DOMINANCE-SUBORDINANCE
RELATIONSHIPS IN THE SOUTHEASTERN
FIVE-LINED SKINK, EUMECES
INEXPECTATUS TAYLOR

By

CLEO DUKE WILDER, JR.

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
June, 1962

ACKNOWLEDGMENTS

I am indebted to the members of my graduate committee ,
Dr. Coleman J. Goin (Chairman), Dr. Archie Carr, Dr. H. K.
Wallace, Dr. J. C. Dickinson, and Dr. Caspar Rappenecker for
their valuable criticisms and suggestions, and for their
patience during the long course of ray graduate program.

For their aid and comments concerning statistical
methods, I wish to thank Dr. Henry Wallbrunn, Dr. John
Howell, and Dr. Vonne Porter (Department of Psychology, Mem-
phis state university) .

I am especially grateful to Dr. Franz Sauer, who
contributed much through his critical reading of this paper.

For his constant encouragement and stimulation
throughout this research. Dr. Robert Haubrich receives my
deepest appreciation.

ii

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ii

LIST OF TABLES iv

INTRODUCTION 1

THE EXPERIMENTAL ANIMAL 10

MATERIALS AND METHODS 17

DESCRIPTION OF CATEGORIES 21

Aggressive Categories 21

Submissive Categories .... 23

Unclassified Category ........ 24

STATISTICAL METHODS 26

DISCUSSION 70

SUMMARY AND CONCLUSIONS 94

LITERATURE CITED 98

BIOGRAPHY

iii

LIST OP TABLES

Table page

1-A Approaches — Females 29

1- B Approaches — Males 30

2- A Neck Arching — Females 32

2- B Neck Arching — Males . , 33

3- A Tail Vibration — Females 35

3- b Tail Vibration — Males 36

;

4- A Touches — Females ..... 38

4- B Touches — Males 39

5- A Bites — Females . 41

5- B Bites — Males 42

6- A Tail-Waves-- Females ......... 44

6— B Tail-Waves — Males 45

7- A Movements Away — Females 47

7- B Movements Away — Males 48

8- a Retreats — Females ........ 50

8— B Retreats — Males 51

9- a* Unclassified — Females ..... 53

9-b Unclassified — Males 54

10-A Weights of Lizards During Observation

Period— Females 55

- iv -

Table

Page

10-B Weights of Lizards During Observation

Period — Males 56

11 Total Contacts for Each Category ....... 57

12 -a Rankings for Each Category of

Aggression and for Total Aggression —

Females ..... 58

12-b Rankings for Each Cagegory of

Aggression and for Total Aggression —

Males 59

13 Rankings for Each Category of

Submission and for Total Submission 60

14 Rankings for unclassified Category 61

15 Kendall Coefficient of Concordance (W)
for Rankings among Aggressive and

Submissive Categories 62

16 Correlations of Rankings in Aggression

and Submission between Sets 63

17 Frequency of All Contacts Compared

by Sets 64

18 Correlations of Rankings between

Aggression and Submission 65

19 Rank Correlation between Ranks for Total
Aggression and Submission and Ranks for

Each Unit of Aggression and Submission .... 66

20 Correlations of Rankings for Unclassified
Category with All Other Categories and

with Total Aggression and Submission 67

21 Correlations of Aggressive and Submissive

Categories with Each Other 68

22 Activity Rankings of Isolated Females 69

23 Distribution of Bites According to Body Region 69

v

INTRODUCTION

Studies on aggression have been made with subjects
from all levels of the animal kingdom. The majority of
these have dealt with vertebrate animals and have been con-
cerned with general or specific aspects of social relation-
ships, including, in many cases, dominance-subordinance
relationships. Such studies with vertebrates, from the
pioneer work of Sch j elderup-Ebbe (1913) until the year 1944,
have been an$>ly reviewed by Collias (1944) , and no attest
will be made to give a general review of that period. Since
the review by Collias, many more studies have been made on
many groups of vertebrates, and also, an increasing number
of workers are turning to the invertebrates.

first evidence of dominance and subordinance in
the Chondrichthyes was provided by the work of Allee and
Dickinson (1954) with the smooth dogfish, Mustelus canis
(Mitchill) , in which the most obvious feature was avoidance
of head-on contacts by the smaller of two individuals.

Investigations in the more widely studied Osteich-
thyes were frequently of a much more specialized nature.
Studies have been made of the relationship of social

1

2

organization with such things as territory and "leadership"
in maze-learning (Greenberg, 1947) in the green sunfish,
Lepomis cyanellus, and with position in social hierarchy
and amount of growth achieved (Allee, et al. , 1948) in the
same species; with the effect of prior residence in the
live-bearer, Platypoecilus maculatus (Braddock, 1949);
with interspecific competition in two species (Salvelinus
iQHfeiqalj.s and Salroo gairdneri) of trout (Newman, 1956) ;
and with juveniles in the Kamloops trout (Stringer and Hoar,
1955) . Other studies have dealt with the livebearer, platv-
poecilus maculatus (Braddock, 1945), cichlid fishes (Baerends
and Baerends-van Roon, 1950) , the Siamese fighting fish,

Betta splendens (Braddock and Braddock, 1955) , and fourteen
species of darters (Winn, 1958) .

The Amphibia have revealed less information about
aggressive activity than any of the five major classes of
vertebrates. Martof (1953) demonstrated a primitive type of
territorial behavior in the green frog, Rana clamitans, but
saw no active aggression. Test (1954) described a challenge
and definite aggression in defense of a territory that seemed
to be a part of the daily life of the frog, Phvllobates
trinitatis. Pearson (1955) observed possible avoidance re-
actions around undefended home sites in the spadefoot toad.

3

Scaphiopus holbrooki. Haubrich (1961) demonstrated under
laboratory conditions a loose type of hierarchical behavior
with several levels of aggression in the South African
clawed frog, Xenopus laevis. Lutz (1960) described fighting
which implied an incipient territoriality centering around
nesting holes in the male smith frog, Hyla faber. Grant (1955)
observed territoriality in two species of salamanders. Bury—
cea bislineata and Hemidactylum scut a turn in captivity. The
The former defended its territory by intimidation and direct
assault, while the latter used intimidation only, and those
animals not establishing territories starved.

Birds, especially domestic fowl, have been widely
used in behavior investigations since the work of Schjelderup-
Ebbe (1913) . Most such recent investigations have dealt with
special aspects of the much-minded avian social hierarchy.

Using domestic fowl as subjects. Potter (1949) con^ared
dominance relations among different breeds; Guhl (1945)
dealt with differences by age and sex; Douglis (1948) com-
pared resident and part-time members of different flocks;

Allee, et__al. (1955) studied the effects of testosterone
propionate on lower-ranking individuals; Banks and Allee
(1957) studied the effects of different flock size on
agonistic behavior; and Banks (1956) , using the more

4

primitive red jungle fowl, Gallus gallus subsp. , studied
general hierarchical relationships . Numerous other studies
have dealt with other birds, both captive and natural popu-
lations. Castoro and Guhl (1958) found differences in ag-
gression and territoriality according to sex in captured
pigeons. Sabine (1949) studied inter- and intraspecific
dominance relationships in winter flocks of j uncos and tree
sparrows. Thompson (1960) demonstrated social hierarchies
in captive flocks of the house finch and found degree of
aggressiveness and size of space inversely correlated in
caged birds. Kilham (1961) observed interspecific aggres-
siveness of myrtle warblers toward woodpeckers and other
birds. Andrew (1957) , on the basis of work with buntings
of the genus Emberiza, questioned the existence of subordina-
tion of aggressive behavior to the tendencies to behave sexu-
ally or to feed, and to certain other drives. Laskey (1950)
described dominance struggles (without territorial defense)
among both male and female cowbirds and considered this to be
closely linked with the winning of a mate. Brain (1949)
found a dominance order correlated with feeding station in
male and female great tits, Parus major. Goethe (1953)
studied social hierarchies in cage-raised herring gulls and
found them to be more similar to chickens than to pigeons.

5

Jenkens (1944) descirbed an interspecific pecking order among
geese. Boyd (1953) described both individual and family
hierarchies in winter flocks of white-fronted geese with the
former determined by the latter.

Both -ield and laboratory studies have been concerned
with aggressive behavior in mammals. Many of these have
dealt with causes of aggression rather than with the nature
Oj. social hierarchies, almost as well known in mammals as in
birds. Scott and Fredericson (1951) studied the effects of
hormones, hunger, and other factors on the fighting of mice
and rats. Fredericson and Birnbaum (1954) found that heredi-
tary differences were correlated with dominance in the social
relations of mice. Denenberg and Bell (1959) found an in-
verse relationship between frequency and relative intensity
of aggression in mice. Clarke (1956) found male voles to be
more aggressive than females and success in fighting to result
in dominance. Chitty (1952) concluded that the most important
factor limiting the growth of vole populations was intra-
specific strife. Balph and Stokes (1959) found a straight-
line dominance order in deer mice, Peromvscus maniculatus.
^®®bt (1948) found the degree of dominance to be inversely
proportional to the amount of fighting on goats. Schein and
Fohrman (1955) described a straight-line hierarchy in a herd

6

of dairy cattle, as did McHugh (1958) in the American bison
in herds of which dominance was depicted as an adaptive ad-
vantage, especially in severe winters. Miller, Murphy, and
Mirsky (1955) studied the effects of conditioning on an
established social hierarchy in a group of rhesus monkeys.
Kummer (1957) found 70 elements of social behavior and
studied the development of individual social position in the
baboon, Papio hamadryas.

Most studies of aggression in Reptilia have been
with lizards, especially in the family Iguanidae. Greenberg
and Noble (1944) conducted a definitive study under semi-
natural conditions of the social behavior of the American
chameleon, Anolis carolinensis, in which aggression was found
to be related to mating and maintaining of territories .

Gordon (1957) found that a similar situation existed in nat-
ural populations of the same species, and further, that a
reduction of aggressive behavior in the winter facilitated
the formation of hibernating aggregations. Ruibal (1960)
analyzed the different patterns of aggression in the terri-
torial and sexual behavior of five species of Anolis in Cuba
and found distinctive patterns of response among them.
Stebbins (1944) described a mild intraspecific competition
in the mountain swift, Sceloporus graciosus crracilis. Evans

7

(1946) described a social hierarchy which included vigorous
fighting in the lizard, Sceloporus grammicus microlepidotus.
ana another (1951) in the black lizard, Ctenosaura pectinata.
in which aggression was primarily in the form of bluff and
display. Lowe and Woodin (1954) described what they consid-
ered to be "male type aggressive behavior" in the lizard,
Pfarynosoma p. platyrhinos, which displayed extraordinary ag-
gression toward any object introduced near it in captivity.
Carpenter and Grubits (1960) described several forms of
aggressive interaction resulting in the dominance of one male
over another in the tree lizard, Urosaurus ornatus. Miller
(1951) observed what he interpreted as a suggestion of terri-
torial defense in the Yucca night lizard, Xantusia viailis.

