zuzunmmm .
x 3.!“
”n53

at: 2.:
, 18 1

gummy. ..

5.... a...

”11!
Qty!
33.11:!
I 2....
‘ n... I.»
K. I 33

.2. 3.2.9.1
bl! 33:35:... ,-
Sillzsltonxlt iffﬂi‘
I :Laix-ilv s: .
.19.? “st-1:535
. 12?]: .l: n
:.. o. . .3» a...
in 57......le
313.433,:le
.ﬁiﬁuuk .ﬂﬂztf
, . , I. ‘26,. a .,
. . Nun-Sis
h»Ki::..hﬂnvl,soan In .. 2 hr»: i... 1...:
It is .31. 033"} A.

5....- a.)
£319,119 «- in!

r - 3....
\z 3 :3.

15.36...

3.1.6....) ..
n. In"...

3:32..
‘%I!

l‘ex‘t
.. . 5?

.1) $1.. 0.1:...
it
, tn .

. . (I);
.2) 9.3.. a. 3.6.3.917 X1. :2.
1.»: :5 (<1.

 

|
II

F,

 

i,

MMLI‘J. f
3.14.!“ :3... 4. . . .

$11.53!;

A; ﬂwwwwﬁﬂﬁﬁé .

.. 3.3.

, 1 ‘ ‘ ll‘ , J . . «rimmtﬁaunwn c
as}... .Q . CMAuhﬂnﬂid‘lwh CRIN; , : .l...».v

7601/

This is to certify that the
dissertation entitled
The causes and consequences of

social rank in the
spotted hyena, Crocuta crocuta

 

presented by

Anne Linnea Engh

has been accepted towards fulﬁllment
of the requirements for

Doctoral degreein Zoology

 

 

    
 

Major professor

ay E. Holekamp
DateJ\ (1ng m1

MSU i: an Afﬁrmative Action/Equal Opportunity Institution 0- 12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE ' DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DatoDue.pss-p.15

THE CAUSES AND CONSEQUENCES OF SOCIAL RANK IN THE SPOTTED
HYENA, CROCUTA CROCUTA

By

Anne Linnea Engh

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Zoology
Program in Ecology, Evolutionary Biology and Behavior

2002

ABSTRACT

THE CAUSES AND CONSEQUENCES OF SOCIAL RANK IN THE SPOTTED
HYENA, CROCUTA CROCUTA

By

Anne Linnea Engh

Social rank can have a profound impact on an individual’s life. Here, I
examined how spotted hyenas attain their ranks, and how the hyenas’ rank
system inﬂuences social cognition, male reproductive success, and parasite
loads. I began by testing predictions of four mechanistic hypotheses proposed to
explain maternal rank “inheritance”, a phenomenon observed in both hyenas and
monkeys. In hyenas, neither genetic superiority nor aggression from adult
females plays a strong role in rank “inheritance”. Maternal interventions and
supportive coalitions appeared to reinforce aggression directed at appropriate
conspeciﬁc targets, whereas coalitionary aggression directed at cubs apparently
functioned to extinguish their aggressive behavior towards inappropriate targets.
Young hyenas and primates thus appear to “inherit” their mothers' ranks by
strikingly similar mechanisms.

Given the similarities between societies of hyenas and primates, I next
tested a prediction of the social complexity hypothesis, that animals living in
similar social environments should possess similar abilities in the realm of social
cognition. Speciﬁcally, I used two tests to determine whether hyenas can
recognize third-party rank relationships. I found that young hyenas in the

presence of food were able to exploit the arrival of potential allies by increasing

rates of aggression toward lower-ranking animals. In addition, hyenas joined in
on-going ﬁghts in support of the dominant contestant, even when that animal was
losing. These lines of evidence suggest that hyenas can recognize relative rank
relationships and strongly support the idea that social complexity drives the
evolution of intelligence in phylogenetically disparate taxa.

In many mammalian societies structured by rank hierarchies, male
reproductive success (RS) is positively correlated with social rank. Hyenas are
unusual in that adult females are dominant to males, and there is relatively little
ﬁghting among males. Using genetic markers, I tested whether R8 was related to
social rank or dispersal status among male hyenas. Dispersal status and length
of residence were the strongest determinants of RS. Immigrant males sired most
cubs, despite the fact that immigrants are socially subordinate to all adult natal
males. High-ranking immigrants did not monopolize reproduction, and tenure
accounted for more of the variance in male reproductive success than did rank.
The lack of reproductive skew among male hyenas may be a result of female
choice of mates, rather than male-male contest competition.

Finally, I examined whether several social or demographic variables were
related to parasite burden in hyenas. For several species of parasite, burden was
higher in higher-ranking hyenas, probably because high-ranking individuals were

exposed to more parasite eggs at kills than were low-ranking individuals.

