less complete and
self-sustaining units or ecosystems, except for the
indispensable input of solar energy. Minor communi-
ties, often called societies, are secondary aggregations
within a major community and are not, therefore,
completely independent units as far as circulation of
energy is concerned. When in this book communi-
ties are spoken of the reference is to major communi-
ties unless otherwise indicated.

Dominance

When a number of species come together to
form communities, each fits into a different niche and
plays a different role in the internal dynamics of the
community. Dominance is the relative control ex-
erted by organisms over the species composition of

18

the coiuimmity. Species exerting tliis iiii])ort;mt con-
trol are called dominatUs. Plants are more frequently
dominant in terrestrial communities than arc animals.
In aquatic communities, animals are relatively more
imiwrtant in this role, although dominance is often
not developed.

Dominance is most commonly expressed in the
rfactions of an organism on its habitat (Clements
and Shelford 1939). Dominants shoulder the full im-
pact of the climate or the environment but modify
this effect for other organisms within the community
by tempering light, moisture, space, and other con-
ditions. Only those other organisms that find these
modified physical conditions tolerable can exist within
the community. Furthermore, dominants are ordi-
narily the most prominent species in the community,
make up its greatest mass of living material, and
serve as the major source of food, substrate, and
shelter for the animals that are present. In a forest
community, trees are dominant. They decrease light
intensity, increase the relative humidity, intercept
precipitation, monopolize most of the moisture and
nutrients in the soil, decrease wind velocity, and
furnish shelter and food for animals. Grasses play a
similar, though less conspicuous, role in prairie com-
munities ; sedges, rushes, and cattails in marsh com-
munities ; sagebrush in the arid habitat of the Great
Basin ; mussels and barnacles on a rocky seashore ;
and so forth.

Sometimes dominance is demonstrated in coac-
tions. direct effects of organisms on each other. In some
fresh-water ponds, carp and suckers may consume
much, perhaps all, of the submerged vegetation. This
coaction thus prevents the plant constituents from
assuming their usual role in the community, and by
so much prevents the occurrence of animal species
that depend directly upon the plants. These fish also
react upon the habitat by stirring up the bottom, from
which they derive organic matter, thereby greatly in-
creasing the turbidity of the water. Penetration of
light into the water becomes poor, greatly handicap-
ping sunfish, bass, and other species which locate food
visually (Table 7-3).

In primeval days, bison on our western great
plains fed on the luxuriant taller grasses more ex-
tensively than on the short grasses, with the conse-
quence that, over extensive areas, short grass species
replaced tall grasses almost entirely. Thus bison
were coactant with and dominant over the composi-
tion and character of the community (Larson 1940).
In a similar manner, when European meadow voles
are numerous they reduce the vigor and prevalence
of the grass dominants in consequence of their feed-
ing and tunnelling in the ground, and angiosperms
which are normally absent or scarce appear (Sum-
merhayes 1941). Overpopulations of the European
rabbit alter the character of the forest by frustrating

re|jroduction of oak, beech, and hornbeam, upon the
seedlings of wiiich they feed to the exclusion of other
species. When introduced into Australia, the Euro-
l)ean rabbit converted grassy areas into desert-like
tracts (Mourliere 1956).

Although animals are more common coactors than
plants. i)lants may occasionally exert dominance in
this way. As a notable example, chestnut blight (a
fungus) virtually eliminated the chestnut tree from
the deciduous forest of eastern North America during
the first few decades of the twentieth century. This
fungus infects the cambium, forms pustules under the
bark, and causes the bark to fall off and the leaves
to wilt. The blight has eliminated the chestnut from
the community, and the conse(|uent ojjening u]) of the
canopy has allowed the extensive invasion of new
species of shrubs, herbs, and animal inhabitants.

Trees are the dominants in a forest community,
but species in the lower stratum of shrubs modify
the habitat still further and even the herbs exert some
control over the physical conditions on the surface
of the ground. A subdominant species must tolerate
the conditions established for it by the dominants;
hut it in turn is a modifier of the community composi-
tion in a secondary manner.

Influence

By influence is meant efTect upon the abun-
dance, health, and activities of other organisms in the
community but not to the extent of directly excluding
species. Influence is conspicuously expressed through
coactions, but it may be effected through reactions
as well. Insects may partially or wholly defoliate a
tree ; a pack of wolves may diminish a population of
deer over winter ; squirrels may bury acorns and nuts
and thereby aid germination of them ; parasitic or
poisonous plants may lower the vigor or destroy the
life of some other plants or animals ; animals may
burrow into the soil and thereby increase percolation
of water and air, a benefit to plants : and all organ-
isms, upon death, add organic matter to the habitat.
These and other actions influence the community,
but unless these influences become extreme they do
not absolutely determine whether or not other species
will occur in the community. Influence, then, is of
essentially the same nature as dominance but is less
vigorous in the modifying role that it plays.

Evaluating and rlaxsifying
animals ecologically

One of the most important, yet difficult, prob-
lems in ecology is evaluation of the roles the different
kinds of animals play in community dynamics.

The biotic community

A basis for classifying species is exclusiveness,
fidelity to the community. A species is exclusive
when it occurs only in a single area, habitat, or com-
munity ; characteristic (selective, preferential) when
it is abundant in one area or community but also oc-
curs in small numbers elsewhere ; and ubiquitous
(indifferent) when it is found more or less equably
distributed in a wide variety of communities. The
terms given in parentheses are synonyms used by
plant ecologists (Braun-Blanquet 1932). Exclusive
species are often rare and of little importance in the
dynamics of a community, but when they are con-
spicuous they often make useful indicator species for
identifying and recognizing community units.

The recognition of characteristic species presents
special difficulties, since one must decide how much
more abundant a species needs to be in one com-
munity to be sure a definite preference over another
is indicated. In a distributional study of breeding
bird populations in Ontario (Martin 1960), a species
was considered characteristic of one type of vegeta-
tion if the species was at least three times more abun-
dant in it than in any other type of vegetation. This
was at population levels of from 1 to 9 pairs per 40
hectares (100 acres). For species reaching popula-
tion densities of from 10 to 100 pairs per 40 hectares,
preference was considered demonstrated if the species
were twice as abundant in one type of vegetation as
in any other. For populations greater than 100 pairs
per 40 hectares, differences of 50 per cent are prob-
ably significant. It seems logical that a stricter test
should be applied to small populations, for errors in
measuring the size of populations and random popu-
lation fluctuations attributable to factors other than
choice produce a relatively greater disturbance in the
data. An experimental study in measuring foliage
insect populations also indicated that populations
differing by a ratio of 3:1 could be accepted as sta-
tistically significant (Graves 1953). When the bot-
tom fauna of two ponds were sampled, true differ-
ences could be detected at minimum ratios between
their populations of 1.9 (Hayne and Ball 1956). A
species, to be termed characteristic, should also be
well distributed through a community, this to be in-
dicated by its occurrence in at least 50 per cent of all
samples taken (Thorson in Hedgpeth 1957).

Another criterion for evaluating species is by
numbers of individuals present. Other things being
equal, a species in time of high population affects
other organisms to a much greater extent than it does
at times of low population. A species that is perma-
nently more abundant than another will consume
more food, occupy more nest-sites, and demand more
space ; hence its influence will be greater. Predomi-
nants are the more numerous constituents of a com-
munity, in contrast to members, which are species
of lesser importance. The dividing line between these
two categories is an arbitrary one.

The time and duration of occurrence of a species
in a community affects the amount of influence it ex-
erts. Generally, the longer the yearly period during
which a species is active, the more important its role
becomes. Species may be classified on a temporal
basis into perennials, those which are active in a
community throughout the year, year after year ;
scasonals. which are present or active only part of the
year : and cyclics, which may be very important some
years but of negligible importance other years, as
evidenced by their wide fluctuations in numbers. Even
though present, a species is usually considered inac-
tive when it is hibernating or dormant or when it is
represented only by eggs, spores, or encysted stages
of its life cycle.

The effect produced on the community by indi-
viduals and species may be modified by the way they
form secondary groupings within the community.
These minor aggregations of plants and animals are
called societies and are of various sorts (Shelford
1932). Layer societies occupy different strata, such
as the subterranean, ground, herb, shrub, and tree
societies in a forest ; local societies are usually parts
of layer societies but are more confined in area, as
groups of animals occupying an ant hill, a rotting
log or stump, or a restricted but distinctive area of
ground ; and seasonal societies include all the organ-
isms at particular times of the year.

Other factors that affect the influence of a species
in the community are the size of individuals, their
metabolism, food habits, and general behavior. A
moose consumes more food than a mouse, and a
warm-blooded mouse more than a cold-blooded sala-
mander of the same size. A carnivore at the top of
several food-chains affects the lives of more different
species in the community web of life than a herbivore
feeding on plants. Burrowing rodents react on the
habitat more than do most birds. One factor may
cancel another. An individual of a perennial species
of carnivorous mammal certainly eats more than an
individual cold-blooded herbivorous insect of small
size and active only during the warm season. Yet
there may be 1000 insects to one mammal, so that in
the aggregate a single species of insect may actually
produce more disturbance than a single species of
mammal. The difficulty of evaluating the relative
effects of species is partly alleviated by calculating
their respective biomasses and energy requirements.

The biomass of a species is the average weight
or volume of an average individual, multiplied by the
total number of individuals present. The computa-
tion of the biomass of each species thus corrects for
differences in size and number of individuals between
species. Because of differences between species in
body moisture and amount of inert substances such as
endoskeleton, chitin, shell, and the like, biomass is
expressed with greater accuracy in terms of dry
weight than wet weight or volume, and is even more

20 Background

accuratf if given in terms of carbon or nitrogen con-
tent, or calories.

.Attempts have recently been made to compute
more significant biomasses of l)ird jwiiulations, using
physiological constants. A biomass comiiosed of few
but large individuals has a lower metabolic activity
than an equal biomass composed of a large number
of small individuals. In one study (Turcek 1956),
the importance of different species in the community
was evaluated in terms of the total body surface area
rather than total weight jiresented by each species,
using the formula .V • 10 • li'""''. where .V is the num-
ber of individuals and W the average weight of tiie
species. In another study (Salt 1957) the number of
individuals was multiplied by the mean weight of the
species raised to the 0.7 power (.V- [/'""). One gets
the best evaluation of the importance or influence of
a species ifi a community where metabolic activity
can be measured directly and expressed in terms of
calories per unit of time (Macfadven 1957. Teal
1957).

Produrtivin

.SLJCCE.SSION

Communities are in ;i more or less continual
process of change (Clements 1910). These changes
result in part from the reactions and coactions of the
organisms themselves and in part from such external
forces as changing i)hysiogra|)liy, changing climate,
and organic evolution. The habitat is usually affected
as well as the community, and as the habitat changes,
new species invade it and become established, and old
species di.sappear. These changes are especially no-
ticeable in dominant s])ecies, since these species exert
a controlling role over the composition and structure
of the community as a whole. The replacement of one
comnnmity by another is succession, and succession
continues until a climax or final stage is reached.

-Succession is a process. The series of steps or
communities comjirising a successional sequence lead-
ing to the climax is the sere. Seres are sometimes
classified according to the predominant force that is
bringing them about. These forces are biolic, cli-
matic, physiofnal^hic. and t/eoloc/ic and their resultant
seres are commonly called hioscres. clisercs, eoseres,
and geoscres.

A characteristic of communities that has be-
come of considerable importance in modern ecological
research is productivity. The number of individuals
or biomass present in a community at any one time
is the standing crop. At the beginning of the year
or reproductive season the standing crop is usually
small, but as reproduction and growth take place
there is an increase in the amount of organic matter
making up the biomass of the community. The pro-
duction of organic matter per unit of time and area
is productivity. Productivity is commonly indicated
on a yearly basis, but it is also possible to measure
monthly, weekly, or daily production. Small standing
crops may have a high productivity and large stand-
ing crops a low productivity, hence average bio-
mass or standing crop differentials between different
communities is not comparative of the productivity
of the habitats in which these communities occur.
The largest standing crop which a habitat can sup-
port without deterioration, or the maximum num-
ber of biomass of animals that can survive least favor-
able yet tolerable environmental conditions during a
stated period of time, is the carrying capacity. Carry-
ing capacity is determined not just by the amount of
food available, but also by shelter, social tolerance,
and other factors (Edwards and Fowle 1955). A
variety of methods are being used to measure pro-
ductivity of different kinds of organisms and of dif-
ferent habitats. It is desirable to indicate productivity
as accurately as possible in descriptions or analyses
of community dynamics.

Biotif

suffi's.ston

Biotic succession is brought about by forces
inherent within the community and in the activities
of the plants and animals themselves. The most im-
portant of these activities are the organismal reac-
tions and coactions that produce modifications in the
iiabitat and interrelations between species. Important
reactions involve filling in of ponds with plant and
animal remains, the addition of organic nutrients to
sterile soil, and the reduction in light intensity by
increasing density of plant growth. With progressive
improvement of the soil and changing light and mois-
ture conditions, a series of new dominants come into
the area. When invasion of new species occurs, in-
tense competition develops ; if the invaders are suc-
cessful, the old species disappear as a new community
replaces the old one.

Contributing factors that may be involved are dif-
ferences in growth and dispersal rates, which are
different for different species. After a forest fire or
logging operation, herbaceous plant growth is im-
mediately stimulated ; since herbs grow rapidly, they
become dominant within a year or two. Shrubs begin
to spread, and tree suckers or seedlings also appear
quickly, but because they require a longer time to
reach maturity several years may elapse before they
gain control of the area.

Succession is considerably influenced by the kinds
of propagules available in the vicinity. Seeds, spores,
and the like are dispersed more or less readily, de-
pending on form. Some kinds of animals roam more

The biotic community 21

FIG. 3-1 Early stages in the pond sere: open water, floating
stage, emergent vegetation, swamp shrubs. Everglades National
Park, Florida.

widely or spread more readily into new areas than
do other kinds. The composition of the community
that develops and the rapidity with which it becomes
established depends, in large part, on the rates at
which different species invade.

When the volcanic island of Krakatoa in the East
Indies blew up in 1883, all plant and animal life on
it and on two adjacent islands was destroyed. The
following year, the only living animal reported was a
single small spider. In 1886, it was apparent that the
pioneer plant stage consisted of a crust of blue-green
algae covering the lava. A few mosses were also
present, as were many ferns, and a scattering of some
15 species of flowering plants including 4 species of
grass. The vegetation stage following the algae con-
sisted predominantly of ferns, but the grasses
had become dominant over most of the island. By
1906, woodland had appeared, which has since de-
veloped into an increasingly luxuriant mixed forest.
Dispersal of seeds, spores, and other propagules was
effected by wind, sea, animals, and man, in that de-

scending order of importance. The order in which
the propagules reached the islands greatly influenced
the succession that occurred ; as vegetation developed
it reacted on the soil and habitat, bringing about con-
ditions amenable to the return of the tropical rain
forest (Richards 1952).

Very little study of animal life was made until
1908, and then only for three or four days. At that
time, 202 species were found on Krakatoa and 29 on
a nearby island. There were no earthworms, snakes,
or mammals present, but there were many spiders
and centipedes, a number of insects, 2 species each
of land snails and lizards, and 16 species of birds. A
more thorough survey in 1921 revealed 770 species
of animals including rats, apparently introduced in
1918, and bats, first noticed in 1920. In 1933, 1100
species were found, including 3 species of earth-
worms, but true forest mammals had not y§t appeared
and many families of forest birds were unrepre-
sented. As with plants, the invasion by animals de-
pended on wind, sea, and man and other animals, in
that descending order of importance. Survival and
establishment of the animal species was correlated
with the stage of vegetation that was reached. It is
of interest, however, that scavenger species appeared
first, then omnivores, herbivores, and finally preda-
tors and parasites. Succession of animals depended
in large part on the speed with which they reached
the island and on their finding proper food and shelter
(Dammerman 1948).

Bioseres may be broadly grouped as priseres and
sttbseres, depending on whether they develop on pri-
mary or secondary bare areas. A primary bare area
is a sterile habitat, such as rock, sand, clay, or water.
A secondary bare area is a denudation resulting from
temporary flooding, fire, logging, cultivation, over-
grazing, or other phenomenon that does not produce
an extreme disturbance of the soil or substratum.
Since in the latter the habitat has already sup-
ported community life, and since the soil or sub-
stratum is already in an advanced stage of develop-
ment, the resulting subsere progresses rapidly and the
early pioneer stages of the prisere do not usually oc-
cur. A subsere will develop following the destruction
of any stage in a prisere, and the species composition
of the stages in the subsere will be influenced by the
particular priseral community that was destroyed.

The early stages or communities that make up the
prisere depend largely on the type of bare area on
which the prisere originates. As succession proceeds,
however, later stages in the various seres in any area
having a relatively uniform and humid climate come
to be more and more alike. The successional devel-
opment from widely diversified communities in ini-
tially different habitats to closely similar or identical
climax communities in habitats that have also become
much alike is called convergence.

22 Background

Hioseres may occur on a small scale in microhabi-
tats as well as in major ones. When hay infusions,
prepared in the laboratory, are seeded with repre-
sentative protozoans, the order of appearance of
maximum or peak fwpulations in the various species
is bacteria and monads, Colpoda. hypotrichates,
Paramecium. I'orticcUa. and Amoeba. Disappear-
ance of species is in the same order, except that
Amoeba precede Paramecium and I'orticcUa. Algae
may come in at the tinal stage, so that a more or less
balanced community is established. The succession of
species appears a result of the higher reproductive
rate of earlier species, and to the fact that the excreta
of at least some forms, especially the hypotrichates
and Paramecium, are toxic to them (WoodruiT 1912,
1913, Eddy 1928).

Another common microsere occurs in the death
and decay of trees (Graham 1925, Ingles 1931,
Savely 1939). The sequence of animal species pres-
ent as decay progresses depends on the species of tree,
the community in which the tree occurs, the climate,
and the geographic locality. The following stages
have been recognized : 1 ) tree dying, but still with
leaves and sap; 2) tree recently dead, bark beginning
to loosen, termites and other insects boring into
wood: 3) wood well seasoned, bark very loose or off.

wood borers still predominant ; 4 ) wood softened and
permeated with fungus : fungus beetles, elaterids, and
IKissalids common; 5) wood largely disintegrated
and crumbly, snails and millipedes, occur. Wilson
(1959), working in New Guinea rain forests, sub-
divides stages Z--' in a difTerent manner, each of which
he names after characteristic insects found: 2)
scolytid, 3) cucujid, 4) zora])teran, 5) passalid, and
6) staphylinid. Each stage also has a significantly
different aggregation of resident ants. I'-ventualiy the
decaying log becomes a part of the forest floor, and
the animal species then present are those in general
occurrence.

Climatic succession

With changes in climate, environmental condi-
tions often surpass the limits of tolerance of estab-
lished plants and animals. The result is the replace-
ment of the existent community by another.

A most interesting clisere is the one that has oc-
curred since the northward retreat of the continental
glacier of Pleistocene time (Sears 1948, Deevey 1949,
Table 21-1 ). .Stages in this clisere may be detected

FIG. 3-2 The types of frees
which have lived and died during
fho pasf few thousand years in
Quebec. Instrument taking bor-
ings of lake bottom for pollen
samples is operated from a boat.
Symbols representing different
pollen grains and the percentage
which each species constitutes in
the total are shown at the left.
The type of climate indicated by
the prevailing vegetation at
the time is shown at the right
(Wilson 1952).

The biotic community 23

and even the relative duration of each stage measured
from an analysis of the number and persistence of
different kinds of pollen grains at various depths in
peat bogs. To make such studies, a core of peat is
obtained from the deepest part of the bog by means
of a special hand auger. The lowest portion of the
core is the oldest ; the most recently formed is at the
top. Samples of the core at various depths are suit-
ably prepared, examined under a microscope, and the
pollen grains identified and counted. The predomi-
nant kind of pollen at any level of the core represents
the probable prevailing species of plants in the vicin-
ity at the corresponding period of time, although the
proportionality between all kinds of pollen and abun-
dance of the various species may not be exact (Davis
and Goodlett 1960). Thus, during the last 20,000
years, the clisere in eastern North America is repre-
sented in simplified form by the following climaxes
and climates, reading downward to present time :

spruce, fir cold, moist

pine cool, dry

hemlock, oak, beech warm, moist

oak, hickory warm, dry

beech, oak, hemlock cool, moist

The climate is conjectured from the relation of
similar flora to climate at the present time. The com-
plete clisere occurs only in regions near the southern
limit of the reach of the glacier. The later stages have
not developed in more northern localities where the
glacier has been gone for a shorter time and where
the climate has not warmed up sufficiently. Climatic
succession actually occurs at all levels in the biosere,
as the serai stages leading to one climatic climax is
replaced by the corresponding stages leading to an-
other climax (Table 7-(y).

land and floodplain forests. Stages in the erosion
cycle, as it occurs in a stream, may be discerned by
examination of habitat and animal life progressively
from headwaters to mouth.

Physiographi

succession-

Changes in the earth's surface bring a change
of communities. The sea alternately inundated and
retreated from the Atlantic coastal plain during the
Pleistocene as ice, in which large amounts of water
were tied up, alternately melted and formed. In
earlier times, the sea inundated much of the conti-
nental interior, and on its recession the eosere of
plant and animal communities which developed cov-
ered vast areas.

Mountain-building brings the replacement of low-
land communities with new ones that invade at higher
elevations. As mountains erode, the eosere progresses
in the opposite direction, until base-level or pene-
planation is attained. The development that the
eosere will undergo with continued erosion is some-
times locally apparent in the difference between up-

Geologic

succession

The evolution of new forms of life and dis-
persal of them through the world entails replacement
of pre-existing forms and gradual change in the com-
position and character of communities. The first or-
ganisms to appear on earth were unicellular forms
confined to the sea (Table 3-1). During the Cam-
brian and Ordovician periods, the marine animal life
differentiated rapidly into a rich variety of inverte-
brate types and the anlage of vertebrates. The Si-
lurian and Devonian periods are noteworthy for the
invasion of fresh water and land by both animals and
plants. Fishes became predominant both in fresh
water and the sea. During the remainder of the
Paleozoic era, a luxuriant flora evolved, especially in
swampy areas. Modern conifers, such as spruce, fir,
juniper, tamarack, cypress, and yew made their ap-
pearance. Amphibians became the predominant ad-
vanced animal types, and a diversified invertebrate
fauna occurred in all habitats.

In the Mesozoic era, the existing land flora of
giant rushes, tree-ferns, and cycads gave way to for-
ests of hardwoods which then spread over the world.
Conifers persisted. The earliest woody angiosperms
probably originated in the Jurassic and included sas-
safras and poplar. The forests soon contained elms,
oaks, maples, and magnolias. Herbaceous angio-
sperms, such as grasses and sedges, appeared towards
the end of the era but did not become important in
North America until the drying up of the interior of
tiie continent in the middle Cenozoic. Although the
Mesozoic is predominantly the age of the giant rep-
tiles that lived on the land and in the water and flew
through the air, less conspicuous types such as the
toothed birds, archaic mammals, and insects were de-
veloping rapidly.

At the beginning of the Cenozoic era, ancient
types of animals, including the great reptiles and
many types of invertebrates, became extinct and mam-
als rose to predominance. It is significant that the
rich development of mammals, birds, and insects came
after the worldwide establishment of the angiosperms
with their rich nutrient seeds, fruits, and grasses.
Pleistocene glaciation brought a major change in the
habitat of these animals, and some large mammals
disappeared. The last stage of the geosere, the Re-
cent epoch, brought the dominance of man. Only
future ages will determine whetlier this stage is cli-
max, or whether new and different types of animal
and plant life will someday evolve to replace man and

24 Background

steps

Bdrth. especlall

appli)

(modifio

Dunbor 1949).

Cosmic era : 3-5 billion years ago; tidal disruption of an ancestral sun and origin of

earth.
Azoic era : formation of a stable cold exterior shell to the earth; origin of oceanic de-
pressions and continental platforms; first formation of water and a thin atmosphere.
Archeozoic era : first plant life - bacteria, marine algae, (elevation of Laurentian Up-
lands and peneplation of continents).
Proterozoic era : marine algae abundant; animal life chiefly sponges and segmented

marine worms, (peneplanation of continents).
Paleozoic era : 500 million years ago.

Cambr ian period: marine invertebrates only.

Ordovician period : marine invertebrates continue predominance; rise of armored

fishes.
Silurian period : first invasion of land by plants; rise of air-breathing scorpions

and millipedes and of fresh-water fishes.
Devonian period : first forests and extensive land floras; diversification of fresh-
water fishes, rise of labyrinthodont amphibians, and increase in land fauna,
especially spiders, mites, and wingless insects.
Mississippian period : increase of amphibians.

Pennsylvanian period : luxuriant swamp floras cosmopolitan in distribution, mostly
of spore-bearing types; fresh-water clams and amphibians abundant and on
land, giant insects, spiders, centipedes, snails, and first reptiles.
Permian period : (elevation of Appalachian and Ouachita Mountains); decline of
ancient flora and rise of conifers; modern insects and advanced types of
amphibians and reptiles appear.
Mesozoic era : 200 million years ago .

Triassic period : (desert climates); plants mostly rushes, ferns, cycads, conifers;

stegocephalian amphibians and dinosaurs numerous, archaic mammals appear.
Jurassic period : reptiles evolve higher and more diversified forms, first toothed

birds and frogs appear.
Lower Cretaceous period : woody Angiosperras spreading over world.
Upper Cretaceous period : (great inland seas, warm climate world-wide); modern

genera of deciduous hardwood trees predominant, sedges and grasses appear-
ing; clams and snails common in fresh-water, culmination of dinosaurs,
toothed birds, archaic mammals (elevation of Rocky Mountains).
Cenozoic era : 60-70 million years ago.

Paleocene and Eocene epochs : (inland seas recede, Appalachian region pene-

plained (Schooley) but later again uplifted, climate warm and humid) hardwood
forests predominant, palms abundant; modern mammals and birds replace
archaic forms.
Oligocene epoch : 40 million years ago (continent peneplained); turtles, alligators, croco-
dile at maximum.
Miocene epoch: 29 million years ago (western mountains becoming elevated, climate

turning drier and colder); grasses disperse over open plains; insects reach full de-
velopment and mammal fauna expands ■
Pliocene epoch : 12 million years ago (continued elevation of western mountains, espe-
cially Sierra Nevadas; lower Great Basin becomes arid); grasslands become exten-
sive and desert vegetation develops in southwest; mammals at maximum and man-
ape changing into man.
Pleistocene epoch : 1 million years ago (continental glaciation); great mammals disappear.
Psychozuic era : 26-30 thousand years ago (glaciers recede).

Recent epoch: man becomes predominant, rise of civilization.
Note: The epochs, Paleocene to Pliocene inclusive, are often grouped and designated the Tertiary
Period, and the Pleistocene and Recent epochs the Quaternary Period.

to continue the succession into the indefinite future.
As higher types evolved in each taxonomic group,
primitive forms mostly died out. The first primitive
wingless insects appeared in the Devonian ; in the
Pennsylvanian there were giant forms of primitive
dragonflies, cockroaches, and grasshoppers. One
cockroach had a wingspread of nearly 12 cm. By the
Permian, these giant forms disappeared and were re-
placed by many modern orders. During early Meso-

zoic, most modern families of insects were estab-
lished, and by the Upper Cretaceous many modern
genera are recognizable.

Of the two great groups of warm-blooded ani-
mals, the earliest mammals had originated by the
Triassic. Modern orders did not become well diflfer-
entiated until very late Cretaceous or early Paleocene,
modern families by the Oligocene, and modern genera
by the Pliocene (Simpson 1953).

The biotic community 25

Toothed birds appeared in the Jurassic and dif-
ferentiation of types went on rapidly as modern
orders were already represented in the Lower Cre-
taceous, modern families in the Eocene, modern
genera in the Miocene, and modern species in the
Pleistocene. The great group of songbirds evolved
and dispersed through the world rather late, perhaps
in the Miocene; since the Pleistocene epoch, evolu-
tion of birds has been largely limited to subspecia-
tion.

An important aspect of the geosere is the dispersal
of new types of animals and plants, for the impor-
tance of a new form depends on the extent of its dis-
tribution and the size of its population. A successful
species saturates the available niches at its center of
origin and thence spreads outward in all directions.
Dispersal continues until an impassable barrier is
reached. Tracing the origin of taxonomic groups,
their phylogenetic relations with other taxonomic
groups, and their dispersal is the subject matter of
zoogeography. This knowledge is desirable for the
ecological interpretation of communities, for it helps
to explain their species composition and the geologi-
cal history of the community itself.

The mechanics of evolution at the species level is
called speciation, and is as much a problem of ecol-
ogy as it is of zoogeography and genetics. Speciation
in animals is initiated only when one population be-
comes isolated from another similar population so
that interbreeding does not occur. This permits vari-
ation and natural selection to proceed independently
in the two groups and to become fixed in the germ-
plasm. Isolation is usually effected by geographic
barriers and generally involves occupancy of new
niches and development of new coactions with other
members of the community. The history of the past
ecological relations of species and of whole communi-
ties is the subject matter of paleoecology. The dy-
namic forces involved in these processes of speciation
are among the determinants of the geosere.

The climax

We have described four types of succession as
if each were entirely independent of the other three.
This is not the case. All types of succession are going
on simultaneously, although the relative importance
of any one varies from one habitat to another. Biotic
succession is most conspicuous, since it proceeds most
rapidly and appears to reach a final, permanent stage
in just a few decades or centuries. The climax stage
of the biosere is undoubtedly more nearly stabilized,
self-maintaining, and in steady state in its particular
habitat than are the serai stages, yet it also is sub-
ject to gradual change over long periods of time. The
climax, as well as the serai stages, changes with cli-
mate, physiographic forces, and evolutionary proc-

esses. However, the clisere usually requires a few
thousands of years before changes in the community
structure or composition become evident. The prog-
ress of an eosere is even slower ; geosere, slowest of
all. The climax is defined as the last stage in the
biosere : no absolute stability or final permanency
should be construed, since it is simultaneously a stage
in the clisere, eosere, and geosere.

The climax may be recognized by the fact that
in a uniform climatic area all seres tend to converge
into it, and by its steady state in respect to structure,
species composition, and productivity. In the climax
community, all species, including the dominant spe-
cies, are continually able successfully to reproduce
and there is no evidence that new and different
species are invading. In serai communities, on the
other hand, the developing new growth, particularly
evident in the dominants, contains many individuals
of invading species which will eventually take over
and replace species already present.

RECOGNITION OF COMMUNITIES

The community as an organic entity

Although the major community or ecosystem
is the generally accepted unit of analysis in syneco-
logical studies, there is a difference of opinion as to
whether the community constitutes a discrete organic
entity. Two different points-of-view are incorporated
in the organismic and individualistic concepts which
are usually associated with the names of F. E. Clem-
ents (1916) and H. A. Gleason (1926), respectively,
and more recently Phillips (1934-1935), Tischler
(1931), and Emerson (1952) on one side and
Bodenheimer (1938), Whittaker (1951, 1952, 1956,
1957) and Curtis (Brown and Curtis 1952) on the
other. Ramensky (1926) stated the individualistic
concept independently in Russia as early as 1924.

The organismic concept considers the community
to be a supraorganism, a complex organism, or a so-
cial organism. As such, it is the highest stage in the
organization of living matter ; namely cell, tissue, or-
gan, organ system, organism, species population,
community. There is emergent evolution, so to speak,
at each higher stage in this hierarchy ; the whole is
more than merely the sum of its parts. Tissues have
properties, characteristics, and functions over and
above those of the individual cells involved ; the
organ, the organ system, or whole organism func-
tions in a way not to be predicted from a knowledge
of the parts of which each consists. The species pop-
ulation has inherent characteristics of density, rate of
natality, rate of mortality, and age distribution, while
the total community has such unique functions as
dominance, cooperation, trophic balance, competition,

26 Background

and succession wliicli arc beyond the cliaracteristics
of the individual organisms of wliich the coninninity
is composed. Tlie community behaves as a unit in its
competitional and successional relations with other
connnunities, in its local and geographic distribution,
in its seasonal activities and res]X5nse to climate, and
in its evolution. Although the community varies in its
taxonomic comijosition and structure in different en-
vironmental situations, this variation is proportion-
ally no greater than occurs in different cells of the
same type or between individuals belonging to the
same species.

The individualistic concept places emphasis on the
species, rather than the community, as the essential
unit for analysis of interrelations, activities, distribu-
tion, and evolution. ICach species responds independ-
ently to the integrated influence of the various factors
of the physical environment and biotic coactions.
The environment may be conceived as a pattern of
gradients with the intensity of the various factors
changing gradually in space from one extreme to the
other. The gradient may be a short one, as from the
subterranean to the tree stratum in a forest or from
the open water of a pond to a nearby swamp or cli-
max forest, or it may be longer, as from the bottom
to the top of a mountain or even from the tropical
to the arctic zones of a continent. The population
density of each species is distributed in a form re-
sembling a normal curve when plotted along the
gradient of a given factor, and the curves of many
species in relation to various environmental gradients
overlap in a heterogeneous, and apparently almost
random, manner. There is seldom agreement be-
tween the limits of the distribution curves of any two
species : species are not in general bound together
into groups of associates which must occur together.
Furthermore, the vegetation and its associated animal
life very often form a continuum of gradually chang-
ing composition and complexity from one extreme of
the environmental gradient to the other. A vegeta-
tional continuum has no sharp boundaries between
individual communities of different types, and the
ecologist must choose the manner in which he dis-
tinguishes these units so as best to suit his interests
and objectives.

These two points-of-view are not necessarily in-
compatible. There is no doubt that each species is
distributed according to its own physiology, its own
complex interrelations with other species, and its own
tolerances, and that no two species are exactly alike
in their responses to the environment. On the other
hand, one can be convinced that the community and
its habitat, collectively the ecosystem, is a functional
system and that every species of necessity occurs in
and as a part of such a system so that its distribution
is importantly modified by these interactions and
community relations.

The community is usually recognized and identi-
fied by its most imiwrtant org.inisms, the dominants
and ])redominants (Shelford 1932). Subdominants
and member species, however, are not usually de-
jjendent on the dominant species directly : rather, on
the environmental conditions that the dominants
establish. DifTerent sjjecies of dominants in adjacent
or related communities may react on the environment
in a manner so nearly the same that subordinate spe-
cies find suitable conditions for existence in each,
although they are usually more characteristic of one
than the other.

The niche that a species occupies is a finite unit
of distribution that can be measured in an absolute
and objective manner. The community-stand is an
actual aggregation of organisms occurring in a par-
ticular locality. In a sense, it is a collection of niches
occupied by a particular set of species, but it is some-
thing that one can see and study in the field. Because
of the great variation in composition and character of
community-stands in different habitats and parts of
the world, they need to be evaluated and classified
in some logical manner for reference purposes. Com-
munity-types are abstract groupings of individual
community-stands which resemble one another and
consequently must be defined rather arbitrarily. Dif-
ferent systems of community-types have been pro-
posed, each designed to emphasize a particular point-

15 20 25 30 35 40 45 50 55 60
ELEVATION IN FEET, lOO's

FIG. 3-3 Continuum of tree (e,b,c.d) and foliage Insect (e,f,
g,h,i,i,k) species in an elevation gradient in the Smoky Mountains,
Tennessee, (a) 7sugo canademii; (b) Haleiio Carolina: (c) Acer
spicatum: [d) Fagui grartdilolia: (e) GropAocep/io/o coccineo; (f)
Coec/7/us sp.; (g) Agalliopsis ryovella; (h) Polypioeui corrupiui;
(i) Anospii rula; (j) Cicadella HaYOiCufo; (k) Oncops/s $p. (after
Whittaker 1952).

The biotic community 27

of-view. We are, in this book, using the biome
system, the various parts and concepts of which will
unfold as we proceed. We will consider the com-
munity as being at least analogous to an organism in
being a functional unit of interacting parts and hav-
ing some degree of structural uniformity. Although
community-types are certainly not highly discrete and
absolute units, recognition and naming of them is one
way of indicating positions in the continua along en-
vironmental gradients that are occupied by particular
aggregations of plant and animal species.

Physiognomy

The gross structure of a community or its phys-
iognomy is an important basis for its recognition. In
terrestrial communities, physiognomy is determined
by the life forms of the dominant plant species and
their spacing. The life forms that prevail in a given
area depend on the climate and sometimes the sub-
strate or other special features of the habitat and give
character to the landscape. The distribution of ani-
mal communities is closely correlated with the struc-
ture of the vegetation, hence these vegetation-types
need to be recognized and defined :

Desert : hot, arid habitats with scattered scrubby
or thorny vegetation or, in extreme cases,
none.

Steppe, plains : semi-arid grassland covered with
short grasses.

Prairie : semi-humid grassland covered with mid-
and tall grasses.

Chaparral: semi-arid areas covered with bushes
and shrubs, usually broad-leaved evergreen.

Savanna : grassland with scattered groves of trees
or shrubs.

Woodland: open stand of small deciduous or
evergreen trees with undergrowth of grass-
land or desert vegetation.

Forest-edge : mixture of trees, shrubs, and open
country, ordinarily occurring as a narrow belt
on the margin of forests.

Forest : closed stand of trees forming a continu-
ous canopy over most of the area.

Deciduous forest : broad leaves fall during cold
or dry seasons.

Broad-leaved evergreen forest : no regular sea-
son of leaf fall, leaves often sclerophyllous,
warm climates.

Rain forest : tall luxuriant forests, often with sev-
eral strata of trees, foliage retained through-

FIG. 3-4 A mixed deciduous-coniferous plant community (after Dansereau 1951). Above, a semi-realistic diagram of the community;
below, symbolic structure of the community depicting life-form, size, function, leaf type, and texture.

"WWW Um^

28 Background

out till' year. cliTiiate contimunisly waiiii and
wet.

Needle-leaved evergreen or coniferous forest :
l'ore.st.>; of ])iiu's. s])riices. t'lrs. larciies, lieiu-
k.cks. ami tlie like.

Forest-tundra: stunted o\w\\ k''"^^''' of conifer-
ous forests in cold climates.

Tundra: extensive flat or gently rolling treeless
areas occurring in cold climates.

Alpine tundra : treeless areas at liiglier elevations
of moinitains.

Bog: wet areas in northern climates containing
sphagnum, heath plants, coniferous trees.

Swamp : wet areas covered with trees or shrubs.

Marsh: wet areas containing sedges, rushes, cat-
tails, and the like.

Inasmuch as animals choose niches in response
primarily to the physical structure of the vegetation
regardless of its taxonomic composition, it is helpful
in describing biotic communities to show the vegeta-
tion structure in as much detail as possible. This
may be done by semi-realistic diagrams or by a sys-
tem of symbols (Dansereau 19.S1).

The 50

per

cent rule

If the primary basis for community recognition
is based on the life-form of the dominants, which on
land is expressed in the physiognomy of the vegeta-
tion and in some aquatic habitats on the life form of
the predominant animals, then the secondary break-
down of community units must be on the basis of tax-
onomic units. Here, the species unit is most useful,
as the species is the smallest taxon having objective
reality and precise interrelations with its environ-
ment.

Two aggregations of species occurring naturally
in different areas or in the same area at different
times are to be considered as distinct communities
when at least 50 per cent of the predominant species
of each aggregation are if not exclusive at least char-
acteristic to the aggregation. This we may call the
50 per cent rule. The recognition of communities
should not be influenced by the presence of rare
species, for such are near the boundary of their habi-
tat or geographic range. It is important to have quanti-
tative information on the size of the populations to
evaluate the importance of each species before com-
munity classification is attempted (Sparck 1935).

The distinctiveness of communities must work in
both directions ; that is to say, 50 per cent or more
of the important species of each aggregation must be
different from the other aggregation. This means
that the two aggregations are more different than
they are alike. If the species composition does not

exliihit the 50 i)er cent distinction, the two aggrega-
tions are considered as belonging to the same com-
nuinity. If the difference ap|)ro;iches but does not
e(|iial 50 per cent, it is often worthwhile to designate
the two aggregations as jacifs if they are serai, or
farialions if they are climax, of the s.une community.
It is i^referable to u.se this criterion for differentiating
communities, in the light of present ecological knowl-
edge, rather than use more involved statistical cri-
teria (Hray 1956). The 50 per cent rule has been
earlier ap])lied for separating zoogcographic regions
(Mavr 1944).

!\ anting lommunities

Since comnnmities are distinguished by differ-
ences in life form and taxonomic composition of the
dominant or predominant organisms, these charac-
teristics are usually used also in naming the com-
munity. Where the habitat is well defined but vege-
tation is largely or wholly lacking, as in many acjuatic
communities, habitat may be used in the terminology.
.Since names are largely a matter of convenience,
they should be short and be derived from some easily
recognized feature of the community or habitat. Very
often the generic names of two, sometimes three,
conspicuous dominants are used to name plant com-
munities ; two or three predominant characteristic or
exclusive animal species, together with the prevail-
ing type of vegetation or habitat, are employed to
name animal communities. In case of some large
communities, geographic names are more convenient.

Large geographic units, differentiated on the basis
of difference in the climax type of vegetation, are
called hiomes. They are specifically named by the
characteristic form of vegetation present : tundra
biome, or grassland biome, for instance.

Secondary communities within the biome can be
distinguished as climax or serai, respectively, by the
suffixes -iation and -ies. An association is a climax
plant community identified by the combination of
dominant species present ; an associes is an equiva-
lent serai plant community. Thus we may speak, for
instances, of the Fagits-Acer association, which is a
climax deciduous forest community, and of the Ca-
lamoyrostis-Andropogon associes, which is a grass
stage in a sand sere (Clements and Shelford 1939).

Animal communities on land are related to differ-
ent life-forms of plants or types of vegetation, but
only seldom to plant communities distingui.shed by
the taxonomic composition of the plant dominants.
Thus animal communities must be analyzed and
named independently of plant communities. A bi-
ociation is a climax animal or biotic community
identified by the distinctiveness of the predominant
animal species ; a hiocics is the serai equivalent. The

The biotic community 29

North American deciduous forest biociation is to
be contrasted, for instance, with the pond-marsh
biocies.

When two plant or animal communities merge,
either by interminghng of species in the same habitat
or by juxtaposition of different communities in the
same region, the resultant transitional state is called
an ecotone. Ecotones occur between consecutive com-
munities in serai development on an area as well as
between adjacent existing local or geographic com-
munities.

SUMMARY

A community is an aggregation of organ-
isms in a distinctive combination of species. The
community and the habitat in which it occurs con-
stitute an ecosystem. Inherent within the community
are forces of dominance which control the species
composition, and of influence which affects the abun-
dance, health, and activities of organisms. Dominance
is exerted primarily through reactions of organisms
on the habitat, influence primarily by coactions of
organisms on one another. The relative importance
of each of the various species within the community
is evaluated on the bases of exclusiveness, abundance,
time of activity, secondary groupings, and influence.
Reproduction and growth brings a production of or-
ganic matter ; the rate at which formation of it takes
place is called productivity.

Communities are constantly changing, the result
of reactions and coactions of the organisms, and cli-
matic, physiographic, and evolutionary processes.
This change is one of succession, an orderly replace-
ment of one community by another until a climax,
especially evident in bioseres, is reached.

The community may be considered as a highly in-
tegrated self-contained organic unit or as merely an
aggregation of independent species whose preferanda
coincide in the same habitat. These are extreme
points-of-view ; an intermediate one is adopted in this
book.

The gross structure of the community is the pri-
mary basis for distinguishing and recognizing it. On
land, this structure is characterized by type of vege-
tation ; in water, by the life-form of the predominant
organisms, which are usually animals. Communities
are then subdivided according to their taxonomic
composition. An aggregation of species is given com-
munity status if at least 50 per cent of the predomi-
nant species are exclusive to or characteristic of it.
Animal communities are named for the type of vege-
tation, life-form of predominating species, or habitat,
depending on which is the most conspicuous feature ;
and secondarily for the predominant two or three ex-
clusive or characteristic species that it contains or
for the geographic area in which it occurs. Biomes
are major geographic community units. Biociations
are secondary climax communities distinguished by
the distinctiveness of their predominant animal spe-
cies. Biocies are the serai equivalents of biociations.

30 Background

4

Background:

Tlic analysis of ecological comiiuinitics must in-
diulf a ineasurcnicnt of animal |jo|)iilations that the
role |)laye(l by i-ach sjifcies may be \iropeT\y evalu-
ated. The ecologist should also be able to determine
i|uantitati\ely the iihundance of species at difTerent
times and different places. It is not sufficient in eco-
logical research to indicate that a species is abundant,
common, or rare; abundance must be expressed in
such objective terms as lend themselves to statistical
manipulations. In spite of their fundamental imjxjr-
tance, available methods for measuring population
size are only moderately satisfactory and are in need
of vast improvement (Balogh 1958, Davis in Mosby
1960).

Indices of abundance are sometimes used; for
instance the number of individuals or songs observed
per hour, per day, or per trip; per cent (frequency)
of samples in vk-hich the species was recorded ; num-
ber of nests, dens, tracks, or fecal pellets per unit
area; amount of food or bait consumed per unit of
time, and so forth. Under certain conditions of uni-
form habitat and weather, random distribution of
individuals, and uniform conspicuousness of the ani-
mals, indices are useful for demonstrating differences
in population size within a single species as functions
of time or space, but they are seldom accurate enough
to allow comparisons between different species.
There have been various attempts to correlate rela-
tive indices with absolute abundance (Hendrickson
1939, Bennett et al. 1940, Cahalane 1941, Baum-
gartner 1938, Emlen ct al. 1949, Eberhardt and Van
Etten 1956), but the results have been usually un-
satisfactory (Clapham 1936, Dice 1952). In most
types of ecological research, the aim should be to de-
termine absolute abundance or the actual number or
biomass of a species in an area of known size. The
difficulty in doing so is no greater than in correcting
relative indices for all the variables that are involved.

STRIP CENSUSES

Measurement
of Populations

Tliis method is one of counting all indi-
viduals of birds and larger mammals seen on each
side of a line of travel over a measured distance.
Sometimes the count is made only of animals ob-
served within a definite distance from the line of
travel. In other cases, the effective width, and hence
the area, over which the animals are being censused
is computed as twice the average distance at which
each species is first observed. This makes possible a
quick survey of large areas in any kind of terrain,
but is subject to inaccuracies of individuals omitted,
especially as the distance from the trail increases
(Hayne 1949a); differences in conspicuousness of
different species or individuals e.\hibiting atypical or
unusual behavior; and variations in visibility as one

31

passes from one type of terrain or vegetation to an-
other.

A variation of this method, often employed for
counting larger animals such as deer, is to increase
the width of the census strip by using a line of many
observers that progresses uniformly over an area of
previously fixed dimension. The animals are counted
as they are driven back through the line or out be-
tween other observers stationed along the boundary
(Rasmussen and Doman 1943). Helicopters may be
efifectively used for counting large animals in open
country (Aldous 1956) ; faster flying aircraft are less
successful (Gilbert and Grieb 1957).

SAMPLE PLOTS

Since it is seldom possible to count all the
individuals present in a large area, it becomes neces-
sary to take sample counts over small areas where
accurate counting of individuals is practical. The
problem then arises as to the number, size, shape, and
distribution of plots required to give reliable infor-
mation on species composition and the mean density
for all the organisms involved. Much work on this
problem has been done by plant ecologists, and their
techniques should also be of use to animal ecologists.