In the field and experimental studies of the six— lined racerun—
ner' Cnemidophorus sexlineatus . Carpenter (1960) observed that
in groups certain animals accounted for most of the aggressive
activity, that smallest lizards were most subordinate, that
dominant lizards tended to be more active, that subordinates
displayed a peculiar submissive posture, and that aggressive
behavior could not be related to territoriality. Fitch (1954)
described vicious fighting associated with the breeding
season in male five-lined skinks, Eumeces fasciatus. but saw

little aggression in females or juveniles. Lowe (1948)

8

declared that territorial behavior is widespread among snakes
and that what has frequently been called a "courtship dance"
between male and female snakes was in most cases a terri-
torial fight between two males. Lowe (1950) observed fight-
ing, including striking, between two male sidewinders, Cro-
talus cerastes, and stated that most aggressive behavior in
snakes is due to competition involving territory, social
discrimination, or sexual dominance. Raney and Josephson
(1954) observed fighting between two snapping turles, Che-
lydra serpentina, after the breeding season, but the sexes
were undetermined. Parrish (1958) observed territoriality
and some indication of a social hierarchy in cpative sea
turtles in large tanks.

Many of the earlier studies of aggression in verte-
brates, as seen in papers discussed above, were concerned with
social hierarchies. This has been especially true of Ameri-
can behaviorists , some of whom (Allee and Dickinson, 1954;
Haubrich, 1961) were interested (at least in part) in demon-
strating such social hierarchies in certain vertebrate classes
for the first time. European ethologists, on the other hands,
have tended to make detailed descriptions of behavior patterns
and to study the factors that underlie their causation, their
genetic basis, their ontogeny, and the survival value and

9

evolution of these patterns (Baerends, 1958). This has been
seen in the studies of releaser mechanisms (Tinbergen, 1942,
1948, 1953), social facilitation (Darling, 1938? Davies,
1953), displacement activities (Lorenz, 1941; Tinbergen,

1953) , and imprinting (Lorenz, 1952) . It is a characteristic
of the latter group that they deal with behavioral units,
most of which are highly stereotyped (Baerends, 1958).

The purpose of this study is fourfold: to analyze

the phenomenon of aggression in a family of non-social lizards
in terms of aggressive units, to see if such aggression shows
any of the characteristics of social hierarchies, to relate
the units to the whole of aggression, and to analyze the
nature of individual differences in aggression in terms of
units and the whole. Finally this study has been conducted
with the hope that such analyses within different taxonomic
groups might lead to a better understanding of the evolution-
ary relationships of the groups as well as of the evolution

of behavior itself.

THE EXPERIMENTAL ANIMAL

The southeastern five-lined skink, Eumeces inexpec-
tatus, was described by Taylor (1932) . It is considered by
Taylor to be a member of the fasciatus group of the genus,
and very closely related to Eumeces fasciatus and Eumeces
laticeps. Eumeces inexpectatus is found from Virginia to
the Florida Keys and west to Louisiana (Conant, 1958) . It
occupies a wide variety of habitats, from mountain woodlands
to dry, open, coastal islands. In addition to having a broad
geographical overlap with E. fasciatus and E. laticeps. it
also appears to overlap these two species ecologically, and
the three may sometimes be found together in limited areas.
However, E. fasciatus does not occur in most of Florida where
the other two are locally abundant. In spite of this local
abundance, it is difficult to collect large numbers of either
species at a given time.

For purposes of this study it was considered necessary
to have a large number of specimens from which to select the
experimental animals. Fortunately, there was available on
Seahorse Key a large population of Eumeces inexpectatus .

This island is located in the Gulf of Mexico, approximately

10

11

five and one-half miles from the mainland and three or four
miles from the fishing village of Cedar Key, Levy County,
Florida.

Seahorse Key is the most seaward of a group of five
major islands (and numerous smaller ones) just off the Gulf
coast of Florida and located almost equidistantly between
the mouths of the Suwannee and Waccasassa Rivers. The island
is approximately one mile long and comprises 104.86 acres
above mean low tides (Wharton, 1958) . The topography of
Seahorse Key is dominated by a long, curving, central ridge
with a maximum elevation of 52 feet. The western end of the
island flattens out into a broad peninsula which probably
does not exceed 10 or 12 feet in elevation. There is no
fresh water, even in the form of temporary pools.

The vegetation of Seahorse Key, according to Wharton
(1958) , resembles a mesophytic mainland habitat. However, it
is very dry, and as Whatton points out, lacks such elements
as magnolia, American holly, and hackberry. It does not pre-
sent the classif picture of a self-perpetuating climax hammock
which can be found on mainland Florida. Wharton classified
the specific plant communities of Seahorse Key ass hardwood
mesophytic and xerophytic forests, mixed hardwood forest,
cherry laurel thicket, palm hammock, and saw palmetto.

12

Eumeces inexpectatus was found to be most abundant in the
first two of these communities.

Seahorse Key supports what has been described as a
"depauperate fauna." Nonetheless, it has unusually lar-ge
populations of two animals, the southeastern five-lined
skink, Eumeces inexpectatus, and cottonmouth, Agkistrodon p.
piscivorus. There are no amphibians. Other reptiles include
the fairly common American chameleon, Anolis carolinensis,
and the somewhat less common ground skink, Lygosoma laterale.
worm lizard, Rhineura floridana. six-lined racerunner,

Cnemi dophor us sexlineatus, red-tailed skink, Eumeces egregius,
and crowned snake, Tantilla coronata. The only mammals which
exert an influence on the island are the Norway rat, the gray
squirrel, and man. One part of the island is seasonally in-
fluenced by a large rookery of herons, ibises, and cormorants.
It is also frequented by crows and other birds from the main-
land.

The population of the southeastern five-lined skink
on Seahorse Key seems to represent a much larger concentration
of skinks than would be encountered anywhere on the mainland.
Because of the physical limits to dispersion and possibly
the somewhat limited number of predators, the population is
very high. These lizards also have an infinite number of

13

hiding places in the very heavy vegetation, the thick leaf
mold, the porous, sandy, friable soil, interlaced with roots,
and the numerous rotten logs. When the lizards are active,
they can be heard scurrying through the leaf mold. On many
occasions such activity was observed from one spot for as
long as an hour. It was, in part, observations of this type
which led to the use of this animal in this study of aggres-
sion.

When active on Seahorse Key, skinks are constantly
foraging for food. They move among the leaves, sometimes
under them for several inches, sometimes on top. At times
when moving like this, they seem almost oblivious to outside
movement. Thus, on occasion, two animals may be seen almost
making contact before either is aware of the other* s presence.
When this happens one lizard usually darts quickly toward the
other and pursues it for several feet. The second lizard
retreats just as quickly, sometimes as far as twenty feet.
Then each usually resumes its foraging activity. Fitch (1954)
describes similar behavior for Eumeces fasciatus. In their
daily movements the skinks also frequent logs and fallen
trees with some using crevices and hollows as retreats.

Most of their activity, however, is on the leaf -covered

ground

14

In the field these animals react to either movement
or sound of humans, sometimes from many feet, by racing up
the nearest tree. If the tree is a cabbage palm, they might
go all the way to the top and hide among the dried and broken
fronds; if it is an oak, they might disappear on the top
side of a limb high up. In almost every case they return to
the ground in less than five minutes if the disturbance does
not continue.

Because of the agility and speed of the skinks, they
were difficult to capture. Trapping by means of hardware
cloth funnel traps set against a drift fence of the same
material was not successful, so almost all the lizards were
caught by hand. This was done in two ways. One method in-
volved the use of numerous nineteen-inch squares of tar
roofing scattered through the woods and left in place per-
manently. A lizard under one of these pieces of roofing
could be caught if the collector acted quickly enough. The
second method was the more successful. If a lizard was seen
on the ground, it was approached carefully and slowly. Usu-
ally the skink dashed for the nearest tree, and there the
collector attempted to grasp it before it climbed out of
reach. Most of the lizards used in this study were captured

in this manner.

15

Most of the contacts observed in the field prior to
this study occurred during the breeding season between mem-
bers of the same sex. In fact, except during the breeding
season in early spring, these lizards are relatively inactive,
with most movement occurring between 10 A.M. and 4 P.M. In
habits this animal is essentially solitary, spending most of
its time inactive and under cover in hiding. Contacts seen
in the field are chance contacts of short duration. In sev-
eral contacts seen between females, one animal always moved
away quickly at the sight of the other. However, among males
vigorous chases and fights of short duration occurred before
one animal retired from the scene. These exchanges were
similar to those described by Fitch (1954) for Eumeces fasci-
atus. The aggressive and submissive behavior would seem to
strengthen its solitary habits. No evidence was ever seen of
any territoriality. It should be lacking in this animal for
the same reason that it is lacking in Eumeces fasciatus
(Fitch, 1954) and Cnemidophorus sexlineatus (Carpenter,

1960) : its habits would preclude the establishment of dis-

tinctive territories such as are found in other animals, in-
cluding lizards (especially Iguanidae) .

The behavioral units, or categories (as described in
a later section) , about which this study is organized, were

16

first observed in the laboratory. It was only after these
laboratory observations that behavior in the field was seen
in terms of units at all. A tail vibration was observed
during several encounters between animals. Tail-waving fre-
quently occurred as lizards were being pursued by the col-
lector. Arching of the neck usually preceded the capture of
food (most frequently beetles, some of which gave their cap-
tors real battles before succumbing) and continuedaas the
predator chewed upon its prey. Of course, the approach and
movement away or retreat occurred during encounters between
skinks. On two occasions skinks were observed fighting.

They bit, clung, turned, twisted, and rolled, and each time,
one retreated from the scene. It appeared that the combatants
wereof the same sex. The beginning of neither encounter was
witnessed. Whereas such field observations contributed to
the total knowledge of the animal, this work is based entirely
on laboratory observations.

MATERIALS AND METHODS

The twenty lizards used in this study were captured
in the late spring or early summer of 1959 on Seahorse Key,
Levy County, Florida. They were kept in captivity for at
least one month before they were used in experiments. Each
lizard was kept in a separate one-gallon jar in which a hand
ful of leaves had been placed. They were fed from one to
five mealworms (larvae of the beetle, Tenebrio sp.) each
day and a little water was sprinkled in the jar every two or
three days. Once a week each jar was cleaned out and the
leaves replenished. During the observation periods each liz
ard was weighed and measured every week or every two weeks.

Throughout their period of captivity lizards were
maintained in the animal room of the Department of Biology,
University of Florida. This room was equipped with an air
conditioner with thermostat which limited the room tempera-
ture to a maximum of approximately 80° F. (26.5° C.). Ob-
%

servations on Seahorse Key had indicated that Eumeces inex-
pectatus was most active between 75° and 90° F. (23.6° and
32.1° C.).

17

18

The general plan of this study was built around the
acquisition of data from a series of observations of paired
lizards treated as described below. It had been determined
in preliminary work that when two lizards were placed in the
space used, they encountered each other often enough to pro-
vide data, but still had sufficient room to "maneuver" and
avoid aggressive action — at least within a short period.