ACKNOWLEDGEMENTS

I received help from many people both in Kenya and in the US while
writing this dissertation, and I am grateful to them all for their help and
encouragement. In particular, I’d like to thank my advisor, Kay Holekamp, for
giving me the opportunity to live and work in the Masai Mara for 2 years. Kay has
been a thoughtful and generous mentor, always willing to share her wisdom and
experience anytime that I needed help or advice. I consider myself very lucky to
have worked with her, and much of the credit for this dissertation should go to
Kay. Committee members Laura Smale, Fred Dyer, and Heather Eisthen
supplied me with useful insights and comments, and I thank them as well.

My friends and family have given me unfailing love and support. My
parents, Helmer and Jane Engh, and my brother, Tom, have always indulged my
obsession with animal behavior and encouraged me to follow my dreams, no
matter how strange they might be. My husband and best friend, Keith Nelson,
has provided me with unconditional love and kindness, as well as a year’s worth
of companionship in the ﬁeld. I am indebted to many other friends, particularly
Terri McElhinny and Micaela Szykman, who befriended me early in my graduate
career and were always there to bounce ideas off of, commiserate with, and have
fun with. Terri frequently dragged me away from my computer, provided me with
pet therapy, and taught me a lot about how to be a good teacher and friend.
Micaela was not only a wonderful friend, but also a great colleague. She patiently

taught me everything that she knew, from how to push-start a very heavy truck to

how to interpret a logistic regression. She saved my but many times both in the
ﬁeld and in Michigan, and her cheerful demeanor brightened my life. Micaela and
Keith each spent a year with me in the ﬁeld, and I cannot imagine a better pair of
people to have lived and worked with.

My labmates were a wonderfully kind group-of people who always
welcomed discussion of behavior and life. I thank Pat Bills, Erin Boydston,
Stephanie Dloniak, Keron Greene, Karen Hubbard, Joe Kolowski, Terri
McElhinny, Eva-Maria Muecke, Keith Nelson, Scott Nunes, Micaela Szykman,
Jaime Tanner, RussVan Horn, Soﬁ Wahaj, and Heather Watts for making my life
in Michigan richer.

Numerous people gave me their help and friendship while I was in the
ﬁeld. John and Peris Keshe and Moses Nairori provided me with a comfortable
and pleasant home in the middle of the African bush. Jakob Bro-Jorgensen, Matt
Walpole, and Vincent Onyango engaged me in great discussions and deadly
aerobics workouts. Dave Acton, Anthony Chefﬁngs, Milton and Elly Kirkman,
Zohar Navon, Andrew and lsla Peart, Nasha Silantoi, Kristofer Zachrisson, and
Kaz Zanella were a welcome relief from the sometimes tedious routines of life in
Fisi Camp. John Watkin, Howard Saunders, Barbara and Nigel Dundas, and the
Foster-Tlam family deserve special thanks for welcoming me into their homes,
pampering me while I was in Nairobi, sending car parts and supplies to the bush,
and bringing great joy (and cheese!) during their visits to camp.

Many people participated in the collection and analysis of data presented

in this dissertation. Several undergraduate students provided me with valuable

assistance and displayed exceptional fortitude in pulling mounds of data out of
notebooks. Katrina Esch, David Greenberg, Erin Siebert, Andrea Jenkins, and
Anna Baumgrass have been an immense help to me; I hope that they have
learned as much from me as l have learned from them. Mike Bruford and
Stephan Funk at the Institute of Zoology in London and the Scribner lab at MSU
worked very hard to identify microsatellites and genotype hundreds of hyenas.
Mike warmly welcomed me into his lab, and Stephan patiently molded me into a
molecular geneticist. Kim Scribner and Scot Libants identiﬁed several new
hyena microsatellites, without which our paternity assignments would have been
much weaker, and Russ Van Horn spent months painstakingly genotyping the
Talek animals at the new loci. Micaela Szykman kindly provided mating data and
association analyses for the paternity chapter. Numerous people on the Mara
Hyena Project collected data that | used in my analyses, including Nancy Berry,
Erin Boydston, Susan Cooper, Stephanie Dloniak, Martin Durham, Josh
Friedman, Paula Garrett, lsla Graham, Tyson Harty, Cathy Katona, Keith Nelson,
Karen Nutt, Gabe Ording, Laura Smale, Micaela Szykman, SoﬁWahaj, Kim
Weibul, and Brad White.