Plot distribution and shape

Sample plots may be distributed either sys-
tematically or at random. Systematic arrangement of
plots of uniform size spaced at equal intervals along
straight lines is often preferred because of its easy
application. However, if the distribution of organ-
isms over the area shows a uniform pattern of vari-
ation, systematic sampling may indicate densities
either too high or too low. Furthermore, systematic
sampling does not permit the assessment of error,
since statistical theory requires that the location of
each sampling unit be independently determined,
whereas in systematic sampling, the position of all
plots is determined by the location of the first one.
The completely random location of sample plots over
an area may be somewhat more difficult to apply in
the field, but the data obtained are just as precise
and have the advantage that the error of sampling
can be calculated (Bourdeau 1953). In order to get
randomly located sample plots, a map of the entire
area is subdivided into numbered plots of the proper
size. The plots to be used are then selected by using
tables of random numbers. If the same number comes
up twice, the duplication should be discarded (Dice
1952). Other plans of sampling, such as stratified
random sampling, may sometimes be preferable.

Where a habitat is perfectly uniform, the shape
of a sample plot is not of great importance, although
square plots are commonly used. Where a habitat is
obviously not uniform a rectangular plot oriented with
its long axis across any observed contour-, soil-, or
vegetation banding will furnish less variable data
than plots that are shorter and wider (Bormann
1953). Circular plots, which possess a smaller
periphery than any other shape, are useful where the
influx and exit of animals must be minimized.

The size of plot suf^cient to include an ade-
quate sampling of the species composition of a par-
ticular local community varies with species involved
and density of populations. Larger plots must be
used for larger organisms, richer fauna, for situations
in which one or a few species are so markedly pre-
dominant that minor species are scattered, and where
population levels generally are low. Since the number
of species included will vary with the size of the area
covered in sampling, some standardization is desir-
able for comparing the species composition of dififer-
ent communities.

A standard size for sampling plots may be deter-
mined empirically (Vestal 1949). If the numbers of
species found on plots of different sizes are plotted
against the logarithms of the plot sizes, a sigmoid so-
called species-area curve is formed. The characteris-
tics of this curve are that an increase in the size of
small sampling plots includes, at first, a considerably
larger number of species, but later a size of plot is
reached, varying with the kind of organisms being
counted, beyond which there is little to be gained by
increasing the area sampled. Two arbitrarily chosen
points on the upper part of this curve, where it is
concave toward the scale of plot size, have ecological
significance. One of these points represents a plot
fifty times the size of the other, containing twice the
number of species of the other. The larger plot is
close to the upper asymptote of the curve and repre-
sents a fair-sized sample plot for practically all pur-
poses. The smaller area, located near the point of
inflection and containing half the number of species,
is the smallest representative area that is sufficient to
identify the community, but hardly usable for any
other purpose. A third point may be identified, mid-
way between these two points on the curve, as the
minimum area large enough to include all important
species and about half of the minor ones. It clearly
defines the community and the approximate ranking
of species in points of number and biomass. The area
this intermediate point represents is five times the
smallest representative area and one-tenth the fair-
sized area.

32 Background

FIG. 4-la. b Spoeles-area curves plotted (above) on arithmetic
and (below) on logarithmic bases, to illustrate the method of
determining sampling area sizes adequate for analysis of com-
munity species composition.

30
25

-

_,''' Fair-sized

5
^20

-

yf Mrnimum area

fe

/

a. 15

-

A\

i,o

-

/

^ Smollesl representalive area

5

„^

X'

1 1

1 1 Mill 1

1 1 1 lllll 1 1 M lllll 1 1 1 1 1 III

10

100 1000 lOP

The sizes of "Smallest representative. "
"Minimum," and "Fair-sized" areas may be de-
termined from a sigmoid curve — of which Fig.
4- lb is an example — by the following technique.

A rectangular sheet of tracing paper is
placed over the graph, the bottom edge of the
paper coincident with the horizontal axis of the
graph. Using the graduations on the log-
arithmic scale as a guide, place a mark on the
bottom edge of the paper by one of the
graduations. To the right of that mark, place
another by that scale graduation which is 50
times the value of the first (on Fig. 4- lb, if
the left mark Is at 2. the right mark should bo at
100). The interval between the marks represents
a 50-fold Increase in area. From the right mark,
draw a vertical line several Inches long. Now
place the sheet over the graph in such a way that
the bottom edge Is parallel to the horizontal
axis of the graph, and the left mark lies on the
curve — the vertical line from the right mark
should be long enough so that it will continuously
Intersect the curve. Keeping the bottom edge of
the paper parallel to the horizontal axis of the
graph, move the paper in such a way that the
left mark traces along the curve. Move the sheet
until the vertical line intersects the curve at a
point which is twice the value of the point at
which the left mark Is resting on the curve, os
6o//> vo/ues ore reoc/ off t/ie veriicat scale (on
Fig. 4- lb. when the left mark rests on the curve
at a point opposite 15 on the vertical scale, and
the bottom edge of the tracing paper is hori-
zontal to the horizontal axis of the graph, the
vertical line intersects the curve at a point
opposite 30 on the vertical scale). The point
established by the left mark is the value of the
"Smallest representative area." and the point
established by the perpendicular line Intersect-
ing the curve is the value of the "Fair-sized area,"
OS both values are read m area unifs off the
horaonial scale. Find, by measuring, the point
on the curve which lies midway between the two
points just found — this is the "Minimum area,"
and its value is read in area units off the hori-
zontal scale (after Vestal 1949).

-ACRES (Units of '

Measurement of populations 33

In ecological sampling, fair-sized areas should be
used wherever possible, but minimum-sized areas are
sometimes acceptable. For evaluation of this and
other procedures, see Goodall (1952). When the
number of species encountered on several randomly-
distributed sample plots is known, it is possible to
estimate statistically the actual number of species
present in the whole area (Evans, Clark, and Brand
1955).

The size of the plot should also be adequate to in-
clude an accurate representation of the population
densities of the various species present. Much of the
difficulty of accurately determining population densi-
ties results from populations being non-randomly dis-
tributed in the space they could occupy (Cole 1946).
To be randomly distributed, populations must have
been scattered by chance rather than coercion, re-
gardless of the proximity or distance one from an-
other. This seldom occurs either with plants or ani-
mals. Plants reproduce by rhizomes, stolons, or suck-
ers or by seeds concentrated near the parent plants.
Animals usually lay eggs or drop young in local areas
or nests, so offspring are at least temporarily con-
centrated. Many animals congregate socially or
form colonies, concentrate on some local food supply,
or are grouped closely together in certain microhabi-
tats because of less favorable environmental condi-
tions elsewhere. Even the attraction of male to fe-
male for reproductive purposes is a variation from
random dispersal. Whenever the occurrence of one
or more organisms in an area increases the likeli-
hood that other organisms will occur nearby, this is
spoken of as contagious distribution. Species may
also exhibit negatively contagious distributions when
they are spaced more regularly than would be ex-
pected by chance, as for instance flocking or colonial
birds where each individual keeps just beyond the
pecking reach of its neighbor.

When small-sized plots are used, contagious dis-
tribution shows itself in an excessive number of plots
containing no individual and of plots containing a
large number of individuals with a corresponding
deficit of plots with intermediate numbers of indi-
viduals. This represents a deviation from the typical
Poisson distribution which is expected with random
distribution (Snedecor 1956). In a Poisson series,
the mean number of individuals per quadrat should
equal the variance according to the formula

S(^-J)2 _

x(n-l) -'

The letter .r is the number in each quadrat, x is the
mean number in all quadrats, and n is the number
of quadrats. If the value obtained is significantly
greater than unity, then contagious distribution is
indicated, if the value is less than unity, then nega-
tively contagious distribution is indicated. For a

reasonably large number of sample quadrats, say 20
or more, a deviation from unity would be considered
significant if it were greater than 2\/2n/{n — 1)^
( Andre wartha and Birch 1954).

With contagious distribution of individuals, the
aggregates themselves are often randomly distrib-
uted, in which case quadrats may be increased in
size until they give a random distribution of aggre-
gates rather than of individuals. The total popula-
tion woidd then be computed by multiplying the num-
ber of aggregates per unit area by the average
number of individuals per aggregate. When aggre-
gation occurs but is not easily observed, then other
procedures must be employed (Cole 1946a, Goodall
1952).

Number

The number of sample plots needed depends
upon the precision desired for the statistical char-
acteristics to be estimated. The degree of precision
required will vary with the trustworthiness of the
data and the objectives of the study. In most sta-
tistical investigations, a range of 20 to 40 replications
is ample (Snedecor 1956: p. 104). Too few replica-
tions may fail to detect important differences, but too
many are unrewardingly wasteful of time and energy.
.Any differences noted between population densities
of different species on the same area, or of the same
species on different areas or at different times, should
be significant at least at the 5 per cent level of sta-
tistical probability. Where the number of samples is
small, the differences must be relatively large to insure
this level of confidence. With ecological studies in
the field, there are often practical difficulties involved
in obtaining a sufficient number of accurate measure-
ments to permit reliance on minor differences in pop-
ulation size. It is best, therefore, to be conservative
in evaluating the importance of differences in popula-
tion densities. Special care must be used in evaluat-
ing the densities of rare species, as such densities are
unlikely to be reliable if based on counts of less than
20 or 30 individuals (Preston 1948).

CAPTURE-RECAPTURE METHOD

Some general methods of calculating popu-
lation densities need to be considered. C. G. J. Peter-
sen, of the Danish Biological Station, working with
fish in 1896 ; F. C. Lincoln, of the US Fish and Wild-
life Service in 1930, trying to estimate the number
of ducks on the North American continent ; and
Jackson (1933), working with insects, all independ-
ently derived a formula for determining the popula-
tion size of various species of animals, much used in

34 Background

recent years (Kicker 1948). Tlie method deiieiuls
first on capturing a fair sample of individuals in a
unit area, marking them in some distinctive manner
(Ecol. 37, 1956: 665-689), releasing them for redif-
fusion over the area, then after a short interval, re-
trajjping the area. The ratio of marked individuals
reca])tured to the total numher marked sliould theo-
retically be the same as the total marked and un-
marked animals captured during the second trapping
is to the total population or :

Total ]ioini]atioii

il numher marked

marked individuals
X total captured

Other formulas make use of accumulating totals of
marked and unmarked individuals during successive
periods of trapping (DeLury 1958).

The greater the percentage of the population
marked and subsequently reca])tured, the greater is
the accuracy of the calculations. However, there are
several possible, uncontrollable sources of error : un-
equal mortality of marked compared with unmarked
individuals ; dispersal of individuals out of the area,
influx of animals from outside : increase by reason of
reproduction, marked animals not becoming randomly
distributed among the unmarked ; marked animals
being recaptured with greater or less ease than un-
marked ones : marks being lost or not reported, and
so forth. Some of these possible errors can be cor-
rected statistically, and a considerable body of litera-
ture has accumulated describing means of so doing
(see especially Bioiiietrika since 1951).

CAPTURE PER LLMT-EFFORT

In a closed or stabilized population, when
the same time, traps, and effort are employed to cap-
ture or count individuals in the same area at difYerent
times and there is no loss or increment in the
original population, and weather and other conditions
remain the same, the number of new individuals cap-
tured or discovered witii each subsequent effort be-
comes less and less, and should eventually reach zero.
When the number of new individuals captured per
unit of effort is plotted against the cumulative num-
ber of animals captured, a straight line results. A
line thus derived from a few catches may be extended
to zero, and the total population of animals in the
area determined (DeLury 1947, Zippin 1958). A
variation of this method is to use the increasing per-
centage of marked animals in the total number cap-
tured at successive intervals of time, as the increase
in these percentages follows a definite trend that
would eventually include the total population (Hayne
1949).

ai'PI.icahon to am.mal groups

Tiie more conspicuous diurnal mammals are
commonly censused by cruising or drives, but noc-
turnal forms, es])ecially mice and shrews, usually have
to be trapped (but see Kmlen ct al. 1957). Snap or
kill traps are commonly used. When set in a variety
of microhabitats they quickly gather specimens to
show the species composition of the community. An
early attempt at estimating abundance was expressed
in terms of the number of individuals caught per trap
per night. At the same site, more animals are usu-
ally caught during the first night than during later
nights: 10 traps set for 10 nights will not capture as
many small mammals as 100 traps set for only one
night, although 100 trap-nights are involved in both
instances. When the trapping procedure is standard-
ized as to location in community, number of traps
used, interval between traps, length of trap lines,
number of nights trapping, and so on as has been
done in the Xorth American Census of Small Mam-
mals (Calhoun 1956), it is possible to follow changes
in relative abundance from year to year. It is not
])ossible to relate such data to the absolute number
per unit area unless the home range of each species
in each locality is known (Stickel 1948).

The next advance in censusing technique was to
confine the location of kill traps to a small area, usu-
ally an acre (0.4 hectare). Enough traps, a hundred
or more, are included to saturate the area to the end

100

200 400 600 800 1000 1200
TOTAL CATCH TO DATE, m

FIG. 42 Total population {K = 1170) calculated by extension
of a straight line through data on successive catches per unit
effort, C(/), plotted against the accumulating total catch, K(i)
(from DeLury 1947).

Measurement of populations 35

FIG. 4-3

Individual ter

itori

s of birds

epr£

senting two

ompet-

Ing specie

s, wood pewee

(stippled) and

least flycatcher,

along a

forest trai

. Note that te

ritor

es of indivi

dual

of the same

species

do not ov

eriap, and that territories of th

e tw

species are

largely

but not er

tirely exclusive

The

outline of

ach

territory is b

ased on

observatio

ns made from

the

numbered

poin

ts. The dates

of the

several dc

ta-collection

rips

are shown

for

two territories only

(Kendeigh

1956).

of capturing all individuals present during a trapping
period of three nights (Bole 1939). Influx and de-
parture of animals, however, disturbs the accuracy
of the measurement. Influx is usually more of a prob-
lem than escape from the area, as the trap bait and re-
moval of captured individuals encourages invasion
(Stickel 1946). Since all animals whose home ranges
approach or overlap the boundary of the trapping
area are likely to be caught, a correction for this error
may be made by considering the census area to in-
clude a surrounding belt equal to one-half of the home
range of each species concerned (Dice 1952). In
order to reduce the boundary of contact with the out-
side area to a minimum, square or circular areas are
used, rather than rectangular or irregular-shaped
areas. Censuses taken in this manner and live-trap
censuses sometimes give comparable results (R. M.
Wetzel 1949, Buckner 1957), but in neither case can
one usually be certain that he has captured all the
inhabitants of the area (Fowle and Edwards 1954).

Live trapping, marking, and release of individuals
is a more trustworthy means of censusing small mam-
mals, but is more laborious and time-consuming
(Blair 1941, Stickel 1946). Traps are usually dis-
tributed grid fashion at intervals of 15 to 20 meters,
over several acres or hectares. Trapping is continued
for a week, or until very few or no unmarked ani-
mals are captured. Marking is commonly toe clip-
ping, ear notching, tattooing, or tags (Taber 1956).
Since the animals are immediately freed, the popula-
tion equilibrium is not greatly disturbed, and influx
of extraneous individuals is negligible. The method
has the further advantage of allowing the determina-
tion of home ranges. Individual animals differ, how-
ever, in the readiness with which they will enter traps
(Geis 1955), and this will affect the accurate deter-
mination of home ranges.

The type of bait used varies with the species be-
ing trapped. Seasonal fluctuations in numbers of ani-
mals trapped may sometimes be due to variability in
the acceptance of bait (Fitch 1954). For mice and
shrews, a paste made of peanut butter, oatmeal flakes,
and raisins is commonly used.

The most recent and promising development of
technique for determining home ranges is the label-
ing of individuals with radioactive material, then fol-
lowing the movements of the freed animals by use of
geiger counters (Godfrey 1954, Pendleton 1956,
L. S. Miller 1957, Harrison 1958). This procedure
has also been used with amphibians ( Karlstrom
1957).

Birds

Airplanes have come into common use for cen-
susing large concentrations of waterfowl. Aerial pho-
tographs are made and enlarged, and individual birds

36 Background

l)iii-])()iiUe(l. Roadsidi' counts, calling-male transects,
indices derived from i>opiilation structure, kill rec-
ords, and a variety of other procedures are used to
inventory upland game species (Hickey 1955).

I'or determining populations of smaller sjiecies
during the nesting season, the spot-map method is
commonly used and censuses thus obtained are prob-
ably reliable within plus or minus 10 per cent, if they
are carefully made (Kendeigh 1944). A sample jjlot
of uniform vegetation of at least 10 hectares {2S
acres) is marked out in a grid with numbered stakes
or tree tags at intervals of not over 50 meters, or the
stakes may be placed along a trail. At least five,
preferably more, daily counts of singing males, fe-
males, and nests are made at suitable intervals
throughout the nesting season. Kach time a bird is
observed it is marked on a map of the plot. At the
end of the season all the spots at which a species was
observed are ])laced together on one map. Since in-
dividual birds are observed most frequently in the
vicinity of their nests and within their territories, the
spots fall naturally into groups so that each group
indicates the presence of a breeding pair or at least
a territorial male. Counting the number of groups of
spots for each species gives the total population for
the area. For the large predators, gallinaceous birds,
or wide ranging species, census plots of much larger
size are necessary than for the smaller song birds, so
that procedures must be adjusted to the conditions
of the habitat and the species involved. For detailed
studies of small populations, the birds should be
handed and color-marked for individual recognition
(Hickev 1943).

Foliage arthropods

In order to determine the insect and spider
composition in the herb, shrub, and tree strata of a
forest, use of a variety of collecting methods is de-
sirable : net sweepings. light traps, bait traps, ad-
iiesive snares, and the like (HofTmann ct al. 1949,
Morris 1960). Some of these methods may be made
semiquantitative to show relative abundance, but
there is considerable difficulty in converting the data
obtained into absolute abundance.

The use of the sweep net can be standardized to
give useful and comparable estimates of population
densities (Carpenter 1936). A series of 48 strokes
of the net through the upper level of the herb stratum
synchronized with one's pace so that successive
strokes do not hit the same plants gives approxi-
mately the same number of individuals as one would
find on the herbs covering one square meter if all
could be captured. The net should have a diameter
of 33 cm (13 in.), the strokes should be about one
meter long (.Shelford 1951a), and comparative sam-

ples should be taken at appro.ximately the same time
of day (Adams 1941). A similar number of strokes
through the shrub foliage may be used, but the con-
version to number of individuals per S(|uare meter
depends on the extent and uniformity of the shrubs
that cover the ground. Inaccuracies involvetl in
sweep net sam|)ling are the result of variations in the
activity of the insects and s])iders ])roduced by
changes in teni])erature, wind, and humidity ; varia-
tions in position of the insects on the plants and hence
exposure to capture ; insects taking flight in advance
of the collector; variations in the height of the herbs;
and variations in the length and rapidity of the
strokes (DeLong 1932. Hughes 1955). Differences
between sexes and species in behavior and life his-
tory will also cause variations in the sampling effec-
tiveness.

Tests on the reliability of population estimates of
single species based on the sweep-net method, made
by comparing the results of two different workers in
the same woods at the same time, showed an agree-
ment within 100 per cent in only 36 per cent of com-
parisons between single weekly collections, but in 74
per cent of comparisons between averages of weekly
collections taken over the entire summer (Graves
1953). This would indicate that variations in popu-
lation estimates obtained by sweep-net samples are
not significant unless a good series of data is ob-
tained, and only then when differences between aver-
ages amount to more than 100 to 200 per cent; i.e.,
when the larger population is at least 2 or 3 times
the size of the smaller. .Xctually. variations in popu-
lation size of the same species of insect or spider at
different times or in different communities m.iy
amount to several hundred per cent, and hence the
sweep net method is useful for quantitative studies.

Sampling of arthropods in the tree canopy is
more difficult. In the absence of wind, small trees
can be jarred or shaken so that released animals fall
on a cloth spread beneath. With proper e(|uipment,
trees may be fumigated with such poisonous sprays
as DDT so that the dead insects fall onto cloths
spread below. To put the data on a comparative
basis, the volume of the space occupied by the foliage
may be measured or estimated, and the number of
individuals per cubic meter calculated. In deciduous
forests of eastern Xorth .America the tree canopy is
commonly about 10 m thick. A useful standard for
comparison with the numbers per square meter of
ground, herbs, and shrubs is the number per 10 nr''.

With taller trees, samples of the foliage for visual
counting of the immature stages of arthropods pres-
ent may be collected with the aid of aluminum pole
pruners and extension ladders, or from trestles or
platforms. Foliage samples, especially of coniferous
species, should consist of entire branches or longi-
tudinal halves, since arthropods may vary in abun-

Measurement of populations 37

dance from the newer apical growth to the older basal
foliage. If the width of the branch at mid-length is
measured, then the length of the foliated part times
the width gives the foliage surface. The total foliage
surface of representative trees is determined from
felled individuals, and the total foliage surface per
unit area may be computed from the known density
of trees. If the arthropods vary in abundance at dif-
ferent levels in the tree, representative sampling must
be taken at each level. Considerable variation in ani-
mal density also occurs from tree to tree so that sam-
pling must be well distributed over the area under
investigation (Morris 1960).

Soil animals

One must resort to a variety of methods to
census the difTerent kinds of animals in the soil be-
cause of great differences in their size, physical char-
acteristics, and behavior (Fenton 1947, Van der Drift
1950, Kevan 1955). The uiegafaitna consists of the
larger millipedes, centipedes, snails, amphibians, rep-
tiles, and small mammals. Mammals must usually he
trapped. For the other forms mentioned, if there are
a half-dozen workers available, a plot 10 meters on
a side (100 m-) may be marked out and the ob-
servers, forming a line at one side, may gradually
work over the plot, turning over all the leaves and
sticks. This gives a good count but must be repeated
in various parts of the community.

FIG. 4 4 Tullgren modification of
a Berlese funnel for quantitative
sampling of soil animals.

For quantitative sampling the fauna of fallen logs
and decaying stumps, it is convenient to mark out an
area 50 meters on a side (0.25 hectare) and then
measure the length of all logs and the height of
stumps. A medium-sized log and stump are com-
pletely torn apart and all animals counted. The total
population for the whole area may then be calcu-
lated.

The inacrofauiia, consisting of the larger insects
and spiders, the smaller millipedes, centipedes, and
snails, and the earthworms may be censused by means
of a steel ring, 7.5 cm wide, having a sharpened edge,
and covering 0.1 m-, which is pressed into the ground
until it is flush with the surface. The litter and top
10-12 cm of the soil, which contain most of the
ground animals, may then be sorted by hand, either
in the field or in the laboratory. Samples brought in
from the field should be transported in paper or
plastic sacks or tight containers and sorted as soon
as possible before the predatory animals in the sam-
ple have consumed prey species.

Lumbricid earthworms commonly penetrate well
below the topmost few centimeters of the soil. Hand-
sorting of considerable amounts of soil is both la-
borious and time consuming, but gives the most de-
pendable results. In one study, small sample plots
were thoroughly soaked with a potassium permanga-
nate solution, and a later check by hand sorting indi-
cated that 80 per cent of the adult and 100 per cent
of the immature worms were forced to the surface
(Evans and Guild 1947). Attempts at earthworm
censusing with other chemicals have been less satis-
factory (Svendson 1955). Driving worms to the sur-
face by a discharge of alternating current from a
probe thrust into the ground has also been tried
(Kevan 1955).

Enchytraeid or pot-worms belong to the meso-
fauna as do the smaller arthropods, such as spring-
tails, symphylans, pauropods, proturans, mites, and
various insect larvae. Enchytraeids may be extracted
quickly and efficiently by putting a layer of sand on
top of a soil sample in a special container, and ap-
plying heat and water from below. This forces the
animals to accumulate in the sand, from which they
may be easily separated (Kevan 1955).

A common method of extracting small arthropods
from litter and soil is by means of the Tullgren modi-
fication of the Berlese funnel. Where possible, the
soil sample should be kept intact as a block and in-
verted into the funnel, bottom side up with an electric
light bulb placed above the sample. Forced to retreat
from the light and heat and the gradual drying of the
soil from the top to the bottom, the animals move
down the funnel and fall into the bottle of alcohol
below. A week or ten days is usually required to ob-
tain all possible animals from the sample. The pro-
cedure is subject to a number of faults, however, and

38 Background

must lie moililicd tor ilifl'ercnt soil tyiK-s ;iiul tax-
onoiuic grou]JS to produce tlu- best results (Mac-
tadyen 1953).

W'itii tlie ftotiition method, the litter or soil is
placed in a large pan, and warm — not hot — water is
l)oured in to thoroughly soak and cover the soil to a
depth of 1-2 cm. The warm water stimulates the
animals to activity, and they come to the surface
where they may be collected with forceps or suction
bottle. For greater efficiency, air may be bubbled
through the material to break up clumped masses ;
chemicals, such as magnesium sulphate, may be added
to increase the specific gravity of the water (Kevan
1955).

The microjaiina consists mostly of microscopic
forms such as protozoans, rotifers, nematodes, tardi-
grades, and turbellarians. To quickly obtain a non-
quantitative sample, a few grams of soil or litter may
be placed on a screen in a glass funnel, and water,
warmed to about 40°C, poured over it. The fluid
collected from the funnel may then be centrifuged.
A slower but more effective method for collecting
nematodes is to wrap the sample in cheese-cloth or
muslin and immerse it in water in a funnel that has
a clamped rubber tube fastened to the stem. Let it
stand for a few hours, and the nematodes will collect
in the stem of this Baerman funnel. They may then
be released into a petri dish for e.xamination ( Kevan
1955).

To obtain a good idea of the protozoans present,
culturing is usually necessary. Edible bacteria are
added to a non-nutrient agar or silica jelly in a petri
dish, small amounts of properly prepared soil dilu-
tions inserted at various points, and the culture in-
cubated for two weeks. Final examination of tlie
culture is made under the microscope, and the pres-
ence or absence of protozoans at the various points
determined. Cultures may also be prepared using soil
extract or hay infusion (Kevan 1955).

Fish

In small streams, a representative section of
known length may be blocked ofT at the upper end by
stretching a net from one bank to the other. Seining
proceeds from the lower end up to the stretched net.
There are limits, of course, to the size and depth of
streams that can be examined in this manner, and
unless care is taken, fish will escape around the ends
or underneath the net. Seines can be used efficiently
in this manner only when the bottom is free of large
stones or other obstacles.

It is sometimes possible to draw long nets over
measured areas of ponds and the shallow waters of
lakes, but the data obtained usually give a relative
abundance only. In deeper water, trammel, gill, or

hyke nets may be set, but each type of gear has limi-
tations with reference to species, locality, and time
of day. Seines, trammel and Fyke nets that catch
fish alive are commonly u.sed, however, in applying
the Petersen method for obtaining absolute abun-
dance.

When fish are removed from a body of water by
gear of some sort or by s|X)rt fishing, the catch per
unit-effort may be a basis for estimating total popu-
lation. Creel censuses of the catch of fishermen are
commonly taken to measure the yield of fish over
periods of time.

In modern practice, artificial ponds are usually
built in such a way that they may be drained and unde-
sired species or surplus populations removed. Fish
may be counted and measured in these operations,
and only those species which are desired for re-
l)lenishing the population restored to the pond.

Small bodies of water, or representative areas of
larger bodies, can be blocked of? with nets and cen-
sused by the use of a poison, such as rotenone (pow-
dered derris root). Fish are killed and float to the
surface of the water where they may be collected and
counted. This method cannot, of course, be used
where the population is to be left undisturbed. The
])oison also kills most zooplankton and some, but not
all, other kinds of invertebrates (Brown and Ball
1942).

A less drastic method, that of shocking, is most
effective in small streams. Two electrodes are in-
serted into the water and the electric charge tempo-
rarily stuns the fish so that they float to the surface
where they can be captured. After the desired data
are obtained, the fish are returned to the water and
recover rapidly (Lagler 1952).

Plankton

Plankton nets are commonly made of silk bolt-
ing cloth ; number 20 or 25 is ordinarily the finest
mesh used. Tow nets are made with a conical bag
attached to a wire frame, to which the tow string is
attached by means of cords. Collections may be re-
moved by turning the net inside out in a jar of water,
or the organisms may be concentrated in a vial
screwed in at the tip of the cone.

For surface plankton, it suffices to use the plank-
ton net as a sieve and pour through it a known quan-
tity of water. The tow net may be dragged behind a
boat either at the surface or submerged to any depth
by means of weights attached to the tow line.

Since the depth at which plankton occur varies
with the time of day, vertical hauls sampling all
depths are preferred. Comparison of plankton popu-
lations at different times or in different areas had
best be made in terms of unit surface area. The Wis-

Measurement of populations 39

'*yr:%

FIG. 4-5 A Fyke fish trap (courtesy Illinois State Natural History Survey).

consin plankton net is especially designed for this
purpose. Closing nets or traps can be made so that
they may be lowered to any desired depth, then closed
and brought to the surface. This enables the investi-
gator to determine at what depths the organisms oc-
cur. The Kemmerer sampler is used extensively for
bringing up known volumes of water from measured
depths for plankton or for chemical analyses. Nan-
noplankton, which passes through the finest tow net,
needs to be filtered out or centrifuged out for quanti-
tative measurement (Ballantine 1953).

Net plankton is ordinarily counted with the use
of a Sedgwick-Rafter cell that holds exactly one cubic
centimeter at a time, and the number present calcu-
lated per unit volume or surface area of the pond or
lake. The volume of water filtered is equal to the area
of the net opening, times the distance pulled, times
a correction factor. No plankton net filters out all the
organisms from the column of water through which
it is dragged. The efficiency of such nets depends on
fineness of mesh, rapidity with which it is pulled, and
the abundance of organisms present. Fine-mesh nets
offer resistance to water flow, which is further in-
hibited as the pores become clogged with organisms.

so that a part of the water column is diverted around
the net as it is pulled. A correction coefficient must
be determined for each net and for each different rate
at which it is pulled. This may be done by comparing
the quantity of catch obtained in the tow net with the
density obtained through use of plankton-traps or the
Kemmerer sampler. Detailed instructions for con-
structing different kinds of nets, and statements con-
cerning the advantages, disadvantages, and possible
errors in the use of different methods are given by
Sverdrup ct al. (1942) and Welch (1948).

Bottom organisms

Dip-nets are commonly used for obtaining mac-
roscopic bottom organisms and those attached to sub-
merged vegetation. In shallow water, bottom or-
ganisms may be scooped out from a bottomless
cylinder covering a known area. We find that four
good scoops with a dip-net are necessary to get most
of the organisms from a cylinder covering 0.2 m^, so
we sometimes consider two scoopsfull with a dip-net
as equivalent to 0.1 m^ when the cylinder is not used.
The Surber swift-water net is standard equipment

40 Background

(courtesy Illinois State Natural History Survey).

for sampling rocky stream bottoms. A frame marks
out 0.1 m^, and a net downstream catches organisms
dislodged as the rocks are removed into a pail, for
closer e.xamination (Fig. 5-2a).

Dredges of various shapes and sizes may be pulled
along the bottom for measured distances to get or-
ganisms in deep-water, but quantitative determina-
tions obtained in this way give population estimates
that are generally too low. The dredge commonly
does not dig sufficiently deep into the bottom ; often
it skips and slides along the surface without picking
up all the organisms that are present. Much more
reliable are the Ekman bottom sampler, on soft bot-
toms, and the heavier Petersen sampler, used also on
sand and harder bottoms (Fig. 6-9) . For microscopic
organisms small core samples are usually collected
and brought back to the laboratory for examination.

The bottom samples obtained in various ways
must ordinarily be washed through sieves to remove
the debris, and the animals put into vials or jars for
identification and counting. The size of mesh to be
used in the sieve depends on one's objectives (Reish
19.=i9). We find four nesting sieves efficient, with the
top sieve having a coarse mesh (2 per inch) and the

lower ones of increasing fineness (10, 20, 30 meshes
per inch) to capture smaller organisms. Suppliers of
limnological and oceanographic apparatus and sup-
plies have been listed by Ryther et al. (1959). It
should also be noted that for on-site studies of ani-
mals under water, increasing use is being made of
piiotography and even television. The bottom may
also be explored at first hand using diving equipment.

SUMMARY

Although determination of relative abun-
dance is sometimes useful in projects of limited scope,
the measurement of absolute abundance is generally
to be preferred. Measurement of absolute abundance
requires the counting of individual animals or meas-
urement of their biomasses on strip censuses or sam-
ple plots. The size, number, shape, and distribution
of sample plots and methods of measuring population
densities present special problems that must be ad-
justed for each habitat and group of organisms con-
cerned. The development of improved methods of
population sampling is one of the major needs of
ecology today.

Measurement of populations 41

Local Habitats,

Communities, and

Succession:

Streams

When rainwater falls on an uneven surface, it col-
lects in depressions. As the water overflows them,
the current erodes a narrow channel that deepens
with each succeeding shower and may eventually
drain the depression. There is usually also a lateral
meandering of the stream, by which a valley is
formed. The site of the headwaters of such streams
is impermanent, and continued erosion forces the
headwaters and the channel farther and farther back
into the upland. The stream is at first a temporary
one, dependent for its waterflow on rainfall runoff,
but when its channel is cut below the level of the
groundwater table the stream becomes permanent,
fed by general, continuous seepage. The headwaters
of such a stream are therefore its youngest portions
physiographically, and the stream is progressively
more aged towards its mouth.

In hilly or mountainous terrain, water may ac-
cumulate in large basins until ponds or lakes are
formed. In the Great Basin of North America, such
lakes have not found an outlet to the sea, and evapo-
ration has left them with a very high salt content.
Ordinarily, however, the water level in such a lake
will rise until it overflows at the lowest point on the
perimeter. Then the waters continue to flow down-
ward until they eventually reach the sea. Streams
springing from fixed headwaters (melting snowfields
and glaciers, springs) carve valleys that are of essen-
tially the same age throughout. Streams less than
3 m (10 ft) wide are usually called creeks or brooks;
rivers are streams 3 m or more wide.

A river system in youth is characterized by val-
leys that are narrow and steepsided ; the flow of water
is usually fast, there are few tributaries, and there
are many waterfalls, ponds, and lakes along its
course. As the river system matures, its valleys be-
come wider, its slopes more gentle, and its tribu-
taries more numerous and longer. Many ponds and
lakes are drained, and waterfalls are worn down to
rapids or riflles. The areas of upland are well dis-
sected, and the land is thoroughly drained. In old
age. the river system has reached base-level. The
upland has been worn down to low ridges between
tributary river valleys, and the region as a whole is
called a peneplain. There are no lakes, ponds, or
rapids, and the flow of water is sluggish (Strahler
1951).

HABITATS

Exclusive of its lakes, the principal habi-
tats in a stream are falls, rapids or rijfles. sand-
bottom pools, and mud-bottom ponds. The character
of the bottom depends primarily on the velocity of
the water current, which, along with the volume of
stream flow, can be readily measured (Robins and

42

Crawford l'-'54). Water tiowiiig at tlif rale of about
50 cni-sec is considered swift- Howiiig ; velocities
greater tlian 300 cm-sec rarely occur. Fast currents
roll or slide pebbles and rocks along the bottom ;
move sand partly by rolling and partly by buoyant
transportation ; and carry line materials, such as silt
and organic matter, in suspension (Twenhofel 1939).

In places where the topograjihic gradient is steep,
the stream bottom will be composed largely of cobble
and boulders too heavy to move, and smaller pebbles
which are trapped by obstructions. This habitat is
called a raf'ids. if extensive and turbulent ; riffles, if
of a lesser order.

When the gradient is less steep and the water cur-
rent thus slower, gravel (particle size 2-64 mm) is
deposited first, then sand (0.06-2 mm), but the finer
materials are carried along. Only when the current
becomes negligible does the suspended material settle
so that silt (0.004-0.062 mm) or mud-bottom pools
or ponds are formed. Clay has a particle size even
smaller (Morgans 1956). These mud-bottomed pools
are the most fertile parts of the stream because of
the presence of organic matter entrained in the silt.
The rate at which o.xygen diffuses into water from the
atmosphere increases as the turbulence of the water
increases : rapids therefore have, often, the highest
oxygen content of a stream's waters. Ordinarily,
however, oxygen is near saturation in all parts of a
flowing, non-polluted stream. In a general way, rif-
fles, sand- and mud-bottom pools represent three
stages in the aging of a stream, and ecological study
of them gives a good idea of what the eosere would
be over a long period of time.

Trout streams do not normally exceed 24 °C max-
imum summer temperature : streams with higher
summer temperatures are more characteristically oc-
cupied by species of Centrarchidae and Esocidae
(Ricker 1934). Streams have been classified into a
variety of different types, using the most character-
istic fish present as a basis (Van Deusen 1954). The
salt content of stream waters depends both in quan-
tity and in chemical nature on the fertility of the land
drained or the rock strata which produce the springs.

STREAM BIOCIES

When quantitative sampling is made of the
invertebrate populations of streams, one finds that
there is a sharp distinction of species found in riffles
and those found in mud-bottomed pools (Table 5-1 ).
The sand-bottom pool habitat has few characteristic in-
digenous invertebrate species, but it is occupied by
small numbers of individuals of species otherwise
occurring abundantly in the other two habitats. The
unstable bottom apparently prevents the development
of a characteristic community. The unionid clams

are really the only invertebrate group to become es-
tablished in this habitat with any degree of success,
although they are not exclusive to it. There are, how-
ever, several fish species (Table 5-2) that find .s.iiidy
pools a favorite habitat, although they depend in
large part upon riffle organisms for their food. Many
lish overwinter in the deeper, more (|uiescent sand-
bottom pools, especially since low water temperature
makes them too sluggish to withstand rapid currents.

Mud-bottom pools form in backwaters of the main
stream, behind natural or artificial dams in the main
channel, or where the current is sluggish. Very often,
a(|uatic vegetation fringes the edges of these pools.
These quiet pools are essentially young stages in the
developmeiU of ponds and support many animal spe-
cies indigenous to ponds. Such pond animals as
afiuatic annelids, dragonfly and damselfly naiads, and
burrowing mayfly naiads commonly occur also on the
muddy margins of streams in which the main channel
has a sand, gravel, or rocky bottom.

The stream biocies consists most typically, there-
fore, of the inhabitants of the rifiies and sand-bottom
pools found throughout the course of the river. The
riffle and pool organisms make up two different fades
in this community. Mud-bottom pools and sluggish
streams are occupied by the pond-marsh biocies, to be
later described.

Plants are not abundant in the stream biocies,
although the upper surfaces of rocks in a riffles may
be completely covered with branched filamentous
algae (particularly Cladophora) , and a few species
of water mosses (Fontinalaceae) may occur. Di-
atoms, mostly sessile forms, may be numerous in
early Spring and again in the Autumn. Dominance
in the true sense, .such as occurs in terrestrial com-
munities, does not exist, although the algae and
mosses passively provide food and shelter for active
forms.

The most characteristic and abundant animal
forms of the stream biocies are the caddisfly larvae,
mayfly naiads, stonefly naiads, fly larvae, crayfish,
snails and clams, sponges and bryozoans, and fish,
each occupying its own particular niche (Berg 1948).
Plankton is mostly absent in swift-running water
(Carpenter 1928, Coker 1954), but may be abun-
dant in sluggish, pond-like stretches of large rivers.
The fishes listed in Table 5-2 are mostly warm-water
fishes. In the colder waters of mountain and northern
streams, the fish fauna changes. Trout, sculpins, and
sticklebacks become the most conspicuous species.
Streams that empty into the ocean may have a special
fauna of anadromous ("upstream") fish, such as sal-
n)on, shad, striped bass, and some trout, that spend
much of their lives in the sea but migrate into fresh-
water streams to spawn, and catadromous fish, such
as the eel, which migrate "downstream" into the sea
to reproduce. There are a few vertebrates other than

Streams 43

FIG. 5-1 Diagrammatic arrangement of streai

of different physiographic age on the

south shore of Lalte Michigan. Each number

shows the location of that pool nearest

a headwaters which first contains these

fish: (I) creek chub; (2) redbelly dace;

(3) blacknose dace; (4) suckers; minnows;

(5) grass pickerel; bluntnose minnow;

(6) sunfish, bass; (7) northern pike, lake
chubsucker, and others (after Shelford 1913).

fish commonly found in streams. Some salamander
species occur only in fast mountain streams ; other
species are more typical of pond-like pools. The
belted kingfisher feeds on stream fishes, and nests in
adjacent clay banks. In the western mountains, the
water ouzel feeds under water on the insect larvae
and naiads of the riffles. Muskrats make their bur-
rows in the stream banks and feed on vegetation and
clams. Mink patrol the streams for the muskrats and
fish that serve them as food. The once-abundant otter
is now absent from most localities. Beaver dam
streams to enlarge the pools in which they build their
lodges and find shelter. Beaver feed on the bark and
cambium of aspen, willow, and other trees and
shrubs occurring on the shores of the stream.

ADJUSTMENTS TO CURRENT

Probably the characteristic of a stream
most critical to the life therein is the current. All
organisms that occur in streams must adjust to it to
maintain constant position. Torrential floods scour

FIG. 5-2 Apparatus for collec
quantitative samples of botton
in streams: above, swift-water net,

;0.l m' (Surber 1936); right,
sampling cylinder for use in pools, covers
0.2 m', has sharpened lower edge.

the stream bed, move rocks and sand, cut new chan-
nels, and destroy entire populations. Recovery after
such catastrophes, however, may take place within a
few weeks or months, especially by those species pos-
sessing short life cycles (Moffett 1936, Surber 1936).
Position is ordinarily maintained by clinging to the
substratum, avoidance of the current, or vigorous
swimming, and requires a good development of ori-
entation behavior.

Clinging mechanisms

The growth form of fresh-water sponges is af-
fected by a number of factors, but in riffles sponges
are usually simple encrustations. In quieter water,
long, slender, finger-like processes may form. The
distribution of species depends both on current and
organic content of the water (Jewell 1935).

Plumatella is a common bryozoan that forms an
encrusting, plant-like, branching colony on tiie under-
side of rocks or fallen trees in swift water. Pectina-
tella. on the other hand, forms a gelatinous spherical
ball, and is more commonly found in ponds or slow-
flowing portions of streams.

Turbellarians, such as Planaria, and swift-water
snails, such as Goniobasis and Pleurocera, and the
limpet Ferrissia. cling to the substratum by means
of flat, slimy, adherent body or foot surfaces, and are
most common on the protected lower surfaces of
rocks.

Mayfly naiads have efficient adaptations which
enable some of them to tolerate currents up to 300
cm-sec (Dodds and Hisaw 1924). The animals cling
to the smooth undersurfaces of the rocks, keeping
their heads toward the current and their bodies par-
allel with it as they move sideways, forward, and
back. The head is flattened, and when pressed firmly
against the substratum the water current exerts a
downward pressure which helps to hold the animal
in position. Compared with forms found in quieter

44 Habitats, communities, succession

TABLE 5-
as detern

Site and distribu
ined by class studii

vertebrate populations

Number per square meter

Sand bottom

Mud bottom

Common name

Classification

Riflles

pool

pool

Caddlsfly larva

Trichoptera

1,006

Mayfly naiads

Heptageniidae, Baetidae

246

Hellgrammite

Corvdalis

46

Riffle beetle larva

Psephenidae

19

Riffle beetle adult

Psephenidae

Limpet snail

Ferrissia tarda

Bryozoan

Pluniatella

Fresh-water sponge

Spongillinae

Flat worm

Planaria

Broad-shouldered water strider

Rhagovelia

Stonefly naiad .

Plecoptera

61

1

Snails

Goniobasis livescens,

Pleurocera acuta

39

White midge fly larva

Tanypus

16

Horse fly larva

Tabanldae

Fingernail clam

Sphaerium

Crayfish

Orconectes propinquus

Damselfly naiad

Zygoptera

Dragonfly naiad

Anlsoptera

Clams (28 species)

Unionidae

++

++

Crayfish

Orconectes virilis

Snail

Physa gyrina

Red midge fly larva

Tendipes

AquaUc annelid

Chaetopoda

134

Burrowing mayfly naiad

Hexagenia

+

139

Water boatmen

Corixidae

AJderfly larva

Slalidae

Flshfly larva

Chauliodes

Crawling water beetle

Haliplidae

Amphipod

Hyalella

Predaceous diving beetle

Dytlscidae

Backswimmer

Notonectldae

Water scorpion

Rawtra

Aquatic isopod

AselUdae

Whlrl-i-gig beeUe

Gyrinidae

Springtail

Podura aquatica

Mayfly naiad

Caenis

Snail

Gyraulus parvus

Snail

Lymneidae

Total taxa

24

14

26

Total individuals

1469+

13+

320+

...Iter, they show a larger thorax and legs, a smaller
alxlomen, absence of hair on the caudal cerci, shorter
middle cercus, and smaller gill lamellae. These modi-
fications enhance body streamlining and reduce the
drag of the water. Furthermore, the legs are articu-
lated in a way which allows the current to press them
hrmly against the substratum. The body itself swings
freely in the current.

Mayfly naiads that occur in quiet waters do not
lia\e these modifications. They commonly spend

most of their time in burrows, dug into the mud.
They come out at night to swim around and search
for food. The abdomen of the mud-inhabiting forms
is thick, with little taper, sometimes bowed vcntrally,
and the three terminal cerci are provided with long
stiff hairs that overlap and make an excellent oar for
swimming.

Stonefly naiads are not limited to stony habitats.
Some species occur in the masses of leaves that lodge
against rocks or along the banks, in the algae grow-

Streams 45

TABLE 5-2 Distribution of predominant fish species in si
habitats of central Illinois (after Thompson and Hunt 1930).

Common Name

Gravel

and sand Mud

bottom bottom

pools pools

Suckermouth minnow +

Banded darter +

Bigeye chub +

Log perch +

Green-sided darter +

Stonecat +

Hog sucker +

Fantail darter +

Steelcolor minnow +

Common shiner +

Channel catfish +

Hornyhead chub +

StoneroUer minnow
Silverjaw minnow
River shiner

Reffin shiner
Rainbow darter
Quillback carpsucker
Smallmouth bass
White crappie

Orangespotted sunfish
Longear sunfish
Green sunfish
Bluntnose minnow
White sucker

Northern redhorse
Shorthead redhorse
Creek chub
Johnny darter
Golden shiner

Creek chubsucker
Grass pickerel
Blackstripe topminnow
Pirateperch
Freshwater drum

Gizzard shad
Highfin carpsucker
Largemouth bass
Bigmouth buffalo
Carp

Black crappie

Black bullhead

Total species 12

ing on the rocks, on sand bottoms, and in small mud-
bottom streams rich in organic matter. The general
form of the body is similar to that of swift-water may-
fly naiads, although the gills are filamentous and lo-
cated at the base of the legs.