Thus, a series of observations of paired lizards allowed an
accumulation of data without the danger of injury or death
to experimental animals.

The space used for making observations was a ten-
gallon aquarium, 19-1/2" long by 10-1/8“ wide. Fitted in
the bottom of the aquarium was a piece of masonite with the
rough side turned up. This was done to facilitate the move-
ment of the lizards, for they tended to slip and slide when
crawling on the smooth slate bottom. A goose-neck lamp with
a 75-watt bulb was positioned so that the bulb was approxi-
mately 24 inches above the center of the aquarium bottom.
Approximately seven feet above the aquarium bottom a 100-watt
bulb in a photoflood reflector was suspended. In addition
to producing a constant source of light for observations ,
these lights also provided a fairly constant temperature
in the aquarium which was slightly above room temperature.

19

The temperature on the floor of the aquarium during observa-
tions ranged from 26° to 29° C.

All observations of skink activity were made with the
observer seated in front of the aquarium. Because of the re-
sponsiveness of the animals to outside movements and sounds,
it was necessary for the observer to be shielded from them
by means of a curtain. Observations were made through a
small eye-level opening in this curtain.

Two groups of lizards were used in this study. The
first group was composed of ten females (No. 11-No. 20) , and
the second group was composed of ten males (No. 31-No. 40) .
Identification of individual animals within each group was
easily accomplished because of the considerable individual
variation in body size, tail length, proportion of regenerated
tail, color pattern, and behavior. Because of these differ-
ences and because they were maintained separately between
observations, it was not necessary to mark them in any way.

Observations with females were made in the late
spring and early summer of 1959? observations with males
were made in the summer of 1959. Ten observation periods
of fifteen minutes each were made for all possible pair com-
binations (45) of the ten animals in each group. Thus, each
pair was observed for a total of 150 minutes. These ten

- 20 -

observation periods were conducted in two sets of five each.
The first set of five observation periods was completed for
all possible pair combinations before the second set of five
was started. No lizard was ever paired with any other lizard
more than once in a twenty-four hour period. Between the
completion of the first set of observations and the beginning
of the second, the animals were given a "rest period" of two
weeks. The procedure was the same for each sex.

The first individual observation period of a set
started as the two lizards were released simultaneously in
the observation space. (at opposite ends) . The remaining
four periods of the set followed consecutively without the
removal of the animals from the aquarium. With allowances
for rest periods for the observer, maintenance of the liz-
ards, and other necessary activities, it was possible to com-
plete from two to five pairings a day in this manner. These
pairings were generally made between the hours of 10 A.M. and
4 P.M. which corresponded with the period of greatest activity

by the skinks in the laboratory as well as in the field.

DESCRIPTION OF CATEGORIES

The recognition of the aggressive and submissive
patterns occurred during a long period of preliminary ob-
servations of the lizards. Specific behavioral traits began
to be recognized as these observations continued.

Aggressive Categories

It appeared to the observer that there were five more
or less distinctive and discreet actions which could be con-
sidered as components of aggression. These were approach,
neck arching, tail vibration, touch, and bite. The responses
of the lizards were helpful in the recognition of these ac-
tions. If one lizard reacted to an overt action of another,
it indicated to the observer that some sort of aggressive
behavior had occurred. These lizards may be able to recog-
nize aggression that the observer could not detect. The evi-
dence indicates that the five actions referred to above are
five categories that show aggression.

Approach

A movement of one animal directly toward another was
an approach, if it continued until the two animals were

21

22

approximately three or four inches apart. If the approached
animal showed any kind of response, it was called an approach
regardless of distance. Occasionally, one or both animals
would move toward the other while climbing up or attenpting
to escape from the container. Neither this nor any resulting
contact was considered aggression.

Neck Arching

Neck arching was characterized by a sharp tunring
of the head downward with the snout to the floor of the con-
tainer. This resulted in a slight raising of the fore part
of the body with a sharp flexure in the raised portion of the
neck. It was always directed toward another lizard. Occa-
sionally the animal would maneuver around, always facing the
other lizard, while keeping the neck arched.

Tail Vibration

The vibration of the tail was accomplished as the
tail lay extended straight behind the animal. It was shaken
very rapidly (sometimes so rapidly that the tail itself was a
blur) for a few seconds at a time and repeatedly. This was
usually done by the approaching animal on its approach, but
was frequently done by the approached animal. It seemed to
be one of the first reactions to the discovery of another

animal at close range.

23

Touch

A touch occurred when one animal made physical con-
tact with any part of another with its own snout. It usu-
ally occurred following an approach and was a definite con-
tact or push.

Bite

A bite occurred whenever one skink snapped at or bit
another animal. Not all attempts to bite were successful.

The misses were originally recorded separately, but since,
as aggression, they all apparently represent the same thing,
the total figures were lumped in the bite category. A record
was kept as to whether the bite was made on the head, body, or
tail.

Submissive Categories

There were three activities which were classified as
categories of submission. It was recognized that these ac-
tivities might be a part of the defnese or escape actions of
the animal. However, because of the individual differences
in performance when these were compared with aggression, it
was felt that the general term "submission” is a useful one

in this study

Tail Waving

The waving of the tail was not at all like the tail
vibration. Instead, the tail was moved back and forth later-
ally in a series of slow undulations while the lizard was
otherwise motionless. It was usually accompanied by a twitch-
ing motion near the tip. This was always done by the submis-
sive lizard of a particular contact or pair.

Movement Away

If, following a contact, one of the skinks crawled
away slowly, a movement away was recorded. Such movements
away occurred not only after "mild" displays of aggression,
but also after vigorous exchanges of bites. It represents
a mild but obvious form of avoidance.

Retreat

The retreat represented a more vigorous form of avoid-
ance than movement away. It was a sudden rush by one animal
away from the other, and usually followed physical contact
between the two individuals.

Unclassified Category

Not every contact between pair-mates could be inter-

preted by the observer as either aggression or submission

25

For example, there were many physical contacts when the
activity of the animal was directed toward escape from the
container. Such contacts produced no response on the part
of either animal to the presence of the other and were re-
corded as unclassified.

STATISTICAL METHODS

Tables 1 through 9 show the row-column (frequency
of category received compared with frequency delivered) rank
correlation. This correlation is the Spearman rank correla-
tion coefficient (rg) . The computation was made through use
of the formula.

s

1 -

6 E d^

N3 - H

where rg = rank correlation, d » difference between rankings,
and N = number of animals ranked. The significance level of
the rank correlation was determined by consulting a table of
critical values of the Spearman rank correlation coefficient
(Siegel, 1956, p. 284) . The same basic formula was used for
the correlations in Tables 16 through 20 and the significance
of these correlations was determined in the same manner.

The data contributing to Table 11 were used to com-
pare the differences in frequency of contacts occurring in
males and females. The test used was the Mann-Whitney U-test
which is used to test whether two independent groups have been
drawn from the same population. This is considered to be one
of the most powerful of the non-parametric tests (Siegel,

26

27

1956, p. 116) . The significance of the results of this test
was also determined from the table of critical values for U.

The Kendall coefficient of concordance was used to
compare the rankings among aggressive and among submissive
categories. These correlations and the levels of significance
are shown in Table 15. The Kendall coefficient of concord-
ance is expressed as

l/12k2 (N3 - N)

where rj « sum of ranks, N » number of animals ranked, and
k= number of sets of rankings (the numerator is usually ex-
pressed as s, which, as inserted in the formula above, is the
sum of the squares of the observed deviation from the mean of
Rj). The significance of the confuted W is determined by
the formula,

® = k(N - 1) W, with df • N-l (Siegel, 1956, p. 236).

Chi square was also used in comparing the location of
bites received (Table 23) with the expected distribution of

random bites

28

TABLES 1-A AND 1-B

Tables 1-A and 1-B indicate the
number of approaches made by the lizard
numbered in the row across the top toward
the lizard listed in the column at the
left. Totals at bottom indicate number
of approaches made; at right, number of
approaches received.

TABLE 1— A

29

r-l

O

v

-column (totals) rank correlation* -.79 (p

30

m

i

H

1

£)

H

O

V

cu

CM

CO

CO

<3

o

•H

4J

ID

H

a)

to

U

O

O

(8

M

TABLES 2 -A AND 2-B

Tables 2-A and 2-B Indicate the
number of neck arches made by the lizard
numbered in the row across the top toward
the lizard listed in the column at the
left. Totals at bottom indicate number
of arches made; at right, nuntoer of
arches received.

32

1

(N

9

3

*

Row-column (totals) rank correlations -.58 (P <C .05)

33

m

i

c»

1

3

Row-column (totals) rank correlation: -.10 (P .05)

34

TABLES 3-A AND 3-B

Tables 3-A and 3-B indicate the
number of tail vibrations made by the
lizard numbered in the row across the top
toward the lizard listed in the column at
the left. Totals at bottom indicate num-
ber of times vibrations made; at right,
number of times vibrations received.

35

co

a

(A

«

'-column (totals) rank correlations .176 (not significant)

TAIL VIBRATION — MALES

- 36 -

Row- column (totals) rank correlation: -.35 (P .05)

37

TABLES 4 -A AND 4-B

Tables 4-A and 4-B indicate the
number of touches made by the lizard num-
bered in the row across the top toward
the lizard listed in the column at the
left. Totals at bottom indicate number
of touches made; at right, number of
touches received.

TABLE 4 -A

38

-column (totals) rank correlations -.43 (P y .05)

TABLE 4-B

39

Row-column (totals) rank correlation: -.49 (P > .05)

40

TABLES 5 -A AND 5-B

Tables 5-A and 5-B indicate the
number of bites made by the lizard num-
bered in the row- across the top against
the lizard listed in the column at the
left. Totals at bottom indicate number
of bites made; at right, number of bites
received.

TABLE 5-A

41

Row-column (totals) rank correlations -.25 (P y .05)

42

CQ

a

Row- column (totals) fank correlation: -.44 (P .05)

TABLES 6 -A AND 6-B

Tables 6-A and 6-B Indicate the
number of times tail-waves were made by
the lizard numbered in the row across the
top when in the presence of the lizard
listed in the column at the left. Totals
at bottom indicate number of times tail
waves made; at right, number "received."

TAIL-WAVES

44

f

vO

f

I

Row-column (totals) rank correlation: .58 (P <. .05)

45

CQ

I

vO

3

h

-column (totals) rank correlation: -.54 (P .05)

TABLES 7 -A AND 7-B

Tables 7-A and 7-B indicate the
number of movements away made by the liz-
ard numbered in the row across the top
when in the presence of the lizard listed
in the column at the left. Totals at
bottom indicate number of movements away
made; at right, number made in presence
of lizard at left.