Finally, I would like to thank all of the organizations that provided
clearance and funding for this research. The Ofﬁce of the President of Kenya, the
Kenya Wildlife Service, the Narok County Council, and the Senior Warden of the
Masai Mara National Reserve gave me permission to live and conduct research
in the Mara. The Mara Hyena Project has been supported by NSF grants
BNSB706939, BN89021461, IBN9309805, IBN9606939, IBN9906445, and

vi

IBN0113170. Funding for my graduate studies and research was provided by a
National Science Foundation Graduate Research Fellowship, a Michigan State
University Distinguished Fellowship, the Department of Zoology, the Program in
Ecology, Evolutionary Biology, and Behavior, and a Barnett-Rosenberg

Fellowship.

vii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ....................................................................... iv

LIST OF TABLES .................................................................................. x

LIST OF FIGURES ............................................................................... xi

CHAPTER 1:

GENERAL INTRODUCTION .................................................................... 1

CHAPTER 2:

MECHANISMS OF MATERNAL RANK “INHERITANCE” IN THE SPOTTED

HYENA .............................................................................................. 13
Introduction ................................................................................ 1 3
Methods .................................................................................... 16
Results ..................................................................................... 22
Discussion ................................................................................. 34

CHAPTER 3:

SOCIAL COMPLEXITY AND THE EVOLUTION OF INTELLIGENCE IN THE

SPOTTED HYENA ............................................................................... 42
Introduction ................................................................................ 42
Methods .................................................................................... 47
Results ..................................................................................... 52
Discussion ................................................................................. 58

CHAPTER 4:

REPRODUCTIVE SKEW AMONG MALES IN A FEMALE-DOMINATED

MAMMALIAN SOCIETY ......................................................................... 65
Introduction ................................................................................ 65
Methods .................................................................................... 68
Results ..................................................................................... 74
Discussion ................................................................................. 87

CHAPTER 5:

SOCIAL AND DEMOGRAPHIC FACTORS INFLUENCE PARASITISM IN A

WILD CARNIVORE .............................................................................. 94
Introduction ................................................................................ 94
Methods .................................................................................... 95
Results ..................................................................................... 99
Discussion ............................................................................... 104

APPENDIX A ..................................................................................... 114

viii

LIST OF TABLES

Table 2.1. Relationships among the sample sets of hyena cubs.
Table 4.1. Patterns of paternity during eight alpha male “reigns”.

Table 5.1. Food consumed by spotted hyenas in the Masai Mara National
Reserve, Kenya (data from Cooper et al. 1999).

Table 5.2. Results of general linear modeling analysis on parasite intensity data.
An asterisk indicates a signiﬁcant relationship.

Table 5.3. Results of logistic regression on parasite prevalence data. An asterisk
indicates a signiﬁcant relationship.

Table A1 Characteristics of microsatellite repeats used to assess paternity in
spotted hyenas.

Table A2. Reaction conditions for 25 pl PCR reactions.
Table A3. PCR reaction cycle conditions for hyena microsatellites.

Table 8.1. Prevalence and intensity (eggs per gram of feces) of parasite eggs
and oocytes in the feces of 70 wild spotted hyenas.

LIST OF FIGURES

Figure 2.1. Hourly rates at which adult females directed unprovoked aggressive
behaviors at cubs, ordered on the basis of the victim’s maternal rank.

Figure 2.2. Hourly rates of at which adult female hyaenas intervened on behalf of
their offspring, ordered on the basis of their social rank.

Figure 2.3. Mean proportion (: SEM) of maternal interventions against high-bom
and low-bom opponents that were effective in food and non-food contexts.
Sample sizes represent numbers of mother-cub pairs observed.

Figure 2.4. Mean proportion (1t SEM) of coalitions that reinforced the existing
dominance hierarchy organized in terms of victim, initiator, and supporter of the
attack. Numbers of cub are indicated above bars.

Figure 2.5. Mean proportion (i SEM) of coalitions joined by cubs in which the
joining cub was higher-ranking than the victim of attack. Coalitions are grouped
according to the age of the victim and initiator of the attack. Numbers of
supporting cubs are indicated above bars.

Figure 2.6. Age-related changes during the ﬁrst two years of life in the hourly
rates (i SEM) at which a mother intervened on behalf of her own cub, a cub
joined another hyena attacking a third animal, a cub was the victim of coalitionary
aggression, and a cub initiated an attack in which one or more conspeciﬁcs
joined. Means for 20 cubs are represented by each data point.

Figure 3.1. Initiation of ﬁghts by hyenas in all contexts. Bars represent the mean
number of instances per hyena in which an individual’s rate of aggression was
higher either before (black) or after (gray) the arrival of a hyena that ranked
dominant to, or intermediate between, each pair of opponents. An asterisk
indicates a signiﬁcant difference (P<0.05).