Caddisfiy larvae occur most abundantly in streams
with medium to swift currents, but some species oc-

cur only in sluggish rivers, in lakes, or in pond vege-
tation. Caddisfly larvae are of especial interest be-
cause of the cases they construct, in which the pupae
also occur later. In some species these cases are
portable. They are made of pieces of leaves, twigs,
sand grains, or stones which are cemented or tied
together with silk that the animals secrete. In stand-
ing or sluggish water, the cases are often large and
made of buoyant plant material, or they may be made
of sand grains, more fragile and slender. In swift
water, the cases are stout, cylindrical, tapered pos-
teriorly, and are usually smaller and more solidly
constructed of sand, small pebbles, or rock fragments
(Dodds and Hisaw 1925). The Hydropsychidae,
Philopotamidae, and Psychomyiidae are unique in
spinning fixed abodes in the form of a finger, a trum-
pet, or a tube. The Hydropsychidae erect a net at
the front end of the tube to catch particles of food
washed down with the current. Some psychomyiid
larvae, particularly Phylocentropus. burrow into sand
and cement the burrow walls into fairly rigid cases.
Some larvae belonging to the Rhyacophilidae are
free-living. Found in algal growth, they crawl
around seeking food, and are provided with large
abdominal hooks as clinging devices to supplement
the legs for clinging. However, they form a stone
case, or cocoon, for pupation (Ross 1944).

The black fly larvae, Simuliidae, are often very
abundant in the swift waters of mountain brooks and
northern streams. The larvae secrete from their sali-
vary glands a delicate silken thread by which they
attach to the rocky substratum, and by manipulation
of which they can move short distances. At the pos-
terior end of the semi-erect body is a circlet of rows
of outwardly directed hooks which, when the muscles
of the disk are relaxed, move outwards and catch on
to a silk web placed there previously by the larva;
the anterior end of the body then swings freely in
the current. There is a fan-like food-gathering organ
on each side of the mouth. Before pupation, the lar-
vae spin a sedentary cocoon. The pointed end faces
the current and the other end, open, faces down-
stream. Out of it, the peculiar gills of the pupa float
in the water (Hora 1930, Nielsen 1950).

The net-veined midge larvae, Blepharoceridae,
are unique in possessing six unpaired suckers on the
ventral side, by means of which they fasten to the
substratum. The original segmentation of the body
is almost obliterated ; it has been replaced by a sec-
ondary segmentation correspondent with the number
of suckers.

Adult riffle beetles (Psephenidae, Dryopidae,
Elmidae) are small in size and are the only coleopter-
ans that live in or near rinming water. The legs are
not fitted for swimming, but rather possess hooked
claws for clutching the substratum. The body is cov-
ered with silken hairs that hold a thin film of air
about it when the beetle is submerged. The larvae

46 Habitats, communities, succession

are disc-shaped and pressed close upon tlie sul)-
stratiim, to wliicli they cHng with tiieir legs and
backward-chrected spines. They are sometimes called
water pennies. When ready to pupate the larvae
crawl out of the water.

■iiiiidinp the current

Diminutive body and appendage sizes and as-
sumption of a stream-line shape keep the amount of
surface exposed to the full impact of the current at
a minimum. The conical shape of the limpet Fcrris-
sia and the flatter cone of water pennies offer little
resistance to water flow. Flat bodies, such as arc
found in many swift-water animals, appear to be not
only an adaptation lowering resistance to current but
also to escape it by enabling the animals to seek
shelter in crevices and underneath stones (Dodds
and Hisaw 1924, Nielsen 1950). Most species, even
those with specialized means of clinging to the bot-
tom, are more abundant on the undersides of rocks
in riffles than they are on the uppersides. Some spe-
cies, however, such as the free-living caddisfly lar-
vae, rotifers, tardigrades, water mites, and proto-
zoans, find shelter within the mass of algae that may
cover the top of the rocks. The hellgrammite, tabanid
fly larvae, and stream crayfishes possess no special
structures for withstanding currents and only occur
in riffles providing protection or lodgement under-
neath and between rocks. Even swift-water fishes,
strong swimmers, take maximum advantage of what-
ever protection is available.

The clams avoid the full force of stream current,
and at the same time retain position, by lodging their
bodies between stones. In pools, they bury them-
selves in an oblique position in the gravel, sand, or
mud. Their posterior ends are directed upstream (ac-
cording to Ur. Max Matteson), and their siphons
usually maintain contact with open water so there
can be circulation through the mantle cavity, for gain-
ing food and oxygen. Clams occurring in pools one-
half to one meter in depth may remain more or less
sedentary, but those occurring in shallower waters
move around considerably, especially in res|)f)nse to
changes in water level and temjjerature.

S,i,

Locomotion of swift-water invertebrates is, in
the main, restricted to short-distance crawling. May-
fly naiads that occur in riffles do not swim, although
related sjiecies frequenting quiet waters do so, regu-
larly. Only the more vigorous fishes can maintain
position in swift currents by swimming, and many of
them do so only when feeding. At other times they
congregate in the pools that occur between riffles.
Salmon and trout are well known for their ability
to swim against strong currents, an accomplishment
of sheer force of powerful, muscular, tails. The sub-
family of darters, Etheostominae, which contains a
variety of brightly-colored small fish, are especially
adapted to live in the riffles. The air bladder of the
darters has become very degenerate, even absent, so
that the specific gravity of the body is increased.

TABLE 5-3 Rheofacti

ic responses

of Invertebrates from riffles and oools (from Shelford 1914).

Velocity of current

4-6 cm/sec

10-12

cm/sec

16-20

cm/sec

Response in percentages

Posi-

Indif-

Nega-

Inac-

Posi-

Indif-

Nega-

Inac-

P^iT

Indif-

Nega-

Inac-

tive

ferent

Uve

tive

Uve

ferent

tive

tive

tive

ferent

Uve

tive

RIFFLES ANIMALS

Crayfish, Orconectes virilis

30

40

28

2

54

8

16

22

78

2

6

14

Snail, Goniobasis livescens

45

27

28

65

22

13

76

7

17

Caddisfly larva, Hvdropsvche sp.

23

26

16

35

18

9

6

67

26

2

4

68

Damselfly naiad, Argia sp.

79

17

4

63

18

4

15

63

4

33

Stonefly naiad. Per la sp.

31

24

3

42

65

6

15

14

61

3

3

33

Mayfly naiad, Heptageninae

25

12

14

49

52

3

45

52

3

45

Water penny, Psephenus sp.

26

32

36

6

67

26

7

74

15

11

Averages

37

23

20

20

55

13

6

26

62

5

3

30

POOL ANIMALS

Damselfly naiad, Calopteryx

maculata

78

22

59

8

33

63

37

Snail, Campeloma subsolidum

51

32

6

80

20

10

90

Burrowing dragonfly naiad.

Macromia sp.

17

36

41

6

12

72

10

6

100

Clam, Anodontoides ferussacianus

16

66

18

17

67

16

100

Fingernail clam, Sphaerium sp.

17

66

17

16

67

17

100

Averages

Streams 47

FIG. 5-3 Mayfly naiads: (a)
adult of Hexagenia limbafa:
(b) naiad of H. //'mfco/o from
quiet water; (c) naiad of
Hepiagenia flaveicens from swift
water (courtesy Illinois Natural
History Survey).

FIG. 5-4 Stonefly: (a) adult;
(b) naiad, hoperia confusa
(courtesy Illinois Natural History
Survey).

Habitats, communities, succession

FIG. B S [a] External (satures of
a cadditfly larva: (b) larva
and case from a weedy lake;
(c) larva and caje from a
spring-fed brook (courtesy Illinois
Natural History Survey).

FIG. 5-6 Immature stages of th
black fly: (a) larva: |b) pupa:
(c) pupa case (Shelford 1913
after Lugger); (d) enlarged
detail of arrangement of hooks
on the posterior end of the
larva (after Nielson 1950)

FIG. 5-7 Water pennies, larva

of the psephenid beetle: (a) dorsal

(Shelford 1913); and (c) larva
of the net-veined midge; showing
the central row of six
suckers (after Mora 1930).

Streams

Their fan-shaped pectoral fins are enlarged, and
project at right angles from the lower side of the
body. When at rest they maintain position by con-
tact with the bottom, fins lodged between pebbles or
the body partly buried. They never float suspended
in the water, as do other fish ; when disturbed, they
dart swiftly from one anchorage to another. The
Etheostominae are confined to North America east
of the Rocky Mountains.

Stream fishes are in general quite sensitive to cur-
rent, and the discontinuous distribution of a species
within the same stream may be closely correlated
with gradient (Trautman 1942, Burton and Odum
1945). The smallmouth bass, for instance, is mostly
absent in southern streams of gradient less than 40
cm/km (2 ft/mi) : is of moderate abundance in
gradients up to 135 cm/km (7 ft/mi) ; is very abun-
dant in gradients of 135-380 cm/km (7-20 ft/mi) ;
and becomes less common again, until it disappears
altogether, in gradients above 475 cm/km (25 ft/mi).
Perhaps streams with very slow current do not pro-
vide suitable gravel nest-sites for spawning, and in
streams with very fast currents they are unable to
maintain position.

Salamanders that live in swift mountain streams
generally have short limbs and toes, reduced size of
fins, smaller lungs in the adult and shorter gills in
the larvae, and relatively few large eggs, which they
fasten to the underside of flat rocks (Noble 1931).

Orientation behavior

Structural adaptations for withstanding or
avoiding current are of no avail without appropriate
behavior responses to make use of them. The rheo-
tactic responses of animals may be tested either in
the field or in the laboratory by means of special ap-
paratus.

When animals from riffles and those from pools
are compared (Table 5-3), it is apparent that, at low
current velocities, responses of the two groups of
animals are nearly the same. The elongate body,
notably of stream animals, brings an automatic turn-
ing into the current much as wind directs a weather-
vane. As the velocity of current is increased, how-
ever, there is a marked increase in the percentage of
riffle animals that face into or move against the cur-
rent, while a very large percentage of pool animals
are swept away by the current or are forced to with-
draw into their shells. Caddisfly larvae, free of their
cases, are not very able to withstand a strong cur-
rent, although within their cases they readily main-
tain position.

Blackfly larvae can tolerate water currents as
swift as 180 cm/sec, and studies indicate that their
clinging to the substratum is a response to current

rather than to any associated factor, such as food or
o.xygen requirement (Wu 1931).

When tested experimentally, 80 per cent of the
stream crayfish Orconectes propinquus were able to
maintain position in currents of 50 cm/sec, but only
about 20 per cent of the pond crayfish O. jodiens
were able to do so (Bovbjerg 1952).

Fishes generally respond to current by showing
nearly 100 per cent positive response, regardless of
whether they be taken from streams or ponds. Since
the response involves a tendency to swim upstream,
other factors must be involved for the fish to main-
tain a constant location in the stream ; otherwise they
would all move to its headwaters.

Some stream fishes, such as the blacknose dace
and the common shiner, can be shown experimentally
to respond visually to landmarks on stream bank and
bottom to maintain their location. Some pool fishes,
such as the sunfish and topminnow, likewise respond
visually, but much more sluggishly, and irregularly.
Darters are entirely unresponsive to visual stimuli,
depending on the tactile stimulus of contact with the
bottom for maintaining position (Lyon 1905, Clausen
1931). Smell may be important to some fish for
orientation. The backswimmer Notonecta (Schulz
1931 ) and whirl-i-gig beetle Dineutus (Brown and
Hatch 1929) have also been shown to use visual ori-
entation in running water.

RESPONSES TO BOTTOM

The segregation of stream animals be-
tween riffles, and sand- and mud-bottom pools may
be, in part, a response to type of bottom. With no
current flowing, the species listed in Table 5-3 were,
in another experiment, given a choice between a hard
bottom and a sand bottom. Eighty-five per cent of
the riffles animals selected the hard bottom, but only
10 per cent of the pool animals did so. Of the pool
animals, all species made 100 per cent response to
sand, except the damselfly naiad, Calopteryx macu-
lata, which divided equally between the two types of
bottom. When the riffles animals were given a choice
between loose stones and a bare bottom, nearly all
individuals selected the stones, and they distributed
themselves among the stones or on top or underneath
in the manner one would expect of them under nat-
ural conditions (Shelford 1914). Stream crayfish,
when given a choice between mud and cinders, ori-
ented 88 per cent to the cinders, while the pond cray-
fish responded 40 per cent to cinders and 60 per cent
to mud (Bovbjerg 1952).

Type of bottom is important to invertebrates for
support and locomotion. Sand bottoms are note-
worthy as unstable and shifting. Insect larvae and
naiads find footing very uncertain ; planarians,

50 Habitats, communities, succession

sponges, and bryozoaiis tiiul no stable anchorage : and
rock-inliabiting snails and limpets are quickly buried.
Clanis, however, tiiid a sandy bottom suitable, if it is
hrmly packed, as they are adapted to burrowing and
plowing their way through a loose substratum. They
are able also to move through a mud bottom, but
where silting is Iieavy they close their valves to avoid
an accumulation of silt within the mantle cavity and
on the gills. The anodontas seem to be the most tol-
erant of mud bottoms.

Some of the mayfly naiads, such as Hexagcnia.
are adapted to burrowing in mud, and the surface of
the bottom in shallow water is often closely dotted
with the openings of their burrows (Hunt 1953).
These burrows are relatively permanent in compact
mud but would quickly collapse in loose sand. The
genus Caenis is peculiar in possessing covers at the
anterior end of the abdomen ; they protect the gills
from becoming clogged with silt. Midge fly larvae
and aquatic annelids exist in mud bottoms : they
would be ground to bits among moving sand parti-
cles. The pond crayfish will burrow into mud down
to water level as a pond dries up, but stream cray-
fish will not do so and consequently suffer high mor-
tality (Bovbjerg 1952).

The bottom is important to invertebrates and
vertebrates alike for placement of eggs. Some caddis-
fly eggs are fastened to smooth rock surfaces in long
strings by a cement-like substance. The eggs of other
species occur in jelly-like masses and may be secured
to plant stems or other submerged objects. Jelly-like
masses of snail eggs are often quite common on the
undersides of rocks in riffles. Some fish, such as the
fantail darter (Lake 1936). make nests in small cavi-
ties under stones, but other species, for instance the
rainbow darter (Reeves 1907), creek chub (Reig-
hard 1908), and river chub (Reighard 1943), build
nests in gravel bottoms in the upper parts of riffles.
Some of the suckers (Reighard 1920) spawn in shal-
low water ; their eggs scatter downstream, finding
lodgment in various riffles.

RESPIRATION

AND OXYGEN REQUIREMENTS

Oxygen is usually ample in streams, often
saturating the water in turbulent riffles. The oxygen
concentration is sometimes low, however, in sluggish
streams and standing pools. The difference in oxygen
tension of the two habitats is reflected in the respira-
tory adaptations of the organisms that inhabit them.
The lamelliform gills of the mayfly naiads in-
habiting mud-bottom pools are larger in size than
those of species inhabiting streams, are doubled in
number on the anterior abdominal segments of some
species, and are almost continuously flicked back and

FIG. 5-8 Apparatus for studies of
rheotaxis. Right, box for use In streams
where water enters at upper end, flows
through center trough, and out lower
end. Controls nnay be run in side troughs
filled with still water. Above, rheotaxis
pan in which current is produced artifi-
cially with a rod or finger. An organisnn's
response is positive when it turns to confront
the current; negative, when it faces down-
stream; Indifferent, when it orients crossways.

forth for better aeration. The gills of naiads living
in riffles, or in waters in which the oxygen content is
high, may have the surface area of the gills reduced
by two-thirds in proportion to body weight, compared
with mud-dwelling forms. They are never flicked,
since the water movement continually brings oxygen
to them (Dodds and Hisaw 1924). Other species do
not flick their gills at high oxygen tensions, but will
do so when tension is reduced. In some swift water
species, there appears to be sufficient oxygen diffusion
through the general body surface to make gills ines-
sential equipment (Wingfield 1939).

Caddisfly larvae have filamentous gills, and there
is some evidence that they increase in number as
body size increases and oxygen content of the water
decreases. It is probable that oxygen also diffuses
readily through the thin skin. A constant current of
water is maintained through their cases by undula-
tions of the abdomen. Stonefly naiads have poorly
developed filamentous gills, located on the thorax,
or have none at all. As a result, they are more sen-
sitive to variation in oxygen supply than are the other
forms mentioned.

Streams 51

The respiratory equipment of pond-inhabiting
animals permits them not only to live in habitats with
lower oxygen tensions but also to survive longer at
high water temperatures. Often, these animals dis-
play relatively low rates of general body metabolism
and oxygen requirement. Such relations between
riffle and pond animals have been observed for may-
fly naiads, caddisfly larvae, isopods, crayfish, and
fishes (Allee 1912-13, Wells, 1918, Fox et al. 1935.
Clausen 1936, Whitney 1939, Bovbjerg 1952), and
to some extent for limpet snails (Berg 1951).

RESPONSES TO STREAM SIZE

Of the species of clams indigenous to Mich-
igan, the 3 commonest are largely limited to creeks,
14 others to medium-sized rivers, and 5 to large rivers
(Van der Schalie 1941 ) . In central Illinois, the num-
ber of species of fish per collection increased from
about 4.5 in streams draining 4 sq km to 15.5 in
streams draining 500 sq km of upland. At the same
time the number of fish decreased from 9 to 2.5 per
sq m of water surface (Thompson and Hunt 1930).
Large species of fish can occur only in stream with
sufficient volume of water to permit freedom of move-
ment ; small fish may find orientation difficult in
large rivers. The preference of fish for streams of
specific size is evident in the tendency for some spe-
cies to travel upstream in times of flood and down-
stream in times of drought.

An increased number of species downstream cor-
relates with greater variety of available niches and
moderate environmental conditions. In many in-
stances the correlation between distribution of species
and stream size, or volume, is not direct but depend-

FIG. 5-9 Clam tracks
1956).

sandy pool (courfesy R.E. Rundus

ent on associated changes in temperature, type of
bottom, fertility, silting, pollution, and other factors.

Headwaters

The headwaters of drainage streams present a
highly variable habitat. During dry periods, pools
shrink and may disappear ; temperature may be very
high in summer and the water largely converted to
ice in winter ; there may be a lack of oxygen, an ex-
cess of carbon dioxide, and a high acidity ; fishes and
other organisms may become greatly overcrowded.
In times of heavy rain, on the other hand, the stream
is swollen, there is considerable erosion of materials
into the stream, and animals are washed downstream.
At all times food is likely to be scarce.

Only the hardiest species can exist under these
conditions. The creek chub is a remarkably hardy
fish ; it may be found in large numbers in shrunken
pools, stirring up the water with tail action and gap-
ing for air at the water surface. Crayfish burrow
into the bottom when the pool dries up. Small snails
may survive desiccation of habitat by crawling un-
der rocks or into crevices, secreting a mucous mem-
brane across the aperture of their shells, and remain-
ing dormant until water returns. The occurrence of
insect larvae and naiads is hazardous, for if the
aquatic stages of their life cycles are characteristically
prolonged, they perish at times of low water or
drought.

Temperature and altitude

In drainage streams the temperature of the
headwaters is variable, but as the water volume in-
creases downstream and becomes more constant, the
range of temperature variation decreases. The head-
waters of spring-fed streams, or of streams arising
at high elevations, usually have a progressive increase
in temperature downstream.

Some species of stonefly and mayfly naiads and
caddisfly larvae are absent from the headwaters of
Ontario streams because the temperature never gets
high enough to permit them to complete their life
cycle. More species are present downstream, and the
headwaters species tend to emerge earlier and earlier
in the summer while the waters are still cold. Still
further downstream, the headwaters species disap-
pear altogether. Species that are limited to the lower
portions of the stream emerge late in the season,
when the waters are the warmest. Closely related
species are thus segregated to different positions in
the stream by temperature tolerances. Headwaters
species have generally a northerly distribution over
the continent and the downstream species a southerly

52 Habitats, communities, succession

distribiiliuii ( klc \':^3S, Spriilcs I'M/). Linear dis-
tribution of tisli in streams may be, in part, a result
of ditTerences in temperature tolerance. Brook trout,
for instance, do best in waters cooler than IQ^C in
\'irginia, wiiiie some varieties of introduced rainbow-
trout prefer waters above 19°C (Burton and Odum
1''45). The altitudinal zonation of various species of
invertebrates and fish in mountain streams is well
detined, and is in large part contingent on differences
in temperature ( I )odds and Hisaw 1025).

Sl,a,,r uml

(if indit iiliials

In tl'.e Tennessee River, riffles snails of the
genus lo show a progressive change in shape from
the iieadwaters on downstream. There is a decrease
in shell diameter, a decrease in globosity, and an in-
crease in number and length of spines (Adams 191 .S).
flowever, the riffles snail Plcitroccra was found to in-
crease in globosity downstream in Michigan (Good-
rich 1937). Some pond snails, such as Lymnaea
staanailis and Galba paliistris. develop a larger foot
and shell aperture when exposed to wave action
(Baker 1919). Primitive types of clams, on the other
hand, such as Fnscoyiaia. Amblcma, Qiiadnila. Plciir-
obcitia. and others, change progressively downstream
from a large, compressed, smooth shell to one that
is shorter, more obese, and sculptured with tubercles
( Ortman 1 920 ) . Some species of clams show no such
changes in shape. In some fish of central Asia (Ni-
kolski 1933), the body changes downstream from a
torpedo-shape to a flatter, longer form. These
changes are probably a result of downstream reduction
of water current, increase in amount of calcium in the
water, and higher temperatures. The formation of
spines and tubercles, for instance, would require an
abundance of calcium and quiet water.

EVOLUTION

In all probability species inhabiting quiet
waters are ancestral to those occurring in running
waters (Dodds and Hisaw 1925, Hora 1930). In-
vasion of stream habitats requires mechanisms for
contending with the force of current, and orientation
behavior for maintaining position. Convergent evolu-
tion has occurred in many kinds of animals under the
influence of current, as shown by similarities in struc-
ture and habits (Shelford 1914a). Inducements to
the invasion of swift waters have doubtless been new
sources of food, escape from enemies, and avoidance
of competition with the abundant life of lakes and
ponds. As adaptations to stream habitats evolved, ani-
mals have largely lost their ability to occupy quiet
waters. They no longer can tolerate the lower oxy-

j;en tension, silt buttums, and the absence of current
which brings them food and o.xygen, and, in some
forms, such as the Hydropsychidae, helps build their
shelters and nests.

I.ll K lil.vrOKIK.S

Tiie life-cycle of stream insects is reniark-
.■d)le for the long duration of the immature stage in
many species and the brief life of the adult. The
naiads of mayflies pass through a number of molts
(20-40), and this immature stage may last from six
weeks to two years. When ready to emerge, the naiad
comes to the water surface or crawls out onto a stone,
molts into a subimago, and flies away. Within a few
minutes, or a period of one to two days at the longest,
the subimago undergoes another molt, miique in in-
sects, into the fully mature adult. The adult insect
does not eat and lives only a few hours or days ; dur-
ing this time reproduction takes place. Mating oc-
curs in flight, hundreds or thousands of individuals
swarming in flight together. The females lay their
eggs almost immediately after mating. In some spe-
cies, deposition is made upon the water surface.
the eggs sinking to the bottom : in other species the
female crawls down into the water and attaches the
eggs, as they are laid, to a rock surface. The eggs
have a viscid surface or filaments and (juickly be-
come attached to submerged objects. Embryonic life
may last 11 to 23 days, at the end of which time the
naiad is fully formed (Needham et al. 1935, Burks
1953. Hunt 1953).

The life-cycle of stoneflies is also 1 . 2, or possibly
3 years long in different species, of which time all
but a brief interval is spent in the water (Prison
1935). Molting into the adult occurs after the naiad
crawls out of the water onto a rock or other project-
ing object, and there is no subsequent molt in the
adult stage. Adult diurnal stoneflies may feed, al-
though the adults of nocturnal species apparently do
not. It is of great interest that many species emerge,
mate. feed, and carry on all essential activities during
the coldest months of the year (Prison 1935). At al!
seasons, the eggs may be dropped into the water while
the female is in flight over the water, or as she alights
on its surface. The eggs are mucilaginous and may
contain surface filaments or hooks.

Caddisfly larvae pupate submerged in cases. As
the pupa approaches the adult form, it leaves the
case ; and. after crawling and swimming, emerges
either upon the water surface or on some protruding
object. Larval life in different species may be as short
as 25 to 80 days, but since overwintering occurs in
this stage it may be greatly prolonged. The pupation
period is ordinarily shorter than the larval period,
and the adults, which probably feed, may live from

Streams 53

several days to a few weeks. Species living in tem-
perate climates have either one or two generations
per year. Some females drop their eggs while in flight
but others crawl under the water to deposit them.
The eggs are laid in masses in either a single-layered,
cement-like encrusting form or in a jelly or gelatinous
matrix that swells in water. Eggs are sometimes de-
posited on objects above water. Usually, 10 to 24
days are required for their hatching (Balduf 1939).

The common hellgrammite of North America ap-
pears to require three years to complete its life-cycle,
of which it spends two years and eleven months as
an aquatic larva. When ready to pupate, the larva
crawls out of the water and underneath some loose
stone or piece of wood. The adults do not eat and
live only a few days. The female lays her eggs in
masses attached to supports situated near water or
to the upper surface of leaves. Upon hatching, the
larvae make their way back into the water (Balduf
1939).

In the crayfish Orconectes propinquus copulation
occurs in cool climates from July to November. Fur-
ther southward, copulation is delayed until Septem-
ber, continues until cold winter weather, and is re-
newed again during March and April. Eggs are laid
beginning in late March or early April and are car-
ried around by the female, attached to her pleopods,
or swimmerets. The eggs hatch in 4 to 6 weeks, and
the young are carried for another week or two before
they become free-swimming. The majority of the
young become sexually mature at the end of the first
growing season in early October (Van Deventer
1937).

The female adult black fly deposits her eggs in a
mass or string on a stone or other object at water
level during late afternoon, usually with only the tip
of the abdomen submerged. If the eggs become ex-
posed to the air they do not hatch ; normally, the lar-
vae appear in four or five days at medium water
temperature of 20°-22°C. The larval stage persists
13 to 17 days, the pupal period a little more than 4
days, and the adult stage a little over a week when the
adults feed, or only 5 or 6 days when they do not
(Wu 1931).

Stream snails attach their eggs in a jelly mass to
the sides of stones during late spring and summer,
and development leads directly to the adult. Clams
of the family Unionidae, however, have a peculiar
mode of reproduction. The sexes are separate, and
fertilization of the eggs takes place in the supra-
branchial chambers of the female. Development takes
place through several weeks in these marsupial gills,
and each egg grows into a minute glochidium. These
larvae are later shed into the water, where further
development requires that the glochidia become at-
tached to the gills, skin, or fins of fish. The larvae
may be parasitic, feeding on nutrients absorbed from

the fish ; this stage may last from 9 to 24 days. Later,
the cyst formed by the fish around the glochidium
weakens, and the young animal escapes to take up a
free-living existence. Breeding occurs from May to
August in Quadrula and Unio, while in some spe-
cies {Anodonta, Lampsilis) breeding does not occur
until late in the summer and the glochidia are re-
tained in the female over winter.

The life history of clams is of special significance
in showing that dispersal depends, to a large extent,
on the movements of the fish to which the clams are
attached. There is evidence that some species of
clams cling to particular species of fish only, so that
distribution of the two forms in the stream is closely
correlated. The fingernail clams Sphaeriidae, on the
other hand, are hermaphroditic and lack the glo-
chidial stage. The fingernail clams are annuals ; the
larger unionid clams may live 10 to 15 years (Coker
et al. 1922, Boycott 1936, Matteson 1948).

Some sponges, and perhaps also bryozoans, are
perennial, although they may become fragmented as
a result of floods or freezing during the winter ; they
may die during times of low water. Both kinds of
animals have vegetative buds, gemmules in sponges
and statoblasts in bryozoans, that become free of the
parent body. The buds are adapted to withstand un-
favorable drought or winter periods, and to germinate
and form new colonies when favorable conditions re-
turn.

The nesting habits of some stream fishes have
already been mentioned. Some of the darters and
dace defend their nests, or small territories around
their nests, against intruders ; other species appear
to not do so. Individuals of territorial species do little
wandering, and it is possible that a darter may persist
through several generations in the same riffles. There
is increasing evidence that some larger species of
stream and pond fishes have definite home areas, and
that the fish population of a small stream with riffle-
pool development may be considered as a series of
discrete, natural units. This has been demonstrated
with tagged individuals for species of bass, sunfish,
suckers, and bullheads (Gerking 1953). Homing
tendencies, however, are developed to varying de-
grees, and some species appear to move around in a
quite random manner (Thompson 1933).

FOOD COACTIONS

The basic food substances for stream ani-
mals are detritus, diatoms, and filamentous algae.
Detritus consists of dead fragments of plants ; par-
tially decomposed, finely divided, plant material ; and
a certain amount of dead animal matter. Plankton,
either plant or animal, is not normally a common
source of food, except in outlets from the lakes and

54 Habitats, communities, succession

ponds from wliich they dorivc ;iiul in thi' skiggisli
waters near tlie inoiitli of the stream. Larger aquatic
[ilants are not characteristic of swift flowing streams ;
tliey occur in sluggish pools. Filamentous algae, how-
ever, may be abundant in riffles, and a rich micro-
flora of diatoms, with scattered protozoans, may
furnish a thin slimy film over the surface of rocks.
Animals are adapted to these food resources as filter
feeders, microflora eaters, or carnivores (Nielsen
1950).

The caddisfly larva Hydropsyche is a fine example
of a filter feeder. This species and related forms con-
struct silken nets at the entrances of their shelters
and strain out food particles brought down by the
current. The anterior legs of some caddisfly larvae
and mayfly naiads are furnished with brushes of hair-
like setae which catch and transfer the detritus to
the mouth as the animal faces the current. Black fly
larvae have a pair of fans at the anterior end of the
body. These fans are of long, curved setae. The larva
folds them periodically, and the mandibles comb or
brush off the detritus that collects. Clams siphon
water through the mantle cavity, and detritus ma-
terial and plankton are carried to the mouth through
the activity of the cilia of the mantle, gills, and labial
palps. Sponges and bryozoans also take detritus into
body cavities for feeding purposes.

Feeding on the microflora and filamentous algae
are planaria, snails, and various insects. Some cad-
disfly larvae have mouthparts specially adapted to
scrape the thin film of microflora from the surface of
rocks. The maxillae of mayfly naiads serve as a comb
or brush with which diatoms are swept up into the
mouth.

Carnivorous species may also be partly herbivo-
rous (Table 5-4). Too, there is apt to be seasonal
variation in food habits and there are differences of
habit between closely related species. Fall and winter
stonefly naiads are largely herbivorous, but spring
and summer forms comprise genera that are either
carnivorous, herbivorous, or omnivorous. Hellgram-
mites are largely carnivorous, feeding on immature
insects. Crayfish are omnivorous ; they appear to
prefer dead and decaying material. The smaller fish,
including the darters, are largely insectivorous, but
also consume some plant material. Suckers, carp, and
catfish feed on bottom debris as well as small living
animals and plants. Young bass and trout are largely
dependent on insects for food, but as they grow larger
they turn also to young crayfish and small fish. The
population density of fishes is ultimately determined,
therefore, by the abundance of invertebrates and,
when fishes rely on vision for finding their food, also
on the turbidity of the water.

The average weight of food in the stomach of fan-
tail darters of all sizes, sampled from October to May
in New York State, was found to be 0.01354 g

( Daiber 1956). If the average biomass of the living
food averages 2.83 g/m- of bottom, then one indi-
vidual of this species could get 209 full meals from
one s(|uare meter if it captured everything that was
there. Similarly, mottled sculpins could obtain 130
meals from a square meter. It would be interesting
to know what actual percentage of the invertebrate
population can be readily captured by fish and how
frequently the fish feed, for correlation with the
density of the fish population. Fish, however, also
depend to a considerable extent, especially in summer,
on small terrestrial organisms that fall, or are washed,
into the stream.

BIOMASS AND PRODUCTIVITY

Of the kinds of animals present in one short
coastal stream in California, the caddisfly larvae were
found to be not the most populous. But when size
was considered, they constituted more bulk than any
other invertebrate group (Table 5-5). The inverte-
brate biomass per unit area of riffles is invariably
much greater than in sand-bottom pools, whether
biomass be computed in terms of wet weight, dry
weight, or volume. However, the abundance of spe-
cies within the riffles depends on whether the stones
are loose or are fastened to the bottom, and on
whether or not they are covered with algae, moss, or
other vegetation (Percival and Whitehead 1929).
The biomass of mud-bottom pools may sometimes
exceed that of the riffles, especially if it contains the
burrowing mayfly naiad Hexagenia (Behney 1937,
Forbes 1928, Lyman 1943, Needham 1932, O'Connell
and Campbell 1953, Pennak and Van Gerpen 1947,
Smith and Moyle 1944). In the mud-bottom Silver
Springs stream in Florida, the dry weight biomass
of plants averaged 809 g/m-, herbivores 37 g/nr',
small carnivores 11 g/m-, and large carnivores 1.5
g/m- (Odum 1957a).

Insect populations in streams vary with the sea-
son (Table 5-6). Peak populations commonly occur
during late spring and again in autumn (Daiber 1956,
Lyman 1943, Needham 1934, 1938, Stehr and Bran-
son 1938). Populations become reduced in summer
because of low water ; in winter, because of low tem-
perature and ice.

Small streams tend to have greater densities of
insect populations per unit area than do large streams.
In New York State, streams up to width 2 m have
biomasses that average 22.2 g/m- wet weight ; from
2 to 4 m, 18.0 g/m-; from 4 to 6 m. 10.1 g/m^; and
over 6 m, 7.7 g/m- (Needham 1934). In small
streams, the distribution of organisms is nearly uni-
form from one side to the other, but in large streams
there is a decrease in density from the sides toward
midstream (Behney 1937). Larger streams actually

Streams 55

TABLE 5-4 Food habits of immature stream insects in Yellow-
stone National Park, Wyoming (Muttkowski and Smith 1929).

Number of Per cent food

specimens types consumed

examined Animal Plant Detritus

Stonefly naiads

80

54

22

24

Mayfly naiads

109

4

30

66

Caddisfly larvae

115

28

54

18

Diptera larvae

20

77

23

contain more organisms, however, in spite of lower
densities per unit area, because they have a much
larger total bottom surface. The reason for this vari-
ation in density per unit area is not clear, but it may
be that per given population of sexually mature adult
insects in the surrounding region, small streams offer
less area than large streams, over which the females
can spread their egg-laying.

The standing crop of fish in Indiana streams
varies from 5.2 to 106 g/m-' (46-939 lbs/acre) wet
weight for minnows, suckers, centrarchids, darters,
and bullheads (Gerking 1949) to 2.7-4.2 g/m^ (2^
37 lbs/acre) for rock bass (Scott 1949). The fish
crop in warm water streams is generally higher than
in cool trout streams, a relation that also holds for
the biomass of invertebrates (Pennak and Van
Gerpen 1947). Fish are usually more abundant in
relatively deep streams than in shallower ones. Brook
trout and three other species in one stream in New
York State averaged 10.9 g/m^ (97.5 lbs/acre), a
ratio of 1 :2.1 to the invertebrate food supply (Moore
etal. 1934).

Of a stream, the richness of a fauna and the size
of the biomass that develops depend largely on the
fertility and chemical composition of the water.
Hardwater streams, with an abundance of salts in
solution, tend to have a large and more varied fauna
than do softwater streams. Calcium salts, in particu-
lar, are required by mollusks for building their shells,
and by crayfish for the exoskeleton. The salts and
organic matter which are basic substances in all
aquatic food chains depend directly on the fertility
of the soil over which the water drains. Streams
draining areas of fertile soil usually have an abun-
dance of stream organisms ; biomasses of both in-
vertebrate organisms and fish in streams occurring in
areas of poor soil are low.

The productivity of insects in Algonquin Pro-
vincial Park, Ontario, was periodically measured
during one summer by collecting, in cages a yard
square, all insects as they emerged from the water
and transformed into adults. The count varied over
different kinds of bottom between June 1 and August
31, 1940, as follows: rubble 6603, gravel 1636, sand
1079, mud 2618 individuals per sq m. Various moun-
tain streams in different parts of the country have

TABLE 5-5 Relation between numbers per m' and biomass of
Insect groups in a riffles of a California coastal stream during
February and March (after Needham 1934).

Total Wet weight

number of

Individuals Per cent Grams Per cent

Insect

43.9
28.0

Caddisfly larvae

and pupae 742 22.2 5.66

Mayfly naiads 1,853 55.5 3.61
Fly larvae and

pupae 343 10.3 1.02 7.9

Stonefly naiads 260 7.8 1.58 12.2

Miscellaneous 137 4.1 1.02 7.9

Totals

3,335

12.89

shown an annual productivity of trout taken by fish-
ermen of 2.2 to 3.9 g/m- (20 to 35 lbs/acre) wet
weight (Surber 1937).

APPLIED ECOLOGY

The chief problems in applied ecology of
streams are those of erosion and silting, pollution,
and maintenance of biotic productivity at the highest
possible level.

Erosion and silting

Stream erosion becomes considerable when up-
land vegetation is so reduced that there is little or no
retardation of runoff from heavy rains. Dredging and
stream straightening for drainage purposes usually
eliminates the riffles habitat. The bare, hard clay that
often emerges as the new stream bottom supports
very little animal life.

Continuous erosion throws a heavy load of fine
silt into the stream. This is detrimental. It makes
the water opaque ; reduces or prevents photosyn-
thesis in algae, water moss, and other plant life ;
handicaps those fish and other animals that depend
on sight for finding and capturing food : and clogs
the filtering mechanism of various invertebrates.
Clams are ordinarily closed less than 50 per cent of
the time, but in silted waters they may stay closed
up to 95 per cent of the time. Clams secrete mucous
to keep the mantle cavity cleansed, but when silting
is heavy this may not be sufficient and mortality will
result (Ellis 1936). Deposition of silt on rock or
sand bottom may bring a considerable change in spe-
cies composition of animals present. During the last
several decades, greatly increased soil erosion in agri-
cultural areas has reduced pan and game fishes in our
streams, and rough fish, such as carp, have taken
their place.

Chronically muddy streams may often be cleared
by reforesting the watershed, and by practicing mod-

56 Habitats, communities, succession

TABLE 5 6 Seas

onal variation

in invortebr

ate populations per

m" in Callforn

a coastal strea

r, (Needham

1934).

Number of

Wet weight

Predominant

Month

Individuals

In grams

species

February

2,862

7.89

Mayfly naiads

March-April

2,324

9.76

Mayfly naiads

May

18,254

52.94

Blackfly larvae
and pupae

August

4,524

19.37

Caddlsfly larvae
and pupae

November

6,531

23.03

Mayfly naiads

erii erosion control in cultivated areas. With slower
runoff, more rainwater soaks into the ground, and
the water table is raised. It is also desirable to main-
tain vegetation on the immediate stream banks to
slow up undercutting. Streambank vegetation is also
beneficial for shading the water and keeping it cool
enough for such fish as trout. If artificial dams are
necessary, they should be small, and located where
the drainage begins in the numerous headwaters of
the streams. Contour plowing, strip planting, and sod
ditches also slow up water movement in hilly areas
and should be practiced.

Pollution

Pollution occurs wlien foreign substances are
introduced into a body of water in amounts sufficient
to change its character and chemical composition.
This type of pollution is of two forms : industrial
wastes, such as those from lead and zinc works, tan-
neries, breweries, paper mills, gas plants, mines,
atomic energy plants, etc. : and organic sewage. In-
dustrial and mine wastes are often acid, and extreme
acidity will kill fish and other organisms. Clams are
greatly reduced or disappear altogether in acid
waters. Industrial wastes contain a great variety of
chemical compounds, including salts of the heavy
metals, and many of them are very toxic to fish.
Young fish and species of small fish appear especially
sensitive, and the polluting materials may cause physi-
cal or chemical injury to the gills without actually
being absorbed into the body (Doudoroff and Katz
1953).

The control of radioactive wastes from uranium
mills and other atomic energy plants has become an
especially serious problem in modern times. No
stream can purify itself of these wastes. However,
they become diluted downstream, undergo natural
decay, settle out in the mud bottom, and are taken
up by organisms. Organisms take up elements at
equal rates whether they are radioactive or not.
Radioactive elements may thus accumulate and be-
come concentrated in organisms to an extent many

thousands of times greater than their concentration
in water. This is of ])otential harm to man (Tsivo-
glou (7 al. 19S7). Fortunately, streams are but little
used at the present time for the dis|X)sal of radioac-
tive wastes. The ecological significance of radioactive
wastes and fallout from atomic explosions has been
summarized by Odum (19.^9).

The introduction of small quantities of organic
wastes may increase the size and productivity of ani-
mal [copulations by adding to the basic nitrogen sup-
ply. The limit of the sewage load that a stream can
carry without harm is, however, low and soon
reached. As fresh organic material oxidizes, carbon
dioxide and toxic gases are released into the stream,
and there is a drastic reduction in the oxygen content.
Fermentation is more rapid in summer than in win-
ter, and may begin in wastes before they are dis-
charged into the stream. The decomposing organic
material continues to be oxidized as it is carried
downstream, and when this action is completed the
stream is again pure (Coker \954).

There have been many attempts to determine the
degree to which a stream is polluted, by means of
chemical analyses of the water. There is difficulty,
however, in evaluating the extent to which each of
the many chemical compounds to be found is harm-
ful to the various kinds of organisms. There is con-
siderable variation in this respect, even between dif-
ferent stages in the life-cycle of the same species.
Furthermore, the sewage load may vary from time to
time, and infrequent heavy loads may wipe out the
animal life in localities where chemical measurements
made at other times do not indicate harmful pollu-
tion.

Various investigators (Richardson 1928, Ellis
1937, Paine and Gaufin 1956, Gaufin and Tarzwell
1956) have attempted to use invertebrate animals as
indicators of pollution. The presence of midge fly lar-
vae Tendipes ripariiis. Glyptotcndipcs. mosquito
larva Cule.v pipiens. rattail maggot, and sludge fly
delimit zones of septic pollution. There are relatively
few species that can tolerate septic conditions, but
those that do may become very abundant. The oligo-
chaete worms Tubifex and Lininodrilus, and certain
midge fly larvae, such as Tendipes plumosiis. indi-
cate low o.xygen. In general, pond invertebrates are
much more tolerant of low oxygen concentration than
are those belonging to stream habitats. Species espe-
cially tolerant of pollution are those that have adapta-
tions for obtaining oxygen at the water surface, such
as the dipteran larvae of Culicidae, Syrphidae, and
.Stratiomyidae, aquatic Coleoptera and Hemiptera,
and pulmonate snails. Gill-bearing species generally
require clean water of high oxygen content. Among
fish, pond species such as carp, bullhead, perch, and
crappie are relatively more tolerant than stream
species.

Streams 57

FIG. 5-10 Schematic diagram of possible stream modifications
affording improved protection and spawning facilities to fish
(after Lagler 1952).

Stone deflector

Fish management

Basic to fish management in streams is the con-
trol of soil erosion and pollution. In clean, clear
streams, both invertebrates and fish can attain high
populations through normal reproduction. Artificial
propagation and release of reared fish into streams
to improve fishing is not necessary except where
habitats have been depleted of breeding stock or
where the fishing pressure is excessive. The artificial
raising and releasing of fish of suitable size for quick-
recapture in sport fishing is expensive but sometimes
justified in highly populated areas. In most regions
the fish manager is better concerned with improving
habitats and letting the fish repopulate them to full
carrying capacity on their own accord.

Stream fishes suitable for sport and food are pri-
marily those inhabiting the pools rather than the rif-
fles. The carrying capacity of streams can sometimes
be raised by artificially increasing the number of
pools without destroying too many of the riffles, the
main source of fish food. The interspersion of ponds
along the stream also increases its fertility, since they
are the sources of plankton, detritus, and dislodged or
escaping organisms. The formation of pools may
often be done inexpensively by making simple log
dams or deflectors. Occasionally, it may be desirable
to haul in gravel from elsewhere to make spawning
beds and to provide artificial log or brush shelters
(Needham 1938, Lagler 1952).

There has been a country-wide practice of intro-
ducing species of fish into streams where they did
not originally occur. The result has been to greatly
mix up and modify the fish fauna ; original primitive
communities no longer prevail. This is unfortunate
for ecological research. The U.S. National Park
Service is, however, attempting to preserve a certain
number of natural stream areas in their original con-
dition, prohibitng fishing therein (Kendeigh 1942a).

SUMMARY

Streams contain riffles, sand-, and mud-
bottom pools. Inhabitants of the riffles and sand-
bottom pools constitute a distinct stream biocies.
Mud-bottom pools are inhabited by species from the
pond-marsh biocies. Animals adjust to the action of
water current by clinging mechanisms, avoidance,
or vigorous swimming. They are generally positively
rheotactic, and several forms maintain orientation to
a particular position in the stream by means of visual
landmarks. Segregation to dififerent habitats depends
largely on differential response to the substratum ;
that is, preference respectively for rock, sand, or
mud. Animals occurring in mud-bottom pools are
usually negatively rheotactic, or become helpless in
strong current. They also have adaptations tolerant
of lower oxygen concentrations in the water. Changes
in the size of the stream, occasioned by various physi-
cal factors, also afifect the responses of animals.
Stream animals have apparently evolved from an-
cestral types that occupied the quiet waters of lakes
and ponds. The life cycles of many stream insects are
remarkable for the long duration of immature stages
and the brief life of adults. Animals have various
adaptations to feed on detritus in the water, on di-
atoms, on filamentous algae, or for being carnivorous.
Density of individuals, biomass, and productivity of
invertebrates are ordinarily less in sand-bottom pools
than in either riffles or mud-bottom pools. Clams and
game fish, however, inhabit sand-bottom pools. Fish
management requires the proper interdigitation of
pools and riffles, as well as control of erosion, silting,
and pollution.

58 Habitats, communities, succession

Lakes are large Ijodies of fresli water, often deep
enough to have a pronounced thermal stratification
for part of the year. Typically, shores are barren and
wave-swept (Muttkowski 1918).

I.akes are formed in youthful stages of river
system develojjment. Water from upland runoff,
groundwater seepage, springs, and melting snow-
fiekls and glaciers collects in basins. As the basins
fill to overflowing, erosion of outlets starts ; as it goes
on, outlets are deepened and water level of the lake
drops. Products of erosion, carried into the basin by
wind and water, and the products of animal and plant
decay accumulate, making the water shallow.

IVIorijhometry aside, the essential distinction be-
tween lake and stream habitats is the characteristic
of water movement ; continuous, rapid flow is the
characteristic of the stream, the lotic habitat. The
lake is a Icntic habitat ; the water is essentially a
standing, quiescent body, although at times wind ac-
tion stirs surface layer and margins into great turbu-
lence. Habitat factors associated with the lentic en-
vironment are uniquely modified to it (Welch 1948,
Hutchinson 1957).

HABITAT

6

Local Habitats,

Communities, and

Succession:

Lakes

Pressure, density, and buoyancy

The pressure imposed on a lake-dwelling or-
ganism is the weight of the column of water above it
plus the weight of the atmosphere. Most lakes have
a ma.ximum depth of less than 30 meters : the Great
Lakes of North America vary from 64 to 393 meters
in depth. Crater Lake in Oregon is the deepest on the
continent, 608 meters (Welch 1952). Ma.ximum
pressures are much less than in the ocean, and or-
ganisms appear to adjust to them readily. The ab-
sence of animal life from deep water is ordinarily a
consequence of low oxygen supply, or low tempera-
ture, rather than pressure.