47

f

6

Row-column (totals) rank correlation: -.65 (P .05)

MOVEMENTS AWAY — MALES

48

Row-column (totals) rank correlation: -.90 (P .01)

49

TABLES 8 -A AND 8-B

Tables 8-A and 8-B indicate the
number of retreats made by the lizard
numbered in the row across the top when
in the presence of the lizard listed in
the column at the left. Totals at bottom
indicate number of retreats made; at
right, number made in presence of lizard
at left.

50

f

oo

a

Row-column (totals) rank correlation: -.74 (P .05)

RETREATS — MALES

51

Row-column (totals) rank correlation: -.44 (p .05)

TABLES 9 -A AND 9-B

Tables 9-A and 9-B indicate the
number of unclassified movements made by
the lizard numbered in the row across the
top when in the presence of the lizard
listed in the column at the left. Totals
at bottom indicate number of unclassified
movements made; at right, number made in
presence of lizard at left.

53

f

m

a

s

s

-column (totals) rank correlations -.61 (P <. .05)

54

03

I

0\

*3

%

-column (totals) rank correlations -.29 (P .05)

55

f

o

H

§

1

vO

m

CN

01

VO

<3*

o

•

•

•

•

•

•

•

IN

rH

IN

CM

CM

IN

IN

IN

rH

rH

rH

rH

H

rH

H

H

CN

CM

in

00

CO

00

Cl

•

•

#

•

•

•

•

II

rH

CV

O

O

01

o

Ov

OV

rH

rH

rH

o

VO

CO

rH

OV

00

CO

•

•

•

•

•

t

•

H

r-

CO

00

01

o

CO

00

rH

in

cv

00

in

OV

in

r-

•

t

•

•

•

•

•

rH

CO

CO

c*

01

o

OV

Ov

rH

u

<D

vO

o

in

IN

o

'3'

00

jQ

VO

•

•

•

•

t

•

•

§

rH

in

vO

VO

p*

00

r-

VO

a

*0

u

nj

01

00

IN

co

co

00

rH

N

in

•

•

•

•

•

*

•

•H

H

CO

CV

O

o

H

Ov

o

rH

rH

H

H

in

Ol

O:

CN

r-

•

•

•

f

•

t

•

rH

CO

00

Ol

©

H

o

CV

r|

rH

rH

in

in

CO

NT

H

VO

co

•

•

•

•

•

•

t

rH

o

o

o

O

rH

o

o

H

H

rH

H

H

H

co

r-«

vO

CM

m

OV

OV

CM

•

•

•

•

•

•

•

rH

00

00

00

Cv

Cv

CO

00

i

o

vO

co

00

CO

o

VO

rH

t

•

•

•

•

•

•

rH

r-

00

00

00

ov

Ov

00

,4

$

•

<

>

4J

•0

T3

£

Xi

Ov

c

j

ffl

c

5H

HJ

•SJ

■U

>

S

c

H

CN

CO

m

vO

<

00

VO

co

in

IN

e'-

en

\

&

56

57

TABLE 11

TOTAL CONTACTS FOR EACH CATEGORY

Females

Males

Approach

1,200

1,414

Neck Arching

417

259

Tail Vibration

619

757

Touch

949

949

Bite*

685

1,283

Total Aggression

3,870

4,662

Tail -waving

160

152

Movements Away

1,045

1,057

Retreat

191

377

Total Submission

1,396

1,586

Unclassified

1,812

1,199

TOTAL CONTACTS

7,078

7,447

*Difference is

significant to

5 per

cent level

(Mann-

Whitney U-Test) .

58

?

(N

3

3

0>

G T-I
C >1

S*

m

in

in

m

O ■—

•

•

*

•

CO

CO

00

CO

00

in

CM

£ ra
1 «

H

CO

H

CO

rl

CO

CM

2

o

4J

•H

«

€

g

a

■H 4J
18 18
IH M

t

0>

«a

<u ^

aa

(8

O

&

Cu

<

co

vO

00

CM

in

<T>

CO

co

m

VO

in

VO

CM

in

oo

m

00

m

o

r- 1

in co

CM

ov

vo

00

in

oo

CO

vo

CM

OV

CM

CV

CO

00

o

H

m

vo

lO CO CV CM

m

oo

CM

CO

H

m

vo

H

00

r-4

<n

o

CM

59

w

tn

a 7~
-H «

j?

2

»W (A
C —

e (o

w c

s

<u

■p

a

CM

co

in

00

ID

01

00

ro

O

in

O)

in

CM

CM

|

i

CM

ro

o

ro

ro

in

CM

m

»

Cl

co

vD

00

GQ

I

CM

9

a

o

frl

c

0

•H +»

« fl

s

CN

00

in

ro

ID

Ol

ID

<n

ro

00

in

CM

X S'

82
a P

vo

m

CO

CM

ro

in is

in

t

oo

o

nj

o

u

8*

»

ro

00

vO

in

o»

in

in

OV

CM

•O

M

rd

N

co

CM

ro

ro

ro

in

ro

vo

ro

r*

ro

00

ro

Ov

ro

O

*

21.5

60

TABLE 13

RANKINGS FOR EACH CATEGORY OF SUBMISSION
AND FOR TOTAL SUBMISSION

Lizard

Tall

Waving

Movement _ ,

_ Retreat

Away

Sum of
Ranks (Rj)

Ranking
by Rj

A.

Females

11

8

8

4

20

7

12

2.5

3.5

6

12

3

13

9

6

8

23

8

14

2.5

5

5

12.5

4

15

7

9

9.5

25.5

9

16

1

3.5

1

5.5

1

17

5.5

1

7

13.5

6

18

5.5

2

3

10.5

2

19

4

7

2

13

5

20

10

10

9.5

29.5

10

B.

Males

31

9.5

10

10

29.5

10

32

7

9

9

25

9

33

5

3

3

11

3.5

34

3

8

8

19

7

35

9.5

7

7

23.5

8

36

1

1

1

3

1

37

4

5

2

11

3.5

38

8

6

4

18

6

39

6

4

6

16

5

40

2

2

5

9

2

61

TABLE 14

RANKINGS FOR UNCLASSIFIED CATEGORY

Females

Males

Lizard

Rank

Lizard

Rank

11

5

31

6

12

4

32

9

13

3

33

5

14

8

34

3

15

2

35

8

16

10

36

2

17

9

37

7

18

6

38

1

19

7

39

10

20

1

40

4

62

TABLE 15

KENDALL COEFFICIENT OF CONCORDANCE (W)
FOR RANKINGS AMONG AGGRESSIVE

AND

SUBMISSIVE CATEGORIES

W

P<

A. Females

Aqqression

Set 1

.871

.001

Set 2

.713

.001

Set 1 +

Set

2

.849

.001

Submission

Set 1

.701

.05

Set 2

.745

.02

Set 1 +

Set

2

.695

.05

B. Males

Aqqression

Set 1

.560

.01

Set 2

.557

.01

Set 1 +

Set

2

.522

.01

Submission

Set 1

.792

.02

Set 2

.659

.05

Set 1 +

Set

2

.806

.01

63

TABLE 16

CORRELATIONS OF RANKINGS* IN AGGRESSION

AND SUBMISSION

BETWEEN

SETS

Rank

Correlation

P<

A. Females

Aggression :

Set 1/Set 2

.730

.05

Submissions

Set 1/Set 2

.564

.05

B. Males

Aggression:

Set 1/Set 2

.712

.05

Submission:

Set 1/Set 2

.885

.01

* Rankings based on sum (Rj ) of rankings in individual
categories .

64

TABLE 17

FREQUENCY OF ALL CONTACTS COMPARED BY SETS

Females

Males

Set 1

Set 2

Set 1

Set 2

Approach

712

488

633

781

Neck Arching

265*

152*

100

159

Tail Vibration

383

236

401

356

Touch

484

465

385

564

Bite

502*

183*

373

546

Totals

2,346

1,524

2,256

2,406

Tail-waving

120*

40*

52

100

Movement Away

558

487

513

544

Retreat

139

52

148

229

Totals

817

579

713

873

Unclassified

657*

1,155*

638

561

*Signif leant difference (5 per cent level or better)
for Mann-Whitney U-Test.

65

TABLE 18

CORRELATIONS OP RANKINGS* BETWEEN
AGGRESSION AND SUBMISSION

Rank Correlation

P<

A. Females

Aggression/Submission

-.782

.01

B. Males

Aggression/Submission

-.645

.05

* Rankings based on sum (Rj ) of rankings of individual
categories .

66

TABLE 19

RANK CORRELATION BETWEEN RANKS FOR TOTAL*
AGGRESSION AND SUBMISSION AND RANKS
FOR EACH UNIT OF AGGRESSION
AND SUBMISSION

Females

Males

P<

rs

P<

Aggr es s ion/Approach

.976

.01

.888

.01

Aggression/Neck Arching

.855

.01

.627

.05

Aggression/Tail Vibration

.842

.01

.576

.05

Aggr es s ion/Touch

.939

.01

.806

.01

Aggression/Bite

.952

.01

.664

.05

Submiss ion/Tail-waving

.867

.01

.818

.01

Submission/Movement Away

.748

.01

.967

.01

Subraiss ion/Retr eat

.815

.01

.888

.01

*Based on sum (Rj ) of rankings for individual cate-

gories

67

TABLE 20

CORRELATIONS OF RANKINGS FOR UNCLASSIFIED CATEGORY
WITH ALL OTHER CATEGORIES AND WITH TOTAL*
AGGRESSION AND SUBMISSION

Females

Males

rs

P<

rs

P

Approach/Uhclas s if ied

.927

.01

-.064

**

Neck Arching/Unclassified

.673

.05

-.009

**

Tail Vibration/Unclassified

.588

.05

.261

**

Touch/Unclassified

.976

.01

-.152

**

Bite/Unclassified

.842

.01

-.409

Total Aggression/

Unclassified

.891

.01

-.139

**

Tail -Waving/unclassi f ied

-.733

.05

,409

**

Movement Away/Unclassified

-.694

.05

.297

**

Retr eat/Unclass i f ied

-.736

.05

.406

**

Total Submission/

Unclassified

-.733

.05

.385

**

* Based on sum (Rj) of rankings for individual cate-
gories .

**

Not significant

63

TABLE 21

CORRELATIONS OF AGGRESSIVE AND SUBMISSIVE CATEGORIES

WITH EACH OTHER

Females

Males

Corr.

*<

Corr.

P<

Approach/Neck Arching

.855

.01

.336

*

Approach/Tail Vibration

.782

.01

.406

*

Approach/Touch

.939

.01

.830

.01

Approach/Bite

.897

.01

.573

.05

Neck Arching/Tail Vibration

.758

.01

.355

*

Neck Arching/Touch

.745

.01

.273

*

Neck Arching/Bite

.764

.01

.445

*

Tail Vibration/Touch

.673

.05

.479

*

Tail Vibration/Bite

.782

.01

-.155

*

Touch/Bite

.927

.01

.482

*

Tail -waving/Movement Away

.586

.05

.706

.05

Ta il -waving/Retreat

.670

.05

.582

.05

Movement Away/Retreat

.434

*

.843

.01

*Not significant.