Figure 3.2. Initiation of ﬁghts by subadult hyenas in food-related contexts. Bars
represent the mean number of instances per animal in which an individual’s rate
of aggression was higher either before (black) or after (gray) the arrival of a
hyena that ranked dominant to, or intermediate between, each pair of opponents.
An asterisk indicates a signiﬁcant difference (P<0.05).

Figure 3.3. Coalitionary support in on-going disputes. Bars represent the number
of times that hyenas ranking dominant to both members of a ﬁghting pair, or .
ranking intermediate between the members of the pair, joined in ﬁghts. Bar color
indicates whether supporters joined on the side of the dominant (black) or

xi

subordinate (gray) contestant. An asterisk indicates a signiﬁcant difference
(P<0.05).

Figure 3.4. Coalitionary support in attacks initiated by the subordinate member of
a ﬁghting dyad (reversals). Bars represent the number of times that hyenas
ranking dominant to, or intermediate to, ﬁghting pairs joined in disputes. Bar color
indicates whether supporters joined on the side of the dominant (black) or
subordinate (gray) contestant. An asterisk indicates a signiﬁcant difference
(P<0.05).

Figure 4.1. Mean (: SEM) reproductive rate of immigrant males at each rank
position. Numbers above bars represent the number of males at each rank
position.

Figure 4.2. Observed paternity of 75 cubs in relation to male social rank
compared with the pattern of paternity predicted by a dominance-based priority of
access model (Altmann et al. 1996).

Figure 4.3. Mean proportion of all 21 adult females present in the Talek clan in
1994 that were encountered per observation session during each of 209 2-wk
long intervals by the immigrant males present in the clan during this period,
ordered on the basis of male rank in the immigrant hierarchy.

Figure 4.4. Distribution of mounting and copulatory behavior across immigrant
male social rank.

Figure 4.5. Total number of cubs sired by 24 immigrant males shown in relation
to their tenure in the Talek clan.

Figure 4.6. Mean reproductive rate (iSEM) for immigrant males during each year
of their tenure. Numbers above bars represent the number of males present
during each year of tenure.

Figure 5.1. Log(x + 1) transformed Spirometra intensity in relation to hyena social
rank.

Figure 5.2. Log(x + 1) transformed Isospora intensity in male and female hyenas.

xii

Chapter 1

GENERAL INTRODUCTION

Although gregarious mammals may gain through protection against
predators (Carl 1971 ), division of labor (Sherman et al. 1991 ), or numerous other
beneﬁts (see Feldhammer et al. 1999), they also face costs associated with
group living. In particular, animals living in groups suffer from increased
intraspeciﬁc competition over critical resources (e.g., Hoogland 1979). When
disputes occur over who controls particular resources, animals typically settle
disputes based on asymmetries in resource holding power (RHP). In stable
social groups where animals interact repeatedly, they may form a dominance
hierarchy (Pusey & Packer 1997). In societies structured by dominance
hierarchies, an individual’s rank within the hierarchy typically has a massive ‘
impact upon access to resources, and ultimately, upon reproductive success
(e.g., Sherman et al. 1991; De Ruiter et al. 1994; Altmann et al. 1996; Holekamp
et al. 1996; Creel et al. 1997 (but see Packer et al. 1995; Gross 1996)). Here, I
investigate the causes and consequences of social rank in the spotted hyena
(Crocuta crocuta). I examine how young hyenas attain their social ranks, whether
they recognize relative rank relationships between conspeciﬁcs, the relationship
between male rank and reproductive success, and how rank inﬂuences parasitic
infection.

Hyena biologists often think of spotted hyenas as baboons (Papio spp.)

with big teeth. Although they are separated by over 90 million years of

evolutionary change and ﬁll very different ecological niches (Lillegraven et al.
1979; Kingdon 1997), spotted hyenas and cercopithecine primates share many
characteristics of their social lives (Holekamp et al. 1999a). Cercopithecine
primates are one of two subfamilies of Family Cercopithecidae, the Old World
monkeys. Subfamily Cercopithecinae consists of 48 species of monkeys,
including baboons, mangabeys (Cercocebus spp.), guenons (Cercopithecus
spp.), macaques (Macaca spp.), and vervet monkeys (Chlorocebus aethiops),
that are distributed throughout Africa and Asia (Feldhammer et al. 1999). With
the possible exception of the guenons (Cords 1987), most cercopithecine
primates live in groups containing multiple adult males and multiple matrilines of
females and their offspring. These cercopithecine social groups are among the
best-studied of all primate societies (Melnick and Pearl 1987).