The density of water varies inversely with tem-
perature and directly with the concentration of dis-
solved substances. Water is most dense at approxi-
mately 4°C. Water becomes progressively less dense
as it is cooled below -)-4°C; ice expands markedly
(i.e., becomes less dense) the colder it gets. It is
because the coldest water is at the surface in
winter that ice forms there, rather than at the bot-
tom. In summer, the coldest waters of deep lakes
are at the bottom. Dissolved salts increase the
density of water; the density of most inland water-
bodies' is much less than that of the ocean. When
great evaporation occurs in a lake having no out-
let, as in the Great Basin, the lake may come to
contain a higher percentage of salts than the ocean.
The few species capable of living in these very salty
lakes include some algae and Protozoa, the brine
shrimp Artemia gracilis, and the immature stages of

59

two brine flies, Ephydra gracilis and E. hians. There
are no fish in the Great Salt Lake of Utah (Wood-
bury 1936).

By the law of Archimedes, the buoyancy of an
object is equal to the weight of the water it dis-
places. Buoyancy varies with the density of water,
and is influenced by the factors that affect density.
Viscosity, the measure of the internal friction of
water, varies inversely with temperature and also in-
fluences buoyancy.

An organism will sink unless it keeps station by
swimming movements, or unless it has special adap-
tations to decrease the specific gravity of the body
and take advantage of any turbulence in the water.
Such adaptations take several forms : absorption of
large amounts of water to form jelly-like tissues;
storage of gas or air bubbles within the body : forma-
tion of lightweight fat deposits within the body, or
oil droplets within the cell ; increase of surface area
in proportion to body mass, which increases frictional
resistance (Davis 1955). When an organism so
equipped dies, the special mechanisms quickly cease
to function, and it sinks to the bottom. If dead or-
ganisms did not sink to the bottom, living organisms,
with the exception of some bacteria, could not exist in
an aquatic habitat.

An interesting phenomenon is cyclomorphosis. a
seasonal change in body form that develops in many
plankton organisms, both plant and animal, including
protozoans, cladocerans, and rotifers. In general, the
summer generations have higher crests, longer spines,
longer beaks, or longer stalks, than do the winter gen-
erations. It is believed that the increased surface area
provided in the summer forms is induced by the
higher water temperatures obtaining then, and may

FIG. 6-1 Cyclomorphosis of Daphnia retrocurva In a Connecticut
lake (from Brooks 1946).

be an adaptation to the decreased buoyancy of the
water at this season, but factors other than temper-
ature also appear to be involved (Brooks 1946).

Light

The daily alternation of light and darkness es-
tablishes a rhythm in the activities of many aquatic
organisms. Light is essential to plant photosynthesis ;
some fish require light by which to feed. Many or-
ganisms orient to light, and some are sensitive to
light of particular wavelengths, notably ultraviolet.
Small, soft-bodied, bottom-dwelling organisms are
particularly sensitive to light, and it is thought that
the evolution of pigmentation, chitinous exoskeletons,
shells, cases, and similar structures may have helped
certain otherwise photosensitive species to survive
in shallow, well-lighted areas (Welch 1952).

A common way to measure the relative transpar-
ency of water is to lower a Secchi disk, a white plate
20 cm in diameter attached to a cord marked off in
linear units, marking the depth at which the disk dis-
appears from sight. The disk is lowered a bit farther,
then raised until it reappears, and that depth marked.
The two depths are averaged. The light intensity at
the depth of disappearance of the disk is usually about
5 per cent of that at the surface (Hutchinson 1957).
Other more exact procedures employ photographic
methods, pyrlimnometers, or photoelectric cells
(Shelford 1929).

The depth to which light penetrates into water
is affected by intensity of the light, angle of ray in-
cidence, reflection at the surface, scattering within the
water, and absorption. Penetration anywhere is re-
duced when the sun is away from the zenith ; is less
in waters at high latitudes ; and is much less in winter
compared with summer. About 10 per cent of the
light falling on Lake Mendota, Wisconsin, during
the spring and summer is reflected ; about 1 5 per cent
during the autumn (Juday 1940). In the unusually
clear waters of Crystal Lake, Wisconsin, measure-
ments with a pyrlimnometer indicated only a small
surface reflection light loss, a penetration of 67 per
cent of full intensity to a depth of one meter, and
10.5 per cent of full intensity at 10 meters (Birge
and Juday 1929). In pure water, red light is ab-
sorbed most rapidly, at a rate of 64.5 per cent per
meter : orange, at 23.5 per cent per meter ; yellow, at
3.9 per cent ; green, at 1.1 per cent ; blue at only 0.52
per cent. Blue penetrates the farthest. Violet is ab-
sorbed at 1.63 per cent per meter. Very little ultra-
violet penetrates the water, and nearly all the infra-
red is absorbed in the first meter (Clarke 1939,
Ruttner 1953).

Suspended material in water produces turbidity,
and reduces light penetration. In western Lake Erie,

60 Habitats, communities, succession

the (lc])tli to wliicli 1.0 i>er cent i)t surface light pene-
trates varies from 9.7 in when tnrbidity is 5 ppm, to
0.8 ni when turbidity is 115 ppm (Chandler 1942).
Since phytopiankton rec|uire light for photosynthesis,
abundance varies inversely with turbidity. Light
penetration is also affected by the abundance of or-
ganisms themselves, both phyto- and zooplankton.

An appreciable amount of light jjasses through
ice in the winter. This enables phyto])lankton photo-
synthesis to continue. In eutrophic lakes many fish
may suffocate when snow overlies surface ice, pre-
venting photosynthesis and, thus, the generation of
oxygen (Greenbank 1945).

The apparent color of water bodies may be the
result variously of sky reflections, the color of the
bottom, suspended materials, or of plants and animals.
Hut apart from these extraneous factors, water often
has an intrinsic color deriving from its chemical
contents. The blue color of pure water is a result of
blue light scattering by water molecules. Iron gives
water a yellow hue. A green color is usually associ-
ated with high concentrations of calcium carbonate.
\\'ater from bogs or swamps contains humic ma-
terials and is often dark brown. Many waters arc
essentially colorless. In a Wisconsin lake showing
practically no color, maximum photosynthesis of
algae occurred at one meter depth on bright days :
some photosynthesis occurred down as far as 15
meters. In a highly colored lake, maximum photo-
svnthesis occurred at 0.25 meter, none at 2 meters
r'Schomer 1934).

Photosynthesis releases oxygen into the water :
respiration and decomposition absorb it. The upper
layer of a lake, where photosynthesis predominates,
is called the trophogcnic zone. Below this zone there
may still be considerable photosynthesis, but oxygen
absorption is greater than oxygen release. The
deeper portion of a lake is called the tropholytic zone.
The two zones are separated by a thin layer where
the oxygen gains from photosynthesis during the day-
light hours are balanced by the respiratory and de-
composition losses during the day and night. This is
the compensation depth, to which generally about one
per cent of the full sunlight at the water's surface
penetrates. The compensation level in a dark-colored
bog may lie less than a meter below the surface ; in
a deep, clear lake it may be 100 m down.

Wind and currents

^\'ind is an important environmental factor of
lakes because of water currents it generates. The
effect of wind action depends largely on the extent of
the exposed water surface, the presence or absence
of protecting upland, and the configuration of the lake
relative to the prevailing wind direction.

Waves may become sizable in large lakes, but the
forward motion of a wave does not involve any great
mass of water. The rate of movement of surface
water is usually less than 5 per cent of the velocity
of the wind. The wave form moves on while the
water beneath undergoes a more nearly cycloidal mo-
tion, except along the shore, where the wave mass
progresses forward and breaks as surf. The water
washes back off the beach as an undertow, only to
he carried forward again by the incoming waves.
Tiie problem of maintaining position here is similar
to the problem of maintaining position in streams. The
depth of wave action in the open lake and along the
shore depends largely on the strength of the wind
(Kuttner 1953).

In summer, surface water is warmed by solar
radiation and its density, weight, and viscosity de-
crease. In deep lakes the warm water piles up on the
exposed shore until, moving down along the bottom,
it encounters colder and denser waters, which resist
mixing. The warm water is then diverted horizon-
tally to the opposite shore. Thus the lake beomes
stratified horizontally into an upper cpilimnion, where
the water circulates and is fairly turbulent, and a
lower hypolnnnion. which is relatively undisturbed.
This difterence in circulation in deep lakes is closely
correlated with differences in temperature and oxy-
gen characteristics : it is of considerable importance
in the distribution of the biota.

Temperatur(>

The thermal conductivity of water is very low ;
but because of the thorough mixing of the waters in
the epilimnion during the summer by wind action, the
temperature is nearly uniform down to the thermo-
cline. The thermocline is the zone of most rapid
temperature decrease, generally involving a drop of

FIG. 6-2 Water cu
Uke.

Lakes 61

32 35

40

TEMPERATURE
45

-■ °F

30 55

50

5

' r

1

1

' / / 1

'l\-

/

/ [: "^"f ^ 1

Nov 2 ''

^ J Aug 6 j -

1/

10

V!

/ Jul 2 L

/[

^ 1/

--"""^ ,

1"=.

May 7 /j

/ y

1

1

/

•■%!

-

Dec 7 1

-

V,

Oct 2 1

25

\\i

f // 1

1

1 Feb 6 1

30

\J

\

\W 1

if /

!«. .|

35

/

/

1 1

/

/

1

-

TEMPERATURE, °C

FIG. 6-3 Vertical temperature distribution throughout a year In a dimictic lake of the second order — Convict Lake, California;
elevation 2308 m (from Reimers and Combs 1956).

at least 1°C per meter of depth (Birge 1904) and
occasionally as much as 7°C per meter. The thermo-
cline, as here defined, equals the metalimnion of some
authorities who have a different conception of the
thermocline (Hutchinson 1957). When the thermo-
cline forms, early in the season, it is close to the
surface. As the season progresses, it sinks lower, in-
creasing the volume of the epilimnion and decreasing
the volume of the hypolimnion. The temperature of
the hypolimnion is fairly uniform, although it de-
clines gradually from the lower edge of the thermo-
cline to the bottom, where it is seldom below 4°C.
During the autumn, the surface water cools and

the thermocline sinks. The epilimnion increases in
thickness until it includes the entire lake. The waters
are then uniform in temperature and density, at all
depths. Even slight winds produce complete circula-
tion. This is the autumn overturn, which may last
for the several weeks, until ice forms.

As surface waters cool below 4°C they no longer
sink, and ice may form. Less dense than the underly-
ing water, it floats. Immediately below the ice, the
temperature of the water is very close to 0°C, but in
one or two meters of additional depth it usually rises
rapidly to 4°C, although in some lakes temperatures
below 4°C occur even at considerable depths.

62

Habitats, communities, succession

As tlie ice melts during the sjiriiig aiul the sur-
face waters warm up, a sf>rin(i overturn occurs wlien
the water at all depths is at the same temperature.
The time and duration of the spring overturn de-
pends on weather conditions ; it may last several
weeks. It often occurs intermittently, however, cor-
responding with changes in weather and water tem-
peratures.

When a lake has two overturns during the year,
it is called a diiitictic lalcc. Such lakes are character-
istic of. but not limited to, temperate climates. In
warm, oceanic climates and in the tropics, the sur-
face waters may not cool sufficiently to permit com-
plete circulation, e.xcept during the coldest period of
winter. Lakes undergoing a single overturn are
called ti-anii monomictic. The temperature of the
water in the hypolimnion of such a lake is never
lower, of course, than the mean air temperature dur-
ing the period of the last complete circulation ; in
w-arm climates this may be several degrees above
4°C. On the other hand, lakes in polar or alpine
regions may never warm above 4°C, and complete
circulation occurs only in the middle of the summer.
These are cold monomictic lakes (Hutchinson 1957).
The three types of lakes were formerly called tem-
perate, tropical, and polar, but this terminology is
undesirable since their geographical segregation is
not precise.

Lakes of the first order are those in which the
bottom water remains at or near 4°C throughout the
year, and while one or two circulation periods are
possible, there is often none. In lakes of the second
order, the temperature of the bottom water rises
above 4°C during the summer, and there are one or
two regular circulation periods during the year.
Lakes of the third order do not develop thermal strati-
fication, and circulation of water is more or less con-
tinuous (Whipple 1927). In general, lakes over 90
meters in depth belong to the first order : those be-
tween about 8 and 90 meters belong to the second
order : and those less than 8 meters to the third order.

The specific heat of water is greater than most
other substances : accordingly, a vast amount of heat
must be absorbed to cause a temperature change.
Temperature change is, in any event, slow. Much of
the energy of solar radiation is lost by reflection from
the water surface. The rest of the radiation is ab-
sorbed by the water, the solutes, and the suspended
material. But much of the diurnal energy increment
may be dissipated by re-radiation at night or in
cloudy weather, by evaporation, and by convectional
cooling. The amount of heat actually retained by a
lake to melt its winter ice and warm it from the
winter minimum up to the summer maximum is its
annual heat budget (Table 6-1). For many dimictic
lakes this is between 20,000 and 40,000 g-cal/m- of
surface ; there is wide variation in different kinds of

TABLE 6 I Monthly change in cumuUtivs heat budget and solar
radiation In the Bass Islands region (depth 7.5 m| of western
Lake Erie, in 1941. A 20.3 cm ice covering formed in mid-Janu-
ary, melted in late March. The maximum heat budget, reached
on July 30. was 19,575 g-cal/cm*. The heat budget was about
15 per cent of the total solar radiation received during the
year (after Chandler 1944).

g-cal/

cm'

g-cal/cm'

Solar

Solar

Heat

radia-

Heat

radia-

Month

budget

tion

Month budget

tion

January

105

3.364

July 18,765

17,291

February

206

5,849

August 18,112

16.375

March

581

10,201

September 16.331

12.737

April

6,405

13,952

October 11,115

6,829

May

12,581

17,156

November 4,350

4,147

June

16,369

15.960

December 1,369

2,599

lakes. The annual heat budget is important in de-
termining a lake's productivity.

OXY

gen

The distribution of o.xygen at various depths
depends upon the presence or absence of a thermo-
cline, the amount of vegetation, and the organic na-
ture of the bottom. The amount of oxygen in water
is only one-fortieth to one-twentieth of that present
in an equal volume of air when the two are at equi-
librium, although their partial pressures are the same.
Diffusion of oxygen from the air into comparatively
sedentary water occurs very slowly ; agitation of the
water increases the surface area and promotes a faster
rate of equilibration.

SAMPLE DATES- "^^^"""^ "^ '^

12 3 4 5 6 7

CUBIC CENTIMETERS PER LITER

FIG. 6-4 Changes in the vertical distribution of oxygen through-
out a year in a dimictic eutrophic lake — Lake Mendota, Wis-
consin (after Blrge and Juday 1911).

Lakes 63

The amount of oxygen released by plants varies
with their abundance and time of day ; photosynthesis
can take place in light. Phytoplankton and the
rooted vegetation restricted to the shore line are im-
portant sources of oxygen to the water. With rapid
photosynthesis in relatively small volumes of water,
the water may be supersaturated with oxygen for
short periods of time.

The oxygen supply of lakes is reduced in various
ways; most notably through the respiration of ani-
mals and plants and the decomposition of organic
matter. As lake waters warm up during the summer,
their capacity to hold oxygen is reduced and oxygen
may be released into the atmosphere. The saturation
capacity of water at 0°C is 10.2 cc per liter, but at
25°C it is only 5.8 cc per liter. In some lakes, decom-
position of organic material at the bottom may deplete
the hypolimnion of its oxygen content for several
weeks during the summer ; perhaps lower than the
level minimal to the support of life. This is called the
summer stagnation period. During the winter, if the
lake is covered with ice and snow, there may be a zvin-
ter stagnation period. The oxygen supply of the deep
waters is renewed with the autumn and spring over-
turns. Before decomposition can proceed very far
there must be calcium in the water. Hence, decom-
position is slow in soft or acid waters.

At temperatures of 15°-26°C oxygen concentra-
tions of less than 2.4 cc/1 (3.5 ppm) are fatal within
24 hours to several species of fish. From 0°-4°C,
oxygen concentrations can decline through 48 hours
to 1.4 cc/1 (2.0 ppm), or even to 0.7 cc/1 liter (1.0
ppm), before the same mortality results (Moore
1942). Some planktonic invertebrates can tolerate
oxygen concentrations as low as 0.2 cc/1 (0.3 ppm)
and, for short periods, even 0.1 cc/1 (0.1 ppm).
Some bottom-dwelling protozoans, annelids, mol-
lusks, and insect larvae may survive actual anaerobic
conditions for periods of days, even weeks. Organ-
isms that tolerate a lack of oxygen do so by creating
an oxygen debt ; that is, the lactic acid and other
breakdown products produced in consequence of
muscular activity simply accumulate until conditions
permit oxidation of them. In true anaerobes these
acid waste products are eliminated from the body ;
no oxidation debt is established.

European workers, principally Thienemann and
Naumann, have devised a classification of lake habi-
tats into three main categories on the basis of fertil-
ity and the amount of oxygen in the hypolimnion
during the summer concentration. The oxygen con-
centration in the hypolimnion is, of course, a reflec-
tion of the fertility of the lake, since it is inversely
proportional to the amount of decaying organic mat-
ter. Dystrophic lakes contain considerable organic
matter but are infertile because the organic matter

does not completely decompose and there is release
of organic acids.

Oligotrophic lakes are usually deep (over 18
meters) with very little shallow water, and little veg-
etation around margins. Bottom contours are V-
shaped ; they are low in fertility, rich in oxygen in
the hypolimnion (orthograde distribution), low in
CO2, and the color of the water varies from blue to
green. The volume of the epilimnion is usually less
than the volume of the hypolimnion. The fish popu-
lation is not large. Characteristic species are lake
trout, whitefish, and cisco. The midge fly larva,
Tanytarsus. predominates. Plankton is not abundant.
The Finger Lakes of New York are of this type.

Entrophic lakes are usually less than 18 meters
deep, the bottom contour is U-shaped, water color
varies from green to yellow or brownish green, and
there are larger areas of shallow waters and more
marsh vegetation. Fertility is high, and because of
rich bottom humus the oxygen content of the hy-
polimnion is greatly reduced during the summer
{clinograde distribution). The COo content is ac-
cordingly high. The volume of the epilimnion is
usually greater than that of the hypolimnion. Plank-
ton is abundant. The midge fly larva Tendipes is
very numerous and the culicid larvae Chaobonis is
usually present. The bottom fauna is rich, and there
is a large fish population in the epilimnion. Charac-
teristic fish species are the largemouth bass, perch,
sunfish, and pike. These lakes occur in relatively
mature river systems ; many lakes in Minnesota and
Wisconsin are of this type.

Dystrophic lakes are bog-like, very rich in mar-
ginal vegetation and organic content. Oxygen is
likely to be scarce at all depths. The water is usually
conspicuously colored, yellow to brown, and may be
acidic because of organic acids and incompletely oxi-
dized decomposition products. Plankton, bottom or-
ganisms, and fish are usually scarce, but blue-green
algae are sometimes abundant. Tendipes may pre-
dominate among the bottom forms, but at times only
Chaoborus is present. Characteristic fish are stickle-
backs and mud minnows. Many lakes of northern
latitudes are dystrophic in type.

All gradations exist between these three types of
lakes, and individual lakes are often difficult to
classify. Oligotrophy is indicated if the loss of oxy-
gen in the hypolimnion during the summer is not
over 0.025 mg/cm-/day ; eutrophy. if it is over 0.055 ;
mesotrophy, if it is between the two (Hutchinson
1957). A lake may change from one type to another
as succession proceeds (Lindeman 1942). Probably
all lakes start as oligotrophic, but as they accumulate
vegetation and decaying organic matter, they change
into eutrophic lakes ; or, if the organic matter does
not completely decompose, into dystrophic lakes.

64 Habitats, communities, succession

Kutropliic lakes may later develop into ponds and
marshes ; dystrophic lakes, into bogs. Biotic succes-
sion is scarcely discernible, however, in very large
or very deep lakes. The Great Lakes, for instance,
will endure until erosion lowers their outlets.

Carbon dioxide and other gases

Carbon dioxide is required by plants for photo-
synthesis. Its presence in lake waters tends to vary
inversely with oxygen. Carbon dioxide is derived
from the atmosphere, the respiration of both animals
and plants, decaying organic matter, ground water,
and bicarbonate salts. It may occur in either the jrcc
state (dissolved CO.), Iialj-boiind state (HCO3),
or fixed state (CO.i). These three states are asso-
ciated respectively with pH values 7, 7 to 10, and
above 10. Algae and some rooted aquatic vegetation
are able to obtain the half-bound CO2 from the solu-
ble bicarbonate salts, thereby converting them into
the less soluble carbonates :

Ca(HC03)2:

: CaCO:, + CO-, + H2O

Mollusks, a few insects, and some bacteria are also
able to precipitate carbonates. Carbonates precipitate
as to make conspicuous marl deposits on the bottom
of some lakes. When marl formation becomes consid-
erable, there is a decrease in lake fertility and a conse-
quent decrease in animal life present, including
bottom-inhabiting organisms.

When there is sufficient free carbon dioxide in
the water derived from sources other than carbonates.
they are converted back into bicarbonates and marl
does not form. The degree of alkalinity of a lake is
measured by the amount of carbon dioxide or acid
required to convert the excess carbonates into bi-
carbonates, yielding neutral water. Soft-water lakes
contain not over 5 cc/1 fixed carbon dioxide :
medium-class lakes contain 5 to 22 cc/1 ; hard-water
lakes may have from 22 to as high as 50 cc/1 ( Birge
and Juday 1911).

Marsh gas (methane) evolves from organic mat-
ter decomposing at the bottom. It rises in bubbles to
the surface of the water. Methane formation may be
extensive during the summer stagnation period.
Methane does not appear to be particularly toxic to
organisms until it is generated in very large amounts.

Hydrof/en sulphide results from anaerobic de-
composition of sulphurous organic matter. It may be
conspicuous in sewage-polluted waters. It is inher-
ently very poisonous.

Nitrogen occurs in water by reason of difTusion
from the atmosphere. When present in excessive
amount it has been known to form bubbles in the
circulatory systems of fish causing death, but this
does not commonly occur in natural waters.

Ammonia may occur naturally in water, a result
of decomposition of organic matter. Ammonia may
also be dumped into streams and lakes from indus-
trial plants, often in concentrations toxic to fish. Fish
are apparently unable to detect the ])resenfe of an)-
monia in water.

Dissnlvvd solids

l-'.-illing r.iin may contain as much as 30 to 40
l)pm of solids, and the runoff dissolves more as it
drains over the upland into streams and lakes. Water
draining off siliceous or sandy soils may contain 50
to 80 ppm of dissolved minerals ; off more fertile
calcareous soils, 300 to 660 ppm. Lake waters com-
monly vary from about 15 to 350 ppm of dissolved
minerals, although in some lakes of the Great Basin,
the total dissolved salts exceed 100,000 ppm. The
ocean contains only 33,000 to 37,370 ppm.

Inorganic salts especially important for plants in-
clude ammonium salts, nitrites, and nitrates as
sources of nitrogen ; phosphates to supply phosphorus
which, with nitrogen and sulphur, are raw materials
for protein synthesis ; silicates, which furnish silicon
to diatoms and sponges : and salts of calcium, mag-
nesium, manganese, iron, copper, sodium, and po-
tassium for proper development of chlorophyll and
growth of plants and, indirectly, of animals. Mol-
lusks require calcium salts for shells. Crayfish and
other arthropods require calcium for the carapace ;
vertebrates, for their skeleton. Absence of these
necessary salts in lake waters limits the kinds and
abundance of animals that can live there. Phosphorus
and nitrogen are the most likely to be deficient.
Nutrient salts tend to accumulate in the deeper waters
and at the lake bottom, but they are brought to the
surface at the autumn and spring overturns. Lakes
in prairie regions tend to have more salts than those
in deciduous or hardwood forests, which, in turn,
have more salts than lakes in coniferous forest areas
(Moyle 1956). The total dissolved content of a lake
is important in determining its general level of pro-
ductivity (Northcote and Larkin 1956).

Little is known about the amount of amino acids,
fats, and carbohydrates occurring in natural bodies
of water and how much of this nutrient material may
be directly absorbed by organisms. Dissolved organic
matter is derived chiefly from plankton remains, and
other dead plants and animals as well as from bot-
tom mud and external sources. In Wisconsin lakes,
there is about 15 mg/1, of which crude protein con-
stitutes 15 per cent, fats or ether extract 1 per cent,
and carbohydrates about 83 per cent. Dissolved or-
ganic material becomes higher, of course, in dystro-
l)hic lakes and peat bogs (Birge and Juday 1934).

Lakes 65

Hydrogen-ion concentration

LAKE BIOCIES

The acidity or alkalinity of water depends on
the ratio between the H+ (or hydronium, H3O + )
and OH- ions. The amount of acidity or alkalinity
is commonly expressed in terms of potential hydro-
gen ions in a pH scale. The values on this scale rep-
resent the logarithm of the reciprocal of the normal-
ity of free hydrogn ions. When the number of H +
ions is equal to the number of OH" ions, the pH
value is 7, the value which represents absolute neu-
trality. All pH values less than 7 indicate a greater
number of H+ ions than OH^ ions, which is to say
the closer the pH value approaches 0, the more acid
the water. Above pH 7, there is a preponderance of
OH~ ions; the higher the pH value, up to 14, the
more alkaline is the water.

The hydrogen-ion concentration of most unpol-
luted lakes and streams is normally between pH 6.0
and 9.0, but extreme values of pH 1.7 and pH 12.0
occasionally occur (Hutchinson 1957). In some
bodies of water, the pH value fluctuates consider-
ably. Hydrogen-ion concentration increases (low pH
values) with active decomposition of organic matter.

In general, aquatic animals can tolerate great
changes in pH, although the range of toleration varies
between species. Mollusks are not ordinarily found
in acid lakes, but some snails can survive pH as low
as 6, and the fingernail clam Pisiduim down to pH
5.7. At the lower pH values, the shells of mollusks
become thin, fragile, and transparent, but it is be-
lieved that the cuticular covering is partially protec-
tive and prevents complete dissolution of the calcium
carbonate by the acid (Jewell and Brown 1929). In
Campeloma snails, the apex of the shell may com-
pletely dissolve, exposing the apex of the visceral
mass. Most fish can tolerate pH 4.5 to 9.5 provided
there is plenty of oxygen (Brown and Jewell 1926,
Wiebe 1931), and many invertebrates will tolerate
even greater extremes. Fish as individuals become
acclimated to certain pH values, and will select those
values when given choice in a gradient. Such ac-
climation of individuals may have an effect on their
choice of natural habitats, although when forced into
a habitat with a different hydrogen-ion concentration,
they change their acclimation. Although the direct
ecological importance of differences in hydrogen-ion
concentrations is doubtful, the measurement of pH
may serve as an index of other environmental con-
ditions, such as the amount of available carbon di-
oxide (with which it varies inversely), dissolved
oxygen (with which it varies directly), dissolved
salts, etc. Sometimes the difference in species of
plankton found in bodies of water with permanently
different pH values, for example in granite and lime-
stone, is very striking (Reed and Klugh 1924).

If we reserve ponds and peat bogs to sep-
arate consideration, there remain two major lake
communities. They differ in species composition,
abundance of organisms, distribution of niches, pro-
ductivity, and physical characteristics. Inasmuch as
these two communities correspond fairly well to the
oligotrophic and eutrophic types of lakes, we may
name them simply the oligotrophic and eutrophic lake
biocies. Various facies of each community, or inter-
mediate types (Deevey 1941) are affected by varia-
tions in the abundance of component species and
correspond to differences in temperature, depth, fer-
tility, and other features of the habitat. The com-
munities that occur in dystrophic lakes ; for instance,
are an impoverished facies of the eutrophic lake
biocies. In spite of taxonomic differences in constitu-
ent species, each lake biocies contains organisms be-
longing to the same life-forms and with similar mores
so they may be discussed together.

Depending largely on their morphological adap-
tations and behavior, aquatic organisms are, for con-
venience, divided into plankton, neuston, nekton, and
benthos, although the differences between the groups
are not precise. Seston is a collective term that in-
cludes all small particulate matter, both living and
non-living, that floats or swims in the water. Plank-
ton are free-floating or barely motile organisms,
either plant (phytoplankton) or animal (sooplank-
ton), that are readily transported by water currents.
Most plankton are microscopically small, although
some forms are visible to the unaided eye. Species
that can be caught with a net are called net plankton
to distinguish them from the minute varieties that
pass through No. 20 silk bolting cloth meshes. The
latter include most protozoan, bacterial, and fungal
forms, collectively called nannoplankton. Organisms
that depend on the surface film for a substratum are
called neuston and are more important in the quiet
waters of ponds than in lakes. Nekton are larger
animals that are capable of locomotion independent
of water currents. Aquatic birds that swim and dive
are included in this group. Benthos organisms are
attached to or dependent on the bottom for support ;
there are sessile, creeping, and burrowing forms.

PLANKTON

Fresh-water plankton (Welch 1952, Pennak
1946, Davis 1955) includes representatives from
the photosynthetic algae, Bacillariaceae (diatoms), I

Myxophyceae (blue-green), and Chlorophyceae
(green), and occasional other form such as Wolffia
among the higher plants ; the non-photosynthetic bac-

66 Habitats, communities, succession

tenia and otlier fungi ; ami among tla- zooplankton,
all classes of Protozoa except Sixirozoa, Rotatoria,
Kntoniostraca (especially Cladocera. Copepoda, and
Ostracoda), some immature Diptcra, the statoblasts
and gemmules of bryozoans and sponges, the rare
fresh-water jellyfish, Crast^edacusta. and occasional
aquatic mites, gastrotrichs, and others. Fresh-water
plankton lack many forms common in the plankton
of the ocean. On the other hand, the rotifers, aquatic
insects, and water-mites are mostly absent from the
sea, and the Cladocera are only poorly represented.
It is probable that plankton evolved from benthonic
forms occurring near the shore (Ruttner 1933), and
many species of groups listed above, notably Ostra-
coda and Rotatoria, are still largely benthonic in be-
havior.

The algae in fresh water may vary in numbers
from hundreds of thousands to tens of millions of
cells per liter ; Protozoa, from thousands to hundreds
of thousands of individuals per liter ; and the rotifers
and entomostracans, from less than ten to hundreds
per liter.

Distribution

Many species of plankton are nearly world-
wide in distribution, particularly those that occur in
the larger lakes. Cosmopolitan distribution and the
many primitive types of the plankton community in-
dicate that its origin is ancient. Some plankton, how-
ever, such as species of the genus Pseudodiaptomns,
have a very limited distribution.

The plankton found in the open water of small
to medium-sized lakes is seldom more than one to
three species of copepods, two to four species of
cladocerans, and three to seven species of rotifers,
although the species change from one time of the
year to another. It is also unusual to find more than
one species of the same genus at the same time.
When two do occur, one of them is usually much
more abundant than the other. It is commonplace to
find that 80 per cent or more of all limnetic copepods
present belong to a single species : 78 per cent of all
cladocerans to a single species, and 64 per cent of
all rotifers to a single species (Pennak 1957).

In any one lake the horizontal distribution of the
plankton may be irregular because of water currents,
inflowing streams, irregularity of shore line, or
swarming of a particular species in local areas. The
vertical variations in the composition and abundance
of species is even more striking. The chlorophyll-
bearing algae require light and are most numerous
in the upper stratum, although diatoms commonly
occur at greater depths (Fritsch 1931). The verti-
cal distribution of zooplankton varies widely with the
species, but it is strikingly affected by light, food,

FIG. 6-5 Common invertebrates found in laites. (a) Copepod,
(b) cladoceran, (e) ostracod, (d) the snail Amnlcola //moso,
(e) the snail Volyafa iricarinafa, [i] the ghost larva Chaoborus
albipei, (g) the fingernail clam Pisidium. (Modified from various
sources, Pennak 1953.)

gravity, dissolved gases, particularly oxygen, and
thermal stratification. Few zooplankton occur in the
hypolimnion of eutrophic lakes during the summer
stagnation period, but occur at all depths during the
spring and autumn overturns.

Diel movements

Several species of net zooplankton exhibit pro-
nounced vertical migrations, moving upward into
surface strata during the night and returning to
greater depths during the day. In some instances
this daily shifting of position may extend to 60 or
more meters, in other instances it may be only a frac-
tion of a meter, and some species do not exhibit the
phenomenon at all (Langford 1938). A common ex-
planation of these movements is that the animals are
negatively geotatic by nature, but that during the day
this drive is suppressed by a negative phototaxism

Lakes 67

and can be expressed only at night (Parker 1902).
An alternative explanation is that zooplankton ac-
tively orient to a band of optimum light intensity and
move up and down at different times to avoid light
of too great or too little intensity (Gushing 1951,
Hardy and Bainbridge 1954).

These diel movements are most widespread
among Cladocera and Copepoda, but other species
are also involved. One of the most interesting cases
is the dipteran larva Chaobonis pnnctipennis that
rests on the lake bottom during the daylight hours
but is often teeming in the surface waters at night.
It appears that the buoyancy of this larva varies with

the size of its two pairs of air-sacs (Damant 1924).
There are a few rotifers, Mysis among the Mala-
costraca, and Ceratiinn among the Mastigophora, in
which vertical day and night movements have been
demonstrated (Pennak 1944).

Seasonal distribution

The different species of plankton vary in their
response to seasonal changes in the physical and chem-
ical nature of the water, in number of generations per
year, and in time of occurrence. Accordingly there
is a marked seasonal variation in total numbers dur-

ZOOPLANKTON PER LITER

25 50 75

5000 4000 3000 2000

1000 1000 2000
ALGAE PER LITER

3000 4000 5000 6000

FIG. 6-6 Vert
oligotrophlc I
Note that the horizontal

distribution of net plankton (left) in an

and (right) in a eutrophic lake, Wisconsin.

different for the two lakes, and

for the algae as compared with the looplankton. The cross-
hatched horizontal belts show the region of the thermocline
(from Birge and Juday 191 I).

68 Habitats, communities, succession

FIG. 6-7 Vertical distribution of three species of copepods in the daytime (stippled) and at night (black) in the oligotrophic Lake
Nipissing, Ontario, on a July day, when the thermocline occurred between 12 and 15 m (from Langford 1938).

ing the year. In larger and deeper lakes, a maxi-
mum population usually occurs between April and
early June, a minimum in August, a second maxi-
mum in late September or October, and the yearly
minimum in late winter, February or March. How-
ever, not all species follow this schedule ; some spe-
cies have a maximum in the spring and not in the
autumn, or vice versa : and some species reach great-
est abundance during the general summer or winter
minimum.

A species can also exhibit alternate increases and
decreases in population at other times ; these, as well
as fluctuations in total plankton, are called pulses.
At times, especially during the summer when the
water is warm, an algal form, most commonly a blue-
green species, may become so abundant that it dis-
colors the water ; these irruptions are known as
blooms. The death and decay of such masses of vege-
tation may so deplete the oxygen supply that great
mortality of fish and other animals results. In some
cases the algae produce chemicals toxic to animals.

The ways in which environmental factors control
seasonal and other changes in population are not all
clearly understood, but it is significant that the max-
ima in total plankton of deep lakes often come at the
times of the two annual overturns, times when food

and oxygen are abundantly distributed at all depths.
But the bimodal curve may also be found in shallow
lakes and ponds that do not possess thermoclines.
In small lakes, however, there is greater irregularity,
and one, two, three, or no maxima may occur at vari-
ous times of the year (Pennak 19-k)). Periods of
high rainfall, which means increased drainage of nu-
trients into a lake, may be a factor of importance
in producing maxima : seasonal changes in water tem-
perature and oxygen tension certainly are important.
There appears to be no relation between the pulses
of net phytoplankton and zooplankton suggesting ex-
clusive dependency of the latter on the former.

Division.<i

The lake bottom can be divided into a littoral
zone and a profundal zone.

The littoral cone extends from the water's edge
to the limit of rooted aquatic vegetation. It may be
subdivided into the eulittoral zone, between high and
low water marks at the water's edge where the beat-

Lakes 69

I SEP I OCT I NOV I DEC | JAN | FEB | MAR | APR | MAY | JUN | JUL

Seasonal plankton populations In western Lake Erie through a year (after Char

I AUG I
dier 1940).

ing of waves is most effective, and the sublittoral
zone, which extends from the lower Hmit of wave
action to the lower limit of rooted vegetation. Where
such vegetation is absent, the sublittoral zone may be
considered the bottom of the epilimnion down to the
thermocline.

The projundal zone is the entire bottom below the
rooted vegetation, or commonly the bottom of the
hypolimnion. The boundary lines between the zones
are variable and change with the depth of the thermo-
cline. The open water of the lake above the bottom
is known as the limnetic zone.

Littoral zone

The bottom of the littoral zone may be rock,
cobble, gravel, sand, or mud. The muddy shallows
of protected bays may have considerable rooted vege-
tation ; they are essentially pond habitats. Differen-
tiation of species distribution is primarily between
the hard bottom and mud bottom habitats ; sand bot-
tom habitats are transitional (Table 6-2). Sand bot-
toms ordinarily have the lowest population of most
species e.xcept clams because they are unstable habi-
tats at best ; indeed, they are often destructive by
reason of the action of sand grains grinding on each
other (Rawson 1930, Krecker and Lancaster 1933,
Lyman 1956). A lake-bottom and a streambed of

similar composition will contain many of the same
kinds of organisms because of the similarity in the
physical conditions of existence. The respective spe-
cies compositions, however, are often different.

Oneida Lake in New York has an unusually high
mollusk population. Baker (1918) recorded 59 spe-
cies and varieties. It is interesting that most of them
occurred on mud and sand bottoms. The highest
populations were 1890 individuals per m- on mud at
depths less than two meters, and 1573 individuals per
m- on sand. On rocks and gravels there were only
656 individuals per m-.

In eutrophic Douglas Lake, Michigan, bottom de-
posits in the littoral areas show zonation down to a
depth of about 18 m. Beginning at the shoreline,
there are belts of barren, wave-washed sand, muddy
sand, sandy mud, and deep-water soft black ooze, in
that order. The average number of macroscopic ben-
thic animals is large, varying in the different types of
bottom from 369 to 1178 to 3822 to 1713 per m^,
respectively. The abundance of animals is related not
only to the nature of the bottom but also to depth
and vegetation present. Where vegetation was scarce
there were only 162 animals per m-, but with increas-
ing density of plants from sparse to common to abun-
dant the population of animals rose to 1531, 2525,
and 4407 per m-, respectively. Vegetation was most
dense at depths of 7 to 14 m in mixtures of sand and
mud. Most abundant animal species in decreasing

70 Habitats, communities, succession

order were the amphipod Hyalella azteca, the dip-
teran larvae Tendipes and Protcnthes, the snail
Amnicola. tubificid worms, and the sphaerid Pisidittm
(Eggleton 1952). For comparison, the depth distri-
bution of animals in an oligotrophic lake is shown
in Fig. 6-10.

There is also a fauna of microscopic animals
inhabiting the bottom. This consists of Protozoa,
Hydra, Rhabdocoela (flatworm), Nematoda, Rota-
toria, Gastrotricha, Oligochaeta, Cladocera, Cope-
poda, Ostracoda, Acarina (mites), and Tardigrada.
These organisms are often very numerous in the thin
organic ooze-film that covers mud bottoms (Bigelow
1928), but may penetrate underlying deposits to
depths of 20 cm. Sand bottoms also support a varied
and abundant microfauna (Pennak 1940, Cole 1955).
In general, number of microfauna species and indi-
viduals varies inversely as the depth of water ; only
a few species remain active in the profundal zone dur-
ing the summer stagnation period (Moore 1939).
Bacteria, a source of food, are abundant in the bottom
at all depths.

Much of the bottom fauna of the littoral zone
consists of immature stages of otherwise terrestrial
insects. The pulmonate snails and water mites have
evolved from terrestrial species. Other aquatic spe-
cies, however, have related forms in the sea and this
may indicate their evolutionary origin. The funda-
mental problem involved in dispersal from the sea

FIG, 6-9 Two types of bottom samplers for measurement of
benthic populations. When open, each sampler covers a known
area. As each is closed, it scoops up the organisms present.
Right, Peterson's bottom sampler for hard or sandy bottoms; left,
Ekman's bottom sampler for soft bottoms and deep water (from
Welch 1948).

into fresh water would be that of osmoregulation, and
the ability to live in fresh water has doubtless con-
stituted a selection factor in the origin of this com-
munity.

I'rofundal zone

In oligotrophic lakes, species characteristic of
the littoral zone are found at much greater depths
than they are in eutrophic lakes, in which the oxygen
supply during the summer stagnation period is re-
duced. The amphipod Pontoporeia occurs only in the
deeper oxygenated cold waters of some northern lakes
(Adamstone 1924). It is a relic from the glacial
period, when it was probably more widely distributed.
The profundal benthos of one oligotrophic lake in
British Columbia increased from 470 individuals per
m~ in January to 1270 in August (Ricker 1952).

The most common bottom organisms are the an-
nelids Tubifex and Limnodrilus , and the insect lar-
\ae Tendipes. Chaohorus, Protenthes. There may
be a few moUusks, Pisiditim, and Musculium for in-
stance, nematodes, and other forms, including a
microscopic fauna (Eggleton 1931).

The midge larvae represent a variety of species.
Fifty species occur in one small lake in Algonquin
Park, Ontario, that has a pH range of 4.6 to 6.6
and thermal stratification in summer with ample oxy-

Lakes 7 1

Number per n

^

TABLE 6-2 She and distribu-

cobble

and

gravel

tion of invertebrate
populations on different

Common name

Classification

sand

mud

types of bottom in the

Midge fly larva

Cricotopus exilis & others

1,000

littoral zone of western

Caddisfly larva

Hydropsyche

70

Lake Erie (after Shelf ord and

Mayfly naiads

Stenonema tripmictatum.

Boesel 1942).

S. pidchellum. S. inter -
punctatum

15

Snail

Physa sp.

13

Water penny

Psephenus contei

7

Sponge colonies

Spongillinae

3

Snail

Planorbida crissilabris

2

Leech

Glossiphonia

2

Snail

Amnicola limosa porata

1

Mayfly naiads

Baetis, Centroptilum

+

Clam

Elliptio dilatatus sterkii

^

Damselfly naiad

Argia moesta

+

Midge fly larva

Tendipes pallidas

+

Bryozoan colonies

Plumatella

32

^

Parnid beetle & larva

Stenelmis creiiata

17

+

Caddisfly larva

Trichoptera

2

1

Clam

Leptodea fragilis

+

1

Flatworm

Platmria

+

+

Snail

Gomobasis livescens

12

1

2

Midge fly larva

Chironomidae

1

2

1

Clam

Amblema costata

+

+

Clam

Lampsilis siliuoidea rosacea

+

1

Clam

Obovaria subrotunda

Clam

Lampsilis ventricosa

Mayfly naiad

Ephemera

Alderfly larva

Sialidae

Midge fly larva

Tendipes flavus

Parnid beetle larva

Stenelmis bicarinatus

Clam

Anodonta sitbglobosa

Clam

Micromya fabilis

Snail

Pleurocera acuta

Clam

Fusconaia flava parvula

Mayfly naiads

Hexagenia occidata. H. rigida

33

Midge fly larva

Tendipes digitatus

Midge fly larva

Procladius cidiciformis

Amphipod

Gammarus limnaeus

Water boatman

A rctocorixa lineata

Midge fly larva

Tendipes decorus

Snail

Valvata tricarinaia

Leech

Herpobdella punctata

Amphipod

Gammarus fasciatus

Leech

Glossiphonia stagnalis

Crayfish

Cambarus argillicola

Mite

Lunnesia imdulata

Midge fly larva

Cricotopus trifasciatus

Clam

Protera alata

Clam

Ligumia nasuta

Clam

Truncilia donaciformis

Total taxa

~22

23

22

Total individuals

1,177

17

49

72 Habitats, communities, succession

geii in tilt' hyixjlimnioii. Of this luiiubcr, Si species
are confined to the littoral and sublittoral zones, 7
to the profundal zone, and 10 occur throughout
(Miller 1941).

The bottom mud of outrojihic lakes commonly
consists of a thin, ui)])cr, brown, detritus layer of
newly de])osited organic matter that has drifted down
from above : a relatively thick gray layer containing
many fecal pellets and much organic matter, as well
as diatoms : and a relatively barren bottom layer. In
England's Lake Windemere, 85 per cent of all bot-
tom organisms occur 6 meters below the surface,
and 100 per cent 12 meters below tlic surface in
the upper layers (Humphries 1936).

Maximum populations of insect larvae in eutro-
phic lakes are ordinarily reached during the winter.
Minimum populations occur during late spring and
summer, both in the littoral and profundal zones,
because many immature insects have completed their
development and emerged, as adults (F.ggleton 1931,
Ball and Hayne 1952). Although relatively few gen-
era make up the bottom fauna, populations may at
times be enormous. Chaoborus larvae alone have
been recorded in populations of 97,000 individuals
per square meter, and Tcndipcs larvae at 26,000 indi-
viduals per square meter (Deevey 1941).

re(|uires special adjustments by organisms. Some
bacteria are truly anaerobic, an(l perhaps some ani-
mals are, too, but most forms simply accumulate an
o.xygen debt that is rejiaid when the autumnal over-
turn takes place. It is of interest that the annelid
worms and those midge fly larvae that tolerate the
lowest o.xygen concentrations ])ossess hemoglobin in
the blood, the i)igment which has the greatest capacity
and efficiency in transporting o.xygen at low tensions.
Tubificid worms extrude farther from their tubes
and wave their tails more vigorously for a time as
the oxygen content becomes reduced. The nightly
excursions of Chaoborus larvae into the oxygenated
epilimnion is certainly an opportunity for replenish-
ment of their oxygen needs. A considerable propor-
tion of the larvae migrate out of the profundal zone
during the spring, and do not return until autumn
or early winter (Wood 1956). The copepods Cyclops
bicuspidata. Canthocamptus staphylinoides. and per-
haps others, encyst and lie on the bottom during the
summer period (Moore 1939), although this action
has not definitely been related to any particular en-
vironmental factors (Cole 1953). Some midge fly
larvae also form cocoons inactive.

NEKTON

Summer stagnaliDn period

The low concentrations or complete disappear-
ance of oxygen in the hypolininion of eutrophic lakes
for periods of several days or weeks in the summer

The nekton of lakes consists principally of
fish. There is an interesting small shrimp, Mysis
relic ta. found in the deeper waters of many northern
lakes of North America that is often included with
the nekton. This species is believed to be a relic of

DEPTH IN FEET
100 150 200 250 300 350 400 450 500 550 600 . 1500

'0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190
DEPTH IN METERS

10 Variation with depth in abundances of

lisms in oligotrophic Great Slave Lake (F

Lakes 73

a marine fauna that happened to get cut off from the
sea in some past geological period, yet was able to
survive as the water became fresh.

Numbers and species of fish are more concen-
trated in the littoral zone of lakes than in the open,
deeper waters of the limnetic zone. Limnetic species,
however, invade shallow waters for spawning. In
deep waters, fish tend to remain close to the bottom,
where their food supply is located, unless there is a
deficiency in oxygen there. Caged fish, lowered to
various depths in a eutrophic lake, did not survive
long below the thermocline (Smith 1925). Fish may,
however, make short excursions into the hypolim-
nion.