69

TABLE 22

ACTIVITY RANKINGS* OF ISOLATED FEMALES

Lizard

Ranking

11

2

12

10

13

3

14

5

15

1

16

7

17

8

18

6

19

9

20

4

*Based on frequency of movement into marked quadrants
during two fifteen -minute observation periods.

TABLE 23

DISTRIBUTION OF BITES ACCORDING TO
BODY REGION

Head

Body

Tail

Totals

Females

46

126

329

501

Males

144

446

679

1,269

Totals

190

572

1,008

1,770

DISCUSSION

In this study the author has attempted to analyze
the phenomenon of aggression by indicating different levels
of intensities of aggressive activity, different expres-
sions of aggression, and individual differences in aggres-
sive behavior. These and other aspects of the relation-
ships among the experimental animals are emphasized in the
following discussion of the individual tables in which the
data are presented.

Tables 1 through 9 show the summaries of all individ-
ual contacts in aggression and submission. In each table,
the lizards are listed from highest to lowest frequency in
the performance of the particular activity. This sequence
occurs from top to bottom and from left to right. This sort
of table gives a picture of differential aggression and sub-
mission since it also shows the number of contacts received
by each animal. A comparison of the totals delivered (bot-
tom row) with the totals received (right-hand column) , as a
row-column rank correlation, should show a negative correla-
tion in aggressive activity is not a random occurrence.
Theoretically, aggressive-submissive contacts could range

- 70

71 -

from complete randomness to a one-sided hierarchy in which
no individual ever shows any aggression toward one ranking
above him. In such straight-line hierarchies as that shown
by the domestic fowl (Guhl , 1945; Guhl and Allee, 1944) and
the jungle fowl (Banks , 1956), the upper right half of
this type of chart would have mostly zeros with only a few
numbers indicating reversals or peck-order violations. In
random aggressive activity one would expect to find high
numbers scattered randomly over the table. The relation-
ships of these skinks are obviously neither of these ex-
tremes.

Inspection of Tables 1 through 9 shows that most
of the higher numbers are concentrated in the lower left-
hand part of the table; however, inspection of the upper
right shows many instances of lower ranking animals deliv-
ering more contacts than are received from higher ranking
animals. These exceptions and variations in aggression
among the higher ranking individuals are apparently suf-
ficient to prevent all row-column rank correlations from
being significant (and one. Table ?A, is even positive).
Those categories showing a significant negative correlation
are approach (male and female) , neck arching (female) ,
tail waving (female) , movement away (male and female) *

- 72 -

retreat (female) , and unclassified (female) . It appears
from these data that not all categories are equally repre-
sentative of differential aggression, nor do males and
females show differences in the same way. This is sup-
ported by a consideration of the data presented in the
remaining tables.

In many papers dealing with social hierarchies or
dominance-subordinance situations (Schjelderup-Ebbe, 1935?
Collias, 1943; Uhrich, 1938? Ginsburg and Allee, 1942?
Maslow, 1940? Crawford, 1942? Evans, 1936; Braddock,

1945) , the authors do not adequately define the hierarchy
or peck-order which they describe or discuss. For instance,
they do not suggest where, between the two extremes, random
aggression (if such a thing exists) and the rigid com-
pletely one-sided type of aggression, the peck-order be-
gins. Should there be an arbitrary line drawn in this ap-
parent continuum of aggressive activity? Masure and
Allee (1934) apparently drew one line when they distin-
guished between the "peck-right" and the "peck- dominance "
type of hierarchy, though the line is not a sharp one.

In their peck-right type the subordinate birds do not gen-
erally peck birds above them in the peck order, whereas
in the peck-dominance type there is a more or less asym-

73

metric peck exchange. Tinbergen (1953) uses a criterion
of learning in which each animal learns "its place," thus
reducing the amount of actual fighting. Haubrich (1961)
implies that a significant correlation between totals
given and totals received (row-column rank correlation)
relates his data to the peck order. If the last standard
is used in this work, then certain categories strongly
resemble the hierarchical relationship while others are
much less clear. However, it is probably best not to
attempt to establish too rigid a set of criteria for the
peck order; and, because of the many differences in ecology,
types and levels of aggression, and possibly in origin of
aggression, it may be more practicable to judge each case by
slightly different standards. This situation is somewhat
analogous to recent discussion (Van Valen, 1960; Reed,

1960; Simpson, 1960) as to whether or not certain extinct
animals should be considered mammal-like reptiles or reptile-
like mammals. The definitions are not as important as the
existence of enough information to generate discussion and
disagreement.

In this study then, on the basis of Tables 1
through 9, it is possible to say that certain expressions
of aggression and submis sion present a pattern which

74 -

resembles social hierarchies in some animals. This pattern,
where present, is best described as a weak peck-dominance
type of hierarchy, which was present under the conditions
of these experiments. As Lorenz (1935) suggests for such
captive groups, there should be no implication of its exist-
ence in nature. Further comment concerning the individual
categories as expressions of aggression and submission will
be made in the discussions of the remaining tables.

Table 11 shows the total contacts for each category
for each sex. These totals are surprisingly similar. The
only category which shows a significant difference in fre-
quency between the sexes is the bite category. The fact
that this category shows the one significant difference is
in accord with the impressions of the observer during the
observation period. Biting among the males was more fre-
quent and more vicious than among the females. It termin-
ated the contact among females much more quickly than
among males, the bites of which frequently occurred in rapid
exchanges amid flurries of activity. Since male lizards
are generally considered to be the more aggressive, the
more belligerent, and the more active of the sexes (Fitch,
1954; Greenberg and Noble, 1944), it might be expected that

- 75

this ultimate in aggressive behavior, the bite, should
show the greatest difference. However, it is somewhat sur-
prising that none of the other aggressive categories shows
a significant difference and the question might be raised
as to whether they represent lower levels of aggression.

Table 11 also shows the relative frequency for each
aggressive category. When looked at from this view, the
table shows the order of categories from the most to the
least frequent as follows!

Females; Approach, Touch, Bite, Tail Vibration,

Neck Arching

Males; Approach, Bite, Touch, Tail Vibration,

Neck Arching

The only difference in the two orders is brought about by
the much-increased frequency of bites among males. Can
this list, then, represent an arrangement by levels of
aggression? It is the belief of the writer that, with cer-
tain reservations, it does. As stated above, the bite is
the most severe form of aggression among these animals.

Since the touch is an actual physical contact (and was
frequently followed by a bite) that almost always elicited
a response by the other lizard, it should be considered the
next most severe level. Thus, of the first three, the ap-
proach is the lowest level of aggression and occurs with the

- 76 -

greatest frequency in both sexes. The position of the
other two categories is not so clear. They occur with
less frequency than the bite, but they certainly are not
higher levels of aggression than the bite. The question
arises as to whether or not one or both of these represent
a challenge in this animal. This will be considered else-
where.

It is also seen in Table 11 that the three sub-
missive categories occur in the same order of frequency
for both sexes. The greatest apparent difference occurs
in the retreat category. The greater number of retreats
on the part of males can best be explained by the greater
aggressiveness and intensity expressed in their contacts.

Tables 12, 13, and 14 summarize the rankings for
all categories of aggression and submission and the un-
classified category. Tables 12 and 13 also show the sum
of ranks (Rj) for aggression and submission and the ranking
based on this sum. It is the final ranking based on the
is referred to as “total aggression" or "total
submission" rankings respectively in this paper.

The sum of ranks (Rj) is used in the procedure for
calculating the Kendall coefficient of concordance (W)
which is shown in Table 15. There was no significant change

77

in W between Sets 1 and 2 as these and the W for the total
observations (Set 1 plus Set 2) were all significant. The
choice of the Kendall coefficient of concordance (W) here
is based on the occurrence of more than two sets of rankings
in both aggressive and submissive categories and its use-
fulness as a measure of the relation among these several
sets of rankings. It merely indicates in this case that
there is a correlation among the several rankings but shows
nothing concerning what is responsible for the correlation.

It is the conclusion of the observer that the element which
is common to the group of five categories is aggression and
that the element which is common to the group of three
categories is submission. It was obvious during observations
that such things as bites, touches, and approaches were fre-
quently stimulating submissive responses. The role of the
tail vibration and the neck arching was much less obvious
to the observer . Since these two categories contribute
two-fifths of the data to the computation of W. , it seems
reasonable to conclude that there are definitely some ele-
ments of aggression involved. This conclusion agrees with
the original observations and assumptions upon which the
observer based his recognition of supposedly aggressive

actions

- 78 -

The information in Table 16 indicates that there
was no significant shift in rankings for aggression or for
submission between Set 1 and Set 2. This simply means that
the lizards, and possibly the observer as well, were con-
sistent in their behavior for these two sets, which were
separated by a two-week period. Although this indicates
that the relationships were the same in the two sets, it
still does not show whether the consistency of the two sets
was due to learning and individual recognition which carried
over to the second or Whether it was due to individual dif-
ferences tdiich might have prevailed regardless of the
length of separation. Table 17 was prepared as an effort
to shed some light on this. It shows the frequency of each
category of aggression and submission and the unclassified
category for each set. For the males, there is no signifi-
cant difference in frequency and all variations are not in
the same direction. For the females, all variations except
in the unclassified category are in the same direction,
namely, a decrease. Furthermore, three of these and the
unclassified are significantly different. This suggests
that some learning or individual recognition did occur.

As indicated above, the position of neck arching as an
aggressive category is not clear, and this is true of its
significance here. It will be discussed elsewhere.

79

Another category showing a significant reduction
in frequency in the second set is the bite. In submissive
categories, tail waving is the only one to show such a
reduction. If the submissive animals of the pair matings
recognized certain aggressive animals and/or learned from
their prior experiences, this could explain earlier
responses to lower level contacts which would thus reduce
the number of bites. It might also reduce the frequency
of tail waving, since this seems to be a response of
lower-level than movement away or retreat. Since the dif-
ferences in the above categories show statistical signifi-
cance, this gives strong support to the suggestion that
at least some learning or individual recognition did
occur. But even in the figures that lack statistical
significance, there is further evidence for this idea.

Every category (except unclassified) has a reduced frequency
in the second set. We would not expect such uniformity of
direction if the frequencies were randomly distributed, so
prior experience must be a factor here. There was no such
uniformity among the males; in fact, most of the categories
show an increase in frequency. There is no indication in
these figures that the males show learning or recognition;
however, the greater aggressiveness, pugnacity, and activity

- 80

and possibly the somewhat looser relationship among males
could conceal this.

The striking increase in the unclassified category
among females is difficult to explain. It is possible
that the energies of the lizards were in part taken up
with random movements which could not be called aggression
or submission or that because of prior experiences of the
lizards, there were included in unclassified contacts as-
pects of aggression and submission too subtle for the
observer to recognize. This relationship was not the salt®
among the males, as they showed an insignificant decrease
among unclassified contacts.