Spotted hyenas, members of Family Hyaenidae, are the most abundant
large carnivores in Africa (Feldhammer et al. 1999). Like most cercopithecine
primates, spotted hyenas live in multimale groups, with multiple matrilines of
female kin making up the stable core of the social group (Kmuk 1972; Frank
1986; Holekamp et al. 1999a). In addition to living in groups of similar
composition and complexity, these two taxa also share many aspects of their
social behavior. In the following chapters, I compare several aspects of the
spotted hyena social system to the social systems of typical cercopithecine
primates.

Cerc0pit_hecine primates

Cercopithecine primates live in a wide variety of habitats, from forests to

savannahs, and feed on a diverse array of vegetation, including grasses, leaves,
fruits, tubers, ﬂowers, and seeds. In addition to their plant-based diet, some
species supplement their diets with insects or meat. Multimale cercopithecine
groups tend to be very cohesive and rarely split into subgroups. On average,
they include 30 to 50 members who cooperatively defend a group territory
(Melnick and Pearl 1987). Male cercopithecine primates usually disperse from
their natal troops, whereas females are philopatric (Pusey and Packer 1987).
Within each troop, individuals can be ranked in a linear dominance hierarchy in
which males are usually dominant to females, often as a result of pronounced
sexual size dimorphism. Male-male interactions are typically more aggressive
and less afﬁliatlve than female-female interactions, and male rank is usually
determined by body size, ﬁghting ability, or age (Drickhamer and Vessey 1973;
Packer 1979; Smuts 1985). In baboons and vervet monkeys, the highest-ranking
males are young to middle aged adults who are in prime condition (Walters and
Seyfarth 1987). Among-females, social rank is determined by maternal rank. In a
process termed “maternal rank inheritance”, young monkeys eventually attain
social ranks just below those of their mothers (Kawai 1958; Melnick & Pearl
1987; Chapais 1992). Although maternal rank plays an important role in
determining social ranks of both male and female juveniles, young males grow
larger than their mothers and begin to outrank most or all adult females in the
natal troop before dispersing (Lee and Oliver 1979; Periera and Altmann 1985;
Walters and Seyfarth 1987). Among female cercopithecine primates, high rank is

usually associated with higher reproductive success (e.g., Gouzoules et al.

1982), probably because high-ranking females have better access to food than
do low-ranking females (Silk 1987). Like females, high-ranking males also tend to
have higher reproductive success than do low-ranking males (e.g., Altmann et al.
1996)

Spotted hyenas

Spotted hyenas are gregarious carnivores found throughout sub-Saharan
Africa (Kingdon 1997). A group of hyenas, called a clan, can range in size from
ﬁve to over 100 individuals, depending on local resource abundance, and
members of a clan cooperate to defend a territory (Mills 1990; Kingdon 1997;
Boydston et al. 2001 ). In contrast to cercopithecine primates, hyenas live in a
ﬁssion-fusion society, and sub-group composition can change quite dramatically
from day-to-day or even hour-to-hour (Boydston et al. 2001). Although often
considered scavengers, spotted hyenas are in fact quite capable hunters, and kill
up to 96% of their prey themselves (Kruuk 1972). Within a clan, as within a
monkey troop, individuals can be ranked in a linear dominance hierarchy in which
social rank determines an individual’s priority of access to food and other
resources (T ilson and Hamilton 1984; Frank 1986).

Hyenas are very unusual compared to cercopithecine primates and most
other gregarious mammals in that adult females are dominant to adult males
(Kruuk 1972; Frank 1986). Like cercopithecine primates, young hyenas inherit
their mothers’ social ranks. By the time they are 18 months old, male and female
hyenas rank immediately below their mothers (Holekamp & Smale 1991; Smale

et al. 1993). Most male hyenas disperse from their natal clans between the ages

of 2 and 5 years (Frank 1986; Henschel & Skinner 1987; Smale et al. 1997; East
& Hofer 2001). Although they retain their maternal ranks as long as they remain
in their natal clans, immigrating males behave submissively to all new hyenas
encountered outside of the natal clan. As a result, immigrant males assume the
lowest rank positions in the hierarchy when they join a new clan (Smale et al.
1997). Aggression among male hyenas is less common than aggression among
females, and immigrant male rank is based on tenure in the new clan after
dispersal (Smale et al. 1997; East & Hofer 2001). Immigrant male rank only
increases when higher-ranking animals either die or disappear, and the alpha
male is simply the male who has lived in the clan for the longest time (Smale et
al. 1997). Although the highest-ranking male is called the “alpha” male, he
nonetheless ranks below all natal animals and can be displaced from resources
even by small cubs (Holekamp et al. 1993).