In a study of fishes in six Wisconsin lakes, Pearse
(1934) found that Usually most fishes per unit area
occur in muddy, vegetation-filled, shallow ponds, but
the characteristic fishes (carp, crap pie, sunfish, dog-
fish) are not the most desirable jar food. Rich eu-
trophic lakes produce considerable quantities of desir-
able fishes (perch, largcmouth bass, white bass, rock
bass). Oligotrophic lakes produce littoral game fish
of good quality and size (smallmouth bass, wall-eyed
pike, pickerel) and ciscoes in deep zvater. The aver-
age catch with gill nets in two oligotrophic lakes was
3.5 per hour ; in two eutrophic lakes, 4.2 per hour ;
and in two shallow lakes or ponds, 5.1 per hour.

In the littoral zone, fish species are segregated
according to the composition of the bottom, as are the
invertebrates. The species living over rock and gravel
bottoms in lakes are mostly different from those in-
habiting similar bottoms in streams, but the mud
bottom forms are nearly the same as in ponds (Shel-
ford and Boesel 1942, Nash 1950).

Amphibians and reptiles do not commonly occur
in lakes except around margins supporting attached
aquatic vegetation, and here pond species occur. Such
pond mammals as the muskrat, mink, and otter are
not typical of lakes as such, although they are fre-
quently found in shallow littoral waters. There are
a number of bird species, however, that occur most
commonly in lakes: American and red-breasted
mergansers, loons, pelicans, cormorants, terns, gulls,
ospreys, bald eagles, and swallows. These species
get their living from the lake, but nest on neighboring
shores or islands. In addition, there are many pond
and marsh birds that occur along vegetated lake
margins.

FOOD CYCLE

The lake is a closely knit ecosystem whose
inhabitants are largely independent of the rest of the
world but very much dependent on each other for
existence. It is almost a microcosm in itself (Forbes
1887), but it depends on the insolation of the sun

for energy, rain and snow for water supply, and on
minerals dissolved out of the surrounding uplands
for the basic nutrient salts essential to the formation
and functioning of protoplasm.

Basic to this food cycle are the bacteria. A few
bacteria occur free-floating in the water. For the
most part, however, they are either attached to algae,
to other plankton organisms, to submerged objects,
or occur on the bottom as part of the benthos (Hen-
rici 1939). Their number varies from one place and
time to another, as do the numbers of other organ-
isms ; they are more abundant in eutrophic than
oligotrophic lakes. Their action is to transform the
dead organic matter into nutrients, especially ni-
trates, that the green plants then absorb.

The phytoplankton are the next link in the food
cycle because of their ability to manufacture carbo-
hydrates with the aid of sunlight and to anabolize
proteins after absorbing nitrogen and other com-
pounds dissolved in the water. Rooted vegetation
around the lake margin is important in this respect,
although in large lakes the proportion of food sub-
stances formed by marginal vegetation is small as
compared to the amount manufactured by phyto-
plankton. In Wisconsin lakes, the daily production
of glucose during clear days in August varies from
14 to 44 kg per hectare (12 to 39 lb/acre) (Manning
and Juday 1941).

Zooplankton feed upon phytoplankton. Protozoa,
bacteria, detritus, and each other. Some species ap-
pear to discriminate in their choice of food, but most
species filter out and ingest all particulate matter,
within size limits, non-living as well as living, with
which they come in contact. The ratio of number
of entomostraca to number of phytoplankton cells
has been found to vary from 1:1,800 to 1:63,000.
Ratios of rotifers to phytoplankton vary from 1 :50
to 1:37,500 (Pennak 1946). The plant cells, how-
ever, are much smaller than individual animals. The
mean ratio of zooplankton to phytoplankton by vol-
ume is commonly about 1:4 (Davis 1958), but in
alpine and northern oligotrophic lakes, the ratio may
be reversed (Pennak 1955, Rawson 1956). In the
nannoplankton. Protozoa depend largely upon bac-
teria, although some forms feed also on algae and
detritus ; a few species prey chiefly upon other proto-
zoans (Picken 1937).

When the plankton dies, it settles to the bottom
and furnishes food for the benthos. The accumula-
tion of dead plankton and other aquatic organisms
on the bottom may be extensive enough to form a dis-
tinctive brownish layer. The benthic midge fly and
other insect larvae, annelids, clams including the
sphaeriids, snails, and bottom-dwelling entomostraca
feed on this detritus layer, on organic matter held in
suspension, and on algal plankton and attached forms.

The variety of their food habits is reflected in

74 Habitats, communities, succession

certain anatomical adaptations of fish. Fish feeding
on bottom matter have soft-lip]ied sucking mouths ;
fisii feeding on plankton have numerous slender gill-
rakers : fish feeding on other fish have large mouths
and sharp teeth. The adults of some fish, such as the
cisco. gizzard shad, paddlefish, and sunfish, consume
large quantities of plankton. The gizzard shad also
feeds on bottom mud, straining organic particles out
of it and grinding them up in a stomach that re-
sembles the gizzard of a chicken. Sturgeon, white-
fish, buffalo fish, carp, catfish, bullheads, suckers,
sunfish, and many others feed largely on bottom
annelids, insect larvae, mollusks, and vegetation in
shallow waters. As many as 354 midge fly larvae
have been found in a single whitefish stomach: 331
were found in a sturgeon stomach (Adamstone and
Harkness 1923). Bass, crappies, perch, pike, gar,
and lake trout feed principally on otiier fish. The
bottom feeders scoop up the bottom ooze indiscrim-
inately; several forms maintain contact with the bot-
tom by means of sensitive barbels hanging from the
chin, but plankton-feeders and carnivorous species de-
pend largely on sight for seizing individual prey.

Young fish of many species live largely on plank-
ton, even though as adults they feed on something
quite different. A 10-centimeter perch requires 130
mg dry weight of food per day during the summer, the
equivalent of about 37,300 Cyclops. The perch would
have to consume Cyclops at a rate of 26 per minute
throughout the day in order to ingest such a total.
A 20-centimeter perch would require 600,000 Cyclops
per day, ingested at a rate of 417 per minute, which
is doubtless beyond its efficiency of intake. By con-
suming only four small fish 0.3 g dry weight each,
the perch could obtain the same energy intake ( Allen
1935).

Most lake-inhabiting birds subsist mainly on fish,
diving for their food. Gulls take only dead fish, which
they find floating on the surface or washed up on the
shore. Swallows skimming over the water surface
consume enoromus numbers of emerging adult midge
flies and other insects.

BIOMASS A.\D PRODUCTIVITY

The dry weight of total organic matter of
seston in 329 fresh-water lakes was found to range
from 0.23 to 12.0 mg/I with an average of about
1.36 mg/1 (Birge and Juday 1934). Of this, living
plankton organisms constituted an amount ranging
from 20 to 80 per cent. The biomass of green phyto-
plankton is usually, but not always, greater than the
zooplankton. The biomass of net plankton may be
only one-third to one-tenth of the total net and nan-
noplankton. Net plankton is generally more abundant
in hard water than in soft water, more abundant in

SIZE OF FISH IN INCHES
3 4 5 6 7

10 15 20

SIZE OF FISH IN CENTIMETERS

FIG. 6-M Change
crease In size (afte

3d habifs of perch
I 1935).

eutrophic lakes than oligotrophic lakes (Rawson
1953). The dry weight of net plankton during the
summer in 18 lakes of western Canada and in 2
lakes of Wisconsin varied from 0.9 to 17.7 mg/1, and
averaged 5.0 mg/m- of water surface area (Rawson
1955).

The biomass of benthos varies with the nature
of the bottom, amount of vegetation, and depth.
When computed for the total bottom of 10 Canadian
lakes exceeding 1 1 m in depth, it was found to vary
from 0.07 to 2.47 g/m-, and average 0.63 g/m- dry
weight, not counting the shells of mollusks (Rawson
1955). The mean of 36 lakes in Connecticut ranging
in depth from 1.1 to 11.1 m varied from 1.09 to
34.8 g/m'-, and averaged 7.5 g/m- (Deevey 1941).
The bottom fauna of various European lakes has been
found to range from 0.69 to 5.65 g/m-.

When lakes of different depths are analyzed, it is
found that the mean biomass of both net plankton and
benthos exist in inverse relation with mean lake depth
(Table 6-3). This may indicate that the morpho-
nietric characteristics of a lake affect its carrying
capacity, a factor additional to those of dissolved salt
content, oxygen content, and temperature.

The biomass of plankton is generally greater than
the biomass of benthos. In addition to the five Cana-
dian lakes listed in Table 6-3, Deevey (1940) found
the ratio between plankton and benthos in five other
lakes likewise to vary from 3.8:1 to 10.0:1. In one
eutrophic lake in Michigan, the standing biomass of

Lakes 75

TABLE 6-3 Interrelations between depth, blomass of plankto
and blomass of benthos in five Canadian lakes (from Raws.
1955).

Average dry Average dry

Average weight of weight of Ratio total

depth, net plankton, benthos,' biomass,

meters g/m^ g/m^ plankton/benthc

11

9.05

2.47

2.3

26

3.65

0.41

4.5

38

3.2

0.45

5.7

69.5

2.6

0.20

7.2

120

0.9

0.07

9.6

'Minus weight of shells in mollusks

fish to benthos was in the ratio of 2.7:1 (Ball 1948).
The measurement of productivity is difficult, and
methods presently in use require several assumptions.
If, throughout the year, the plankton population of
Lake Mendota, Wisconsin, should replace itself
every two weeks, then the annual productivity is 624
grams ash-free, dry, organic matter per square meter
of water surface. Of this amoimt, 585 g would come
from phytoplankton and 39 g from zooplankton. The
benthos reproduces less rapidly, nekton, still less so.
The annual productions of bottom fauna and fish m
Lake Mendota is estimated at 4.5 and 0.5 g/m- re-
spectively, and the large aquatic vegetation at 51.2
g/m2 (Juday 1940). Disregarding the large aquatic
plants, the ratio of productivity between plankton and
benthos is approximately 139:1: between benthos
and nekton, 9:1 ; and between plankton, benthos, and
nekton taken together, 1248:9:1. The ratio of annual
productivity between plankton and benthos is much
higher, therefore, than is the ratio of their biomasses
or standing crops. No attempt was made in this study
to determine the standing crop of fish.

■h'

1

fly (fr.

2 (a) Larva, (b) pupa,
Shelford 1913 after Johar

(c) adult of a midge

In a detailed study of the net productivity of the
benthos in the Russian Lake Beloie (Borutsky 1939),
it was found that the standing crop increased during
the year by 125 per cent. Of this total biomass, 55
per cent died without being eaten by other organisms
or was replaced by the small biomass of new eggs be-
ing laid: 14 per cent was consumed by other organ-
isms, chiefly fish : and 6 per cent emerged as adults
that subsequently left the lake ecosystem. The re-
maining 25 per cent constituted the standing crop of
the following year. However, this standing crop was
only 56 per cent of what it was the year before, so
these percentages are not representative of stabilized
populations.

It was estimated that in Costello Lake, Ontario,
the standing population of midge fly larvae was re-
placed during the 135 days of summer eight or nine
times in the epilimnion, and two or three times in
the hypolimnion. Consumption of larvae by fish was
small in shallow waters but amounted to 50 per cent
of the standing crop in deep water (Miller 1941).

LIFE HISTORIES

Although most midge fly larvae. Chirono-
midae, are aquatic, some forms live in decaying or-
ganic matter, under bark, or in the ground. The
earliest larval stage is a wiggler, which may be car-
ried by the current or may squirm about from place
to place. Later, this wiggler larva becomes sluggish
and builds a case or tube, open at both ends, by ce-
menting particles of sand, debris, or silt about itself
with mucous from its salivary glands. Construction
is accomplished in about three hours. The larvae
e.xtend themselves from these cases for feeding, and
in some species may even move the cases to better
feeding areas. The larval period is the longest part of
the life cycle: it lasts at least two months (Mac-
donald 1956). Most of the pupation period, which
is probably less than a week, is spent in the larval
case, but towards the end the pupa swims to the sur-
face of the water. At this time it is preyed upon ex-
tensively bv fish. The adult imago struggles out of
the case and flies off. The adult lifespan is probably
short, as there is no evidence that they feed. They
may occur in immense swarms in the evening. Eggs
are laid in masses of several hundred or in sticky
gelatinous strings that float at the surface attached
to some object or sink to the bottom. The eggs hatch
in a few days, and the cycle is repeated (Cavanaugh
and Tilden 1930, Johnson and Munger 1930). Some
larvae (e.g. Procladius. Tanypus) are carnivorous
rather than herbivorous or saprophagous. They do
not build cases, but roam over the bottom. The num-
ber of generations varies in different species from
two per year, to one per year, or one in two years.

76 Habitats, communities, succession

and dcpciuls in part on the deinh and IcniptTatiirc of
their habitats (Miller 1941).

The tiibificid 2i-oniis do not leave the lake k)ttoni.
Many of tiieni occnr in tubes or cases, similar to the
habit of midge fly larvae. It is evident by the pres-
ence of sexually mature adults and the reproductive
cocoons that reproduction occurs principally during
the periods of autumn and, especially, spring over-
turns. There follows a large increase in numbers of
small immature worms (I"!ggleton 1931 ).

It is obviously impossible to include a descrip-
tion of the life history of all sjiecies. The cisco or
lake herring has been selected to illustrate the life-
history of a lake fish (Cahn 1927, Fry 1937).

The cisco spends much of the year in deep water,
feeding very largely on plankton, strained out by
specialized gill rakers. Because its food habits require
a large volume of water to be passed through its gill-
rakers, the fish swim almost constantly, usually in a
constant and definite direction, in schools of from
twenty to several hundred individuals. During hot
summers the cisco may leave the cool, deep, but oxy-
gen-poor carbon dioxide-rich waters and ascend into
the epilimnion. There they are sometimes killed in
large numbers by temperatures higher than they are
able to tolerate. The fish spawn in November or De-
cember, when the water temperature drops to 4°C.
For this purpose the fish move into water only one to
three meters deep, or even up into rivers. The males
precede the females by two to five days. When the
females arrive, several males consort with each.
When she is ready to spawn, the female descends to
within 20 cm of the bottom and sheds about 15,000
eggs. At the same time, the accompanying males dis-
charge sperm, and fertilization is completed. The
eggs are viscous and become attached to rocks or
bottom debris. Xo nest is made and no further atten-
tion is paid to the eggs. Incubation may last 10 to 12
weeks: hatching normally occurs in late March. The
young fish later return to deep water and reach breed-
ing condition in three years. Doubtless the slow rate
of development in this species is related to the low
temperature of the habitat. After spawning is com-
pleted, the adults may remain in shallow water until
water temperatures reacli 20°C. This temperature is
above their preferendum, although they can tolerate
temperatures up to at least 25 °C.

APPLIED ECOLOGY

Applied ecology involves the management
of lakes and the control of their resources for man's
benefit. Aside from their use in transportation, in
industry, and as sources of drinking water, lakes are
of importance to man for fishing, swimming, sight-
seeing, and boating. For swimming, clean, clear

water with a sand bottom is desirable. Sewage and
industrial wastes must be diverted or eliminated for
reasons of health and the ajjpearance of the water.
.\lgal growth, when excessive, can be controlled with
cop|)er suljihate ; and rooted vegetation can be re-
duced by sodium arsenite treatment. When chemical
treatment of water is limited to low concentrations
.administered with discretion, there is generally no
great harm to fish ; some invertebrates, such as midge
tly larvae, mayfly naiads, and fresh-water shrinij), are
adversely affected (Machenthun 1958).

\\'here there is excessive erosion of the surround-
ing upland, silting may render the waters of small
lakes turbid, decreasing the growth of algae, a basic
food substance for lake organisms. The rapid ac-
cumulation of silt on lake bottoms covers up bottom
organisms, clogs the gills of mollusks, and generally
reduces the lake's productivity. The obvious remedy
is the control of erosion at the source.

The management of large lakes to the end of in-
creasing fish productivity is difficult because of the
area and depth of water involved. Where commercial
fishing is commonly practiced in large lakes, a care-
ful yearly catch record for each species should be
maintained. This will suggest regulations such that
annual cropping will not exceed annual production.
To improve fishing and increase productivity there
must be an increase in a lake's carrying capacity.
Carrying capacity depends on maintenance of good
chemical and physical characteristics of the water, an
abundance of food, plenty of breeding areas, and ex-
clusion of exotic predators. The drastic decline in the
annual yield of lake trout in the Great Lakes is at-
tributed to invasion by the predaceous sea lamprey.
Artificial fertilization of lakes presents greater prob-
lems than it does for ponds, but may eventually prove
practicable (Hasler and Einsele 1948).

Pollution is usually a local problem in large lakes.
A moderate pollution of organic wastes may in fact
fertilize a lake and produce an increase in the plank-
ton and bottom organisms which serve as fish food.
Excessive pollution, however, must be controlled, as
it interferes with the use of the lake for recreation
and as a water supply.

The smaller the lake the more practicable becomes
management of the habitat. The water level may be
manipulated, by damming, to increase the area of
shallow water available for spawning at certain sea-
sons, or lowered at other times to permit growth of
marginal vegetation or prevent spawning of unde-
sirable species. Artificial shelters or spawning areas
may sometimes be created, yielding a significant im-
provement (Hubbs and Eschmeyer 1938). In gen-
eral, rearing small fish in hatcheries for later release
has not proven economically practicable. Any pro-
posed introduction of exotic species should be investi-
gated with considerable skepticism.

Lakes 77

SUMMARY

Important factors in aquatic habitats are
pressure, density, light, current, temperature, oxygen,
carbon-dioxide and other gases, dissolved solids, and
hydrogen-ion concentration. Of special importance in
many lakes is the occurrence of a thermocline that
divides the water into an epilimnion and a hy-
polimnion. The hypolimnion retains a low tempera-
ture throughout the year, and in some lakes becomes
deficient in oxygen in late summer. These differences
in temperature and oxygen greatly affect local and
seasonal occurrence of organisms. Lakes are classified
several ways on the basis of physical characteristics ;
biologically, only two distinct communities, the oligo-
trophic and eutrophic lake biocies, are distinguish-
able.

The life-forms of lake organisms are chiefly plank-
ton, benthos, and nekton. Zooplankton exhibits diel
movements to greater depths in the daytime and gen-
eral dispersal, including movements toward the sur-
face, at night. Peak populations are commonly
reached in late spring, and again in autumn ; low

points occur in summer and winter. Benthos de-
creases in abundance from the littoral to the profundal
zone. Profundal animals in eutrophic lakes are ad-
justed in various ways to survive the low oxygen
late summer stagnation period. Nekton includes
aquatic birds as well as fish.

The lake is a closely knit ecosystem, largely inde-
pendent of the rest of the world except for its solar
energy, inflowing water, and mineral salts. The base
of food-chains is composed of detritus, bacteria, and
phytoplankton, then zooplankton and small benthic
organisms, and finally fish and birds. All dead or-
ganisms, as well as their excreta during life, decom-
pose so that their nutrient substances start the food-
cycle over again. The biomass and productivity of
the three life-forms usually rank, from high to low :
plankton, benthos, and nekton.

The life-cycle and behavior of lake organisms are
closely adjusted to the various environmental situa-
tions available. Control or management of fish pro-
duction by man is difficult, except in lakes of small

78 Habitats, communities, succession

7

Local Habitats.

Communities, and

Succession:

Ponds,

Marshes, Sivamps,

and Bogs

Pond is a jMspular term for lakes of the third
order that are small, shallow, and, when mature, have
rooted vegetation over most of the hottom. Tiiere is
no clear distinction between ponds and lakes of the
first and second orders. The littoral zone of eutrophic
lakes, for instance, is pond-like in habitat and or-
ganisms. Floating and emergent vegetation com-
monly occurs around the margin of ponds to form
extensive tracts of marsh. The pond habitat may
originate as a shallow basin, as a large pool in a
stream, as the result of the filling in of a lake, or
from a stream dammed by beavers, man, or landslide.
Slow-flowing rivers are essentially elongated ponds,
and have a similar fauna (Kofoid 1908, Richardson
1928). Because of the slight water movement in
ponds, the surface film becomes an important micro-
habitat for some species. Pondwater temperature is
often uniform at all depths, but during warm sunny
weather, i)onds well protected from the wind may
show considerable stratification, not only in tempera-
ture, but also in oxygen content and other character-
istics (Wallen 1955). Daily and seasonal variations
in temperature may be great because of the .small
volume of water. Ice forms earlier and lasts longer
in ponds than in lakes, freezing shallow ponds to the
bottom in severe winters. Light penetrates to all
depths, encouraging growth of vegetation except in
high turbidity. Young ponds may have rocky, sandy,
clay, or mud bottoms ; in mature ponds, there is or-
dinarily an accumulation of organic matter and silt.

The dissolved o.xygen content of ponds varies
widely from temporary supersaturation when there is
excessive photosynthesis of plants to near depletion
when decomposition predominates. Oxygen content
is often highest in the spring ; very low in late sum-
mer ; and sometimes low again under the winter ice
cover. Oxygen content is usually higher during day-
light hours than during the night because of the
daytime photosynthetic cycle of plants. Oxygen may
become so low at night as to become critical, espe-
cially for fish. Decomposition of organic matter
evolves carbon dioxide and, at times, considerable
methane, hydrogen sulphide, and other gases. In
ponds, as in lakes, there is as wide variation in hy-
drogen-ion concentration. As ponds mature and ac-
cumulate humus, pH value decreases.

PLANT SERE

The plant hydrosere, or pond sere, typically
contains the following stages and characteristic spe-

Submerged vegetation: water weed, pondweed,
milfoil, hornwort, naiads, buttercup, bladderwort, eel-
grass, and the herb-like alga Chara.

79

FIG. 7-1 Pond animals (from Shelford 1913).

Representative animals of the emerging associa-
tion, a, the common newt; b, the common pond
snail; c, a predaceous diving beetle.

An Amphipod

(Right) Representative animals of the submerged
vegetation, a, a viviparous snail; b, a green sun-
fish above a yellow perch, both juvenile; c, a
shrimp; d, a winter body, or statoblast, of the
gelatin-secreting polyzoan.

80

Floating vegetation (Fig. 2.S) : water lily, poiul
lily, ])oiul\vcc(l, smartweed, duckweed, and water
hyacinth. All c.vcept the last two are rooted in the
nnid, often at depths of two to three meters, and may
ha\e rhizomes from which long petioles extend to
the leaves floating on the surface. Duckweeds and
water hyacinths are unattached floaters, and cover
the surface extensively in some localities.

Emergent vegetation (marsh) The dominant
s])ecies are: cattail, reed, bulrush, bur-reed, swamp
loosestrife, wild rice, and sawgrass. They invade
waters of over a meter depth, but in shallower waters
or in secondary succession they are replaced by a
sedge meadow composed of sedge, rush, and spike
rush.

Swamp shrubs: huttonbusli, alder, dogwood,
swamp rose, and sometimes slirubby willow and Cot-
tonwood.

Swamp forest: red and siKer niaples. elm. ash,
swamp white oak, and pin oak.

Succeeding stages depend on the climate of the
region. In arid regions the swamp forest may be
poorly developed, and grassland or desert vegetation
may come in quickly. In the mesic climate of the
Kastern states, an oak-hickory associes follows the
swamp forest, replaced in turn by a climax of sugar
maple-beech or mi.xed mesophytic forest. The hydro-
sere in the broad-leaved evergreen climax area of
southeastern North America brings in cypress and
a number of other unique species.

\'egetative debris and animal remains, together
with inwashed silt, fill the basin gradually, reducing
the depth of water and allowing vegetation to en-
croach on the periphery. As this process continues,
the succession is effected. Ultimately, open water en-
tirely disappears as the ground stratum is built up
above the water table, and climax vegetation replaces
all other types.

ANIMAL SERE

Animals characteristic of marshes and
ponds constitute a distinct pond-marsh biocies which
extends into sluggish or base-leveled streams and
the littoral zone of eutrophic lakes. Most animal spe-
cies are not restricted to a single stage or community
of the plant sere, but commonly occur in several
stages in varying abundance and for various activities.
Fish, for instance, feed in open water but spawn in
shallow water among the emergent vegetation. Sub-
merged, floating, and emergent vegetation represent
diflferent levels or strata in a single biotic community,
and each stratum has about the same degree of dis-
tinctiveness as forest community strata.

With the invasion of swamp shrubs and with the
grotmd level well above the water table most of the
year, jx^nd and marsh species largely disappear, re-
placed by many characteristic new species. This ani-
mal comnnmity represents the swamp facies of the
deciduous jorcst-cd(/e biocies. which will be discussed
later. The swani]) forest is often quite open at its
outer margin, and forest edge species remain com-
mon. But as this forest develops a closed canopy and
drier ground stratum, it is invaded by the swamp
facies of the deciduous forest biociation. In the pond
sere there is a succession of animal adaptations from
aquatic to amphibious to terrestrial.

POND-MAHSU BIOCIES

Neuston

The supraticustoti. organisms which move on
top of the surface film in pursuit of most of their life
activities, consists of the water striders Gerridae,
V'eliidae, Mesoveliidae : the water measurers Hy-
drometridae ; the whirligig beetles Gyrinidae ; the
springtails Collembola ; some spiders ; and occasional
other forms. The gyrinids of several species com-
monly occur in social groups (Robert 1955). They
are remarkable for having each eye divided by the
margin of the head so that the upper portion looks
into the air and the lower portion into the water.
Several of the forms listed have long legs that dis-
tribute the weight of the body over a large area of
surface film. The portions of the legs or body that
contact the surface film are water repellant.

The undersurface of the water film supports an
injraneuston of Hydra, planarians, ostracods, cla-
docerans, snails, and insect eggs, larvae, and pupae
(mosquitoes, certain kinds of midge flies, and so
forth). For all except the insects, however, the use
of the surface film in this manner is usually transi-
tory. Some cladocerans, such as Bosmina and
Daphnia. occasionally break through the surface film
from below, fall over onto their sides, and cannot
return.

The species of plankton found in ponds differ
somewhat from those in lakes (Klugh 1927), but the
transition from lake species to pond species is a grad-
ual and progressive one. Protozoa and Rotatoria are
usually more abundant in ponds than in lakes.

Ponds, marshes, swamps, and bogs

TABLE 7-1 Development (ecesis)
of invertebrate bottom
populations (calculated In
number per m') in strip-mine
ponds of different ages, as
determined by studies conducted
in October, through three years.
The ponds I and 8 years old had
no rooted vegetation; the ponds
2! and 30 years old had a
little vegetation in protected
coves; the pond 80 years old
had submerged, floating, and
emergent vegetation. Range of
pH: 7.1 to 8.5; locality, near
Danville, Illinois.

Classification

A

jge of pond

Common name

lyr 8yr

21 yr

30 yr

80 yr

Red midge fly larva

Tendipes

16 7

25

Damselfly naiad

Zygoptera

2 23

51

Caddisfly larva

Trichoptera

1

3

Backs wimraer

Notonectidae

+

3

Whirl-i-gig beetle

Gyrinidae

+ +

Alderfly

Sialis

23

Water boatman

Corixidae

13

Burrowing mayfly naiad

Hexagenia

21

262

1

Dragonfly naiad

Anisoptera

9

27

Crawling water beetle

Haliplidae

5

*

Clam

Unionidae

5

+

White midge fly larva

Tanypus

3

10

Aquatic annelids

Tubificidae, Lumbriculidae

4

2

Ghost larva

CItaoborus

+

17

Springtails

Podura aqmtica

+

+

Crayfish

Orconectes propinquus

Fly larva

Diptera

Water spider

Arachnida

3

Fingernail clam

Sphaeriiim

2

Mayfly naiad

Caenis

Snail

Phvsa gvrim

58

Amphipod

Hvallela

105

Other mayfly naiads

Ephemerida

21

Aquatic isopod

Asellus communis

5

Flatworm

Planaria

1

Limpet snail

Laevapex

1

Snail

Gyraulus parvus

17

Snail

Helisoma trivolvis

9

Water scorpion

Ranatra

1

Leech*

Hirudinea

5

Snail

Lymnaeidae

2

Fingernail clam

Pisidium

1

Shrimp

Palaemonetes

1

Water strider

Gerris
ta

+

Total ta;

5 3

16

27

28

Total individuals

19+ 113+

284+

132+

371+

Although plankton distribution does not vary with
depth as much in ponds as in lakes, seasonal fluctua-
tions are as extensive and similar in nature. Eddy
(1934) lists 15 perennial species of zoo- and phyto-
plankton that may be found in ponds throughout the
year, 2 seasonal species which reach their peak of
abundance between December and April, 4 between
February and June, 12 between March and Decem-
ber, and 5 between July and September.

Benthos

Subaquatic animals dwell not only on the bot-
tom but also on the stems and leaves of submerged
plants. Aquatic plant species have many kinds of in-
sects, amphipods, mites, and snails using them for
the food, shelter, or reproductive sites denied them

in the mud bottom below. Other kinds of insect lar-
vae and oligochaetes are more abundant in the mud
than on the plants. The undersurface of lily pads
often contains many small organisms, including Pro-
tozoa, Hydra, flatworms, rotifers, and snails. The
biomass of animals varies directly with the biomass
of vegetation, and the quantity of invertebrates is
especially great on those plants possessing finely dis-
sected leaves (Gerking 1957). Very few species
found in lake bottoms are not found in ponds, but
the pond-marsh biocies contains many species not
found in lakes.

The variety of species and number of individuals
found in the bottom fauna increase with the age of
the pond from the time the pond is formed until at-
tached vegetation becomes excessive (Tables 7—1,
7-2). Coincident with the development of the bottom

82 Habitats, communities, succession:

fauna is an increase in variety and abundance of
plankton ( F.ddy 1934) and fish. The construction
of a beaver dam in a small Ontario river changed tiie
riffle habitat into that of a pond and brougiit a reduc-
tion in mayfly naiads, stonefly naiads, and caddisfly
larvae within two years. Other stream animals fell
from 68.7 to 1 5.6 per cent of the total population while
midge fly larvae increased from 31.3 to 84.4 per cent
( S|)rules 1940). Shallow ponds develop more rapidly
than deep ones, and nnid bottom ponds develop more
rapidly than sand- (Shelford 1911) or rock bottom
|x)nds (Krecker 1919). The increase in number of
species and individuals in ponds depends on an in-
crease in the variety of microhabitats, types and
amount of food, and vegetation. With the develop-
ment of the pond into a marsh there is generally an
increase in humus and an increase in bacteria effect-
ing its decomposition, carbon dioxide, and marsh
gases. O.xygcn and ])H decrease.

Two predominantly terrestrial orders of insects,
Coleo])tera and Hemi])tera, have invaded the pond
comnnmity but are not found in lakes except those
wiiich have [wnd-like margins. The Coleoptera are
rei)resented by three families of diving beetles,
Haliplidae, Dytiscidae, and Hydrophilidae, and by
the whirligig beetles, Gyrinidae. The haliplids are
herbivorous : the dytiscids are jjredacious ; some hy-
drojiliilids and gyrinids are predators, others are
scavengers. The a(|uatic bugs or Hemiptera are the
Corixidae. which feed on the bottom ooze, the No-
tonectidae, which prey upon small Entomostraca, the
Xepidae. the Helostomatidae, and the Naucoridae,
which are all carnivorous ; and the Veliidae, Meso-
veliidae, Gerridae, and Hydrometridae, which are
Ijrobably both carnivores and scavengers. Some of
tiiese species, as already noted, are usually found on
the surface film, I)ut they may occasionally dive and
cling^ to submerged vegetation. The true diving

TABLE 7-2 Succession of dragonfly and damselfly naiads i

Lake Erie (after Kennedy 1922).

Forest-edge

Lake biocies

Pond- marsh biocies

biocies

Lake mar-

gin with

Lake mar-

floaUng

Old marsh

gin with

and emer-

with invading

Dragonfly and

Open submerged

gent vege- Young Mature

shrubs and

damselfly naiads

lake vegetaUon

taUon pond pond

trees

Gomphus plagiatus
Gomphus vastus
Neurocorduiia yamaskinensis
Macromia illinoiensis
Argia moesta

Enallagma carunculatum
Enallagma exsulans
Enallagma ebrium
Ischnura verticalis
Tramea lacerate

Anax Junius
Enallagma signatum
Libellula luctuosa
Libellida pulchella
testes rectangularis

Leucorrhinia intacla
Erythemis simplicicollis
Plathemis lydia
Nehalennia irene
Pachydiplax longipennis

testes forcipatus
Sympetrum obtrusum
Sympetrum vicinum
Sympetrum rubicundulum
Enallagma hageni

testes uncatus
testes unguiculatus

Ponds, marshes, swamps, and bogs

83

FIG. 7-2 The beetle Dryops freshly submerged, crawling along
a stem, encased by a bubble of air (after Thorpe 1950).

forms, especially the beetles and some of the hemi-
pterans, have evolved oar-like legs for rapid pro-
pulsion.

Respiratory adaptations

Air-breathing aquatic insects, as well as the
pulmonate aquatic snails Lymnea, Helisoma, Gyra-
nlus, Physa, Laevape.v, have evolved special mech-
anisms and behavior for respiration. Most species
rise to the surface of the water at intervals to re-
plenish their supply of air. Insects are so buoyant
that they must cling to the vegetation or some other

object to maintain a submerged position. As soon as
they let go of the substratum, they float to the surface
and must return by swimming. Snails commonly
creep to the surface along plant stems or other suli-
merged objects, or suddenly emit mucous threads
that float them to the surface (Dr. Max Matteson,
personal communication). They find their way to the
surface, at times of oxygen need, by negatively geo-
tactic behavior. After they have obtained a fresh
supply of oxygen, they become positively geotactic
(Walter 1906). Pulmonate snails probably also ab-
sorb some oxygen from the water : indeed, some spe-
cies appear never to come to the surface. The gill-
bearing or branchiferous species of snails are seen to
be segregated into rather distinct niches when their
iiabitat relations are analyzed in detail (Baker 1919).
Diving beetles carry a bubble of air beneath the
elytra, and the entire body of Dryopa is enclosed in
air. The hemipteran notonectids and corixids carry
a bubble over the ventral surface of the body, trapped
there by hair-like setae. The spiracles of the tracheal
system open into these bubbles. The body surface
encompassed by and setae holding the bubble are
water-repellant, or hydrofugoits. The fresh air-
bubble contains 21 per cent oxygen and 78 per cent
nitrogen, the same proportion as the atmosphere.
The nitrogen dissolves into the water very slowly.
The carbon dioxide given off by the insect passes
quickly into the water. As the insect uses up the
oxygen in the bubble, it shrinks. The oxygen content
of the bubble may be reduced to one per cent, or less,
before the insect rises to the surface for a fresh
supply; in water containing little or no oxygen, ris-
ing may occur every three or four minutes. If the
water contains ample, however, oxygen will diffuse
into the bubble as rapidly as it is used, and perhaps
three times as fast as the nitrogen diffuses out. Un-
der these conditions backswimmers, Notonecta, have
survived for nearly 7 hours without coming to the

""^

■^^

FIG. 7-3 Dragonfly niches (after Needham 1949): ( I ) on sand,
Macromia; (2) in sand, Gomphm; (3) in muck, Libellula, Neuro-
cordulia; (4) on massed W,7e//o, damselflies; (5) on tips of
Websferio, Enallagmo laurenfi; [b] in open tangles of blad-
derwort, Erythemis and damselflies; (7) in fallen brown leafage,

Pachydiplax longipennis: (8) at sides of ditch, Tefragoneuria,
Cetifbemis, Eryfbrodlplax: (9) on invading roots of woody plants,
Argia fumipenn'n; (10) at water-line rooted green plants, dam-
selflies; (II) in rafts of fallen pine needles, aquatic Hemipfera
that are enemies of dragonfly naiads.

Habitats, communities, succession

surface. The Imlilile is really a physical gill niecha-
iiism, hut functions only as long as tlie nitrogen
present provides an adeciuate surface for oxygen dif-
fusion. The insect's trip to the surface is as much to
get a fresh supply of nitrogen as it is to get a fresh
supply of oxygen (Wolvekamp 1955).

The air-breathing respiratory mechanisms of
other aquatic insects are equally remarkable. In
many larvae, Pytisciis. Culicidae. and other Diptera,
and in tlie a(|uatic Hemiptera, only the terminal ab-
dominal spiracles are functional. The tracheal trunks
of mosquito larvae and Dytisciis larvae among others,
store considerable air so that the animal may remain
submerged for long ])eriods. Ranatra and other
N'epidae have long respiratory tubes extending from
the tip of the abdomen so that they can cling to vege-
tation well below the water surface yet respire di-
rectly into the air.

Dragonfly naiads ]Hini]i water through the anus,
in and out of an enlarged rectum. The walls of the
rectum are abundantly supplied with a network of
tracheae for interchange of gases directly with the
water. In the larvae of midge flies, black flies, and
corixid beetles, the general body surface is richly
supplied with fine tracheae for exchange of gases di-
rectly with the water. The anal papillae of midge
flies and mosquito larvae are not respiratory in func-
tion, as formerly supposed, rather they serve for
osmoregulation. Tracheal gills, plates, or filaments
are found on many immature insects, Ephemeridae.
Plecoptera, Zygoptera. Trichoptera, Neuroptera, and
some Diptera, and effectively increase the area of
surface available for oxygen absorption. The larvae
of the beetle Donacia and certain Diptera including
mosquitoes have a unique ability to puncture the
walls of submerged plants and collect air from the
intercellular spaces (Miall 1934).

TABLE 7 3 Diff.rences In
individuals of fish present
(Cahn 1929).

species

In two

composition
similar-slied

and number of
Wisconsin ponds

Species

Pond
With
carp

Pond

without

carp

Carp

Shorthead redhorse
White crappie
Bigmouth buffalo
Northern redhorse

5,891
66
17

1
14

10

Walleye pike
Bowfin

Northern pike
Rock bass
Bluegill

4
7
3
1
2

20
340
380
940
1,220

Longnose gar
Pumpkinseed
Yellow perch
Black crappie
Largemouth bass

30
610
680

730
1,120

Total species

10

11

Total individuals

6,006

6,080

tom-feeders — suckers, bullheads, bufifalo, and carp;
the last, a species introduced from Europe. By feed-
ing on the submerged vegetation and stirring up the
bottom they may control the habitat and the compo-
sition of species present in the community. This con-
dition, however, does not last indefinitely. Vegetation
encroaches on the margins of ponds, and the fish are
gradually eliminated because of the disappearance of
suitable breeding sites. The mudminnow, bowfin, and
bullhead are usually the last to disappear before the
pond becomes a dry marsh (Shelford 1911).

Amphibians and reptiles

Terrestrial invertebrates

The terrestrial insects found in marsh vegeta-
tion are in the main adult mosquitoes, midges,
dragonflies, damselflies. mayflies, and alderflies, whose
immature stages live submerged. On bare ground
around ponds may be found toad bugs, shore bugs,
springtails, tiger beetles, and sometimes ground
beetles and pigmy locusts. Spiders become numerous
throughout the vegetation, and the snail Succinea ap-
pears. In addition to these true marsh and pond spe-
cies, invertebrates belonging to the forest-edge biocies
may occasionally be found.

Fish

Fish are often very abundant (Table 7-^). In-
cluded among the species that occur are several bot-

Salamanders and frogs are basically aquatic
animals, although they show varying degrees of adap-
tation to terrestrial life. Siren and A'cctiirus have
permanent external gills and spend all their lives in
the water. Most other forms lay their eggs and pass
through their earlier development in water, but the
adults are air-breathing and wander over the land.
Since their skins must be kept moist, they are con-
fined to the vicinity of water, to humid climates, or
to damp humus. A few species, such as Plcthodon
cincreiis and P. glutinosns. lay their eggs in the cavi-
ties of well-rotted logs and seem largely non-depend-
ent on standing water. The ability of salamanders
and frogs to live temporarily away from water ap-
pears positively correlated with thickness, cornifica-
tion, and relative impermeability of the skin. The
aquatic tadpoles and larvae are scavengers or herbiv-
orous in their food habits, the adults feed on in-

Ponds, marshes, swamps, and bogs 85

TABLE 7-4 Populations of br

ceding birds

in units of

pairs per

40 hectares (100 acres) In

marshes of

northern O

io (after

Aldrich 1943).

Swamp

Swamp

Bird species

Marsh

shrubs

forest

Virginia rail

22

Least bittern

12

Short-billed marsh wren

10

Florida gallinule

5

Sora

3

Mallard

3

KlUdeer

1

Long-billed marsh wren

78

3

Red-winged blackbird

113

144

Swamp sparrow

68

49

4

Song sparrow

8

49

12

Yellow warbler

80

Traill's flycatcher

80

Eastern kingbird

24

American goldfinch

21

Tree swallow

7

Catbird

31

4

Green heron

7

1

Yellowthroat

28

13

Robin

3

4

Red-eyed vireo
Black-capped chickadee
Northern waterthrush
House wren
Ovenbird

Downy woodpecker
Eastern wood pewee
Tufted titmouse
White-breasted nuthatch
Blue jay

Rose-breasted grosbeak

Crow

Scarlet tanager

Yellow-shafted flicker

Crested flycatcher

2
2
2
2
2

Cardinal

Wood thrush

Veery

Hairy woodpecker

Black billed cuckoo

1

1
1

1

Red-shouldered hawk
Eastern bluebird
Brown-headed cowbird
Prothonotary warbler

1
1
1
1

Total species

11

13

30

Total individuals

323

526

121

sects, earthworms, or other animal matter that they
catch on land (Noble 1931).

Reptiles are terrestrial. Desert reptiles never go
to water. Painted, geographic, and snapping turtles
bask in the sun on the shore or on protruding logs,
but quickly plunge into the water to escape danger,
to cool ofif, or to feed. The alligator and musk and

soft-shelled turtles spend nearly all their time in
water. The soft-shelled turtle is able to utilize dis-
solved o.xygen in the water and hence has evolved
special readaptation to water. Like other turtles,
however, they lay their eggs on land, placing them
in holes excavated in sand, loose soil, muck, or de-
caying stumps or logs. Water snakes give birth to
living young that enter the water immediately. Water
snakes feed on insects, small fish and amphibians,
crayfish, or whatever other animal food they can find.
The food of turtles is similar to that of snakes ; some
species are also scavengers. The cottonmouth moc-
casin is a prominent poisonous snake in southern
marshes and swamps. The massasauga rattlesnake
occurs in wet areas to the north.

Birds

Bird populations are high, and the pond-marsh,
swamp shrub or forest-edge, and forest communities
are especially clearly defined (Table 7-4). There is
an abundance of nest-sites and food, but the aquatic
and terrestrial species exploit different niches to
avoid competition as much as possible. The aquatic
species feed in all stages of the plant sere, beginning
with the open water, but nest for the most part in the
emergent vegetation (Beecher 1942). Herons com-
monly feed in shallow water but nest in tree-top col-
onies. Grebes, cormorants, and terns feed on fish in
the open water ; the herons, egrets, and bitterns get
fish in water shallow enough for them to wade in ;
cranes and coots are omnivorous ; ibis, stilts, snipe,
and rails probe around in the mud for invertebrates ;
avocets sweep their curved bills back and forth
through the water, catching aquatic insects ; gallinules
eat seeds, roots, and soft parts of succulent plants as
well as some invertebrates ; most ducks feed on sub-
merged and floating vegetation and attached animal
organisms ; song birds inhabiting the marsh feed
chiefly on insects captured outside the water.

Mammals

One of the most characteristic mammals of the
marsh is the muskrat. A well-developed marsh may
contain one of their haycock-shaped lodges on each
acre (2.5 per hectare), with perhaps five animals per
lodge during the autumn. The diet of the muskrat is
largely the leaves and roots of marsh vegetation, al-
tiiough they also feed to some extent on crayfish,
clams, snails, and sluggish fish. Overpopulations of
two or three lodges per acre (5.0 to 7.5 per hectare)
may lead to "eat-outs" or local destruction of the
marsh vegetation (Dozier 1953).

The mink is probably the most common mam-

86

Habitats, communities, succession

maliaii predator of the marsh ; it is an enemy of the
nniskrat. Foxes, raccoons, and coyotes may invade
ilie marsh wlien the water level is low. The otter
jireys on fish and crayfish of the marsh ; it has now
heen exterminated from most of its former range.

The beaver makes its own marsh habitat by dam-
ming small streams, flooding the surrounding low-
land. Here it builds its large lodge and feeds on the
bark and twigs of adjacent aspen, willow, and Cot-
tonwood, and on the roots of aquatic plants. When
the supply of aspen and other food is exhausted, the
colony disappears, the dam decays, the water level
subsides, and marsh vegetation invades. After some
years the pond is converted into a heaver meadow.

In the southern Atlantic seaboard and Gulf states,
the herbivorous rice rat is common to marsh vege-
tation. In northeastern Ohio, the meadow mouse at-
tains populations averaging 58 per hectare (23/acre)
in the marsh and persists in smaller numbers in the
swamp shrub and swamp forest (Aldrich 1943).
The smoky shrew averages 8 per hectare (3/acre) in
the marsh. The cinereous shrew and short-tailed
shrew are found in marsh vegetation and are common
in the swamp-slirub stage (30 and 52 per hectare, 12
per acre and 21 per acre, respectively). AH three of
the insectivorous shrews are also found in the swamp
forest. The white-footed mouse increases in numbers
from the marsh through the swamp-shrub into the
forest stages (2,5-22-32 per hectare, 1-9-13/acre).
Moles, chipmunks, and squirrels also occur in small
numbers where the ground is drier.

FOOD CHAINS IN PONDS

Many animals in ponds depend for food on
floating phytoplankton, bacteria, and bottom detritus.
Ponds, however, unlike lakes have additional pro-
ducers of organic matter in the rooted pondweeds.
Attached to the submerged pondweeds is a peri-
phyton composed of bacteria, diatoms, and green and
blue-green algae. This periphyton is important as
food to many small crustaceans, immature insects,
oligochaete worms, and snails (Frohne 1956). Pond-
weeds are consumed by insects, ducks, and herbivo-
rous mammals. Filamentous algae are also more abun-
dant in ponds than in lakes, and is a source of food
to various immature insects and frog tadpoles. Creep-
ing predators among the pondweeds and algae are
leeches, dragonfly and damselfly naiads, and water
mites. Small swimming predators are the dytiscid
beetles, most of the hemipterans, and the swimming
leeches Erpobdella and Macrohdclla. At the top of
the food chains, feeding on all these small animals,
and often on plants as well, are the fish and other
vertebrate groups (Lindeman 1941).

.SEASONAL (:ii.\N(;ks
AND TEMJ'OHAKV PONDS

Seasonal changes are greater in a pond
than in a lake, a conse(|uence of the smaller volume
of water. Because of decreased rainfall, increased
evaporation, and continuous seepage or drainage,
])onds often become greatly diminished during the
summer, or tiie 0])en water may entirely disappear.
As the volume of water shrinks, water temperature
rises, and the oxygen content and pH decline. Ani-
mals must either adjust to these conditions of the
jxjnd or disappear altogether.