Whereas this study has been concerned with the na-
ture of aggression, it does not diminish the importance of
the submissive categories, which have a distinct reality
by themselves. Although the observer may have used each
type (aggressive or submissive) of category as an aid in
the recognition of the other, ranking by total submission
could stand alone as an indicator of aggression among these
animals. This is true because rankings by total submission
have a significant negative correlation with rankings by
total aggression, as shown in Table 18. These submissive
categories may in some ways be even more realistic than

- 81

aggressive categories, for they indicate a recognition of
aggression on the part of the lizard. What the animal
actually recognizes may not be visible to the observer,
but the submissive reaction of the animal is.

The results seen in Table 19 were compiled in an
effort to see if certain categories of aggression or sub-
mission contributed a disproportionate share to the rankings
for total aggression or total submission. Thus, we have
a comparison of the whole with each of its parts. All
correlations, of course, are significant, but it should

be noted that the three highest correlations in aggression

\

for each sex are with approach, touch, and bite. This is
consistent with other data presented and with the general
impressions of the observer concerning the relationship
of these three aggressive categories and the other two.

The rank-relationship of the unclassified category
with all other categories and total aggression and submission
is shown in Table 20. For the females, every comparison
with aggression has a significant positive correlation,
while every comparison with submission has a significant
negative correlation. For the males, the direction of cor-
relation is reversed in all but one category, but none is
significant. If the unclassified is unbiased and truly

82

represents no aggression or submission, one would expect no
pattern of correlation with either group of categories. But
with the females the pattern is as if the unclassified is an
aggressive category itself. One explanation for this is that
the observer, in his conservative appraisal of pair-mate con-
tacts, included in unclassified much that should have been
called aggression. If this is true, then the absence of such
a correlation with the males indicates that the observer
"learned’' to discriminate between these contacts better after
112-1/4 hours of observations with the females. Whereas this
might have reduced the frequency of aggressive contacts re-
corded for the females, the relationship among the categories
remains the same, as Table 20 indicates for the females.

When the Kendall coefficient of concordance (W) was
used (Table 15) , there was a significant correlation shown by
each sex as discussed above. However, Table 21, showing the
correlations of individual categories with each other, pre-
sents a different picture, especially in aggression. In
females, every aggressive category is significantly corre-
lated with every other, while among the males, only two such
correlations are significant, and the average for all is not
significant. This was at first quite puzzling, especially
since w is supposed to show essentially the same thing
(Siegel, 1956, p. 229) as the average of all paired cor-

relations. However, an analysis of the rankings of indi-
vidual animals in each pair of categories reveals a prob-
able answer to this puzzle. This analysis shows that
three animals differ in their paired rankings by five or
more positions more frequently than any other animals.
Lizards No. 31, No. 36, and No. 40 shifted five or more
positions in ranking four, six, and six times respectively.
Such changes in rankings by just a few animals would have
a strong effect on the degree of correlation between pairs
of rankings. Since the technique for calculating the Ken-
dall coefficient of concordance uses the rankings in each
category only one time, as well as the sum of these rank-
ings, the shifts described above are not effective. In
other words, W gives an over-all picture of the several
rankings, which is believed to represent a "continuity" of
effect by the element of aggression. Individual rank
correlations in this case bring out certain individual
variations or exceptions to the over-all picture, which
probably help to reduce somewhat the value of W, There
were no such individual exceptions in aggression among
females. The particular relationship among these lizards
may help to point out a "weakness" in the use and inter-
pretation of W that could be of importance in certain types

of correlations.

84

The smaller number of categories in submission
perhaps prevents a similar picture to that in aggression
in Table 21, or else the animals do not vary as much in
their submissive actions. The only correlation that is
not significant (W is significant for each sex - Table 15)
is that between movement away and retreat for females.

One would expect these two to have the closest relation-
ship. A check of the individual animals reveals that No.

17 and No. 19 change ranks by six and five positions
respectively, and thus contribute enough to the calcula-
tions to prevent a significant correlation. Possibly these
two animals have different "thresholds" in their reactions
to aggression. No. 17, reunking first in movement away,
dropped to seventh in retreats. Inspection of Tables 4A
and 5A reveals that No. 17 ranks second in touches received
and sixth in bites received. This animal may have responded
quickly to touches by moving away and thus generally failed
to receive bites from which it might have retreated. On
the other hand. No. 19 ranks low in all aggressive categories
given or received and high in only one submissive category,
retreat. Therefore, this animal must have been relatively
inactive, but with a quick threshold of response (in the
form of a retreat) to early signs of aggresiion. This de-

/

85

duction is in accord with the observer's own memory of the
individual .

In the discussion above of Tables 11 and 17, refer-
ence was made to the relationship between approach, touch,
and bite on the one hand and neck arching and tail vibration
on the other. There seems to be a consistent difference
between these two groups. An examination of Tables 11 and
17 reveals that the frequence of each member of the first
group is higher than that of each member of the second group.

In Table 19 the correlations involving each member of the
first group are higher than any in the second. In Table 21
those correlations between members of the first group above
are higher than any other correlations. All of these
relationships are true for both sexes. Neck arching and
tail vibration have here been considered as two contributors
(categories) to the over-all picture of aggression. But
might they be more than this? The question arises as
to whether or not one or both of these categories might
represent a challenge in this animal, comparable, for instance,
to the challenges described for Anolis and other Iguanidae
(Evans, 1936; Greenberg and Noble, 1944? Noble and
Bradley, 1933; Petit, 1928). During the observations

there were times when neck arching and tail vibration

86 -

appeared to be "preliminaries" for other aggressive activ-
ity. But this was not consistently so. There were times
when the observer got the impression that both activities,
especially tail vibration, occurred as a "disturbance "
(defense posture?) concomitant to aggressive activity,
or as a reaction (but not submissive) to aggression. For
instance, there were numerous occasions when both animals
vibrated their tails after separation from a particularly
vigorous contact. Neck arching is associated with still
another type of activity in this animal, feeding. It is
a position almost always assumed just before the skink
lunges for an item of food, in the laboratory or in the
field. Many times the lizards arched their necks at each
other as they seemed to be maneuvering for an advantageous
position. This posture seems to be associated with ag-
gression, but is not necessarily involved as a part of a
particular sequence of aggressive activity. Since neither
of these actions consistently occurred before the approach,
touch, or bite, these explanations for tail vibration and
neck arching are better supported by the data than is the
suggestion that one or both are challenges.

When a correlation is made between rankings for
average weight (Table 10) and rankings for over-all

87

aggression (Table 12) , the correlation for females is
found to be .48, which is not significant; the correlation
for males is .69, significant at the 5 per cent level.
Differences in weight in both sexes can be thought of as
representing differences in age or physiological condition.

In the males these differences are considered to be factors
which affect the relationships of the animals in aggression.
This probably is not the case for females. But in the more
vigorously expressed aggressive activities of the males,
it can be said that the heavier lizards tend to be the more
aggressive. Similar conclusions have been reached in studies
of crayfish (Bovjberg, 1956), lizards (Carpenter, 1960;

Evans, 1936; Fitch, 1940), chickens (Schjelderup-Efcbe,

1922) , mice (Ginsburg and Allee, 1942) , and primates
(Maslow, 1940) .

At the conclusion of observations with the females,
each animal was observed for two 15-minute periods while
isolated in the same container used for paired observa-
tions. The bottom of the container had been marked into
four equal quadrants. Every move made by a lizard from
one quadrant to another was recorded. The rankings of the
females from the greatest to the fewest number of moves
is shown in Table 22. The correlation of this ranking

88

with that for over-all aggression is .697, which is signifi-
cant to the 5 per cent level. This indicates that in the
females the most active animals tend to be the most aggres-
sive. This is further evidence that the physiological
condition of the individual animal affects the degree of
aggressiveness which it shows. It was the opinion of the
observer that the same thing would be true for the males,
but unfortunately unavoidable circumstances prevented the
conducting of such experiments with the males.

At the conclusion of all observations with the males
the observer noted some subjective impressions of individual
animals. These lizards had such distinct behavioral
differences that, after the many hours of observations,
each appeared to have its own "personality" (Schjelderup-
Ebbe, 1953? Pardi, 1946). This subject characterization
of the male lizards is shown below as recorded by animal
number.

31. Occasionally spends fairly long periods
in corners, but quite active when moving.
Usually follows approach and touch with a
quick direct bite.

32. Fairly active. Vibrates tail a lot when
moving around and when approached or when
approaching. When contacting another lizard,
has tendency to push and poke rather than
bite. Doesn't bite often and when it does

89

it is usually a tail nip, but responds to
other larger aggressive lizards with a
vigorous bite.

33. Spends most of time in corner, though in
later observation periods was more active.
When approached and/or touched by a lizard
its size or smaller, frequently arched
neck and bit, but never pursued. Always
moved quickly away when approached, touched,
etc., by larger or more aggressive lizard.

34. Very active, but still for long periods only
to burst into action. Frequently bites and
pursues vigorously. Much of time spent in

• moving widely, oblivious of other animals,
frequently crawling right over them.

35. Arches neck frequently. Moves around mod-

erately but stays still for long periods.
Frequently approaches and touches without
biting. Defends self by returning bites
against some lizards but quickly runs from
others. Rarely vibrates tail? rarely bitten
on tail. (Note: this animal had the short-

est tail, having previously lost most of it
and regenerated approximately 46 mm.)

36. Moves around moderate amount. Vibrates
tail frequently. When pressed, retaliates
with a bite before running. Against sub-
missive individuals is fairly aggresiive
but does not pursue advantages.

37. "The bluffer." Charges close to or into
another animal when approached to within a
few inches from front or side only. Does
not move around much, but more stirred to
action by movements of others. Always runs
when pressed. Only against No. 36 did it
chase, attack, and bite viciously.

38. Large. Moves around constantly seeking
escape. Frequently blunders right by other
lizards. Vibrates tiil frequently while

90 -

moving. Arches head and bites when it fi-
nally notices other lizard or when "provoked."

39. Stays in one spot most of the time, rarely
moving except when forced by dominant animal .
Others, it simply snaps at when they disturb
it too much.

40. Approaches any animal repeatedly, regardless
of how repulsed. Touches with nose but almost
never bites, though occasionally vibrates tail.

In the early stages of observations of these skinks,
it was observed that a large proportion of the bites were
being delivered to the tail rather than the head or body

of the recipient. Pitch (1954) had observed that among

/

fighting males of Eumeces fasciatus most serious wounds were
usually in the sacral region or base of the tail or both."
Hunsaker (1962) refers to tail-biting as ritualized be-
havior in Sceloporus and says that it is about the only
type of body contact occurring during fighting. In these
observations the location of bites was recorded when pos-
sible. The distribution of 1770 bites received, or 89.9
of the total, is shown in Table 23. When these data are
subjected to the chi-square test, with the assumption of
randomness as to location, the distribution is found to
differ significantly (1 per cent level) from random occur-
rence. In other words, a greater number of bites was
received on the tail than might have been expected by

chance .