High-ranking female hyenas, like high-ranking female monkeys, enjoy
higher reproductive success than do lower-ranking females. Compared to low-
ranking females, high-ranking females begin to reproduce earlier, have shorter
interbirth intervals, and have higher cub survivorship (Frank et al. 1995;
Holekamp et al. 1996). Over the course of a lifetime, this difference can result in
a 6-fold difference in reproductive success between the lowest- and highest-
ranking females (Holekamp et al. 1996). Among male hyenas, the relationship
between social rank and reproductive success was unknown at the start of this
study, although anecdotal observations suggested that reproductive success was

strongly skewed toward the highest-ranking immigrant males (Kruuk 1972; Frank

1986; Mills 1990).
Similarities between the social svstems of hyenas and cercopithecine primates
Besides similarities in the basic structures of their societies, hyenas and
cercopithecine primates exhibit many behaviors that are strikingly similar in both
form and function. As described above, in both taxa, youngsters inherit their
mothers’ social ranks, and female dominance relations are extremely stable
across a variety of contexts and over long periods of time (Holekamp & Smale
1991). Among hyenas, non-littermate siblings assume relative ranks that are
inversely related to age in a pattern of “youngest ascendency’ like that seen in
many cercopithecine primates (e.g., Horrocks & Hunte 1983). In both primates
and hyenas, kin associate more closely than do non-kin, and individuals direct
afﬁliative behavior towards kin more frequently than towards non-kin (Seyfarth
1980; Seyfarth & Cheney 1984; East et al. 1993; Holekamp et al. 1997). Both
hyenas and cercopithecine primates have been shown to reconcile after ﬁghts
(Aureli & De Waal 2000; Wahaj et al. 2001), and in both taxa, high-ranking
animals are preferred over lower-ranking individuals as social companions
(Seyfarth 1980, 1981; Holekamp et al. 1997). In addition, patterns of greeting
behavior in Crocuta follow primate patterns of social grooming (East at al. 1993),
in which individuals are more likely to direct afﬁliative behavior towards high-
ranking than low-ranking non-kin (Seyfarth & Cheney 1984). Finally, triadic and
more complex interactions (e.g., coalitions) appear to play important roles in the
acquisition and maintenance of social rank in Crocuta (Mills 1990; Zabel et al.

1992; Holekamp & Smale 1991; 1993; Smale et al. 1993; Jenks et al. 1995), as

they do in many cercopithecine primates (Cheney 1977; Walters 1980; Datta
1983, 1986; Harcourt 1988, 1992; Pereira 1989; Chapais et al. 1991; Chapais
1992)

Here I sought to compare several aspects of hyena social life to primate
social life, and to assess the inﬂuence of social rank on particular aspects of
hyena biology. In Chapters 2 and 3, | asked whether hyenas and cercopithecine
primates share similar mechanisms of maternal rank inheritance, and whether
they exhibit similar abilities with respect to social cognition. In Chapters 4 and 5,
respectively, I examined the impact of social rank on male reproductive success,
and on the intestinal parasite loads of individuals. All of the data in this
dissertation were collected from a single clan of hyenas, the Talek clan, which
defends a 70 km2 territory on the northeastern edge of the Masai Mara National
Reserve in Kenya. The research presented here would not have been possible
without the efforts of many other people. Much of the behavioral and
physiological data presented throughout the dissertation were collected by Kay
Holekamp, Laura Smale, and numerous ﬁeld assistants and graduate students
working on the Mara Hyena Project. UndergraduatesKatrina Esch, Anna
Baumgrass, Erin Siebert, and David Greenberg assisted me in extracting
relevant data from archived notes for Chapters 2 and 3. The data presented in
Chapter 4 are the result of several people’s work. The labs of Mike Bruford and
Kim Scribner were critical in identifying hyena microsatellites, and Russ Van
Horn genotyped many Talek animals. In addition, Micaela Szykman provided

data used in analyses of association patterns and copulatory behavior. I use the

term “we" in each data chapter to indicate that my dissertation research was, in
fact, a collaborative effort.
Overview of t_he chapters

In Chapter 2, I investigated how young hyenas attain their social ranks.
One of the most striking similarities between hyenas and cercopithecine primates
is that both have a system of maternal rank inheritance (Holekamp and Smale
1991). When a hyena is born, it ﬁghts viciously with its littermates to establish
dominance over its sibling or siblings (Frank et al. 1991). As the cubs mature,
they begin to develop dominance relationships with cubs born to other mothers
within the social group. At ﬁrst, these dominance relationships are unrelated to
maternal rank, body size, age, or sex, but by the age of 8 months, cubs have
social ranks in relation to those of their peers that are isomorphic with those of
their mothers. By the time they are 18 months old, young hyenas are treated by
adult clan members as if they have social ranks identical to those of their
mothers (Holekamp & Smale 1993; Smale et al. 1993). Here, I sought to assess
whether young hyenas inherit their maternal ranks via the same processes as do
young monkeys by testing predictions of four mechanistic hypotheses. In both
taxa, coalitionary support, both from relatives and non-relatives was extremely
important. The remarkable similarity between the rank inheritance of hyenas and
primates raised the question of whether they operate using similar cognitive
mechanisms, or whether they arrived at the same basic social patterns via

different mechanisms.