Under the winter ice, active pond life is slight
because of the low o.xygen and pH, but all groups
increase in numbers as the temperature rises during
the spring. In one small Chara-cattail pond near the
south end of Lake Michigan, the snails Amnicola and
Hclisoma dcflectiis. the amphipod Hyalella azteca,
the isopod Lirccus danielsi, the back swimmer Plea
striata, and diving beetles Haliplidae attained maxi-
mum populations during April and May, but then
declined in numbers through August, increasing again
in the autumn. On the other hand, the water strider
Gerris. mayfly naiads Ephemeridae, damselfly naiads
Agrionidae, dragonfly naiads Libellulidae, and the
snails Physa and Helisoma parvus had highest popu-
lations during June, July, and August (Petersen
1926). Maximum populations of Protozoa are also
attained during the warm period of the year (Wang
1928).

Fish sometimes suffer from lack of o.xygen in
the winter when the ice cover lasts a long time. Dur-
ing the summer thermal stratification often forces
them out of the deeper stagnant water. When photo-
synthesis is curtailed at night or in cloudy weather
mortality may become high.

Bird nesting is ordinarily completed by the time
marshes dry up in late summer. At this time ducks
concentrate in the remaining deeper bodies of water,
and other marsh species start their southward migra-
tion.

Periods of drought are times of stress and in-
creased mortality for muskrats, since the lack of
water interferes with their normal locomotion and
predation upon them by invading terrestrial species
increases. Intraspecific competition becomes intensi-
fied as animals become crowded together in the
shrinking habitat. Many individuals undertake over-
land journeys to new areas (Errington 1939).

Many invertebrates have spores or eggs resistant
to the effects of desiccation, enabled thus to pass over
the period during which the pond is dried up. These
include representatives of the protozoans, sponges,
hydras, turbellarians, nematodes, annelids, bryozoans,
rotifers, mollusks, crustaceans, and insects (Mozley

Ponds, marshes, swamps, and bogs 87

1932, Kenk 1949). It is an interesting experiment
to collect top soil from a dried-out pool, place it in
an aquarium with fresh water, and see what comes
out of it (Dexter 1946). When water returns to the
pond the hatching of plankton organisms and their
growth to reproductive maturity is very rapid. The
life cycle of some species appears shorter than could
be readily sustained in a normal year ; perhaps this
is an adaptation to survive extreme years (Table
7-5 ) . During years when the pond does not fill with
water at all, eggs and cysts remain in a dormant con-
dition ; some have hatched in a year of good condi-
tions after they had continued dormant for several
poor years. Some, but not all, species of crayfish sur-
vive dry periods by burrowing down to the water
table.

Among the more interesting inhabitants of tem-
porary ponds are the phyllopods. The fairy shrimp
is not found in permanent ponds, except those with
a wide, shallow shore that dries out during the sum-
mer. Shrimp nauplii develop quickly from the egg,
usually in January or February after the ice melts,
but sometimes as early as November if the pond be-
comes filled with water after an autumn dry period.
In the spring, adults mature in three or four weeks,
egg-laying takes place forthwith, and the species may
be gone by late May. The period when the pond is
dry is passed through in the egg stage, and either
the drying or freezing of the eggs facilitates their
hatching (Weaver 1943).

Some toads, Bufo spp., Microhyla olivacea, and
spadefoot frogs lay their eggs, after a warm spring
rain, in temporary pools rather than permanent
ponds. Development is rapid, and metamorphosis of
the tadpoles may be completed in a month's time,
before the pool evaporates (Bragg et al. 1950).

LIFE HISTORIES

On hatching from the egg, copepods first pass
through six free-swimming naupUus stages by a
series of molts, during which the small compact ani-
mal possesses only three pairs of appendages : then
through five copepodid stages, when additional ap-
pendages are added ; and, finally, into the adult form.
Both sexes occur regularly.

Ostracod eggs also hatch into nauplii, but these
already possess a shell like that of the adult. Several
molts are required, however, before maturity. Some
species always reproduce sexually ; others are par-
tially or always parthenogenetic.

The reproduction of the cladocerans is of special
interest. Most of the time only females are present,
and eggs develop parthenogenetically during the sum-
mer into more females. The thin-shelled eggs are
held in a brood pouch on the dorsal side of the body.

and the young are well grown before they are set
free. There are no free-swimming larvae. After a
number of generations, the number varying with the
species, and as the pond begins to dry up in the sum-
mer or winter, conditions reach a point where there
is a crowding of females, an accumulation of ex-
cretory products, and a decrease in available food.
Parthenogenetic male as well as female eggs are then
produced. The resulting males are usually smaller
than the females, but, subsequent eggs are fertilized
by them and a thick shell is formed around them.
These ephippial eggs are produced in smaller num-
bers and are very resistant to drying and freezing.
When the pond again becomes filled with water, the
ephippial eggs develop into reproducing females to
start the cycle over again (Pennak 1953).

The life-cycle of the rotifers bears some resem-
blance to that of cladocerans. A few species are vi-
viparous, but in most forms development of the egg
takes place outside the body and is direct into the
adult form. Two kinds of females are not distinguish-
able by external characters. One kind, amictic, pro-
duces large diploid eggs that are never fertilized and
only develop parthenogenetically into more females.
The other kind of female, mictic, occurs only at critical
times of the year and produces smaller haploid eggs.
If not fertilized, these small eggs develop into males ;
if fertilized, they form the thick-walled winter eggs
which, under subsequent favorable conditions, de-
velop into females. The males are usually small com-
pared with the females ; they lack an alimentary
tract, and consequently live only two or three days.
Females live one to three weeks or longer. The
production of males appears to be periodic and is
often correlated with a change in type or amount of
food, or degree of crowding. Males have never been
seen, and may not occur in, some groups (Pennak
1953).

The amphipod Hyalella aztcca breeds only dur-
ing the warmer months of the year. The male carries
the female on his back for 1 to 7 days before copula-
tion occurs. Oviposition follows copulation by 12 to
24 hours. The incubation period is 21 days, and the
female may carry the young another 1 to 3 days in
her brood pouch. A period of 24 to 36 days elapses
between successive broods, and each brood is larger
than the last. The females may live into a second
summer and reproduce again. The young on hatch-
ing in the spring have all the adult appendages and
can reproduce later in the summer (Gaylor 1921).

In the spring, aquatic Hemiptera commonly glue
their eggs to submerged vegetation. Some species
insert their eggs into incisions made in leaves or
stems. The young emerge directly into the water and
resemble the adults except that they do not acquire
wings until after several molts. In the Sialidae of the
Megaloptera, eggs are deposited in masses on leaves

Habitats, communities, succession

TABLE 7 5 Monthly changes in the ostracod fauna of temporary po
(after Hoff 1943).

nfral Illinois. The ponds

Ostracod species

March March April April May May June June July
9 17-20 1 15 4 18 9 22 10

Cyphcercus reticuatus
Cypria turneri
Candona simpsoni
Candoiia fossulensis
Candotia distincta

Candoiia indigena
Candoiia biatigulata
Cypria iiiaculala
Cypria opthalmica
Cypridopsis vidua

Candona suburbana
Cypria obesa

or bare ground near water ; on hatching, the larvae
proceed into the water. Here they stay for a full
year, after which they leave the water and pupate
for several months in a hollow that they scoop out
of moist earth. The adult does not live over winter.
Dragonflies may fasten their eggs to plants below or
above the water surface, puncture leaves or stems
for egg insertion, oviposit eggs in the bottom, or
may scatter them through the water and over the
bottom. The naiads hatch out in about three weeks
and are of varied forms and sizes. Dragonfly naiads
may be divided into three groups on the basis of their
habits ; the climbers that crawl through the vegeta-
tion ; the sprawlers that lie half buried in the mud
with legs extended and backs covered with silt : and
the burrowers. They all undergo several molts under
water, some forms living 1 1 months in this stage.
For their last molt they crawl up the stem of some
plant or onto a rock on the shore, molt, and emerge
as adults. They live for a few weeks only. Adult
dragonflies commonly feed on adult mosquitoes, and
the naiads feed to some extent on the mosquito larvae
(Needham and Westfall 1935).

Aquatic beetles commonly attach their eggs to
water plants or bore holes into plant tissues to hold
them. Some hydrophilid beetles make floating silk
cocoons containing many eggs, anchoring these co-
coons to surface plants. Beetle larvae live only a few
weeks before they leave the water and pupate in
characteristic mud cells that they build for them-
selves. Pupation varies from a few weeks to several
months, depending on the temperature, before emer-
gence of the adult occurs. The adults are the chief
survivors of the winter but sometimes eggs or larvae
live through it, too (Miall 1934, Balduf 1935, Rice
1954).

Mosquitoes reproduce abundantly in marshes,
ponds, or even in small pools, tree holes, or other
water-holding depressions. Some species of mos-

quitoes lay hard-shelled chitinous-covered eggs on the
ground which are capable of withstanding freezing,
extreme heat, and drought but hatch very quickly
after being covered with warm water. In water, eggs
may be laid singly or in rafts. The adult female
Ciilex vexans stands at the margin of the pool or on
some floating object and deposits as many as 300
eggs. The individual eggs are cigar-shaped and are
placed vertically to form a floating raft. They fit to-
gether so snugly that the surface film of water does
not penetrate between the eggs, and the surface of
the raft is dry. The larvae hatch in 12 to 28 hours
and hang head down from the surface film. Vibrat-
ing vibrissae continually sweep food particles into
the larval mouth. The respiratory tube at the pos-
terior end of the body penetrates the surface film and
also prevents the body from sinking. At other times
the larvae may suspend themselves from the surface
film, dorsal side uppermost, and feed on floating ma-
terials (Renn 1941). After 3 or 4 molts (5 to 8
days), the larva changes into a quite differently-
shaped pupa, which hangs from the surface film by
two respiratory tubes proceeding from the thorax.
The winged adult may emerge in 2 days. Some spe-
cies may have seven broods per year.

Sexual and other behavior of mosquitoes varies
considerably among species (Horsfall 1955). Ordi-
narily only the female mosquito bites, this to obtain
the blood nourishment necessary for egg-laying. The
male feeds only on plant juices. Studies made on
marked individuals of Aedes vexans showed that 73
per cent of the individuals confined their activities
to within a radius of 5 miles (8 km), but that 19
per cent traveled 5 to 10 miles (8-16 km) and some
even to 16 miles (26 km) away from the point of
marking (Clarke 1937). Other species, however, ap-
pear not to have such wide ranges.

The pulmonate snail Physa gyrina lays its eggs
in the spring when water temperatures reach 10°-

Ponds, marshes, swamps, and bogs

12°C; thereafter the adult population dies. Snails
born in the spring may reach sexual maturity by
autumn but oviposition is normally delayed until
spring because of cold weather. The life span is
usually 12-13 months, but may be prolonged if de-
velopment is interrupted by aestivation resulting from
the drying up of the pond during the summer
(DeWitt 1955).

Many warm-water pond fish, such as the black
bass and sunfish (Breder 1936), spawn in nests or
redds prepared in shallow water by removing all
debris and vegetation over circular areas of one-half
to one meter diameter. There is some preference for
gravel and sand bottoms when they are available.
The male remains to guard the several thousand eggs
during the few days required for their hatching, and
the fanning movements of his tail and fins doubtless
help to aerate them. He may also guard the young
until they can take care of themselves. Both bullhead
parents guard the egg masses and keep them con-
tinually agitated for aeration ; they may even suck
the eggs into their mouths and expel them forcibly.
The adults keep the young in compact groups by
swimming about them. The European carp may
spawn promiscuously a half-million or a million eggs
during the early spring. The eggs settle in the water
and adhere to the roots and stems of vegetation there.
The eggs are not guarded, and the young are left to
care for themselves.

Salamanders commonly hibernate in humus, un-
der logs, Qr in other nooks or crevices on land. They
usually emerge during the first warm rains of early
spring and proceed to the nearest pond, there to lay
their eggs. The males deposit their spermatophores
on submerged leaves or twigs from whence the fe-
male picks them up for fertilizing the eggs. The eggs
are laid in jelly-like masses and require several days
to hatch if the temperature is low. The eggs of
Ambystoina maciilatitm (Gilbert 1944) and A. tex-
anum (Burger 1950) hatch more successfully and at
a faster rate if they contain unicellular green algae
within the capsule. These algae apparently create a
symbiotic relationship for oxygen and carbon dioxide.
Larval salamanders possess gills, but in all but a few
forms these are later absorbed and the adult returns
to land. A. tigrimim sometimes breeds while still re-
taining the larval gills, and never leaving the water.

Frogs commonly hibernate in the mud at the bot-
tom of ponds, although some forms, including toads,
hibernate in the soil on land. In the spring the males
go to small bodies of water where their loud choruses
attract the females for mating purposes. The jelly-
like masses or strings of eggs require only a few days
to hatch, but the tadpole stage lasts longer. Meta-
morphosis in toads that lay their eggs in temporary
ponds takes place rapidly, but in other species, such
as the bullfrog, adults do not occur until two years
after the eggs are laid (Wright and Wright 1933).

Practically all species of birds characteristic of
northern latitudes that nest in the marsh are migra-
tory, as the freezing of the water and drying of the
vegetation eliminate their food supply. Nests are lo-
cated in a variety of situations : on floating masses
of plant debris built above the water level, typical
of grebes, terns, gulls, black-necked stilt, and ducks ;
in plant material, placed in tufts of vegetation or
formed into platforms, or nests attached to cattails
and other emergent plants well above the water level,
typical of cranes, gallinules, rails, avocets, snipes,
bitterns, ibises, marsh wrens, swamp sparrows, and
blackbirds ; in swamp shrubs, typical of flycatchers
and yellow warblers ; in holes in trees, typical of tree
swallow, prothonotary warbler, and wood duck; in
the tops of trees in adjacent forests, typical of herons,
cormorants, egrets, and wood ibis.

The muskrat is one of the most conspicuous and
important mammals of both salt and fresh-water
marshes as well as river banks. Along rivers, the
animal lives in burrows that it excavates well back
in the bank. In marshes, it constructs a dome-shaped
lodge, as high as one meter, by heaping up freshly-
cut marsh vegetation. The lodge is hollow and dry
within, the floor is placed well above the water level.
The lodge has several underwater entrances and ex-
its. In it the animal cares for its young and finds
protection from enemies and weather in both winter
and summer. In addition to lodges, the muskrat con-
structs shelters, where it may feed out of sight of
enemies, and breathing holes, called push-ups,
through the winter ice.

BIOMASS AND PRODUCTIVITY

In a pond in Iowa, the average summer
population of bottom invertebrates in water 0.5 m
deep averaged 3819 individuals, 1334 mg/m-; in
water 1.5 m deep, 1540 individuals, 1370 mg/m^. In
the shallow water the most important components of
the biomass were, in descending order: snails (shells
removed), midge fly larvae, annelid worms, and the
amphipod HyaleUa. In the deeper water the biomass
was mostly midge fly larvae (Tebo 1955). Produc-
tivity of the midge fly Tanytarsiis, one generation per
year, averaged 7.5 g/m^ in a Michigan lake (An-
derson and Hooper 1956). By mooring a floating
cage over open water throughout the season in an
English pond, a total of 8988 midge flies and other
insects per square meter were caught as they emerged
from the bottom mud. In shallow water, where the
vegetation was thicker, a total of 5979 individuals per
square meter were captured, a total which included
fewer midge flies and more dragonflies and caddis-
flies (Macan and Worthington 1951).

Average standing crops of fish in backwaters and
oxbows may be almost 500 lbs/acre {S7 mg/m^),

90

Habitats, communities, succession

in inidwestern Nortli American reservoirs almost 400
lbs/acre (45 mg/m'-), in other reservoirs and ponds
200-3(.X) lbs/acre (23-24 mg/m-), in warm-water
lakes 123-150 lbs/acre (14-17 mg/m-). and in trout
lakes less than 50 lb/acre (5.7 mg/m-). There is no
tendency for the standing crop to decrease with in-
crease in size of the body of water (Carlander 1955).
Biomass varies with the fertility of the pond and the
food supply. Ponds and lakes receiving water that
drains over fertile soil will have more basic food sub-
stances than drainage from poor soils brings. The
presence of certain species depends also on suitable
breeding sites (Shelford 1911).

Biomass is further affected by the food habits of
the fish species present. In fertile ponds in Alabama
containing species feeding largely on phytoplankton,
the median biomass of fish was 925 lbs/acre (105
ing/m-) : in ponds with fish feeding largely on in-
sects, 550 lbs /acre (62 mg/m-) : and in ponds with
fish feeding largely on other fish, 175 lb/acre (20
mg/m-) (Swingle and Smith 1941). It was esti-
mated that about five pounds of food (2.26 kg) are
required to produce one pound of fish (0.45 kg).
The same ratio has been found characteristic of ponds
in Michigaiv (Hayne and Ball 1956). Hence, the
biomass of animal life decreases with each additional
link in the food chain.

In two small Michigan ponds where it was pos-
sible to tabulate the entire fish population, the benthic
production (at least, that portion used as fish food)
during one growing season was calculated at about
17 times the standing crop when fish were present.
This equalled 811 lbs/acre (92.0 g/m-). The pro-
ductivity of the fish during the same period was 181
lbs/acre (20.5 g/m-), giving a ratio of 4.5 :1 (Hayne
and Ball 1956).

When the standing crop of fish remains the same
year after year, its productivity is indicated by the
number or biomass harvested. In northern Wiscon-
sin the maximum annual yield of desirable food fishes
is about 21 per cent of the mean standing crop: in
central Illinois it is about 50 per cent : in southern
Louisiana, 118 per cent (Thompson 1941).

The productivity of ponds and marshes for verte-
brates other than fish has been measured in a few
localities. In northwest Iowa, redhead ducks annually
produce about 1.4 young per hectare (56/lOOacre) ;
ruddy ducks, 0.6 young (24/lOOacre) ; (Low 1941,
1945). Xine species of ducks in the Bear River
marshes of L'tah average 16 voung per hectare (640/
lOOacre) (Williams and Marshall 1938). In Idaho,
nine species of ducks produce over 22 young per
hectare (880/lOOacre) and Canada geese about 0.1
young (Steel et al. 1956, 1957). On a well-developed
marsh, it is generally possible to remove two-thirds
of the muskrats each year and still reserve sufficient
brood stock for a sustained annual crop. This is
about 2.5 muskrats per lodge (Dozier 1953).

POND AND MARSH MANAGEMENT

The maintenance and control, throughout
the year, of the water level of ])onds and marshes is
important for increased productivity. This may often
be accomi)lished by danuning the outlet. It is also
im|)ortant to retard the plant succession which, if left
alone, will eventually bring alxjut the total disap-
jiearance of the habitat. This may be done by cutting,
burning, use of chemical sprays, flooding, and ditch-
ing. Small ponds, called pot holes, with a good mar-
gin of marsh vegetation, or a marsh interspersed with
numerous small areas of open water, give the highest
yield of waterfowl and other birds, and muskrats.
The abundance of waterfowl is often proportional to
the extent of the pond margin rather than the acre-
age of emergent vegetation. In the Louisiana coastal
marshes, the highest sustained yield of muskrats
(14.5/hectare/yr or 580/lOOacre/yr) is in areas with
Scirpits amcricaniis (O'Neil 1949).

Artificial ponds are easily constructed (Ander-
son 1950, Musser 1948) and are an asset to farms
as a source of food and recreation as well as water
for domestic animals. Such ponds are commonly
stocked with bluegill and largemouth bass, although
other combinations may be used. High rates of re-
production bring the fish population up to full carry-
ing capacity within one or two years. If the pond
were stocked with an herbivorous fish only, such as
bluegills, normal reproduction would soon become so
excessive that a dense population of stunted fish
would be present. Using a prey-predator combina-
tion in proper proportions, the predator (largemouth
bass, for instance) will consume the excessive ofT-
spring of the prey species, and the average size of
the remaining fish will be increased. The develop-
ment of aquatic vegetation in these farm ponds is dis-
couraged since it allows too many prey individuals
to escape the predator. The fertility of poor ponds
can be increased by applying fertilizer encouraging
the abundant growth of bacteria, plankton, and bot-
tom organisms providing fish food (Howell 1941).
The control of turbidity is also important. Clear
ponds with less than 25 ppm turbidity may have 12.8
times more plankton and 5.5 times more fish than
ponds with a turbidity exceeding 100 ppm (Buck
1956).

Repeated stocking of ponds with artificially prop-
agated fish is undesirable as there is more trouble
in controlling overpopulation than underpopulation.
The productivity of a pond is determined not by the
number of fish introduced but by available food sup-
ply. The available food supply is divided between the
individuals present. One study showed that 6500
bluegills per acre (16,250 per hectare) averaged 25.5
grams each, 3200 per acre (8000 per hectare) aver-
aged 51.0 grams each, and 1300 per acre (3250 per
hectare) averaged 104.9 grams each (Swingle and

Ponds, marshes, swamps, and bogs 91

FIG. 7-4 Increase In weight of bluegills upon removal from an
overpopulated pond to a pond of lower fish populations (from
Bennett, Thompson, and Parr 1940). The cross-hatched portion
represents the circumstance of overpopulation; a = autumn,
b = summer, c = spring.

Smith 1942). When the available food supply must
be apportioned to a relatively large population,
growth of individuals is retarded, but with small
populations there is more food available per indi-
vidual, and growth is surprisingly rapid (Fig. 7-4).
In Europe, fish, chiefly carp, are raised for food in
small artificial ponds and are handled in much the
same manner as other domestic animals (Snieszko
1941). Problems involved in stocking and maintain-
ng suitable fish populations in small artificial ponds
n various parts of North America are summarized
n the Journal of IVildlife Management (16, 1952,
233-288).

A problem involved in the management of ponds
and marshes is the control of mosquitoes. Oiling the
water surface will kill mosquitoes, but also renders
the habitat unsuitable for other organisms. Mos-
quito larvae and pupae are good food for such min-
nows as Fundidus and Gambusia that regularly feed
at the surface. Stocking of these fish species will
often keep mosquitoes under control. Elimination of
aquatic vegetation in the shallow marginal areas will
do away with hiding places and leave the larvae
more exposed to fish predators.

Because ponds and marshes produce great num-
bers of fish, muskrats, and waterfowl, and are a
source of recreation for hunting, fishing, boating, and
swimming as well, the actual economic value of main-

taining such areas is often greater than it would be
if they were drained and planted to crops (Bellrose
and Rollings 1949). Proper management of them
is therefore a challenge to applied ecologists.

BOGS

Characteristics

Bogs or moors typically develop in the hydro-
seres of cold northern regions ; while marshes and
swamps, which are markedly different than bogs
(Dansereau and Segadas-Vianna 1952), are charac-
teristically southern in their location. Bogs commonly
develop into a coniferous forest climax ; swamps suc-
ceed to deciduous forest or other southern climax
types. Several thousand years ago, when glacial cli-
mates gripped the northern states, extensive bogs
developed and have persisted as relic communities
in spite of the warming of the climate. These bogs
are slowly being replaced by pond-marsh species at
equivalent serai stages bv cliseral succession (Table
7-6).

Bogs occurring in the Great Lakes region are
ordinarily small in area and have little or no drain-
age. There may be oxygen present in the open water
of the larger bogs, but it is characteristically in very
low concentration, at most, in small bogs or in the
marginal zones. Bog water has a distinct brown
color ; a low nitrogen content ; a low temperature
beneath the surface; a low pH, at least in the mar-
ginal vegetated zones ; and a low dissolved salt con-
tent.

A false bottom is characteristic of bogs. It con-
sists of finely divided plant material of a light brown
color, held suspended in the water at varying depths
below the surface. This false bottom may extend
downward several meters before a true solid bottom
is reached. The material disperses on slight disturb-
ance and may render all the open water turbid. Ordi-
narily, the surface waters are quiet and clear. Dead
vegetation does not completely decompose : as it ac-
cumulates, it becomes compressed to form peat.

The plant bog sere

In the early stages of development of the bog,
organic detritus may accumulate mostly in the deep-
est portions (Potzger 1956). As time goes on, how-
ever, a definite concentric-circle zonation of vegeta-
tion is established around the margin. As peat
accumulates, each zone encroaches on the next inner ;
the inmost shrinks until all open water disappears.
The area becomes finally covered with climax forest
(Dachnowski 1912).

92 Habitats, communities, succession

Cold climate

Warm climate

Stage

Bog

Pond

Floating vegetation

Pond and water lilies,
or absent

Pond and water lilies

Emergent vegetation

Sedge mat

Marsh: cattail, reeds,
bulrushes

Low shrubs or heath

Leatherleaf, labra-
dor tea, bog
rosemary

Absent

High shrubs

Mountain holly,
chokeberry

Buttonbush, alders

Swamp or bog forest

Tamarack, black spruce

Soft maple, elm. ash

Climax forest

Hemlock, pine, white
cedar or spruce, fir

Oak, hickory or
beech, sugar
maple

TABLE 7 6 Relation of blotic
succession to climatic succession
in ponds and bogs. Vertical
succession from open water to
cjimai forest is taking place in
both the pond and the bog, but
as the climate gets warmer, there
is simultaneously a horijontal
succession from the various stages
in the bog sere to equivalent stage
in the pond sere.

In some bogs (Gates 1942, Dansereau and Se-
gadas-Vianna 1952) the first plant stage may be com-
posed of floating vegetation {Ntiphar, Nymphaea.
Potaiiwgeton. Spart/aniiini) , but floating vegetation
is often absent and the first stage is a sedcjc-mat com-
posed of sedges, cottongrass, and buckbean. The
rhizomes of the sedges grow out into the water and
become so interlaced that they form a floating mat.
At the water edge the mat may be very thin, but
towards shore it may become as much as a meter
thick. Since the mat floats on open water it jars
easily, hence the name quaking bog — one must watch
liis step that he does not break through. Sphagnum
moss is not essential for the formation of a mat, but
it invades the mat quickly and helps bind it together.
Sphagnum persists into the shrub and bog-forest
stages following. Interesting insectivorous species
such as the pitcher plant and sundew are common,
as are various members of the orchid family.

The next plant stage is dominated by low shrubs.
which encroach on the floating mat. The leatherleaf,
bog rosemary, laurels, labrador tea, sweet gale, and
cranberries are important species.

A high shrub stage commonly follows the low
shrubs at such time as the mat becomes thicker or
grounded. Common shrub species are holly, willow,
chokeberry, alders, and dwarf birch.

The first tree of the bog forest to invade the
shrubs is commonly the tamarack, but this species is
now less common than formerly because of fire, log-
ging, and the depredations of the sawfly larvae
Lygaeonematus erichsonii. Black spruce may either
invade the shrubs directly or follow the tainarack.
Later, the northern white-cedar may become domi-
nant and persist for a very long time, but the ultimate
fate of the bog, upon addition of upland soil or lower-
ing of the water table, is to be covered with the
climax forest of the region.

Animal life

In bogs that have a large body of open water, or
an inflow of water entraining oxygen, and in which
the pH is not extreme, invertebrate life comparable to
that found in ponds and marshes occurs. True bogs,
however, have little oxygen and a low pH, and many
pond species do not appear. Mollusks are character-
istically absent ; sphaeriids may persist but their shells
l^ecome very thin. Bottom organisms in general are
poorly represented because of the tenuous physical
nature of the substratum.

Desmids predominate among the phytoplankton,
although dinoflagellates, Chlorophyceae, and Myxo-
phyceae are common. Rotifers and a variety of Pro-
tozoa are the principal zooplankters (Graaf 1957).

The chief fish found in acid waters in Michigan
are the brown bullhead, northern pike, bluegill, yellow
perch, and mudminnow (Jewell and Brown 1929).
The mudminnow may be found in waters almost de-
void of oxygen since it is one of the few species that
can live indefinitely by gulping air at the surface.

Amphibians and reptiles are not characteristic of
bogs, although the leopard frog is sometimes numer-
ous on the sedge mat of bogs in Minnesota (Marshall
and Buell 1955). Marsh birds are few in species and
in no bogs do populations approach the magnitudes
found in southern marshes. The muskrat and beaver
persist into northern Canada.

In general, the productivity and economic value
of bogs is very low compared with ponds and
marshes. Liming experiments, calculated to improve
productivity, are being made. Calcium combines with
the humic colloids which then flocculate and fall to
the bottom. This clears the water, light penetrates
deeper. pH is raised, and greater algal, zooplankton,
and fish growth is induced (Hasler et al. 1951 ). Peat
is a special bogs product of importance in northern

Ponds, marshes, swamps, and bogs 93

FIG. 7-5 Plant sere at Bryant's Bog, Michigan, from open water through a narrow broicen mar stage or seage,
leatherleaf, a high shrub stage (in middle rear) of holly, to tamarack and black spruce (courtesy R.E. Rundus).

it stage of sedge, a low shrub stage of

Europe. It is cut out in blocks, dried, and used as
fuel.

It is doubtful if the aquatic fauna of bogs is suf-
ficiently distinct or unique to constitute more than a
facies of the pond-marsh biocies. It is succeeded,
however, by a distinct shrub biocies that differs from
the deciduous forest-edge community. The shrub bi-
ocies is replaced by coniferoits forest biociafions.

SUMMARY

Ponds differ from lakes in that they are
generally small and shallow, and, when mature, have
rooted vegetation over most of the bottom. Bogs are

limited to northern regions, contain a northern type
of vegetation, and are generally acid and deficient in
oxygen. As the climate of northern regions slowly
warms, stages in the bog plant sere are replaced by
corresponding stages in the pond plant sere. The
pond sere consists of six or more plant stages but
only three animal stages : pond-marsh biocies, decidu-
ous forest-edge biocies, and deciduous forest biocia-
tion. These animal communities correspond with the
types of vegetation in the plant sere, but not with the
plant communities identified by taxonomic composi-
tion of the plant dominants. The animal community
in bogs is an impoverished facies of the pond-marsh
biocies.

The pond-marsh biocies contains plankton, ben-

94 Habitats, communities, succession

FIG. 7-6 Profile of a bog pUnf sere (from Dansereau and Segedas-Vienne 1952).

BOG FOREST

I 1

Water Folse bottom Sedge peat

Sphagnum peot Woody peot

m Enmi]

mzi

Parent rock Altered rock Humus loyer Live sphagnum Mesic mosses

B- horizon A-horizon

thos, and nekton, as do lakes ; in addition, neuston is
present. Species that constitute these life-forms are
mostly different from those in lakes. In ponds pul-
monate snails replace the gilled snails of lakes, and
clams are of lesser importance. Air-breathing adult
beetles and bugs, mostly absent from the lake biocies,
are often abundant. Adult stages of aquatic insects
and terrestrial forms occur in the surrounding marsh.
Fish spend most of their lives in the ponds, but go
into the marshes to reproduce. Amphibians, reptiles,
birds, and mammals are usually numerous.

Food-chains in ponds and marshes are based in
part on detritus, bacteria, and phytoplankton, true
also of lakes ; and, in part, on rooted plants, the
periphyton that covers them as well as other objects

in the water, and filamentous algae. Biomass and
productivity are usually greater in ponds than in
lakes. Ponds, however, may become stagnant during
dry periods, especially in late summer, with great ad-
verse affect upon their carrying capacity. Pond and
marsh management for high economic yield of fish,
waterfowl, and muskrats requires control of the water
level and control over plant succession ; an incidental
problem is mosquito control. The unique adapta-
tions and behavioral adjustments of animals to meet
the critical periods of summer stagnation and winter
freezing, characteristic of ponds, are most interesting.
Ponds and marshes are available to all ecologists for
the study of the life-cycles and adjustments of animals
in the pond-marsh biocies.

Ponds, marshes, swamps, and bogs 95

The origin of life was undoubtedly in the sea.
Physiological adjustments were necessary before or-
ganisms could occupy fresh water. Although some
organisms may have become air-breathers and in-
vaded land habitats directly from the sea, most evolu-
tion of terrestrial forms has doubtless come from fresh
water. Relatively few major groups of animals have
been successful in this invasion of land, the most
notable being oligochaete worms ; gastropod mollusks ;
many arthropods, especially the insects and spiders ;
reptiles : birds ; and mammals.

ADJUSTMENTS TO THE TERRESTRIAL
HABITAT

Living on land presents many problems.
Our present concern is to analyze the ways in which
animals have met these problems and to trace the suc-
cession of communities in the extreme terrestrial
habitats of rock, sand, and clay.

Gravity

8

Local Habitats,

Communities, and

Succession:

Rock, Sand,
and Clay

In water, organisms counteract gravity by
means of various flotation and swimming mech-
anisms. Fluid buoyancy permits water-dwellers to
attain huge size ; consider the whale. A land animal,
on the other hand, must support its entire weight.
Some terrestrial animals gain a modicum of support
by burrowing into the soil : others drag their bodies
over the ground surface. But the supportive advan-
tage they thus gain is costly in other directions, for
they are slow moving and relatively helpless before
predators. The animals best adapted to terrestrial life
have evolved appendages in the form of legs or wings
that not only raise the body above the ground but are
also the means of more or less rapid locomotion and
adroit movements over the surface or through the air.
Terrestrial adaptation has involved the development
of a tough body covering to hold fluids and internal or-
gans in place ; a skeletal framework to give permanent
shape to the body and, as a system of levers, to fur-
nish means of locomotion ; and powerful muscles to
lift and move the heavy body. Gravity thus limits the
mass of land animals ; dinosaurs, mastodons, and
elephants approach the maximum practicable size.

Moisture

In sharp contrast to aquatic forms, terrestrial
animals are not constantly enveloped with a continu-
ous watery medium, with the limited exceptions of
protozoans, nematodes, and other small organisms
living in moist soil.

96

Land animals, lacking constant contact with the
water medium, iue faced with the prohlems of obtain-
ing water and preventing excessive water losses from
the IkkIv.

Water lieconies available to animals in varying
amounts in tlie forms of rain, snow, hail, frost, and
fog. Whatever the form it arrives in, the significant
things are the amount of free, liquid water added to
the substratum, accessible to plants and animals, and
the humidity of the air. Considerable amounts of
moisture are lost to organisms as run-ofT water flow-
ing into streams, by evaporation back into the air.
and as water bound in snow and ice.

Kvaporation of water from the earth's surface or
from the bodies of organisms increases as tempera-
ture rises, air movement (wind) accelerates, and the
amount of moisture already in the air decreases.
When measured as grains per cubic foot or as milli-
meters mercury pressure, the actual amount of mois-
ture vapor in the air is known as absolute humidity.
This measurement is of less ecological importance
than is relative humidity, the ratio of amount of
water vapor actually in it to the quantity required to
saturate the air at existing temperature and baro-
metric pressure. Relative humidity is easily deter-
mined by means of sling or cog psychrometers, and
may be continuously recorded with temperature by
bygrothermographs.

The evaporation rate of water is more closely re-
lated to saturation deficit than it is to relative hu-
midity. Saturation deficit is a quantity which cannot
he directly measured, it must be calculated. It is that
additional amount of moisture required to saturate air
under prevailing temperature, relative humidity, and
barometric pressure conditions, commonly expressed
as grains per cubic foot or as millimeters of mercury
pressure.

The most exact and desirable measurement of
water evaporation is the vapor pressure gradient ob-
taining between the organism and the surrounding
air. The gradient is positive if water molecules leave
the organism at a rate faster than the rate at which
the organism is absorbing them from the air, and
negative if the reverse is true (Table 8-1). The de-
termination of gradient magnitude involves the meas-
urements of body temperature of the organism and
air, permeability of body membranes, and rate of air
movement over the body surface (Thornthwaite
1940).

Water is obtained by a land animal by various
devices. There may be some direct absorption
through body surfaces such as occurs in the toad in
moist soil and in some beetle larvae in moist air :
this device is important in only a few species. Large
mammals frequently travel several miles each day to
water holes to imbibe drinking water. Many, but not
all, birds require drinking water ; some species, for

c

^ ir;^

FIG. 8-1 Apparatus for measur-
ing weather factors, (a) rain
gauge (courtesy Friez Instrument
Division); (b) cog psyehrometer;
(c) hygrothermograph; (courtesy
Friei Instrument Division); (d)
Livingston spherical atmometer.

instances quail and partridge, get it as morning dew
on vegetation. Butterflies may frequently be observed
drinking water from small pools. An important
source of water is the jree water in food, particularly
in succulent vegetation and in the blood and body
fluids of animals. Desert animals depend almost en-
tirely for water on that contained in their food and
on metabolic ivater, liberated when fats and carbo-
hydrates, and to a lesser extent, proteins, are oxidized
in their bodies.

Water is lost from the body through the skin and
lungs as insensible moisture and perspiration. Rapid
and largely uncontrolled loss of moisture through the
skin of amphibians, snails, annelids, and insect larvae
is a limiting factor confining these animals to moist

Rock, sand, and clay 97

TABLE 8-1 Evaporation and condensation on a free water sur-
face in relation to temperature, relative humidity, saturation
deficit, and vapor pressure gradient (after Thornthwaite 1940).

Water Water

^nctoT evaporates condenses

Water temperature
Air temperature
Relative humidity
Vapor pressure of
saturated air

16°C
16°C

70%

133 mm Hg

16°C
27°C

70%

263 mm Hg

Vapor pressure of

air 70% saturated
Saturation deficit

93 mm Hg
40 mm Hg

184 mm Hg
79 mm Hg

Vapor pressure of

the water
Vapor pressure gradient

133 mm Hg
+40 mm Hg

133 mm Hg
-51 mmHg

habitats or to activity only in times of high humidity.
Adult insects and other arthropods, reptiles, birds,
and mammals have evolved body surfaces of chitin
and waxes, scales, or cornification of the surface
layers of the skin that largely prevent uncontrolled
loss of moisture. Moisture loss through the respiratory
surfaces in these forms remains considerable, how-
ever. Water is also lost with the feces, although in
some species much water is reabsorbed by the large
intestine before the feces are ejected. The amount of
water removed from the body by excretory organs,
particularly the kidneys, varies directly with the
amount of water intake and inversely with the amount
lost through other devices. The kidneys are critical
to maintenance of proper concentration of salts in the
blood and body fluids. It is important that water in-
take balance water loss. Organisms are very sensitive
to disturbances in body water balance, and this factor
is very significant in determining the type of niche
which a species comes to occupy.

In order that animals could exist in terrestrial
habitats, they had to acquire the ability to carry on
reproductive activities in the absence of water. The
chief reproductive adaptations involve the following
(Pearse 1930) : internal fertilization; a shell covering
the egg to conserve moisture and salts: food pro-
vision to the embryo and young, by yolk in the egg
cell, placenta in the uterus of the mother, or direct
feeding by the adults ; reduction in number of young
with more efficient parental care ; reduction or elimi-
nation of free-swimming larval stages; greater seg-
regation of species into different niches to avoid
interspecific disturbances.

Temperature

Aquatic animals are not ordinarily subjected to
temperatures below freezing and are in a relatively
stable temperature environment, but terrestrial spe-
cies are exposed to highly variable temperatures that

may reach extremes of about — 68°C and +55°C.
No single species is required to withstand such a
wide range of temperatures, however. Optimum and
tolerance limits vary from one species to another,
inasmuch as each inherits a specific degree of acclimati-
zation.

No aquatic species has evolved control over its
body temperature. Aquatic warm-blooded mammals
and birds are derived from terrestrial forms. An abil-
ity to maintain a constant body temperature has sur-
vival value in terrestrial habitats, however, and
consequently physiological mechanisms for homoio-
fhcrmisin developed independently in birds and mam-
mals. All other land organisms are poikilothermal;
that is, they have no physiological mechanism for
maintaining a constant body temperature. Some
poikilotherms, such as bees, have developed special
behavior patterns, that enable them to maintain fairly
constant conditions in the hive by cooperative
efforts ; some lizards, snakes, and turtles are able to
exert some control over their body temperatures by
moving into and out of sunlit areas.

Rates of activity, food consumption, metabolism,
growth, and other physiological functions increase,
to a certain limit, with rise of body temperature.
Homoiotherms maintain a continuous high rate of
functioning because of their constant high body tem-
peratures, but the rate at which poikilotherms func-
tion varies with the temperature of their habitats.

Latitudinal distribution of both poikilotherms and
homoiotherms is often limited northward and south-
ward by the extremes of temperature that they can
tolerate. The rate of energy exchange in poikilo-
therms is so directly dependent on the amount of
heat in the habitat that the total growth and repro-
duction of a species may be determined by the extent
to which it can accumulate developmental heat tmits
during the year (Shelford 1929: Chap. 7) ; the prin-
ciple is similar to the principle of heat budgets in
lakes. The closer a region lies towards either Pole,
the shorter the growth season is, and distribution may
be limited not by extreme low temperatures as such,
but by accumulation of heat energy insufficient to per-
mit completion of life cycles.

The relation of energy balance in homoiotherms
to air temperature is even more complicated (Ken-
deigh 1949, Seibert 1949). In cold regions an ani-
mal may require all the energy its food provides it
simply to maintain its own existence, no surplus
available to meet the high demands of reproduction.
Under such conditions, a species cannot become per-
manently established in a region. A warm-blooded
animal requires a range of temperature that is com-
fortable and in which it can ingest and metabolize
food at a rate sufficient to maintain normal body tem-
perature, sustain physical existence, and carry on
reproductive activities, too. We can thus speak of
existence energy and productive energy, concepts es-

98

Habitats, communities, succession

sential to uiulerstaiuling the relation of an organism
to tlie temperature of its environment.

All organisms outside the tropics must adjust to
meet tlie critical winter season. In those species ac-
tive throughout the year, there is an increase in re-
sistance to cold, hrought ahout, in part, by dehy-
dration of body tissues (Payne 1927), or by
increase in density of plumage (Kendeigh 1949) or
fur (Sealander 1931). Those species incapable of
maintaining activity in situ over winter either migrate
to more favorable regions or hibernate, or the adults
die. Many invertebrates survive the winter in re-
sistant egg or larval stages.

Oxygen

Dry air at 760 mm Hg pressure contains ap-
proximately 21 per cent oxygen, 0.03 per cent carbon
dioxide, 78 per cent nitrogen, and traces of other
gases. Oxygen is thus much more abundant, con-
stant, and available at all times in air than it is in
water. Oxygen availability seldom becomes a critical
factor for land animals, with the occasional exception
of forms that live in the soil or invade high altitudes.
Although terrestrial organisms have evolved sim-
ple moist chambers, branched tracheal systems, or
complicated lungs to replace the gills found in many
aquatic forms, the fundamental requirement of moist
membranes for the exchange of oxygen and carbon
dioxide between body fluids, tissues, and the sur-
rounding medium remains the same. The skin still
serves this purpose in some terrestrial forms — an-
nelids and some amphibians — but in most forms the
moist membranes are within the body. Internal place-
ment decreases the loss of water through evaporation.
The evolution of an ability to take oxygen directly
out of the air apparently preceded the actual invasion
of land, and may have been induced in the pond and
marsh habitat when oxygen dissolved in the water
became reduced or absent during summer stagnant
periods (Pearse 1950). The evolution of internal air-
breathing organs was probably concurrent with the
evolution of mechanisms to prevent excessive water
loss from the exposed surfaces of the body.

Solar radiation

Solar radiation takes the form of an endless
procession of waves. The length of a light wave
from crest to crest, or trough to trough, determines
its character in respect to energy and color ; the
height of the wave determines its intensity. Wave-
length is commonly expressed in millimicrons (Im^ ^
0.000001 mm = 10 Angstrom units, A). All wave-
lengths have a velocity of 299,340 kilometers per
second. The solar spectrum varies from 51 m^i, which

is the shortest ultraviolet radiation, to 5300 m\i, which
is the longest infrared radiation. The spectrum vis-
ible to man is between 390 m^i and 810 m^. Consid-
erable ultraviolet is absorbed by the atmosphere, and
the ultraviolet wavelengths re.'iching the earth's sur-
face are mostly between 292 mji and 390 m^. Color
perception by man is as follows : violet, 390- 422 m^ ;
l)lue, 422-492 m(i ; green. 492-535 mn ; yellow, 535-
586 mn : orange. 586-647 mn; and red, 647-810 mpi.
The longer waves are rich in heat energy ; the shorter,
in actinic energy.

Solar radiation may be measured with a ])yrheli-
ometer, by which readings are given in terms of heat
energy (g-cal/cm-'/sec). Results are not accurate for
the shorter wavelengths. Photoelectric cells accu-
rately measure intensities in the shorter wavelengths ;
readings are given in foot-candles. The Macbeth il-
luminometer measures total sunlight in foot-candles
by visual comparison of observed intensity with a
standardized and calibrated light source set within
the instrument ; accuracy is limited by sensitivity
of the human eye. With any photometric instrument,
the measurement of intensity of any portion of the
spectrum requires the use of calibrated color filters
that screen out everything but the desired wave-
lengths.

Different wavelengths have different effects on or-
ganisms. Green light is reflected by plants ; little is
used in photosynthesis. Some early experiments on
tadpoles, fish, snails, and other forms (Davenport
1908) indicate that there is an increasing growth rate
in different wavelengths, in the following order :
green, red, white, yellow, blue, violet. The physio-
logical basis of this phenomenon is not known. An
excess of infrared may produce overheating of the
animal. Ultraviolet in large concentrations is harm-
ful to most animals, but in lower intensities is bene-
ficial to elaboration of vitamin D. Evidence indicates
that ultraviolet combined with rainfall is important
in controlling numbers of some mammals, forest-edge
birds, and insects. In general, terrestrial organisms
are exposed to much higher intensities of solar radia-
tion than are aquatic organisms and have evolved
horny or chitinous body coverings, hair, or feathers,
that function in part to protect internal structures
from lethal concentrations. Long-range vision has de-
veloped only in land animals and is correlated with
the high light intensities characteristic of terrestrial
habitats.

Diurnation

Animals may be divided into diurnal (day-
time), crepuscular (late evening and early morning),
nocturnal {nighi) , s.nd arhythmic (irregular) species.
.\nimals that occupy microhabitats where tempera-
ture and light changes are negligible at most tend to

Rock, sand, and clay 99

I- 115

o
^100

85

75

70

INCHES OF RAINFALL
14 16
I \

.^''^-' .-b'

35 40

CENTIMETERS OF RAINFALL

FIG. 8-2 An ultraviolet-hydrogram for February populations of
pronghorn antelope In Yellowstone Park for the years indicated
In italic numerals. It appears that the number of young produced
in any year has been determined in a sensitive period two

Septembers earlier; hence, the ultraviolet data given for each
year Is for the second preceding September, and rainfall Is
for that September through the August following. Inclusive
(modified from Shelford 1954).

be arhythmic ; cave crayfish, log-inhabiting beetles,
moles, shrews, and some ants, for instances. The
microfauna of the soil is probably arhythmic. About
two-thirds of the mammal species occurring in both
temperate deciduous and tropical rain forests are
nocturnal. Birds are predominantly diurnal, except
that owls are nocturnal and goatsuckers crepuscular.
The majority of amphibian and reptile species are
nocturnal ; some frogs and lizards are diurnal.

Among invertebrates, there is considerable vari-
ation. The drosophilid flies are crepuscular, having
pronounced peaks of activity at dawn and dusk
(Taylor and Kalmus 1954). The major period of
activity of many nocturnal animals occurs during the
first half of the night period, although they often
possess a secondary pre-dawn period of activity.
With diurnal animals, the major period of activity
usually comes during the first portion of the day, al-

100 Habitats, communities,

succession

tliough tlu're may he a secondary prc-dusk period of
activity (Calluniii 1^)44-46).