91

The significance of the above distribution appears
to be related to other behavioral and morphological adapta-
tions of this animal. The survival value of autotomy has
been widely discussed for many groups in the animal kingdom.
One of the submissive traits described for this animal is
tail waving, already described as a slow undulatory twitch-
ing movement. This submissive category can be considered
a behavioral adaptation which is associated with the morpho-
logical one of autotomy, since animals (e.g., predators)
do respond to movement and would be attracted to the moving
part rather than to the more vulnerable head or body. These
two adaptations might be important in intraspecific competi-
tion as well as in predator-prey relationships. Table 23
indicates that biting might also be related to these two.
Although fighting can disrupt the population if it results
in frequent deaths, deaths would not occur if most of the
bites were directed to the tail.

Intraspecific aggression can serve to disrupt a
population. Thus, the adaptive value of strong hierarchies,
which exist in animals such as the jungle fowl (Banks,

1956) and other birds, is in the limitation which they
place on aggressive activity. Since Eumeces inexpectatus
does not have social organization in the field, the

92

disruptive effect of intraspecific strife could be strong.
However, the evidence from this study indicates that this
disruptive effect is reduced not only by the action of the
animal in waving its tail, not only by the fact that the
tail may be "given up" while the animal escapes, but also
by the tendency of aggressive animals to bite the tail
rather than the body or head.

No assertion is made in this study that Bumeces
inexpectatus is a truly social animal. If one accepts
the criteria of Allee (1951) that, "aggressive behavior
can be regarded as asocial insofar as it acts to prevent,
dicrease, or disrupt the integration of vertebrate
societies, and it can be regarded as social insofar as it
furthers the integration and organization of such soci-
eties," then this animal is, in nature, asocial. In fact,
this work, as does Evans' (1936) study of Anolis, emphasizes
anti-social behavior, especially of the male. The nature
of the contacts is such that they tend to disrupt and dis-
perse rather than to integrate.

This study shows that even solitary lizards which
normally display no evidence of territoriality and which,
at the most, show only differential aggression in the
field, can in pair-matings in a confined and limited space

93 -

display aggressive behavior that resembles hierarchies in
some animals. An attempt is made to compare individual
expressions of behavior "which contribute to the behavioral
Whole as well as to the pattern of differential aggression.
It shows that there is much individual and some sexual
variation in the expressions of aggression and that this
variation may be related to a number of different factors.
It indicates that even though aggression may intrinsically
be disruptive in its intraspecffic expressions, it may at
the same time have distinctive adaptive values.

SUMMARY AND CONCLUSIONS

1. The Southeastern five-lined skink, Eumeces
inexpectatus Taylor, displays certain behavioral activities
(neck arching, tail vibration, approach, touch, and bite)
which are interpreted as aggression, and others (tail wav-
ing, movement away, and retreat) which are interpreted as
submission. Another activity (unclassified) was not
identifiable as either aggression or submission.

2. There are no significant differences between
sexes in frequency of contacts except in bites. The males
deliver bites nearly twice as often as do the females in
the same time period.

3. When compared, the rankings within the cate-
gories of aggression and of submission are, respectively,
correlated. There are no significant changes in rankings
between the two sets of observations.

4. There are no significant changes between Sets
1 and 2 in frequencies of contacts among males. Among
females, the frequencies of neck arching, bite, and tail
waving show a significant decrease, while the unclassified
category shows a significant increase. This suggests the

- 94

- 95 -

possibility of learning; however, verification or rejection
of this would require further investigations with both
sexes .

5. For both males and females, comparison of rank-
ings in aggression and submission shows significant correla-
tions, This indicates that the categories reflect a dom-
inance- subordinance r el ationship ,

6, When the ranking for each category of aggression
and submission is correlated with the total ranking, it
appears that no one category contributes a disproportionate
share to the total. The touch, approach, and bite categories
have the highest correlations with the total in each sex.

7, In females the rankings for all aggressive
categories and for all submissive categories have a signif-
icant negative correlation respectively, when compared with
the rankings for the unclassified category. None of these
correlations is significant in the males. This difference
is considered to be due to an inclusion by the observer of
■any unrecognized aggressive contacts in those identified as
unclassified.

8. When rankings between every possible pair of
the aggressive categories are compared, all are found to be
significant among the females, only two among the males.

- 96 -

Of such correlations in submissive categories, only one is
not significant. The exceptions to the significant correla-
tions are shown to be due to the erratic behavior of two or
three animals.

9. It is shown that approach, touch, and bite
have a closer relationship with each other than with neck
arching or tail vibration.

10. The possibility that neck arching and tail
vibration might be considered as challenges is weighted. This
isrrejected in favor of the conclusion that they represent
the agitation and disturbance of the animals When they are
involved in pair-mate contacts.

11. The heavier males tend to be the more aggres-
sive, but this is not true of the females.

12. A ranking based on number of movements by iso-
lated females, when compared with rankings for aggression,
indicates that the more active females tend to rank higher
in aggression.

13. A subjective description of the male lizards
is found to reflect individual differences recognizable
after long periods of observation, and these are compatible
with objective conclusions.

14. A record of the location of bites received

- 97 -

reveals that a greater number than would be expected by
chance is received upon the tail. This is considered to be
a behavioral adaptation related to the morphological adap-
tation of autotoray.

LITERATURE CITED

Allee, W. C. 1951. Cooperation among Animals, with Human
Implications, Schuman, New York. 233 pp.

Allee, W. C. , and J. C. Dickinson. 1954. Dominance and

subordination in the smooth dogfish, Mustelus canis
(Mitchill) . Physiol. Zool., 27(4): 356-364.

Allee, W. C. , D. Foreman, E. M. Banks, and C. H. Holabird.

1955. Effects of an androgen on dominance and sub-
ordinance in six common breeds of Gallus gallus.
Physiol. Zool., 28(2): 89-115.

i i. ■ y, ,.v, -45. ;

Allee, W. C. , B. Greenberg, G. M. Rosenthal, and P. Frank.

1948. Some effects of social organization on growth
in the green sunfish, Lepomis cyanellus. Jour. Exp.
Zool., 108: 1-19.

Andrew, R. J. 1957. Influence of hunger on aggressive be-
havior in certain buntings of the genus Emberiza.
Physiol. Zool., 30(2): 177-185.

Baerends, G. P. 1958. The contribution of ethology to the
study of the causation of behaviour. Acta Physiol.
Pharmacol . Neer landica , 7:466-499.

Baarends, G. P. , and J. M. Baerends-van Roon. 1950. An

introduction to the study of the ethology of cichlid
fishes. Behaviour, Supplement 1: 1-243.

Balph, D. F., and A. W. Stokes. 1959. Notes on the behavior
of deer mice (Peromyscus maniculatus rufinus) . Proc.
Utah Acad . Sci . , Arts and Letters . 37: 55-62.

Banks, E. M. 1956. Social organization in red jungle fowl
hens (Gallus gallus subsp.). Ecology, 37: 239-248.

Banks, E. M., and C. W. Allee. 1957. Some relations between
flock size and agonistic behavior in domestic hens.
Physiol. Zool.. 30(3): 255-268.

98

99 -

Bovjberg, R. V. 1956. Some factors affecting aggressive be-
havior in crayfish. Physiol. Zool., 29s 127-136.

Boyd, H. 1953. On encounters between wild White-Fronted
Geese in winter flocks. Behaviour , 5(2) : 85-129.

Braddock, J. C. 1945. Some aspects of the dominance-

subordination relationship in the fish Platypoecilus
maculatus . Physiol . Zool . , 18(2): 176-195.

. 1949. The effect of prior residence upon dominance

in the fish, Platypoecilus maculatus. Physiol. Zool.,
22: 161-169.

Braddock, J. C. , and Z. I. Braddock. 1955. Aggressive

behavior among females of the Siamese fighting fish,
Betta splendens. Physiol. Zool., 28(2): 152-172.

Brain, Anne D. 1949. Dominance in the great tit Parus major.
Scottish Nat., 61(3): 144-155.

Carpenter, C. C. 1960. Aggressive behaviour and social

dominance in the six-lined race-runner ( Cnemidophorus
sexlineatus) . Animal Behaviour, 8(1/2): 61-66.

Carpenter, C. C. , and G. Grubits III. 1960. Dominance

shifts in the tree lizard (Urosaurus ornatus — Iguani-
dae) . Southwestern Nat., 5(3): 123-128.

Castoro, P. , and A. M. Guhl. 1958. Pairing behavior of
pigeons related to aggressiveness and territory.

Wilson Bull.. 70(1): 57-69.

Chi tty, D. 1952. Mortality among voles (Microtus agrestis)
at Lake vymivy, Montgomeryshire in 1936-9. Philos .
Trans . , B236, pp. 505-552.

Clarke, J. R. 1956. The aggressive behavior of the vole.
Behavior , 9(1): 1-23.

Collias, N. E. 1943. Statistical analysis of factors which
make for success in initial encounters between hens.

Am. Nat.. 77: 519-538.

. 1944. Aggressive behavior among vertebrate animals.

Physiol. Zool.. 17: 83-123.

100

Conant, R. 1958. A Field Guide to Reptiles and Amphibians.
Houghton Mifflin Co. , Boston. 366 pp.

Crawford, M. P. 1942. Dominance and social behavior, for
chimpanzees, in a non-competitive situation. J.

Comp . Psychol . , 33s 267-277.

Darling, F. F. 1938. Bird Flocks and the Breeding Cycle.

The University Press, Cambridge. 124 pp.

Davies, J. L. 1953. Colony size and reproduction in the
grey seal. Proc. Zool. Soc. London, 123(11):

327-332.

Denenberg, V. H. , and R. W. Bell. 1959. Relationship

between social reactions and avoidance conditioning.
Phys . Zool . , 32s 51-56.

Douglis, M. 1948. Social factors influencing the hierarchies
of small flocks of domestic hens The interaction be-
tween resident and part-time members of organized
flocks. Physiol. Zool., 21s 147-182.

Evans, L. T. 1951. Field study of the social behavior of
the black lizard, Ctenosaura pectinata. Amer. Mus.
Novitates, #1493, Mar. 5, 1951: 1-26.

. 1946. social behavior of the lizard, Sceloporus

grammicus microlepidotus . Anat. Rec. , vol. 94, pp.
53-54 (abstract) . 4

. 1936. A study of a social hierarchy in the

lizard, Anolis carolinensis. Jour. Genet. Psychol.,
48: 88-111.

Fitch, H. S. 1940. A field study of the growth and behavior
of the fence lizard, tftiiv. Cal. Publ . Zool., 44(2):
151-172.

. 1954. Life history and ecology of the five-lined

skink, Eumeces fasciatus. Univ. Kan. Publ., Mus.

Nat. Hist., 8(1): 1-56.

Fredericson, E., and E. Ann Birnbaum. 1954. Competitive

fighting between mice with different hereditary back-
grounds. Jour. Genet. Psychol., 85(2): 271-281.