Two different types of selection pressures have been hypothesized to
favor the evolution of large brains and great intelligence in primates. The ﬁrst
hypothesis suggests that intelligence has been favored in primates by selection
pressures associated with complexity in the physical environment, particularly
that confronted when navigating through a three-dimensional arboreal world
(e.g., Povinelli & Cant 1995; Povinelli & Preuss 1995) or when ﬁnding and
obtaining food (e.g., Jerison 1973; CIutton-Brock & Harvey 1980; Milton 1981 ).
The second hypothesis suggests instead that key selection pressures have been
imposed by complexity associated with the labile behavior of conspeciﬁc group
members (Jolly 1966; Humphrey 1976; Byme & Whitten 1988). Because the
societies of spotted hyenas and cercopithecine primates are so similar, I asked in
Chapter 3 whether their abilities were also similar in the domain of social
cognition. Speciﬁcally, I sought to test one prediction of the social complexity
hypothesis, that animals living in similar social environments should possess
similar intellectual skills. Although the social complexity hypothesis is supported
by interspeciﬁc comparisons among primates (reviewed in Byme and Whiten
1988), the generality of the hypothesis is limited by a lack of information about
the cognitive abilities of non-primates. In fact, some primatologists hypothesize
that primates differ qualitatively from other animals, and that certain mental
abilities, such as recognition of third-party relationships, are unique to primates
(T omasello and Call 1997). The ability to recognize third-party relationships, that
is, relationships in which the observer is not directly involved, has been

demonstrated in several species of cercopithecine primates (Cheney and

Seyfarth 1980; Silk 1999). Here I used several tests to assess whether hyenas,
too, could recognize third-party relationships based on the social ranks of
conspeciﬁcs. I found that hyenas were indeed able to distinguish the relative
ranks of other clan members, thus providing strong support for the social
complexity hypothesis.

Social dexterity and morphological traits both appear to affect reproductive
success in a number of primate species (De Ruiter et al. 1994; Altmann et al.
1996; Bercovitch & Numberg 1997). In chapter 4, I examined the role of social
rank in determining the reproductive success of male hyenas. Numerous studies
have assessed the effect of social rank on reproductive success in male
mammals, and most have found that reproductive success among males is
positively correlated with rank, which, in turn, is largely determined by ﬁghting
ability (e.g., CIutton-Brock 1988; Creel et al. 1997; Girrnan et al. 1997). In
contrast to most other mammals, spotted hyenas exhibit some striking sex-role
reversals. That is, among adult hyenas, females are larger and more aggressive
than males, and females dominate males in all dyadic contexts (Kruuk 1972;
Frank 1986). In addition, ﬁghting among adult male hyenas is less frequent and
less intense than is that observed among adult females. (Frank 1986; Kruuk
1972; Mills 1990), suggesting that male-male contest competition over mates
may not be as important in hyenas as it is in many other mammals. We found
that, while rank was correlated with reproductive success, tenure was a much
better predictor of a male’s reproductive success than social rank per se. In

contrast to what was observed in baboons (Altmann et al. 1996), among hyenas,

10

male reproductive success did not ﬁt a priority of access model, in which both
male rank and the number of females simultaneously in estrus were taken into
account. Also, the highest ranking male observed with a fertile female was no
more likely to be the father of her cubs than were other males. It appears that in
this female-dominated species, high-ranking males are unable to monopolize
access to reproductive females, possibly as a consequence of female choice.

Finally, I examined the effects of social rank on parasite burden in Chapter
5. Although parasites and pathogens appear to have substantial impacts on
ﬁtness of wild animals, only rarely do studies incorporate social variables into
studies of parasitism. One of the few studies to examine parasite load in relation
to social rank (Hausfater & Watson 1976) obtained the surprising result that high
social rank was correlated with high parasite load among male baboons. Here, I
tested whether several social and demographic variables inﬂuenced the load or
prevalence of intestinal parasites in hyenas. For each parasite species, a
different suite of social and/or demographic variables inﬂuenced parasitism.
Similar to Hausfater and Watson (1976), I found that high-ranking hyenas had
higher loads of some parasites than did low-ranking hyenas. That different
factors were associated with high parasite prevalence or intensity of infection for
each parasite species probably reﬂects differences in the transmission biology of
that species.