A ciassiticatiqn of the diel activities of aiiinials in
respect to controlling influences can he made (O.
Park in Allee ct al. 1949: 558):

I. Periodic activity. Regularly most active for
a specific ])eriod in the diel cycle.

1. Exogenous type. .Activity rhythm directly
induced and controlled hy periodically re-
current environmental conditions.

2. Endogenous type. .Activity rhythm more
or less inde]H"ndent of obvious factors
in the enviroinncnt : persists even under
controlled, apparently uniform environ-
mental conditions. It is possible, however,
that obscure environmental factors may
still be regulative (Brown 1959).

a. Habitual activity. An endogenous
rhythm that has become established as
tlie result of previous experience of the
individual.

b. Inherent activity. .An endogenous
rhythm that is inherited.

3. Composite type. .A rhythm pattern that
is partly endogenous but is accentuated
when the animal is exposed to environ-
mental conditions periodically recurrent.

II. Aperiodic (arhythmic) activity. Xo con-
sistency between individuals of a species in
exhibiting an activity pattern relating to a
specific time of day or night.

Endogenous diel rhythms have been demonstrated
in Coelenterata. Platyhelminthes, Echinodermata,
Crustacea, Insecta, Cyclostomata. Pisces, Amphibia,
Reptilia, Aves, and Mammalia, although they are not
often inherent. Probably the great majority of species
have rhythms of the composite type. Periods of rest
or sleep alternating with activity seems to be a funda-
mental protoplasmic requirement (Park 1940).

Carnivorous, herbivorous, and omnivorous noc-
turnal animals occur throughout the metazoans. It
is of interest that nocturnal animals are less fre-
quently gregarious and social than are diurnal forms.
Adjustments for night activity involve development
of luminescent organs such as fireflies (Lampyridae)
possess : infrared-sensitive vision, suggested for some
insects and birds but not proven for owls (Dice
1945) : increase in visual acuity by modification of
eye structures (Walls 1942) ; keenness of smell dis-
played by some mammals ; and increased sensitivity
to sound, remarkably developed in bats. Bats have
evolved a radar system, called echolocation, whereby
the animals emit ultra high frequency sound waves,
which are reflected from objects back to die ears
(Griffin 1953). Color vision, well developed in some

diurnal insects, lisli, rejjtiles, amphibians, birds, and
mammals, is largely lost in those nocturnal species
active at such low intensities of light that colors
would be indistinguishable anyway. Correlated with
loss of color vision is restriction of body coloration to
blacks and whites or intermediate shades. Nocturnal
.inimals escape such diurnal predators as reptiles,
liirds, and hymenoi)terous insects. There are noc-
turnal ]5redators, to be sure, but jjredation pressure
at night ajjpears to be less intense than during the
day.

There is decreased competition for food and
shelter when some species are active by night, others
l)y day, over the same range. Butterflies are pre-
dominantly diurnal; moths, nocturnal. .Animals with
moist skin, like snails and amphibians, suiTer less
evaporation of water from their bodies at night, when
the relative humidity is higher and the temperature
is lower. There is some belief (Clark 1914) that noc-
turnal forms are derived from originally diurnal
forms; an adjustment, perhaps, to avoid competition
from aggressive diurnal sjiecics occupying the same
niciies.

Seasonal variations (aspection)

In tropical rain forests there is very little sea-
sonal variation in the number of species active and
size of populations attributable to length of day, tem-
perature, and humidity, for these factors are nearly
uniform throughout the year. In other parts of the
tropics, however, there are definite wet and dry sea-
sons, and a considerable change in numbers and ac-
tivities of animals correlates with the seasonal varia-
tions in vegetation and food supply. In temperate

24

i 90° :

~

22

: North Pole i

-

20

: ;

-

^

18

: /\ !

_

Q

1 /6o\ i

tr

16

-

^

14

/ ^^-~^^ N. :

_

1-

12

i^^>--'^30^^<s^

i

^^i 0° >

N^^"""'^

_i

10

" / Equator

\.

CO

(r

8

-/ \ \

\^ -

3

/ \ :

?

6

-

4

1 i

-

2

i \

-

'■■ ■

|JAN|FEB|MAR|APR|MAY|JUN|JUL|AUG|SEP|OCTjNOV|0EC|

FIG. 8-3 Monthly variation in daily photoperiods at
latitudes of the Northern Hemisphere (after Boggs 1931 ).

Rock, sand, and clay 101

TABLE 8-2 Ecological seasons (Ma

Aspect Sector

Characteristics

Early November to late March.

Hiemine Deciduous trees nearly bare,

herbs mostly dead except for
winter-green species (Beatley
1956); Insects, other inverte-
brates, some mammals, going
into hibernation or dormancy;
last migrant birds disappear.

Hibernine Deep winter condition, little
animal activity evident except
winter resident birds in shel-
tered locations and a few
mammals.

Emerginine Some buds swell and subterra-
nean sprouts begin to appear
above ground, earliest migrant
birds appear, animals begin-
ning to emerge from hiber-
nation.

Early April to late May.

Prevernine First appearance of flowers
both of herbaceous and tree
species; mammals and perma-
nent resident birds begin re-
productive activities; sala-
manders go to ponds and lay
their eggs; all insects, snails,
and other invertebrates come
out of hibernation.

Vernine Deciduous trees now fully foli-

ated, early spring flowers re-
placed by species that tolerate
shading; bird migration reaches
its peak; insects and inverte-
brates become abundant in all
strata.

and arctic regions, seasonal differences in length of
day and temperature become increasingly great the
closer the region lies toward a pole.

Correlated with seasonal changes in climate are
adaptive adjustments of metabolism and energy bal-
ances, regulation of breeding time, change in food
habits, and migration or hibernation. Birds breed in
the spring and early summer, since lengthening daily
photoperiods stimulate maturing of the gonads
(Burger 1949). Photoperiodism also controls the
breeding time of some mammals, fish, and inverte-
brates as well as plants. However, in some species,
say trout and deer, shortening rather than lengthen-
ing photoperiods are stimulating, and such species
regularly breed during the autumn.

In deciduous forests, seasonal differences in the
development of the foliage greatly affect animals.
When trees are bare, sunlight penetrates to the forest
floor more readily than when foliage is in full devel-
opment. Foliage is important because it is protective
cover from weather and offers refuge and conceal-
ment from predators ; to many species it is a direct
source of food.

Four main ecological seasons, or aspects, may be
recognized ; each aspect is divisible into secondary
periods, or sectors (Table 8-2). These periods are
best developed in the temperate deciduous forest but
also occur in modified form in other communities as
well. The beginning and end of any aspect cannot be
set with exactness, since aspects vary from year to
year, with latitude and type of community.

Substratum

Early June to middle August.
Reduced number of flowering
herbs but vegetative growth at
maximum, birds at height of
nesting.

Deciduous forest becomes hot
and dry, many ground plants
dry up; birds quiet and entering
molt, molluscs aestivate, foli-
age insects attain maximum
populations.

Middle August to early No-
vember.

Fruits and nuts ripen, autumn
flowers come into bloom, birds
at height of southward migra-
tion, mammals reach maximum
populations but invertebrates
decreasing.

Foliage of deciduous trees
changes color and falls, insect
and spider populations shift
from higher strata to the
ground.

The substratum greatly influences the kind of
plants and animals that occur in the pioneer stages
of succession. Bare rock presents one extreme physi-
cal habitat, sand another, and clay yet another. The
substratum affects animals indirectly in terms of the
kinds of plants it supports and the variety of niches
it affords. Differences between early sere stages not-
withstanding, later ones tend to be more and more
alike so that convergence occurs. In temperate hu-
mid regions, where the seres pass through several
stages, the climax communities of all seres are very
much alike regardless of the type of bare area on
which they originated.

ROCK SERE

Plant communities

Stages in the plant sere on bare rock are
lichens, mosses, annual herbs and grasses, shrubs.

102 Habitats, communities, succession

E

2

i

be

i

k

a
2

1

10

.2

i

I 1

1

1

1 £
< m

is

1 '3.

1

go

c

3

If

= i « ^ E S S

ll

11

U

ii

1 1

i w

1 P

1 /^

1 1^

1 i

.!./

1P /

■a "Z

3 -a)

§ 1
< w

!V

1

J s < '^

/

— »3

-V

S

5
1
1

1

s

-1—

4I

1-

II

♦ ° g

1

If

r

0-5.

03

iCS^fe

FIG

8-5 Early crustose

lichen,

foliose lichen, and moss

stages

on

rock

with ferns in a crev

ice

(cou

rtesy R.E. Rundus).

and forest. The species composition of each stage
varies with the chemical nature of the rock, the pre-
vailing climate, and the locality.

In the first stage, various kinds of lichens com-
pete for a foothold, but crustose types usually pre-
cede foliaceous types. Mosses and such fruticose
lichens as Cladonia follow foliaceous lichens; or may
initiate the sere, telescoping the earlier lichen stages
(Keever, Costing, and Anderson 1951).

FIG. 8-6 (a) a tardigrade, Echiniscu:
oibounae, occurring in moss, possessing
long filaments (Heinis 1910). (b)
ant-lion adult (Shelford 1913).

"f

Lichens and mosses soak up moisture in wet
weather. They derive mineral nutrients from the un-
derlying rock. Carbon dioxide secreted from the rhi-
zoids forms a weak acid with water and dissolves the
binding material of the small rock particles. Rhizoids
may penetrate rock for several millimeters. These
plants trap windblown dust and obtain nitrogen from
organic compounds in it. When the plants die, they
become an addition to the accumulation of organic
matter. Herbs, grasses, ferns and later stages invade
to continue the crumbling of the rock and buildup of
soil. Freezing and thawing of water may crack the
rock, and in these cracks wind- and water-borne soil
lodges and supports plants. Once shrub and tree
roots get started in crevices, their growth exerts a
powerful force further splitting and crumbling the
rock.

Animal life

Animal life in the pioneer plant stages on rock
is scanty. Ants and spiders roam over the bare rock,
and insects of various sorts may stop there, tempo-
rarily. Spiders may construct webs and nests in rock
crannies or amongst the lichens. Some tardigrades
find preferred niches in lichens. Mosses offer a some-
what more substantial microhabitat, but only those
animals that can tolerate great extremes of flooding,
dessication, heat, and cold can survive. Such forms
are found in the rhizopod protozoans, nematodes,
bdelloidid rotifers, tardigrades, copepods, small in-
sects, and mites (Heinis 1910). They often have spe-

04

Habitats, communities, succession

cial means of attaclinient to keep tlieiii from being
l>lo\vn away by the wind, such as strong claws or
cement glands on the feet, long bristle-like threads to
entangle among the moss filaments ; stickers or spines
covering the eggs. Since wet periods are often too
short to permit complete development, all stages must
he tolerant of desiccation, at which time activities
and growth are largely suspended.

.Animal life in general and land snails in particu-
lar are usually more abundant in vegetation (grass-
land and forests) established on calcareous soils
derived from limestone than in the vegetation estab-
lished on soils derived from sandstone, granitic, or
volcanic rock. Calcium carbonate is a mineral essen-
tial to the metabolism of most animals and for build-
ing such skeletal structures as bones and shells.
Snails are less numerous in the grass stage than in
the later, moister forest communities that develoji in
tiie succession.

SAND SERE

Plant communities

Sand is the product of mechanical pulverization
of various rocks. It is deposited by wind and water.
Where extensive areas of sand occur, strong winds
pile the sand into shifting dunes. These dunes have
a characteristic shape as the sand grains are blown
up a long, rather gentle windward slope and swept
over the crest onto a steep lee slope. Moving dunes
may engulf whole forests ; they eventually move on,
leaving the denuded trunks of trees that they have
smothered. The dunes continue to move until they
reach the shelter of some other dune, get beyond the
full force of the wind, or until invading vegetation
covers the surface and anchors them down. The most
successful sand-binding plants are the grasses Am-
mophila. Calamoznlfa, and Agropyron, willows, sand
cherry, and cottonwoods. Willows and cottonwoods
will survive even when almost buried. Each succeeding
stage ties the sand down more firmly, but any break
in the vegetation occasioned by a blowdown of trees
or disturbance by man may invite the wind to start
moving the exposed sand, and change the partially
anchored dune again into a moving one. Only when
the pine stage or the black oak stage is reached does
the dune become relatively secure from the wind.

The plant sere on the south shore of Lake Michi-
gan consists essentially of the following stages
(Cowles 1899) :

Lower beach: \\'ashed by summer storms and

devoid of vegetation.
Middle beach: Washed only by severe winter

storms : comparatively dry in summer ; upper

limit marked by driftwood and debris. Scat-
tered annual plants present.

Calamagrostis-Andropogon associes (upi)er
beach) : This is where the dunes begin to form.
In this early developmental stage (associes)
grasses are dominant, particularly Calama-
(jrostis lonc/ijolia. Amiropogon scoparius,
Agropyrum dasystachymn, Amniophila are-
iiaria. Elymiis canadensis; various biennial and
perennial herbs make their appearance. The
sandbur grass occurs extensively in some areas.

Prunus-Cornus associes : The coinmoner .shrubs
are sand cherry, chokecherry, red-osier dog-
wood, creeping juniper, and the frost grape
vine. Shrubs may invade the grass directly but
become more common in the following tree
stages.

Populus-Populus associes: The first tree stage
in the southern portion is made up principally
of the eastern cottonwood, and in the northern
portion, of the balsam poplar. The trees com-
monly occur in open stands with grasses and
shrubs forming the lower strata. The habitat is
essentially forest-edge. The shrub and cotton-
wood stages are often missing so that the sere
progresses from the grass directly to the pine
or black oak stage.

Pinus-Pinus associes: Jack pine, red pine, and
eastern white pine may invade one after an-
other, commonly forming mixed stands. North-
ern white-cedar and eastern redcedar also oc-
cur ; the former, more commonly northward.
Succession to this stage is mainly contingent on
stabilization of sand in dunes, and more efficient
utilization of water resources. For succeeding
stages to emerge, soil must develop by deposi-
tion of humus. The floor of pine forest is cov-
ered with a carpet of needles, although patches
of bare sand still occur. As the sere advances,
all bare areas become covered with a layer of
humus.

Quercus velutina consocies: Black oak often
forms a nearly homogeneous stand that may
persist for a long time.

Quercus-Carya associes: Black, white, and, to
a lesser extent, red oaks are commonly mixed
with shagbark and bitternut hickories and, in
moist habitats, American basswood.

Fagus-Acer association: When soil humus and
moisture become sufficient, American beech and
sugar maple invade the sand to form the final
climax stage.

In other localities, the taxonomic composition of
the communities, especially the later stages, differs
considerably. The character of the climax varies ac-
cording as climate and geography, but perhaps the

Rock, sand, and clay 105

.^f^^

FIG. 8-7 Sand sere at Ludington State
Park, Michigan, (a) the lower beach
(at right center) is washed by ordinary
waves; the middle beach (in center)
contains driftwood left by heavy
storm waves; the upper beach (at left)
has a sand dune well anchored by
grass and sand grape (light areas),
shrubs (dark areas), and Cottonwood
trees, (b) grass stage, showing
blowouts devoid of vegetation; a mixed
pine stage is shown in the distant
background (courtesy R.E. Rundus).

v^*?;.

:i?

^'m

sere is as complete and as complex in the Lake Mich-
igan region as it would be anywhere.

Habitat

The sand dune habitat is characterized by ex-
treme fluctuations in physical conditions, generally
resembling those of a desert (Chapman ct al. 1926).
Temperatures, especially at the ground surface, are
very high during bright sunny days ; relative hu-
midity is very low. Evaporation from spherical at-
mometers is 2.5-3 times higher than in forest hab-
itats at the same time of day. At night the ground
surface temperature may be even lower than that of

the air since there is little or no surface covering to
prevent rapid heat radiation.

Correlated with the diurnal changes of tempera-
ture, relative humidity, and light, the kinds of animal
active on the sand during sunny days are quite dif-
ferent from those active on cloudy or rainy days and
at night. When the temperature of the sand nears
50°C, all insects leave the surface. Some climb
grasses to get off the ground, others enter their bur-
rows. Insects flying above the sand can select an
optimum temperature from widely different tempera-
tures merely by changing their elevation only a few
inches. They make hurried landings when entering
their ground burrows. The female velvet-ants are
usually among the last to retreat into their burrows

106 Habitats, communities, succession

in the morning and the first to leave them in tlie eve-
ning. I'.xperiments sliow that they of all insects in
this habitat are the most tolerant of the high temper-
atures. Animals living here must either be physio-
logically tolerant of extreme heat or possess behavior
jvitterns that enable them to avoid it.

Grassh

uppers and other Orthitptc

There have been detailed studies of a few spe-
cial groups of animals occupying the Lake Michigan
sand dunes. Three species of wood roach, 2 species
of walking-stick, 20 species of short-horned grass-
hopper, 13 species of long-horned grasshopper, and
6 species of field cricket occur in various stages
of the sand sere in the Chicago area (Strohecker
1^37). A breakdown of this list shows that 7 species,
all short-horned grasshoppers, occur in the grass and
Cottonwood stages : of these, one species is not found
in the pine stage, and the other 6 species disappear
by the time the black oak stage is reached. Eight new
species of orthopterans, including 4 short-horned
grasshoppers, enter the sere at the pine stage, but
only 5 species persist into the black oak stage. Alto-
gether there are 23 species of orthopterans listed for
the black oak forest, an increase of 18 new species.
There are only 25 species of orthopterans listed for
the climax, but this includes 4 species of camel
crickets which for the first time can find their proper
niches under logs, and a katydid that appears in the
trees. The greatest change in species composition
within the sere occurs at the black oak stage upon
the disappearance of 67 per cent of the species present
in the earlier stages and the appearance of 78 per
cent of the species as new forms. Of that 78 per cent,
61 per cent persist through all later stages. The
change in species composition at this stage can be
correlated with the development of a canopy of foliage
and the resulting reduction in light intensity and
soil temperatures.

The community or niche restriction of the
short-horned grasshoppers appears to be deter-
mined either by soil conditions or by the vegetation
(Tsely 1938a). Before short-horned grasshoppers lay
their eggs in the ground, the female tests the soil
with her ovipositors until she finds soil of proper
conditions. Experimental studies show that in cer-
tain cases soil texture is the critical factor in the
choice of the egg-laying site, while in other cases soil
structure or degree of compaction is most important.
Soil conditions appear particularly important for the
sub-family of band-winged grasshoppers : for other
groups, vegetation is of greater significance.

In an experimental study of how an available
choice between foods may affect distribution (Isely
1938a), one-half of 40 species of short-horned grass-

20 25 30 35 40 45 50 55
TEMPERATURE, °C

rature gradient on a sand dune, on a rainy
nny day (after Chapman e/ o/. 1925).

hoppers showed a feeding preference for grasses and
one-half for broad-leaved herbs. The latter group in-
cluded the spur-throated grasshoppers. Four species
were restricted to feeding on a single plant species ;
30 species confined themselves to a few plant species
only, and usually of a single plant family at that ;
only 2 species fed on a wide variety of plants. In
several instances grasshoppers starved in cages, when
there was an abundance of fresh plant materials pres-
ent that were palatable for other species, because their
own preferred food species were absent.

All five species of false katydids studied in Texas
(Isely 1941) confined their choice of food to related
species of broad-leaved herbs or forbs, refusing
grasses ; adults showed a marked preference for the
flower parts and tender fruit pods. Two species of
shield-backed grasshoppers were wholly carnivorous.
The flower-feeding false katydids disappeared from
the prairie in late spring and early summer as the
flowering plants passed their peak, but the insect-
feeding grasshoppers persisted to the end of July or
initil temperatures became too high for their comfort.

Anis

.•\nts cannot get established on the beach be-
cause of its unstable character and are scarce even
in the grass and cottonwood stages because of the
shifting character of the dunes (Talbot 1934). Spe-
cies found are crater-formers Lasius niger neoniger
and Pheidolc bicarinata. and two species of Cainpo-
notus that find protection under the occasional log

Rock, sand, and clay

07

family found

ch sand sere plant stage except the oak-hickory (afte

Spider family

Middle
beach

Grass
stage

Beech-
maple

Web-builders

Ariopidae

2

5

3

5

26

20

Micryphantidae

2

2

2

8

3

Theridiidae

1

1

2

5

10

13

Dictynidae

1

1

4

3

2

Linyphiidae

2

5

7

Agelenidae

1

3

8

Ciniflonidae

1

1

Hahniidae

2

1

Mimetidae

1

1

Uloboridae

2

Total

5

11

6

17

59

58

Per cent all species

in stage

29

35

34

40

35

48

Non-web-builders

Lycosidae

9

4

2

1

24

11

Gnaphosidae

2

1

3

11

6

Salticidae

1

7

5

10

29

17

Thomisidae

7

4

10

22

14

Clubionidae

1

1

1

14

7

Anyphaenidae

1

1

3

Dysderidae

1

Oxyopidae

1

1

Pisauridae

6

4

Total

12

20

12

26

109

63

Per cent all species

in stage

71

65

66

60

65

52

Number individuals

of all species in

herb stratum per

8

6

10

18

24

that occurs. On hot dry days these ants withdraw to
several inches below the surface and emerge only in
the cool of the evening.

In the pine stage, the slight mixture of humus in
the sand is decidedly favorable, food is more abun-
dant and varied. Of 18 ant species found, 9 live in
patches of open sand with no shelter, 6 require sand
with some protection above it (logs, bark, needles),
and 3 are strictly log-inhabiting forms. Monomorhim
minimum and Paretrechina parvula are characteristic
species.

In the black oak community, 29 species occur of
which only 6 live in scattered open areas of sand.
These 6 species are quickly crowded out when there
is development of a complete leaf covering over the
ground. Formica paUide-fulva, which was becoming
important in the pine community, is the predominant
ant in the black oak stage. Its nests are invariably
found under pieces of bark or branches lying on
rather open ground.

As the sere advances into the white and red oaks
stage, open areas of sand disappear, humus and mois-
ture increase, logs in all stages of decay occur, the
whole area becomes shaded, and the daily extremes
in temperature and humidity typical of the open dunes
are considerably curtailed. Species of ants character-
istic of the early stages disappear, and forms that are
found in mesic deciduous forests generally predom-
inate, although there are only six species found here
that do not also occur in the black oak community.
Formica truncicola obscuriventris is the most numer-
ous species. The number of colonies and variety of
species reach maximum in the oak stages.

In the climax beech-maple community, the num-
ber of soil-dwelling forms is reduced, perhaps be-
cause of the thick rich humus, although log-inhabiting
forms are numerous. Lasius niger alienus americaniis
and Aphoenogaster fnlva aquia picea are the only ants
abundant in the deep woods ; ants are more numerous
in the forest-edge than in the forest-interior.

108 Habitats, communities,

succession

Altlunigli different species reach peaks of abun-
dance at different points in the habitat gradient pro-
ceeding from open sand to dense forest, the nature
of the substratum divides the species into two major
grou])s: tliose tliat tolerate and reacli tlieir greatest
abundance in the sandy areas where vegetation is
scattered, and tliose that are limited by sand and re-
quire humus in the soil or the microhabitat of decay-
ing logs. The transition or ecotone between these
two ant communities comes at the pine and black oak
stages. F.xperimental studies of six species in the
genus Formica indicate that physiological differences
occur, and that some s])ecies are able to invade places
of low relative humidity that others cannot.

Spiders

In the sand dunes on the south shore of Lake
Michigan and in adjacent areas, 228 species of spiders
are to be found (I.owrie 1948). The number of fami-
lies represented, the number of species involved, and
the abundance of individuals per unit area increase
as plant stages in the sere succeed one another (Table
8.3). Probably because of the greater diversification
of the vegetation, the availability of logs, the increase
in number of strata, and the consequent greater va-
riety of niches, spiders, like ants, are represented by
a larger number of species in the oak communities
than in the earlier stages of the sere or in the climax.
It is of significance that up through the black oak
stage new species appear in each succeeding stage with
very few dropping out. In the beech-maple climax,
however, 51 per cent of the spider fauna occurring in
preceding stages are no longer found, while 79 per
cent of the species are either new with this stage or
came in at the black oak stage and remained. Up to
the black oak stage the species composition of the
spider population shows ecesis, but with the advent
of deciduous forest, the change in the fauna composi-
tion is sufficiently extensive to indicate succession of
distinct communities.

There is also a change in the mores of spiders
as the sere advances. Small lycosids that hide during
the day under driftwood or other debris and run over
the sand at night hunting for insect prey washed up
by the waves are most characteristic of the beach.
The permanent population is small. A burrowing
spider, Gcolycosa zvrightii. is usually common. The
burrows in which the spiders stay during the day may
be easily spotted on the beach and through the grass
and Cottonwood stages. Web-building species are at
a disadvantage in the early stages of the sere, how-
ever, because of the general lack of vegetation to
which their webs may be anchored and because of
the destructive effect of unchecked wind. With the
appearance of grasses, a substratum in which spiders

can l)uil(l webs becomes available. In later stages, the
percentage of web-buiUlcrs increases considerably as
stratification jirogresses and the forest furnishes a
scaffold.

Other animal life

Strong offsliore winds often blow insects out
over the water where they are forced down onto the
surface and washed ashore. Windrows of such in-
sects many thousands of individuals representing a
wide variety of species, are sometimes to be seen.
Dead fish washed up on shore are fed upon by flesh-
flies and histerid, dermestid, and rove beetles. The
tiger beetles Cicindcla hirticollis, and C. ciiprascens,
a white ground beetle and other carabids, shore bugs,
digger-wasps, robber flies, and other insects and
spiders come down from higher ground to feed on
tiie scavenger species and those washed up bv the
waves (Shelford 1913, Park 1930). The tiger beetles,
ground beetles, digger-wasps, and sand spiders build
their burrows and larval stages far enough back to
escape the summer waves. Termites feed on buried
wood that is decaying or on the undersides of logs
that have drifted ashore. The piping plover and
spotted sandpiper place their nests in the middle and
upper beaches. At night, the toad, opossum, raccoon,
and the deer mouse come down to scavenge
whatever is available. The light coloration of many
of the insects and spiders that occur on sand is doubt-
less an adaptation for concealment (Hart 1907).

The kinds of animals occurring in the grass,
shrub, and cottonwood communities are similar ex-
cept that new species invade with each successive
plant stage. The white tiger beetle Cicindela lepida
first appears on the upper beach and reaches maxi-
mum populations in the cottonwood stage, as do the
digger-wasps, robber flies, and sand spiders. Another
tiger beetle, Cicindcla jormosa. occurs in the ecotone
between the cottonwood and the pine stages. Snout
beetles, spittle bugs, and miscellaneous other insects
are occasionally very numerous. Some 592 species
and varieties of beetles have been taken from various
stages of this sere (Park 1930). Fifty species were
found to occur in the cottonwood stage, 23 in the
conifer stage, and about 200 in each succeeding forest
stage. The occurrence of bees is dependent to a large
extent on the variety and abundance of flowers, but
the number of species in each plant stage increases
up to the black oak and then declines to the climax
(Pearson 1933).

Vertebrates are not usually numerous on sandy
flats or dunes away from the water's edge. The
vesper and lark sparrows occur among the grasses,
the prairie warbler and chipping sparrow are found
among the shrubs, and the kingbird is conspicuous

Rock, sand, and clay 109

FIG. 8-9 Burrows made !n sand
by arthropods, (a) burrows of
a digger wasp, Microbembex
monodonfa. (b) a digger wasp,
Bembex spinolae, and a cross-
section sketch of its burrow
(Sheiford 1913). (c) the white
tiger beetle and its burrow
(Sheiford 1913). (d) excavated
burrow of a sand spider (courtesy
R.E. Rundus). The upper portion,
shown with a stick in it, is
intact; the lower portion, in the
shadow, is broken open.

U

nil

in the trees. Tracks of the prairie deer mouse
are frequently to be seen on the sand. Fowler's toad
and the hognose snake are the only amphibian and
reptile that regularly occur. The grass, slirub, and
Cottonwood stages ordinarily occupy relatively nar-
row belts parallel to the lake shore. Extensive sandy
areas inland may have a larger variety of species
present (Vestal 1913).

The pine community in the sere is not so well de-
/eloped around the south end of Lake Michigan as

it is northward. The coniferous forest penetrates
southward from the north, and some northern ani-
mals move with it. Nesting birds are represented by
the slate-colored junco, red-breasted nuthatch, black-
throated green warbler, blackburnian warbler, and
myrtle warbler, all belonging to the boreal forest
biociation. Forest-edge and deciduous forest birds
also occur. The red squirrel is a characteristic boreal
mammal that occupies this stage, and the white-tailed
deer browses on conifer foliage, especially the white
cedar. The six-lined racerunner and blue racer snake
appear. Among the invertebrates are the bronze tiger
beetle C. scntellaris and the ant-lion.

With the advent of the black oak and later forest
stages, most species requiring open areas or depend-
ing on patches of bare sand disappear. Although the
bronze tiger beetle remains abundant in the black oak
community, it, as well as the other tiger beetles,
disappear in the higher plant stages. Only the green
tiger beetle C. sexguttata is in the climax, a species that
requires bare spots on the forest floor, but not sand.

Reptiles are not common in the sand sere around
Lake Michigan, but elsewhere around the world liz-
ards and snakes are quite characteristic of sandy
habitats. They are remarkable in showing a variety
of structural and behavioral adaptations specific to
locomotion in sand and for protection of their sense
organs and body openings from sand (Mosauer
1932). The sidewinder rattlesnake, for instance, has
evolved, in addition to the usual undulatory lateral
movement of snakes, a rolling sidewise movement
that involves spiral contractions of the body and ap-
plies vertical rather than lateral pressure to the sand.
Sand offers the snake an unstable footing — lateral
undulations alone do less to propel the snake forward
than to merely push sand aside. Sand provides firm
footing only if it is pushed down upon, hence the
effective, if singular, action of the sidewinder.

110 Habitats, communities, succession

TABLE 8-4 Percentage location

of ovipoiitor

holes

and larvae In different

solli, eiperl

mental conditions (from

Tiger beeUe

Holes or

Sand and

Niche under

species

Larvae

Number

Sand

humus

Humus

Clay

natural conditions

CincindelUi

Holes

69

40%

50%

7%

3%

Wet sandy beaches

hirticolUs

Larvae

50

56

42

2

CincindelUi

Holes

141

34

48

2

16

Adults on sandy

tranquebarica

Larvae

129

7

75

2

16

ridges covered
with vegetation,
larvae on sand
or clay

Cincindella

Holes

117

8

22

70

Clay soils, oak-

sexguttata

Larvae

93

15

53

32
On level

hickory forests,
prefer leaves on
ground
Adults on sand or

Cincindella

Holes

51

2

23

clay, larvae en-

purpurea

Larvae

47

2

23

tirely on clay

linibatis

On slope
75
74

banks

FIG. 8-10 The sidewinder rat
snake and the track it makes
(Mosauer 1935).

Rock, sand, and clay 1 1

Life history of tiger beetles, Cicindelidae

The intimate adjustments of a species to its
habitat and the manner in which it selects a particular
stage in the sere may be illustrated by briefly describ-
ing the life-histories of tiger beetles.

Adult tiger beetles are bright-colored, alert, swift
fliers. They are frequenters of bare ground. Both
adults and larvae feed predatorily on ants, sowbugs,
centipedes, spiders, beetles, flies, dragonflies, butter-
flies, and larvae of various forms. Tigei beetles com-
monly dig shallow burrows in the soil for shelter.
They reach sexual maturity after several warm days
in spring or early summer after they have emerged
from hibernation. They copulate on warm, humid
days when there is an abundance of food and sun-
light. After laying their eggs, they die.

The female deposits one egg at a time, and lays
up to 50 in all, in small vertical holes, 7-10 mm deep,
which she makes with her ovipositor. The female
tests soil with her ovipositor until she locates soil of
the required characteristics. Hatching occurs in about
two weeks.

The larvae are elongated, yellowish, and grub-
like. Anteriorly directed hooks, spines, and bristles
on the dorsal side of the larval body prevent the lar-
vae from being pulled out of their burrows by the
larger prey on which they feed. At the site of the
ovipositor hole the larva excavates a vertical cylin-
drical burrow 8-50 cm deep in temperate climates,
much deeper in colder northern regions. Most of the
time the larva stations itself at the top of its burrow
with its mandibles extended, and with its head and
prothorax just closing the round opening. It grabs
passing prey and carries it ofi to the bottom of its
burrow to devour it ; larger prey are eaten at the en-
trance. Inedible parts are cast out on the surface of
the ground around the burrow entrance. After feed-
ing 3-^ weeks, the larva closes the mouth of its bur-
row with soil and goes to the bottom to molt.
The second larval stage lasts five weeks or
longer, after which there is another molt. The
last of the larval stages closes the entrance to
its burrow in late August or September and goes to
the bottom to hibernate over winter (some species
hibernate in the second larval stage). The larva
comes out of hibernation in late spring and feeds
until summer. Then it closes the entrance of its bur-
row and constructs a side chamber in which it pu-
pates. The adult emerges in late summer and feeds
until October. It then digs a hole in which to hiber-
nate over winter. Two years are commonly required
to complete a generation, although in various species
the interval between successive generations may be
one to four years, depending in part on regional tem-
peratures.

The niche requirements or serai stage preferred
by difl^erent species are rigid and appear determined.

in large part, by the character of the type of soil a
species finds suitable for deposition of eggs and larval
growth. Studies performed under experimental con-
ditions demonstrate the nature of these requirements
(Table 8-4) but suggest no physiological explanation
(Shelford 1908, 1911, 1915: Balduf 1935).

CLAY SERE

Plant communities

Erosion or calculated removal of overlying ma-
terial may leave bare areas of clay exposed. In clay
above pH 4.5 annual plants, of which smartweed is
particularly important, appear within a few weeks to
two years ; the higher the clay pH, the quicker the
appearance of vegetation. Within two to five years
thereafter sweet clover invades and develops nearly
complete dominance over large areas. Sweet clover
is a biennial, and an exotic species unimportant in the
sere in some parts of the country (Bramble and Ash-
ley 1955). Prior to its introduction, this stage in
the sere on bare clay may have consisted of the
perennial grasses still found in small scattered
patches, or it may not have been well developed.
A shrub stage seldom takes dominance over extensive
areas, but thickets of raspberries and blackberries,
smooth sumac, trumpet creeper, and various other
species succeed the sweet clover and grass stage. The
first trees begin to invade early in the sere, but they
are scattered and slow of growth, and do not attain
dominance for 25 to 30 years. The tree stage is com-
monly made up of eastern cottonwood, American
sycamore, silver maple, and American elm. Willows
occur in wet spots. Herb species of the first two plant
stages disappear, for the most part, in the shrub
stage. The herb stratum now consists largely of wood
nettle. Advanced forest stages of oaks, hickories,
basswood, and sugar maple will likely invade in the
future : as they occur now in adjacent areas.

Ani

i/ life

The number of invertebrate species tends to in-
crease as the sere advances, although not always
regularly. In a study of a formerly strip-mined area
(Smith 1928), 18 species were found to be important
in the annual stage, 41 species in sweet clover, 40
species in shrubs, 32 species in the early forest stage,
and 67 species in the upland climax. More species
would be found in advanced stages because of the
greater variety of niches then available. Thus, in the
initial bare area there is only the ground stratum ; in
the annuals and sweet clover stages there are the
ground and herb strata ; in the shrub stage there are

112 Habitats, communities, succession:

ground and shrub strata. The herb stratum is \woT\y
develoiHul or absent ahogether. In the forest there
are tlie ground, lierb, shrub, and tree strata. Since
the early forest is on a floodplain, the ground is fre-
quently swept by floods, and the shrub stratum is
poorly represented. The climax forest has all strata,
richly developed, and possesses the greatest number
iif animal species. There is an increase in the abun-
dance of individuals per square meter with the pro-
gression of the sere: annuals, 268: clover, 531;
shrubs, 532: early forest, 748: climax, 2445 (David-
son 1932).

Beetles, spiders, ants, and mites are the most
abundant animals in the annuals stage, and along with
aphids remain most abundant also in the sweet clover
community. Grasshoppers are fewer in number but
especially characteristic of the first three stages : they
practically disappear in the forest. Earthworms are
absent in the amiuals and scarce in the sweet clover,
as are the springtails ; as the amount of soil humus
increases with the development of the sere, both
groups become more and more numerous. Snails first
appear in the sweet clover stage and increase in im-
portance in the forest stages.

The first two stages are not sufficiently extensive
to support a distinct bird fauna, but they are quickly
invaded by scattered shrubs and trees. A forest-edge
habitat is thus established and is occupied by forest-
edge birds (Brewer 1958). Beginning with the early
forest stage, these forest-edge species are replaced by
the forest bird community. The composition and
stucture of these two communities will be discussed
in Chapter 9.

The first small mammal (Wetzel 1958) to invade
the annuals and sweet clover stages is the prairie
deer mouse. It attains populations as high as 22
per hectare (9/acre). It persits until the shrubs and
trees have become well established. Its place is taken
in advanced stages by the woodland white-footed
mouse. The prairie vole prefers the grassy areas
and is found under briars and other shrubs. Peak
populations are about 18 per hectare (7/acre). The
short-tailed shrew invades the sweet clover stage but
does not establish a stable population until the shrubs
come in ; it persists into the climax forest. Wood-
chucks commonly occur throughout the early stages
of the sere, but mostly disappear in the forest. The
cottontail rabbit is common in the early stages, and
the fox squirrel invades with the first trees.

FLOODPLAIN SERE

A stream continuously deepens its chan-
nel, thus lowering the water table of the surrounding
land. At times of flood, the stream overflows its
banks. The flow rate of water declines as the water
passes over vegetated areas, and there is a deposition

of silt which may sometimes amount to several inches.
In a valley, the lowland area between the river and
the bluffs on each side is called the floodflain. In the
course of time, the river meanders back and forth
across the floodplain, cutting new channels and aban-
doning old ones, and frequently leaving a sequence of
terraces between its present channel and the sur-
rounding upland. A study of these terraces com-
monly shows a variety of i)lant communities that
constitutes the plant sere.

Plant communities

Gravel, sand, or silt is deposited on the inner
side of river bends. Attached aquatic vegetation may
occur in the water. On land, such herbs as smart-
weed, cocklebur, ragweed, beggar's ticks occur. At
some bends small sand dunes may occur, displaying
their characteristic plants and animals : usually this
stage is narrow at most, and may be entirely absent.
On sandy islands in the river or on sandy shores, the
sandbar willow often forms dense, shrubby thickets.
The first tree stage is ordinarily black willow mixed
with eastern cottonwood, and sometimes silver maple.
On the floodplain of the Canadian River in Okla-
homa (Hefley 1937), the sere proceeds next to an
edaphic subclimax of either tall grass prairie or elm-
oak. The climatic climax on the surrounding upland
is mixed prairie. The normal sequence of stages in
this region has become considerably modified by the
extensive ecesis of the exotic tamarisk tree, intro-
duced from Asia.

On the Mississippi floodplain in western Tennes-
see (Shelford 1954b) the mature cottonwood-willow
associes contains an abundance of vines of several
species that form such tangled masses as to be al-
most impenetrable. The next stage is one in which
sugarberry, sweetgum, American elm, and American
sycamore predominate: several other species are
present in small numbers. This leads to an oak-
hickory stage that includes a complex variety of spe-
cies, and eventually to the regional climax of western
mesophytic forest. Cypress becomes part of the com-
position of the early floodplain forest around the edge
of small oxbow ponds or other standing water. The
schedule for this sere, the time from the start to the
beginning of dominance by each successive plant com-
munity, has been estimated as follows : sandbar wil-
low, 3 years : cottonwood-willow, 35 years : sugar-
herry-sweetgum. 82 years : early species of oaks and
hickories. 260 years : intermediate species of oaks and
hickories. 350 years ; early climax of oaks and tulip-
tree, 440 years : full development of the climax, 600
years.

Elsewhere in the eastern United States, the cot-
tonwood-willow stage gives way to a narrow zone of
sycamore. Two or three species of elm .white ash.

Rock, sand, and clay 1 ]3

TABLE 8-5 Distribution of annelid worm spe
Original nomenclature revised by W. J. Harma

the Sangamon River floodplain forest of central Illinois (Goff 1952).

FamUy

Ruderals

WUlow-

silver

maple

SUver
maple -
elm

Elm-bur
oak

Elm-
shingle
oak

Oak-hickory
upland

LWmbricus terrestris
AUolobophora iowana
Bimastos tumidus
Octolasium lacteum
Henlea urbanensis

Henlea moderata
IHplocardia singularis
Fridericia agilis
Fridericia sima
Friderica tenera

Lumbricidae
Lumbricidae
Lumbricidae
Lumbricidae
Enchytraeidae

Enchytraeidae

Megascolecidae

Enchytraeidae

Enchytraeidae

Enchytraeidae

++

-

-

^l

and boxelder follow; then a mixed forest that in-
cludes black walnut, butternut, black maple, Ohio
buckeye, red mulberry, American basswood, tuliptree,
and hackberry ; next an oak-hickory stage ; and finally
the beech-sugar maple climax. The herb and shrub
strata are usually well developed in mature floodplain
forests. Telescoping or skipping of stages is not un-
common in this sere, since variation in ground level
or in height of terraces is considerable and the tran-
sition between heights is often abrupt. The later
stages occur only on the very oldest terraces and may
be hard to find at all.

Animal life

In the bare areas, in the herbs, and among the
invading trees occur such beetles as Heterocerus
pallidus and Bembidion laevigatum that feed on the
algae and detritus present on the shore. They make
their burrows in sand. Fly larvae, a cocklebur weevil,
a cocklebur mirid, and a cocklebur fly also occur.
The tiger beetles Cicindela hirticollis, C. cuprascens,
and on slightly higher ground C. puncUdata, prey on
the ground species and may even dig them out of
their burrows. Spiders, ground beetles, and rove
beetles invade from higher stages. In the herb
stratum and in the shrubby growth of willows, adult
midge flies and other flies are sometimes very abun-
dant. Tarnished plant bugs, 12-spotted cucumber
beetles, and other insects of open area habitats are
present, and there is invasion of various species from
the forest itself (Hefley 1937, Shelford 1954b).

The animal life of the floodplain forest is much
the same as that of the deciduous forest in general
(Chapter 9) and does not need to be discussed here
except for its unique features. Annelid worms make
their appearance in the ruderal stage, become very
abundant in the moist soils of the elm-ash and mixed
floodplain forests, then decrease in numbers in the
drier soils of the late serai stages. They occur mostly

in the first 5 to 10 cms below the surface in moist
soil, but up to 30 cms or more in dry soil. During
the winter they keep below the frostline, and in very
dry weather they roll up in small knots and aestivate.
Ten species occur in the floodplain of the Sangamon
River in central Illinois, and each species has its par-
ticular range of moisture requirements between the
river's edge and the upland forest (Table 8.5).

Snails and slugs are moisture-loving animals and
occur in large numbers and great variety in flood-
plain forests; it is not hard to find 15 to 20 species
with a little searching. Mesodon thyroidns is a com-
mon snail, and on a floodplain in central Illinois an
average population of 6.3 individuals per m^ was
found during the autumn (Foster 1937). This
amounts to a biomass of living flesh (shell excluded)
of 15.8 g/m^ (141 lbs/acre). Siiccinea ovalis on
another old Illinois floodplain (Strandine 1941) aver-
aged 6.5 individuals per m^ in September with a
biomass of only 0.878 g/nr (7.84 lbs/acre). Snail
flesh is an important source of food for such small
mammals as the short-tailed shrew.

Effects of flooding

Animals living on floodplains must usually tol-
erate flooding of their habitats almost yearly, and in
years with heavy precipitation, often several times
annually. All land except the surrounding bluffs may
be flooded. Leaves are swept up from the forest floor
and piled with other debris against shrubs and the
bases of trees. Herbs and shrubs may be damaged,
sometimes killed.

Observations during time of flood (Stickel 1948)
showed that emergent brush, the bases of trees, and
debris rafts supported masses of insects, spiders,
millipedes, snails, and amphibians. Debris rafts were
refuges for box turtles and pine-mice as well. Snakes,
turtles, and amphibians were also seen swimming or
floating in the water. No white-footed mice were

114 Habitats, communities, succession

fouiul in tlic tlood, hut the size of the population per
unit area, determined by live-trapping immediately
after the flood, was tiie same as it was immediately
before the flood, and a number of tagged individuals
were found surviving. This species readily climbs
trees and may well have passed the danger period
arboreally. Tagged box turtles were found on the
identical home ranges they had occupied before the
flooil. This flood lasted only a few days. Severe
flooding persisting for long periods is known to have
virtually exterminated species of small mammals from
wooded floodplains in grassland areas (HIair 19v^9).
[--arger mammals, such as rabbits, opossums, and
foxes, quickly leave flooded areas and may be tem-
(xjrarily concentrated around their margins. Squirrels
and raccoons easily obtain refuge in the trees, but
if the flood persists for some time, they may have
trouble finding food. W'oodchucks normally spend
considerable time in underground burrows and may
be trapped there by flood waters (Yeager and Ander-
son 1944).

Invertebrates in the soil are also affected by flood-
ing. A gradually rising water-table may eventually
displace all the air from a soil, and the arthropods
are killed. Many earthworms leave their tunnels
when these are inundated and are killed. Some spe-
cies that regularly exist in areas subjected to frequent
flooding, however, are not injured. Crane fly larvae
are flood-resisting. In normal flooding, bubbles of
air trapped in the soil provide suflScient oxygen for
at least the smaller arthropods (Kevan 1956). The
eggs of some floodplain mosquitoes are laid just above
the water level of pools during late summer or
autumn and must be flooded the following spring be-
fore they will hatch.

SUBSERES

Abandoned fields

When farmland is abandoned, succession back
to natural vegetation and ultimately to the climax is
rapid, since the soil is already relatively fertile and
does not need a great deal of conditioning. On the
Great Plains the subsere proceeds rapidly through
stages of annual herbs, several of which may be ex-
otics naturalized from other continents ; mixed annual
and perennial herbs ; a short-lived perennial grass ;
dense stands of triple-awned grass ; finally, the climax
of short grasses. This last stage may be attained in
10 to 20 years.

Small mammals and grazing domestic aninrals
retard the succession by feeding on grasses while
avoiding the herbs. Sheep have the opposite effect,
preferring the herbs. Harvester ants denude the veg-
etation in a circle around their mounds and consume

a considerable amount of the available seed supply.
Ant coactions may be of very great importance when
we consider that the population of a mound may aver-
age 10,000 individuals and the inimber of mounds
per hectare range from to 10 (0 to 4/acre) in the
annuals stage, 7 to 28 (3 to 11 /acre) in the mixed
annual and perennial herbs, 12 to 52 (5 to 21 /acre)
in the first grass stage. 40 to 142 (16 to 57/acre)
in the triple-awned grass, and to 32 (0 to 13/
acre) in the final stage (Costello 1944).