101

Ginsburg, B. , and W. C. Allee. 1942. Some effects of con-
ditioning on social dominance and subordination in
inbred strains of mice. Physiol. Zool.. 15: 485-506.

i Goethe, Friedrich. 1953. Sociale Hierarchie im Aufzuchts-
schwarra der Silbermowen. Zeitschr. Tierpsychol. ,
10(1): 44-50.

Gordon, R. E. 1957. The biology and biodemography of

Anolis carolinensis carolinensis Voigt. Dissert.

Abst. , 17(3): 698.

Grant, W. C. , Jr. 1955. Territorialism in two species of
salamanders . Science, 121(3135): 137.

Greenberg, B. 1947. Some relations between territory,

social hierarchy and leadership in the green sun-
fish, Lepomis cyanellus. Physiol. Zool., 20:

267-299.

Greenberg, B. , and G. K. Noble. 1944. Social behavior of
the American chameleon, Anolis carolinensis Voigt.
Physiol. Zool., 17: 392-439.

Guhl, A. M. 1945. Some observations and experiments on
social behavior in the domestic fowl. Trans .

Kansas Acad. Sci., 47(3): 337-341.

Guhl , A. M. , and W. C. Allee. 1944. Some measureable ef-
fects of social organization in flocks of hens.
Physiol. Zool.. 17(3): 320-347.

Haubrich, R. R. 1961. Hierarchical behavior in the South
African clawed frog, Xenopus laevis Daudin. Animal
Behaviour, 9(1-2): 71-76.

Hunsaker, D. 1962. Ethological isolating mechanisms in

the Sceloporus torcruatus group of lizards. Evolu-
tion, 16: 62-74.

Jenkens, D. W. 1944. Territory as a result of despotism

and social organization in geese. Auk, 61(1): 30-47.

Kilham, Lawrence. 1961. Aggressiveness of migrant myrtle
warblers toward woodpeckers and other birds. Auk,
78(2): 261.

102

Kummer, Hans. 1957. Sociales verhalten einer mantelpavian-
gruppe. Beiheft zur Schweizerischen Zeitschrift fur
Psychol . , 33 : 1-91.

Laskey, Amelia R. 1950. Cowbird behavior. Wilson Bull.,
62(4): 157-174.

Lorenz, K. 1935. Der Kumpan in der Uhwelt des Vogels.

Jour, f. Qrnithol.. 83: 137-213, 289-413.

. 1941. Verglelchenda bewegungsstudien an anatinen.

Jour, f. Qrnithol. , 89 (Festschrift Heinroth) :

194-294.

. 1952. King Solomon* s Ring. Thomas Y. Crowell Co.,

New York. 202 pp.

C. H. , Jr. 1948. Territorial behavior in snakes and
the so-called courtship dance. Herpetologica , 4(4):
129-135.

C. H. , Jr., and K. S. Norris. 1950. Aggressive be-
havior in male sidewinders Crotalus cerastes, with
a discussion of aggressive behavior and territoriality
in snakes . Chicago Acad. Sci. Nat. Hist. Misc. , 66 :
1-13.

C. H. , and W. H. Woodin III. 1954. Aggressive Be-
havior in Phrynosoma. Herpetologica , 10(1): 48.

Bertha. 1960. Fighting and an incipient notion of
territory in male tree frogs. Copeia , 1960(1):

61-63.

Martof, B. S. 1953. Territoriality in the green frog, Rana=
clamitans . Ecology, 34: 165-174.

Maslow, A. H. 1940. Dominance quality and social behavior
in infra-human primates . J. Soc. Psychol. , 11 :
313-324.

Masure, R. H. , and W. C. Allee. 1934. The social order in

flocks of the common chicken and the pigeon. Auk. 51:
306-327.

Lowe,

Lowe,

Lowe,

Lutz,

103

McHugh, T. 1958. Social behavior of the American buffalo
(Bison bison bison) . Zoologica (N.Y. ), 43(1):

1-40.

Miller, Malcolm R. 1951. Some aspects of the life history
of the Yucca night lizard, Xantusia vigilis. Copeia.
1951(2): 114-120.

Miller, R. E., J. V. Murphy and I. A. Mirsky. 1955. Modifi-
cation of social dominance in a group of monkeys by
interanimal conditioning . Jour. Comp, and Physiol.
Psychol. . 48(5): 392-396.

Newman, M. A. 1956. Social behavior and interspecific

competition in two trout species . Physiol . Zool . ,

29: 64-81.

Noble, G. K. , and H. T. Bradley. 1933. The mating behavior
of lizards; its bearing on the theory of sexual se-
lection. Ann. New York Acad. Sci.. 35 (Art. 2):
25-100.

Pardi, L. 1946. Sui fenomani di ,'dominazione', nelle societa
animali. Hist. Nat. (Rome), 1(1): 9-16.

Parrish, F. K. 1958. Miscellaneous observations on the be-
havior of captive sea turtles. Bull. Marine Sci.

Gulf and Caribbean. 8(4): 348-355.

Pearson, P. G. 1955. Population ecology of the spadefoot
toad, Scaphiopus h. holbrooki (Harlan) . Ecological
Monographs . 25: 233-267.

Petit, G. 1928. Sur le Chalarodon madagascariensis Peters.
Bull. Soc. Zool. de France, 53: 401-405.

Potter, J. H. 1949. Dominance relations between different
breeds of domestic hens. Physiol . Zool . , 22(3):
261-280.

Raney, E. C., and R. A. Josephson. 1954. Record of combat
in the snapping turtle, Chelydra serpentina. Copeia ,
1954(3): 228.

104

Reed, C. A. 1960. Polyphyletic or monophyletic ancestry

of mammals, or: what is a class? Evolution. 14(3):

314-322.

Ruibal, Rodolfo. 1960. Inter- and intraspecific behavior of
lizards. Anat . Rec . . 137(3): 390 (abstract).

Sabine, W. s. 1949. Dominance in winter flocks of j uncos
and tree sparrows. Physiol. Zool., 22: 64-85.

Schein, M. W. , and M. H. Fohrman. 1955. Social dominance
relationships in a herd of dairy cattle. Brit. J.
of Animal Behavior, 3(2): 45-55.

Schjelderup-Ebbe, T. 1913. Honsenes stemme fidrag til
Honsenes psykologi naturen, 37: 262-276.

. 1922. Beitrage zur Socialpsychologie des Haus-

huhns. Zeitschr. f. Psychol., 88: 225-252.

. 1935. Social behavior of birds? cited in G. A.

Murchinson, A Handbook of Social Psychology. Wbr-
chester, pp. 947-972.

Scott, J. P. 1948. Dominance and the frustration-aggression
hypothesis. Phvs . Zool . , 21: 31-39.

Scott, J. P. , and E. Fredericson. 1951. The causes of
fighting in mice and rats . Physiol . Zool . , 24 :
273-309.

Siegel, S. 1956. Nonparametric Statistics for the Behavioral
Sciences. McGraw-Hill Book Co., Inc., New York.

312 pp.

Simpson, G. G. 1960. Diagnosis of the classes Reptilia and
Mammalia. Evolution, 14(3) : 388-392.

Stebbins, R. C. 1944. Field notes on a lizard, the mountain
swift, with special reference to territorial behavior.
Ecology , 25: 233-245.

Stringer, G. E. , and W. S. Hoar. 1955. Aggressive behavior
of the underyearling Kamloops trout. Canadian J.

Zool. . 33(3): 148-160.

105

Taylor, E. H. 1932. Eumeces inexpectatus : a new American

lizard of the family Scincidae. Univ. Kans. Sci.

Bull. , XX, 1932 t 251-258.

Test, P. H. 1954. Social aggressiveness in an amphibian.
Science, 120(3108): 140.

Thompson , W. L. 1960. Agonistic behavior in the house

finch. II: Factors in aggressiveness and sociality.

Condor, 62(5): 378-402.

Tinbergen, N. 1942. An ob j ectivistic study of the innate
behaviour of animals. Series D. Bibliothica Bio-
theoretica. V. 1. part 2. E. J. Brill, London,
pp. 39-98.

\J . 1948. Social releasers and the experimental

method required for their study. Wil. on Bull. ,

60: 6-52.

\/ . 1953. Social Behaviour in Animals, with Special

References to Vertebrates. Methuen and Co. , Ltd. ,
London. John Wiley and Sons, Inc., New York. 150 pp.

Uhrich, Jacob. 1938. The social hierarchy in albino mice.
Jour. Comp. Psychol., 25: 373-413.

Van Valen, Leigh. 1960. Therapsids as mammals. Evolution,
14(3): 304-313.

Wharton, C. H. 1958. The ecology of the cottonmouths

Agkistrodon piscirorus piscivorus Lacepede of Sea
Horse Key, Florida. Doctoral Dissertation, Univer-
sity of Florida, Gainesville.

Winn, H. E. 1958. Comparative reproductive behavior and
ecology of fourteen species of darters (Pisces-
Percidae) . Ecological Monographs, 28: 155-191.

BIOGRAPHY

Cleo Duke Wilder, Jr., was bom in Macon, Georgia,
on September 24, 1925. He spent his public school years in
Paris, Tennessee, graduating from high school in 1943.

After two semesters at Tulane university, he was
drafted into the United States Naval Reserve in April, 1944.
In June, 1944, he was assigned to the V-12 officers' training
program at the university of North Carolina. He received his
discharge and commission as Ensign, USNR, in June, 1946.

He received his Bachelor of Arts in English from
the University of North Carolina in June, 1948, and his
Master of Science in zoology from the university of Tennes-
see in August, 1951.

After serving as Instructur in Biology at Presby-
terian College from September, 1951, until August, 1953,
he entered the Graduate School of the university of Florida
where he held an assistantship in the Florida State Museum
and several graduate and teaching assistantships in the
Department of Biology and C-6 program. He held a fellow-
ship in the School of Arts and Sciences in the summer of
1956.

- 106

107

In the year 1957-58, he was employed as Sixth Grade
Teacher, Alachua Elementary School, Alachua, Florida. He was
Assistant Professor of Biology, Memphis State University,
from September, 1959, until June, 1962. He accepted an

• .* • * / ■ t

appointment as Assistant Professor of Zoology at Virginia
Polytechnic Institute effective September, 1962.

In 1950 he married the former Betty Rhea and they

t '

have two sons.

Cleo Duke Wilder, Jr., is a member of Sigma Xi, Phi

/ • i • •

Sigma, American Society of Ichthyologists and Herpetologists,
Herpetologists' League, American Society of Mammalogists ,
Society for the Study of Evolution, The Wilderness Society,
Tennessee Academy of Science, Association of Southeastern
Biologists, and American Association of university Professors.

This thesis was prepared under the direction of the

chairman of the candidate's supervisory committee and has
been approved by all members of that committee. It was sub-
mitted to the Dean of the College of Arts and Sciences and to
the Graduate Council, and was approved as partial fulfillment
of the requirements for the degree of Doctor of Philosophy.

June, 1962

Dean, College of Arts an!
Sciences

Dean, Graduate School

SUPERVISORY COMMITTEES