Six plant stages are recognized in the sere that
develops in the mixed prairie region of Oklahoma :
an initial stage of mixed herbs; three intermediate
stages involving different proportions of triple-awned
grass ; a subclimax ; and the climax of Andropogon
and Bouteloiia grasses. The insect population con-
sists of 293 species representing the following orders,
ranked in decreasing abundance : Coleoptera, Hemip-
tera. Homoptera. Diptera, and Orthoptera. There
was a greater variety of species and a greater abun-
dance of individuals in the intermediate stages than in
either the early or the climax stages, probably be-
cause of the greater variety of plant species present
in the intermediate stages. Of the 144 species of in-
sects in the climax, 58 per cent entered the sere in its
initial stage, 15 per cent in the second stage, 12 in the
third, 5 in the fourth, 2 in the fifth, and only 8 per
cent were limited to the climax itself. The ecesis of
the mature animal community was therefore a grad-
ual and progressive one. On the other hand, many
species that were present in the early stages did not
persist into the climax community (Smith 1940).

Succession in abandoned fields of the southern
.Atlantic and Gulf states is of special interest. During
the first year, crabgrass and horse-weed, annuals, pre-
dominate. During the second year an aster and a rag-
weed, and in the third year the perennial broomsedge
grass, become dominant. The grass is invaded quickly
by loblolly and shortleaf pines which form closed
stands in some areas in as little as 10 to 15 years.
These pines do not reproduce in their own shade.
They mature in 70 to 80 years, and are replaced by
tlie climax oaks, hickories, beech, and sugar maple
which take complete dominance by the time the area
is 150 to 200 years old (Oosting 1942).

The bird succession (Table 8-6) shows a change
from grassland to forest-edge to forest species with
an increase of both species and number of pairs as
the sere progresses. A great diversity of species in-
iiabits the pine forest since several forest-edge spe-
cies persist, while pine warbler, brown-headed nut-
liatch, solitary vireo, and yellow-throated warbler
are particularly characteristic of it. Pine-mice and
meadow voles are common in the grassy stages and
pine-mice, cotton-mice and golden mice in the for-
ested stages.

In Michigan, the sere in abandoned fields passes
through the following stages: annuals-biennials;

Rock, sand, and clay 1 15

FIG. 8-11 Stages in the strip-mine
plant sere in east central Illinois,
(a) sweet clover, aster, ragweed,
after 7 years, (b) same area 10 years
later, trees invading, (c) silver
maple-cottonwood-sycamore flood plair
forest after about 40 years (Wetiel
1958).

perennial grasses ; mixed herbaceous perennials ;
shrubs ; and finally three tree stages, the first reached
in 21 to 25 years. Prairie deer mice are at their most
abundance during the early stages, meadow voles in
the intermediate grassy stages, and the woodland
white-footed mouse and short-tailed shrew in the
shrub and tree stages. Such game species as ring-
necked pheasants, bobwhite, and cottontail rabbits
are common on abandoned farmlands but give way to
another group of game species, including white-tailed
deer, ruffed grouse, and gray squirrels, when the
forest stages become established (Beckwith 1954).

Pastures

Pastures in northern Ohio contain a sod prin-
cipally of blue grass. With light grazing, this sod
will resist invasion of other species for a long time,
but with heavy grazing, resistance is weakened and
unpalatable herbs, briars, and hawthorne come in.
The latter two species are armed with prickles or
thorns discouraging animal browsing. When they be-
come dense enough they kill the grass beneath them.
Eastern redcedar may establish itself in horse pas-
tures, but not in cattle pastures ; cattle browse it but

116 Habitats, communities, succession

horses will not. In tlie middle of protecting tliickets
of briars, hawtliorns. and redcedar snch deciduous
trees as elm, ash, tuli])tree, sycamore, and oak come
in. After a few years they grow beyond the reach of
animals, shade out the briars, hawthorns, and red
cedar, and establish a forest dominance. Where left
undisturbed by man, tiie succession of native vegeta-
tion will thus bring about the elimination of domestic
animals from the area and replacement with the biotic
climax natural for the region.

In western areas too dry for deciduous forest,
overgrazing reduces the vigor and abundance of the
taller climax grasses, and the short grasses that are
less easily grazed are favored. Unpalatable herbs,
sagebrush, cacti, and mesquite may also replace
grasses over extensive areas. Although native animah
such as the bison and pronghorn may have heavily
grazed the original prairie in locally arid regions, the
result was less drastic than that produced by the
heavy concentrations of grazing stock on our farms
and ranches at the present time. When the most
favored vegetation was reduced, native animals com-
monly dispersed into other areas so that the carrying
capacity of the land was not critically reduced.

Burns

Prairie fires, frequently started by lightning or
by Indians, were doubtless important in preventing
deciduous forest from succeeding grassland in parts
of the middlew^est. More lately, fires are started by
careless campers or travelers. Fires are especially
destructive in coniferous forests, as the clinging dry
needles encourage crown as well as ground fires to
develop. Many thousands of square miles of forests
are burned over annually.

The extensive pure stands of longleaf pine on the
coastal plain of the southeastern states are probably
a consequence of ground fires that regularly occurred
at intervals of 3 to 10 years before white men came.
The terminal bud of the longleaf pine is well pro-
tected by a thick covering of green leaves, one of
several characteristics that make the species ex-
tremely fire resistant (Chapman 1932). Fire de-
stroys all seedling hardwood trees as well as other
species of conifers.

When coniferous forest is destroyed by fire, the
first trees to invade are usually quaking aspen, paper
birch, and sometimes balsam poplar. These forests
cover extensive areas in Canada and southward on
the Rocky Mountains. Jack pine in the north and
lodgepole pine in the western mountains either come
in with the deciduous trees or succeed them. The
cones of these two trees take several years to open
and shed the seeds held within, and may not do so
at all unless heated by forest fires. Aspen and pine
are eventually replaced by the climax forest. In many

TABLE 8 6 Breeding bird pairs per 40 hectares (100 acres) in
sere developing on abandoned fields, Georgia Piedmont region,
averaged fronn two stations in horb-shrub (I, 3 years old),
three stations in grass-shrub-tree (IS, 20,25 years old), four
stations in pine forest (25, 35, 60, 100 years old), and one
station in oak-hickory (over 150 years old) (condensed from
Johnston and Odum I9S6).

Grass- Oak-

Herb- shrub-tree Pine hickory
Bird species grass (forest-edge) forest climax

Grasshopper

sparrow 20

8

Eastern meadow-

lark 8

6

Yellowthroat

11

Yellow-breasted

chat

7

Prairie warbler

4

Catbird

1

Indigo bunting

1

American goldfinch

+

Bobwhite

+

Field sparrow

36

4

Rufous -sided towhee

9

13

Bachman's sparrow

5

Pine v/arbler

5

43

White-eyed vireo

3

Mourning dove

+

Cardinal

6

15

23

Summer tanager

2

14

10

Chuck-wills-widow

+

•^

Brown-headed

nuthatch

Brown thrasher

Solitary vireo

Yellow-throated

warbler

Pileated woodpecker

Hooded warbler

11

11

Carolina wren

10

10

Ruby-throated

hummingbird

10

Blue -gray

gnatcatcher

13

Tufted titmouse

15

Eastern wood pewee

3

Blue jay

5

Carolina chickadee

5

Crested flycatcher

6

Red-eyed vireo

43

Yellow-throated

vireo

3

7

Wood thrush

2

23

Yellow-shafted

flicker

1

3

Hairy woodpecker

1

5

Downy woodpecker

1

5

Yellow-billed

cuckoo

+

9

Black and white

warbler

a

Acadian flycatcher

5

Kentucky warbler

5

Total species 2

18

30

22

Total pairs 28

104

163

224

Rock, sand, and clay

117

FIG. 8-12 Plant succession In
abandoned fields In Virginia,
(a) annual herbs invading old
cornfield, (b) broom-sedge
grass, (c) invasion of young pine
trees.

FIG. 8-13 The Tillamook Burn
in coniferous forest in Oregon,
1944 (courtesy U.S. Forest
Service).

118 Habitats, communities, succession

«S^~^'

western areas, particularly in the Sierra Nevada,
chaparral may occur in dense stands after fires and
persist for a long time.

Since the burn subsere in coniferous forest com-
monly includes the aspen-birch associes, many of the
typical animals in this stage are deciduous forest and
forest-edge species, although there is a penetration of
coniferous forest species as well. The birds and mam-
mals are not generally very numerous in the aspen-
birch community, but ground invertebrates may be
more abundant here than in the poorly decomposed
acidic ground duff found in the coniferous climax.

ANIMAL COMMUNITIES

Although the plant communities that make
up the stages of the different land seres and subseres
we have described are numerous and varied, the num-
ber of distinct animal communities that can be clearly
recognized are few. Actually, we can distinguish in
eastern North America only the animal communities
of grassland, forest-edge, deciduous forest, south-
eastern evergreen forest and coniferous forest. Each
of these communities varies in the different habitats
of rock, sand, and clay, and in the various subseres,
but the variations are of minor significance and are
best treated as facies of the larger communities.

by one of spruce as
frees now forming ir
reach nnaturlty (cou
Forest Service).

ndergro
>sy U.S.

SUMMARY

For terrestrial living, animals must actively
support themselves against gravity, obtain water, and
prevent excessive water losses from the body. They
must be equipped to endure a wide range of fluctuat-
ing temperatures, to secure oxygen, to endure in-
tense solar radiation, adjust to diurnation (day and
night) and aspection (seasonal changes), and yet
maintain close contact with the substratum.

Succession occurs on all primary bare areas, such
as rock, sand, clay, and floodplains, and in such
secondary bare areas as abandoned fields, pastures,
and burns. In humid regions, all seres converge to
the same climax community. There are normally
more plant than animal stages in any sere. The suc-
cession program of animal communities correlates
with the succession program of vegetation-types or
life-form of the plant dominants, not with plant com-
munities identified by the taxonomic composition of
the plant dominants. In eastern North America, we
can distinguish only the grassland, forest-edge, de-
ciduous forest, southeastern evergreen forest, and
coniferous forest terrestrial animal communities.

Rock, sand, and clay 1 19

We have seen that succession of animal communi-
ties in humid climates passes through three terrestrial
stages before attaining climax : grass, shrubs and
scattered trees (forest-edge), and forest. In arid cli-
mates, the climax may be reached at the first or sec-
ond stage. It is important for us to examine each
community in more detail, therefore, if we are to gain
an understanding of the ecology of animals prevailing
locally in different parts of the world.

VEGETATION

Local Habitats,

Communities, and

Succession:

Grassland,
Forests, and
Forest-edges

Grassland vegetation differs from forests in
that the above-ground vegetation is completely re-
newed each year. Grasses may be divided into three
categories on the basis of height: tall grasses (1.5-3
meters tall ) , such as big bluestem and slough grass ;
mid grasses (0.5-1.3 meters tall), such as little blue-
stem and needle grass : and short grasses ( less than
0.5 meter tall), such as buffalo grass and grama
grass. The taller grasses grow in wet habitats, the
short grasses in arid habitats. Most native grasses
are bunch grasses in that they grow in clumps with
the areas between the clumps either bare ground or
occupied by other species. Broad-leaved herbs occur-
ring between the dominant grasses are called forbs.
A few species are sod formers in that their growth is
continuous over the ground surface. The leafy aerial
parts of perennial grasses die in the winter or in dry
season, leaving the underground stems or rhizomes
to propagate the plant the following year (Weaver
and Fitzpatrick 1934).

Forests are composed of trees growing sufficiently
close together to dominate the entire area of ground
surface. In cold climates, forests are needle-leaved
evergreen ; in intermittently warm, moist climates,
they are broad-leaved deciduous ; and in continuously
warm, moist climates, they are broad-leaved ever-
green. In spite of these secondary differences in life-
form, the structure and internal dynamics of all forest
communities are quite similar. Useful methods for
measuring the density of trees per unit area are de-
scribed by Cottam and Curtis (1956).

Between forests and open country, the trees are
often widely spaced and do not completely dominate
the area : open-country shrubs and grasses become
interspersed. This transition area is usually narrow
around the margins of a mature forest, but where
succession is occurring, large areas of shrubs con-
taining small or scattered trees are essentially forest-
edge in character. Likewise, in agricultural areas,
hedge and fence rows, or narrow strips of trees and
shrubs along streams, are really edges without the
adjacent forest. Essentially, forest-edges provide, in
close proximity, forest, shrub, and open ground habi-
tats which animals take advantage of in a variety of
unique ways.

20

Deciduous trees slied their foliage in the autumn,
are bare over winter, antl obtain new foliage in the
spring. Coniferous trees, on tlie otlier hand, retain
their foliage throughout the year, although old dried
leaves fall a few at a time at all seasons. Diflferences
in the size, shape, and structure of the leaves are ini-
|K)rtant to many animals. The lack of foliage in de-
ciduous forests during the winter permits a greater
light penetration to the forest floor, more wind circu-
lation, and relatively lower temperatures than in
coniferous forests. During the summer, deciduous
forests generally have higher hut more variable tem-
peratures and lower relative humidities than do co-
niferous spruce and fir forests (Blake 1926, Dirks-
Edmunds 1947). Pine forests, however, commonly
develop in habitats that are warm and dry.

As shade producers, the deciduous and coniferous
trees do not vary as groups, but only as individual
species (Weaver and Clements 1938) :

Deciduous trees Coniferous trees

Hcaz'y shade producers
Sugar maple Yew

Beech Spruce

Basswood Hemlock

Firs
Thujas

Medium shade producers
Elms Eastern white pine

White oak Douglas-fir

Northern red oak
Ash
Black oak

Silver maple
Bur oak
Birches
Poplars
Willows

Light shade producers

Ponderosa pine
Tamarack
Lodgepole pine

It is interesting that light shade producers are spe-
cies found in the early stages of succession while the
heavy shade producers are mostly climax species.

There is an important difference between decidu-
ous and coniferous forests in the nature of the de-
composing dead leaves that fall from the trees. De-
composition of broad leaves is rapid and relatively
complete to form a rich humus that mixes gradually
with the mineral soil beneath. Needle leaves decom-
jxjse slowly and form a somewhat acid humus sharply
defined from the underlying mineral soil. Humus
formed in humid grasslands is similar to but richer
than that of deciduous forest : in arid grasslands it is
poorly developed. The nature of the humus and litter
affects the number and kinds of animals that occur
in the soil.

In grassland tiiere are three strata of vegetation :
subterranean, composed of roots and other under-
ground plant parts as well as bacteria, fungi, and
algae, i/routid. including the surface litter, and herb.
tlie stems and leaves of the grasses and forbs. The
forest not only has these strata, but also one of shrubs
and one or more of trees. Animals characteristically
limit their major activity to one or more of these
strata.

HABITAT

Grassland, forest-edge,

and forest-interior compared

.\t the L'niversity of Illinois, no significant dif-
ference in mean monthly temperatures, calculated bi-
hourly day and night, has been found between the
interior of a virgin oak-maple forest and an adjacent
open grassland. In the forest, however, the daily
extremes are not so great; i.e., the maximum mid-
afternoon temperature is not as high, nor the mini-
mum night temperature so low, as in the grassland.

Relative humidity during a summer day in Iowa
was found (Aikman and Smelser 1938) to average
20 per cent lower in grassland than in a shrubby
forest-edge, and .^ to 8 per cent lower in the forest-
edge than in the forest-interior. There is less differ-
ence between the three habitats, however, at night.
Rate of evaporation, as measured with Livingston
atmometers, is inversely correlated with humidity,
being greatest in grassland and least in the forest-
interior. Daily changes in relative humidity between
day and night tend to vary inversely with the tem-
perature, except when there is rain.

During four years at the University of Illinois
woods, rain gauges recorded 88.8 cm (35.5 in.) per
year in the adjacent grassland, upon which full pre-
cipitation fell, and 70.1 cm (28.0 in.) throughfall
(the amount reaching the ground) under the tree
canopy of the forest. There was variation of through-
fall from spot to spot in the forest, depending on the
location of openings in the canopy and drip-points
from the leaves and stems. Stem-fiorv of water down
the tree trunks was not measured. Throughfall and
stem-flow together make up the net rainfall. In a
shortleaf pine plantation in southern Illinois (Bog-
gess 1956), the net rainfall over three years averaged
91.2 per cent of the total rainfall. Interception, the
amount of rainfall presumably evaporated back into
the air, was 100 per cent in very light rainfalls but
less than 5 per cent of rainfalls exceeding 5 cm
(2 in.).

In a beech-maple forest in northern Ohio, which
bordered on an open field, wind velocity at a distance
245 meters (about 800 ft) inside the west margin

Grassland, forests, and forest-edges

21

-MONDAY v/ TUESDAY-

ai Mt BI

-WEDNESDAY-
SI

-THURSDAY-

xn Mt

- FRIDAY -

xn

\

fc

FIG. 9-1 A weekly chart fr

hygrofhermograph placed at shrub level in a decic

entral Illinois.

was reduced to a minimum of 10 per cent when the
trees were in leaf and 25 per cent when not (Williams
1936). With a protective edge of shrubs, the wind
velocity would doubtless have been decelerated more
quickly.

Summer light intensities are much less under
foliage than out in the open. Noontime illumination
under shrubs in Iowa averaged 26 per cent of full
sunlight; within the forest interior, 6 per cent (Aik-
man and Smelser 1938). The forest floor is not uni-
formly illuminated because small openings in the
canopy admit sun-flecks of varying intensity. In the
Cottonwood, pine, black oak, and sugar maple stages
of the sand sere at the lower end of Lake Michigan,
the percentages of the forest floor shaded during the
midday hours were 68, 87, 75, and 90 per cent, re-
spectively (Park 1931.). There may be some change
in the quality of light that filters through the forest
canopy, as there is of intensity, as some wavelengths
are used more than others in photosynthesis ; green
is transmitted or reflected and not absorbed. Where
a stand of trees abruptly confronts an open field,
light penetrates laterally under the forest canopy and,
the typical edge configuration reversed, the light per-
mits shrubs to extend 40 meters or more into the
interior.

Vertical gradient

There is a gradient in microhabitat factors from
above the grasses down to the ground. In one study
of virgin prairie (Weaver and Flory 1934), light
intensity varied from 100 per cent in full sunlight to
25 per cent at one-half the pile depth of the grasses

to 5 per cent at the base of the stems, and, of course,
zero per cent in the subterranean stratum. The rela-
tive humidity above the grass was 20 per cent ; in the
grass, 31 per cent. The wind velocity above the
prairie grasses was 14.5 km/hr (9 mph) ; at the top
level of the grasses, 6.0 km/hr (3.7 mph) ; at the soil
surface, zero. The rate of water evaporation from
white spherical atmometers was 55.3 cc/day above
the grasses, 33.3 cc at top surface of the grasses, 15.1
cc at one-half the pile depth of the grasses, and only
13.4 cc just above the soil surface. The temperature
gradient varies with the height of the grass and be-
tween day and night.

The vertical gradient of temperature in a decidu-
ous forest in central Ohio varies with the season and
with the height of macroclimatic temperature (Table
9-1). In the summer, the greatest extremes of tem-
perature occur in the canopy, but at other levels, both
above and below the ground, summer daily mean
temperature is more stable than at any other season.
Because the canopy largely controls the air tempera-
ture beneath it, there is little or no thermal stratifica-
tion between it and the ground. Summer soil tem-
peratures are always lower at 1.2 meters below the
surface than at the surface. For comparison, air tem-
peratures in a coniferous forest in Wyoming during
July and August averaged 12.3°C at 0.1 meter above
the ground and 7.6 °C 0.1 meter below the surface
litter (Fichter 1939).

During the winter, temperatures in deciduous for-
ests are lowest near the ground and more uniform at
all higher levels than during the summer, since the
absence of a canopy permits greater turbulence, hence
less stratification, of the air. Soil temperature at 1.2
meters depth is generally higher than surface tem-

122 Habitats, cominunities, succession

21 22
°C

FIG. 9-2 Gradient of air temperatures
medium height at night (after Waterhou

and
1955).

23 24 25

)ve (a) tall grass and (b) short grass on

26 TALL
GRASS

SHORT 10
GRASS

perature ; beneath the litter in the central Ohio area
temperatures do not usually go below freezing
(Christy 1952) . A covering of snow gives added pro-
tection against freezing of the leaf litter. Another
study (Wolfe et al. 1949) revealed differences be-
tween temperatures above and below a snow covering
2 to 10 cm deep during a period of two months aver-
aged 8.9°C, and on one occasion reached 15.5°C.

Relative humidity decreases from the ground
stratum upwards. In a young elm-maple forest in
Tennessee, the relative humidity from mid-February
to mid-August averaged 77.9 per cent at the surface
of the leaf litter, 75.2 per cent in the herb stratum
0.5 meter above the ground, 72.5 per cent in the
shrub stratum at 0.9 meter above ground, and 67.4
per cent in the trees at 7.6 meters above ground
(Adams 1941). In this same forest, the rate of
evaporation between May and November in the four
strata respectively averaged 29.4, 60.7, 72.8, and
99.2 cc per week. In a spruce-fir forest in Wyoming,
the average weekly evaporation at 0.1 meter was 50.5
cc, at 1 meter 7S.2 cc, and at 3 meters 103.9 cc
(Fichter 1939).

In the Tennessee elm-maple forest mentioned a

nd (c)

2 13

in grass of

moment ago (Adams 1941), the average daily mid-
summer light intensities measured with a MacBeth
illuminometer for ground, herb, shrub, and tree (be-
neath the canopy) levels were respectively 52.3, 60.3,
60.4 to 76.2 foot-candles; in early May, before the
foliage was fully developed, intensities of 65.8, 78.3,
104.4, 119.1 foot-candles were measured. Under the
leaf litter and in the soil, the light intensity was, of
course, zero. Above the trees it was doubtless several
thousands of foot-candles. Ma.ximum light intensities
from the sun occasionally reach 15,000 foot-candles.
There are, therefore, three distinct sections in the
vertical gradient : below ground surface, between
ground and tree canopy, and above the canopy.

Ground insects, millipedes and isopods, when
placed in experimental gradients, show a prefer-
endum for lower light intensity, higher humidity, and
lower temperature than do insects taken from the
lierb or shrub startum (Table 9-2). In grassland,
motile organisms can quickly vary the microhabitat
to which they are exposed by changing their vertical
position only a few centimeters, and they do shift in
position as the gradient varies at different times of
day or from day to day. To obtain an equivalent

May

- September

November - December

TABLE 9-1 Vertical

Elevation

Minimum

Maximum

Range

Minimum

Maximum

Range

gradient of temperatu

Macroclimate

10.0°C

34.4'>C

24.4°C

-19.4°C

IB.SOC

37.7''C

in a beech forest in

+25.0 meters

7.8

31.7

23.9

-17.8

13.3

31.1

central Ohio (Christy

+18.9

8.9

30.0

21.1

-17.2

13.3

30.5

1952).

+6.1

8.9

30.0

21.1

-16.1

13.9

30.0

+ 1.5

8.9

28.9

20.0

-21.1

12.2

33.3

Surface of

leaf litter

9.4

28.3

18.9

- 8.3

8.3

16.6

Under leaf

litter

12.2

22.2

10.0

10.0

10.0

-0.15

12.2

18.9

6.7

2.8

12.8

10,0

- 1.2

10.6

16.1

5.5

4.4

13.3

8.9

Grassland, forests, and forest-edges 1 23

TABLE 9 2 Re
strata.

Mixed species

snts conducted in the field to establish light orientation of arthropods taken

Light inten-

sity gradient Control

Number of Inter- (no gradient)

experiments Strong mediate Weak Left Middle Right

Grassland animals from herb stratum 5

Forest animals from herb and shrub strata 7

Grassland animals from ground stratum 5

Forest animals from ground stratum 8

47% 34% 19% 40% 31% 29%

46 21 33 33 30 37

24 27 49 34 23 42

18 23 59 34 29 37

change in microhabitat in the forest gradient requires
a shift of several meters in vertical position. Experi-
ments show, however, that each forest animal species
occupies a stratum approximating its preferendum for
a particular microhabitat, in response especially to
the relative humidity factor (Todd 1949).

between these microclimates and the prevailing
macroclimates of the region must be demonstrated.

THE GRASSLAND COMMUNITY

Slope exposure

Microclimatic diiferences between North- and
South-facing slopes are great. South-facing slopes
receive a greater amount of solar radiation and are
commonly exposed to the prevailing winds. As a
consequence, both air and soil temperatures are
higher on South-facing slopes than on North-facing
slopes : relative humidity is lower, soil moisture lower,
and the rate of evaporation is higher. The differences
between the two slopes are most marked close to the
ground, increasingly less so at higher levels (Cantlon
1953).

The vegetation on the protected North-facing
slopes is usually more mesic in type and more luxuri-
ant than on the exposed South-facing slopes, and
there is a deeper organic leaf litter on the ground.
Types of vegetation characteristic of arid habitats
penetrate humid climates on South-facing slopes ;
mesic vegetation penetrates the relatively arid cli-
mates obtaining on North-facing slopes. Southern
types of vegetation invade boreal climates on the
warm South-facing slopes, and boreal vegetation in-
vades southward on North-facing slopes. In moun-
tain areas, vegetation characteristic of lower altitudes
penetrates higher on South-facing slopes, and vege-
tation of the upper altitudes penetrates farthest down-
ward on North-facing slopes. Animals are locally
distributed in a similar manner, partly as a direct
response to the climate, partly to the differences in
vegetation. Many other differences in microclimate
occur in various parts of the forest and forest-edge
(Wolfe et al. 1949), and in grassland.

In studying the distribution of animals in relation
to climate, it is obviously not sufficient to consider
only the macroclimate. Animals respond to the micro-
climate of their particular niches, and the relation

Invertebrates

Snails, earthworms, and myriapods are not nu-
merous in grassland because of the dry habitat. In-
sects, however, are abundant: some 1175 species and
varieties have been listed for different grassland plant
communities in Iowa (Hendrickson 1930). These
belong principally to the orders Orthoptera, Hemip-
tera, Homoptera, Coleoptera, and Diptera. Ants,
bees, and wasps (Hymenoptera) are also numerous.
Spiders make up about 7 per cent of the total
arthropod population in grassland. In one study
made in Nebraska (Muma 1949), 111 species were
collected from 128 hectares (320 acres) of mixed
high and low prairie containing some shrubs. Less
than a dozen species were web-builders ; there is a
lack of suitable web-building sites in grasslands. The
vast majority were wandering cursorial forms. In re-
gard to strata in this prairie, 45 species were re-
stricted to the soil and litter, 30 to the herbs, 1 to the
shrubs. Thirty-five species occurred in two or more
strata. The total population for the area was least in
the spring and greatest in the autumn. Peak popula-
tions in the ground stratum were reached during the
winter, however, because of the presence of many
hibernating immature forms. Similar seasoiial fluctu-
ations occur with other invertebrates, although the
peak populations of insects are usually attained dur-
ing the summer (Shackleford 1939, Fichter 1954).

In grazed pastures and in grassy meadows in
New York State, invertebrates average 777 individ-
uals per square meter (Wolcott 1937). Of this pop-
ulation, ants make up 26 per cent, leafhoppers 15
per cent, other insects 34 per cent, spiders 9 per cent,
millipedes 9 per cent, sowbugs 2 per cent, snails and
slugs 2 per cent, earthworms 2 per cent, and large
nematodes 1 per cent.

124 Habitats, communities, succession

SoiiK- insects show siiuctiiral ailaptatioiis for liv-
ing in grassland (Hayes 1927). May beetles in for-
ested regions commonly feed at night on the foliage
of trees and have well developed wings, but closely
related species in grassland areas feed on low grow-
ing ])lants during the day and are flightless. The de-
velopment of pilosity and thick integuments in some
insects appears to be an adaptation to prevent evapo-
ration. Prairie May beetles pu]»te in the spring
rather than autumn, probably in correlation with
their change in food habits, and adults appear in mid-
summer rather than late spring.

An insect niicrohabitat of special interest is the
dung of the larger mammals. Bison formerly oc-
curred in frequency one to 10-20 hectares. Inasmuch
as the output of each animal is about 2^ droppings
per day. the number of these microhabitats available
was considerable. Some 83 species of arthropods have
been collected from cow dung, mainly beetles and
flies, but including annelids, nematodes, and proto-
zoans. There is a regular succession of insect species
breeding and maturing. The microsere is completed
in about eight days, the length of time required for
the droppings to dry. The first species that arrive
are the obligatory breeders on dung. They have the
shortest life-histories, and remain for the shortest
time. Predatious and parasitoid species prey on the
coprophagous ones. The greatest variety of species
is present at the middle of the microsere. but the
composition of species varies with the season. Spe-
cies disappear as the dung disintegrates into the gen-
eral surroundings (Mohr 1943, Laurence 1954). A
comparable niicrohabitat and succession occurs in
carcasses of dead animals (Chapman and Sankey
1955).

Verlebrates

Table 9-3 g'nes a representative sampling of
small mammal populations found in grassland, al-
though it is to be expected that the species composi-
tion and size of populations will vary locally and from
year to year. The mores of grassland mammals,
which show how they are adjusted in behavior to
live in this community, are tabulated in Table 9-4.

Birds are not numerous in grassland. In north-
western Iowa (Kendeigh 1941b), grasshopper spar-
rows, western meadowlarks bobolinks, ring-necked
pheasants, marsh hawks, and short-eared owls aver-
aged less than one pair per hectare (2.5 acres).
Prairie chickens and sharp-tailed grouse formerly
occurred where now is to be found only the intro-
duced pheasant. The eastern meadowlark predomi-
nates over the western meadowlark in the wetter and
smaller pastures east of the Mississippi River.
Vesper sparrows and horned larks occur in short

grasses, but usually not in climax areas with dense
tall grasses. Upland plovers. Henslow sparrows. lark
buntings, and longsi)urs are common locally.

Some fourteen species of snakes are generally dis-
tributed over the prairies (Car])enter 1940). To the
east, the blue racer, massasauga, bullsnake, and garter
snakes are frequently found. The prairie rattle-
snake is increasingly common westward. The lizards
Cnemidoplwnis sexlineatiis. Sceloporus undulatiis,
and Holhrookia maculata. commonly occur in grassy
areas at forest-edges. The horned toad is found in
arid habitats.

The most characteristic amphibian of grassy areas
is the toad. All species breed in the ephemeral bodies
of water resulting from the rains of spring and sum-
mer. One species. Bujo cognatiis. will not breed un-
less it rains, even though bodies of water are present.
During the hot. dry weather of later summer, the
toads retreat to burrows in the earth or to other
shelter until favorable conditions again return (Bragg
and Smith 1943).

Grazing food coactions
and range management

Since the vegetative productivity of grasses is
very high, herbivorous animals, especially large mam-
mals, are favored in the grassland community (Ren-
ner 1938). Unlike trees and shrubs, the terminal bud
on grasses lies close to the ground and is not ordi-
narily injured by grazing. Meristematic tissue lies at
the base of the leaves so that when the terminal por-
tion of the leaf is eaten ofif, the leaf keeps on grow-
ing. Actually, lateral branching at the base of the
grass stem is stimulated by grazing, and a thicker
and more succulent growth with less fiber is pro-
duced. Productivity of grass is reduced if the herbage
is removed more than two or three times during the
growing season. However, total protein production
is not diminished, for frequent clipping results in an
increased ratio of leaves to stem, and leaves are much
richer in protein content. Light to moderate grazing

TABLE 9-3 Population of small mammals per hectare (2.5 acres)
In mixed prairie of western Kansas (after Wooster 1939).

Mammal species

Prairie meadow-mouse
13-striped ground squirrel
Prairie white -footed mouse
Harvest mouse
Little shrew
Short-tailed shrew
Black-tailed jackrabbit
Cottontail rabbit

7.6
6.8
2.8
2.4
1.3
0.7
0.1
31.3

Grassland, forests, and forest-edges 125

TABLE 9-4 A tab

(Carpenter 1940).

Pro-

DaUy

Sea-

duction

period

Social

Food

sonal

of

of

life

Stratum

habits

acUvlty

young

acUvlty

s

1

Mammal species

i

i

s

1 "

If

1

3
5

a

X

i

1

1

1

g
a

h

1

1
1

1

5

1
1

1

2

1

Bison

~

X

~

X

Pronghorn antelope

X

X

X

X

X

Wapiti

X

X

X

X

X

White-tailed deer

X

X

X

X

X

X

Mule deer

X

X

X

X

X

X

Cottontail

X

X

X

X

X

Whlte-taUed jackrabbit

X

X

X

X

X

Prairie-dog

X

X

X

X

X

X

X

Prairie white-footed mouse

X

X

X

X

X

X

Prairie meadow-mouse

X

X

X

?

X

X

X

Jumping mouse

X

X

X

X

X

X

Pocket mouse

X

X

X

X

X

X

Harvest mouse

X

X

X

X

X

X

Franklin ground squirrel

X

X

X

X

X

X

13-llned ground squirrel

X

X

X

X

X

X

Pocket gopher

X

X

X

X

X

Richardson ground squirrel

X

X

X

X

X

X

Wolf

X

X

X

X

X

X

Coyote

X

X

X

X

X

Badger

X

X

X

X

X

Bobcat

X

X

X

X

X

Skunk

X

X

X

X

X

Weasel

X

X

X

X

X

Red fox

X

X

X

X

X

Swift fox

X

X

X

X

X

Shrew

^

^

X

X

^

can, therefore, be carried with full or nearly full pro-
ductivity. Heavy grazing, however, should not be
permitted. In addition to reducing herbage produc-
tion, heavy grazing may destroy seed stalks prior to
the dropping of the seed or so weaken the plants
physiologically that seed is not even produced. The
growth of underground rhizomes and vegetative re-
production is retarded when photosynthetic activity is
reduced. The best pastures are those in which graz-
ing animals do not consume more than 70 to 80 per
cent of the total herbage productivity of the grasses
(Stoddart and Smith 1943). As a rule, not more
than about 60 per cent of the current forage volume
and 25 per cent of the flower stalks should be har-
vested by grazing animals. Overgrazing always
brings about a reduction in abundance of the more
palatable species and an increase in the less desirable
ones with the consequent deterioration of the range
and the productivity of the community (Weaver and
Tomanek 1951, Kucera 1956). The carrying capac-
ity of grassland or the largest number of animals

that can be supported without deterioration of the
range varies with the type of grasses involved, the
climate, and the soil (Table 9-5).

Although often overlooked, invertebrates consti-
tute one of the three important groups of grazing ani-
mals. Individually they may not consume appreciable
amounts, but in the aggregate they produce a very
significant effect. The total biomass of insects in a
New York pasture amounted to 3.2 g dry weight per
square meter. This is to be compared to 14.5 g for
the dry weight of cows per square meter that the pas-
ture was supporting. Feeding experiments showed
that in one pasture where grazing by the cattle was
moderate and the vegetation was ample, the insects
ate more of the grasses and clovers than the cows
did, but in another pasture which was being over-
grazed and in which the vegetation was short, the
cattle ate more than the insects did (Wolcott 1937).
Grasshoppers and Mormon crickets are sometimes
very destructive in the arid west. In one area in
Montana, a population of 25 grasshoppers per square

126 Habitats, communities, succession

FIG. 9 3 Forest-edge at Wil
and herbs in foreground, bri

am Trelease Woods, University of Illinois
rs and shrubs in nniddle, forest in backgr

meter destroyed enough forage on three acres during
one month to support one cow for a month ( Stoddart
and Smith 1943).

Rodents and rabbits consume very considerable
amounts of grasses and other herbs and cause great
damage at times of high populations. In a study per-
formed in Arizona (Taylor 1930), grazing by
Gunnison's prairie dogs alone consumed 87 per cent
of the total grass production and grazing by cattle
and rodents combined, 95 per cent. In California,
Beechey's ground squirrels eliminated 35 per cent of
the green forage by the end of the season, pocket
gophers 25 per cent, and kangaroo rats 16 per cent

(Fitch and Bentley 1949). Since these various
rodents have food preferences of grass species simi-
lar to those of cattle, there is obviously severe com-
petition between them, especially in times of drought.
When rodents are not overly abundant, they have
some beneficial effects in fertilizing, aerating, and
mixing the soil.

Among big-game mammals, bison and wapiti are
largely grass-eaters, especially during the summer
season. Food consumption of bison is about equal to
that of cattle, but wapiti eat only about half as much
per individual. In Yellowstone Park it has been esti-
mated that wapiti may utilize 67 per cent of the avail-

TABLE 9-5 Carrying capacity of

grasslands for big game and livestock (fr.

sources, compiled by Petrides 1956).

Location

Game or livestock

Number/mi* Biomass/mi
(2.6 km^) Pounds Kg

Oregon

Tanganyilca, Africa

Montana

Arizona

Western U.S.

Western U.S.

Nairobi Nat. Pk., Africa

Nairobi Nat. Pk., Africa

Antelope (64%), mule deer (36%) 9 1,000 454

Bush country game 10 3,300 1,498

Bison (50%), mule deer, elk, bighorn 21 14,000 6,356

Bison 17 17,000 7,718

Cattle, ave. all grassland types 20 20,000 9,080

Cattle, tall grass prairie 28 28,000 12,712

(1) herbivorous big game 85 28,000 12,712

(2) herbivorous big game 134 47,700 21,656

Grassland, forests, and forest-edges 1 27

FIG. 9-4

nterior

of a ten

nperate

deciduous

forest

of sugar m

aple,

basswood

and A

Tierican el

m

in Wiscon

sin (coi

rtesy U.S

Forest Service).

^^'lil'^^M^

111^

Sl^if^ ^OBluHll^BrvPl^fviib ^ 1 1

^V|^

^^^p

IBKll^MM»-r)tii loB^^a^^'' r^^^^

'h^^^ss^KSSSB^BBS^KB^

r^ .- ?

^ 1

»- . ^ ^

able grass forage, 47 per cent of forbs, and 30 per
cent of browse. Browse and forbs are used more
than grasses by pronghorn antelope and deer (Stod-
dart and Smith 1943). Because of difference in food
preferences, competition between the latter big-game
species and cattle, although significant, is not as great
as is sometimes supposed. Furthermore, deer and
wapiti are able to graze steep slopes and other areas
which cattle ordinarily do not (Stoddart and Ras-
mussen 1945). Competition between deer and wapiti,
sheep and goats is more direct, however, because
sheep and goats also feed largely on forbs rather
than on grass.

Overgrazing produces a change both in the kinds
and numbers of animals present (Table 9-6). This
is correlated with the change from mid-grasses to
short grasses to weedy perennials. The short-horned
grasshoppers increase in variety of species with this
change, but in other orders of insects, the number of
species present in overgrazed pastures either remains
the same or declines. There is generally an increase
in population level of all groups of arthropods, except
beetles, with overgrazing until the pasture deterio-
rates to such an extent that erosion becomes severe,
then there is a decline in abundance of all groups ex-
cept the Hymenoptera and Lepidoptera.

Meadow voles, cotton rats, and cottontails are
less numerous in overgrazed than in undisturbed
grassland, but other rodents and lagomorphs increase
in abundance. Tall grass in ungrazed pastures
hinders the vision of jack rabbits, kangaroo rats,

prairie dogs, and ground squirrels. Some rodents are
benefited by the larger and more numerous seeds
of the annual weedy species, and pocket gophers find
more tap- and bulbous-rooted plants in deteriorated
range (Bond 1945). Increased populations of insects
and rodents are a result, not a cause, of overgrazing.
If grazing by larger mammals is eliminated, suc-
cession back to thick grassland will occur in spite of
the smaller animals, and prairie dogs and ground
squirrels may actually be eliminated from the area
(Osborn and Allan 1949).

In the luxuriant native prairie of early days, there
was seldom overgrazing by such large mammals as
bison, antelope, and wapiti, although this sometimes
occurred in the more arid Great Plains. Insects and
rodents occurred in populations that were in equi-
librium with their food supply, and overpopulations
of the species were held in check partly by the vege-
tation itself and partly by predatory birds, mammals,
and reptiles. The most important of the larger preda-
tors were the hawks and owls, coyotes, foxes,
badgers, black-footed ferrets, bullsnakes, and rattle-
snakes (Shelford 1942). In California, it has been
estimated that these predators eliminate about half
of the annual increase of ground squirrels (Fitch
1948). Because coyotes and wolves occasionally took
calves and lambs, they were systematically killed by
ranchers ; many other predators suffered with them.
With the elimination of these predators, one of the
checks on the rodent population was removed at a
time when increased grazing by livestock rendered

128 Habitats, communities, succession

this control even more desirable. Damage done to the
range by increased populations of rodents and rabbits
has undoubtedly been much greater than the mone-
tary value of an occasional killed lamb, calf, or
chicken. In the great grasslands of the West, where
human populations are low, there would be advan-
tage not only in reducing the amount of grazing by
livestock to the carrying capacity of the land but in
restoring balanced populations of herbivorous and
carnivorous species.

THE FOREST-EDGE COMMUNITY

Grassland animals are usually restrained
from penetrating forests in the same way that true
forest animals are restrained from penetrating grass-
land, although the home ranges of these species may
overlap at the forest margin and in shrubby areas.
Since shrubs are especially numerous at the forest-

FIG. 9-5 Interior of a virgin
coniferous forest of Engeimann
spruce in Colorado (courtesy
U.S. Forest Service).

edge and animals have an opportunity to make use
of these as well as both grassland and forest, the
forest-edge biocies is well developed for some groups
of animals. There are probably no soil or small
ground animals characteristic of the forest-edge.
There are some foliage insects that find their pre-
ferred niches here. Many insects of grassland and
agricultural crops that overwinter as adults migrate
into the forest-edge to hibernate. Since many game
species of interest to man reach their greatest abun-
dance on the forest-edge, he has become impressed
by this so-called edge effect. When total populations
of all species are measured, however, the density of
birds (Table 9-11 ) or mammals is not always higher
than in the forest. When tw^o forest types come in
contact, for instance different deciduous forest types
or deciduous and coniferous forests, there is no con-
sistent change in the density of animal species
(Barick 1950 ) . The forest-edge is the preferred nest-
ing site of many birds (Johnston 1947).

Slightly

Heavily

Normal

Properly

over-

over-

Severely overgrazed

TABLE 96 Relative

Order

prairie

grazed

grazed

grazed

and eroded

abundance (per cent of
total specimens collected)
of various orders of

Coleoptera

29

27

19

14

11

Hemiptera

17

11

22

36

14

arthropods in normal and

Homoptera

21

24

22

26

8

overgrazed grasslands in

Hymenoptera

9

11

6

30

45

Oklahoma (after Smith

Diptera

19

22

23

30

6

1940).

Orthoptera

15

16

34

20

15

Lepidoptera

11

13

22

17

38

Arachnida

25

21

25

21

9

Total

19

19

25

24

13

Grassland, forests, and forest-edges 1 29

TABLE 9.7 Size of ar
May to September,

populations in forest and forest-edges,
isive of nnesofauna and microfauna of

Taxonomic
group

Deciduous Coniferous

forest, forest, Chaparral,

central Ill.» Utatf Utatf

Number

per hectare

Shrews, mice

chipmunks

62

31

87

Squirrels,

cottontails.

raccoons, etc.

1

20

+

Birds

12

24

25

Snakes,

lizards

+

+

+

Frogs, toads,

salamanders

+

+

Number per square meter

Snails, slugs

79

+

1

Spiders

158

16

10

Harvestmen

12

Pseudoscorpions

10

4

Sawflies, wasps,

bees, etc.

22

5

20

Ants

141

17

142

Flies

100

20

14

Moths, butter-

flies

8

+

Beetles

165

5

20

Leafhoppers,

aphids

82

27

27

True bugs

40

3

10

Thrips

131

1

Psocids

1

Lacewings

1

+

Crickets,

roaches, etc.

9

1

Insect larvae

307

4

20

Centipedes

67

5

3

Millipedes

31

1

2

Sowbug

24

'Including and extendi
^Hayward 1945
'Hayward 1948

Iford 1951 (a, b)

THE FOREST COMMUNITY

Since the censusing of each group of ani-
mals furnishes special problems, there have been no
studies of total animal populations in single forest
communities. By Table 9-7, however, it appears that
the ratio in numbers of individuals per hectare be-
tween different animal groups is of the order : 1 bird,
3 mammals; 13,000 snails and slugs, 20,000 centi-
pedes, millipedes and sowbugs ; 35,000 arachnids ;
and 225,000 large insects. The mesofauna would
number in the tens of millions (Table9-8), and the
microfauna in numbers so large as to be scarcely con-
ceivable. In general, the number of individuals rep-
resenting a species varies inversely with the body

size characteristic of the species. There is, however,
considerable variation in population levels both geo-
graphically and temporally. We must give special
consideration to each of these various groups of ani-

Soil animals

Some animals, geobionts (Table 9-8), spend
all their lives in the ground ; certain protozoans, flat-
worms, nematodes, annelids, tardigrades, snails, milli-
pedes, centipedes, some spiders, mites, pseudoscor-
pions, true scorpions, many small wingless insects,
some beetles and other winged insects, and a few
mammals are examples. Other animals, geophils, live
in the ground only as eggs, larvae, or pupae, such as
do many flies and beetles ; in cocoons, as do some
moths : or for hibernation, as do many beetles and
bugs.

Soil animals are most abundant in undisturbed
virgin areas. In a longleaf pine forest suffering fre-
quent burning, the number of small animals in the
humus layer was reduced to one-fifth and the num-
ber in the top 5 cm of the mineral soil was reduced
to one-eleventh of the number in unburned areas
(Hey ward and Tissot 1936).

Some 250 species of flagellate, amoeboid, and
ciliate protozoans have been recorded in the soil
(Sandon 1927), but only a few species are limited in
distribution exclusively to the soil since they also
occur in freshwater habitats. Many species occur in
practically worldwide distribution. Flagellates may
range from 100,000 to 1,000,000 or more individuals
per gram of soil ; amoebae, from 50,000 to 500,000 ;
and ciliates, from 50 to 1 ,000 (Waksman 1952) . Over
150 species of rotifers are known as ground inhabi-
tants, and about one-third of these species have been
found only in the soil. They feed on organic material
and, to a lesser extent, on nematodes and proto-
zoans. Nematodes may occur to the extent of 1,000
to 10,000 individuals per cubic centimeter. Most of
these forms belong to the Anguilluliformes and are
more or less worldwide in distribution. They com-
monly possess mucous glands in the skin, the secre-
tions of which aid locomotion. These nematodes are
very resistant to desiccation and will quickly become
active when moisture is added to soil that has been
dried out for years. Tardigrades occur regularly,
sometimes abundantly ; they too are very tolerant of
desiccation (Kiihnelt 1950). Land planarians are not
common except in moist tropical regions. Some of
these soil animals are detritus-eaters, some bacterial
and algal feeders, some partly carnivorous, and some
partly parasitic on plant roots.

The majority of these small organisms are active
only in soil water, present as a thin film lining the

130 Habitats, communities, succession

surfaces of the soil particles. Swimming forms are
necessarily very small ; often, they appear dwarfed
compared to the size they have been brought to in
cultures. Xematodes are somewhat less restricted in
their movements. They can distort the surface of the
water film by means of muscular movements, and
thereby bridge intervening air spaces to the next soil
particle. Amoeboid organisms and hypotrichous cili-
ates usually accommodate their shapes to irregulari-
ties of the solid surfaces over which they crawl and
can become larger in size but still remain in the water
tilm. The variety of micro-habitats in the soil accom-
modating the large number of species that occur in-
cludes spaces between surface litter, caverns walled
off by soil aggregates, root channels, fissures, and
pore spaces between individual soil particles. These
micro-habitats vary in size, temperature, and moisture
conditions (Birch and Clarke 1953).

Most of the insects, as well as