Each of these chapters examines the causes and consequences of social
rank in the loves of spotted hyenas. In Chapters 2 and 3, I found patterns of rank

inheritance and social cognition that were nearly identical to those observed in

11

cercopithecine primates who lice in similar social groups. These data suggest
that hyenas and cercopithecine primates have undergone convergent evolution
and raise the question of whether the neural mechanisms regulating these
behaviors are also similar in these taxa. The results of Chapter 4 suggest that,
unlike what is observed in most mammalian societies organized by dominance
hierarchies, reproductive skew among male hyenas Is not very strong. High-
ranking male hyenas are unable to monopolize reproductive opportunities,
probably because the pattern of female dominance characteristic of hyena
societies diminishes the ability of males to control reproduction. Finally, in
Chapter 5, I found that high-ranking individuals had higher loads of some

parasites than did lower-ranking individuals.

12

Chapter 2
MECHANISMS OF MATERNAL RANK “INHERITANCE” IN THE SPOTTED

HYENA

INTRODUCTION

A variety of factors affect dominance status in gregarious vertebrates
These include genetic inﬂuences (e.g., Craig et al. 1965; Moss et al. 1982;
Kikkawa et al. 1986), body size (e.g., CIutton-Brock et al. 1982), age (e.g.,
Goodall 1986; Pusey & Packer 1997), tenure in the social group (e.g., Sugiyama
1976; Smale et al. 1997), and political machinations (de Waal 1982). In many
cercopithecine primates, maternal rank is an extremely important determinant of
offspring social rank. That is, during early ontogeny, the ranks of young primates
come to be highly correlated with those of their mothers (Macaca fuscata: Kawai
1958; M. mulalta: Missakian 1972; M. sylvanus: Paul & Kuester 1987; M.
fascicularis: de Waal 1977; M. arctoides: Estrada 1978; M. nemestn'na: Bernstein
1969; M. radiate: Silk et al. 1981a,b; Papio cynocephalus: Hausfater 1975; P.
anubis: Johnson 1987; Cercopithecus aethiops: Horrocks & Hunte 1983;
Theropithecus gelada: Dunbar 1980). As in these myriad primate species, the
social ranks of young spotted hyenas (Crocuta crocuta) are also strongly
inﬂuenced by the ranks of their mothers (Frank 1986; Holekamp & Smale 1991;
1993; Smale et al. 1993; Jenks et al. 1995). The process by which youngsters
assume positions in the dominance hierarchy adjacent to those of their mothers

is called maternal rank ‘inheritance’ by primatologists (e.g., Harcourt & Stewart

13

1987). Although social rank is relational and is not literally inherited, we adopt
this same terminology here. Further, following Chapais (1992) we assume that
maternal rank ‘inheritance’ includes processes of both rank acquisition and rank
maintenance.

The social lives of spotted hyenas are comparable in many respects to
those of cercopithecine primates (Frank 1986; Holekamp & Smale 1993; Smale
et al. 1993; Holekamp et al. 19993). Like most primates, Crocuta are long-lived
mammals that bear small litters requiring a long period of nutritional dependence
on the mother. Like cercopithecine primates, Crocuta live in permanent social
groups, the members of which cooperatively defend communal territories. Social
groups in both taxa contain multiple adult males and multiple matrilines of adult
female kin with offspring, including individuals from several overlapping
generations. Males in both taxa typically disperse from their natal groups
whereas females are usually philopatric (Cheney & Seyfarth 1983; Henschel &
Skinner 1987; Pusey & Packer 1987; Mills 1990; Smale et al. 1997). Adults of
both taxa can be ranked in a linear dominance hierarchy based on outcomes of
agonistic interactions, and priority of resource access varies with social rank
(Tilson & Hamilton 1984; Andelman 1985; Frank 1986). In both Crocuta and
various Old World primates, female reproductive success is strongly correlated
with social rank (Altmann et al. 1988, 1996; Frank et al. 1995; Holekamp et al.
1996). In both taxa, members of the same matriline occupy adjacent rank
positions in the group's hierarchy, and female dominance relations are extremely

stable across a variety of contexts and over periods of many years. Juvenile

14

hyenas of both sexes acquire ranks immediately below those of their mothers
(Holekamp & Smale 1991, 1993; Smale et al. 1993), and nonlittermate siblings
assume relative ranks that are inversely related to age in a pattern of ‘youngest
ascendency’ like that seen in many cercopithecine primates (e.g., Horrocks &
Hunte 1983). In both primates and hyenas, kin associate more closely than do
non-kin, and individuals direct afﬁliative behavior towards kin more frequently
than towards non-kin (Seyfarth 1980; Seyfarth & Cheney 1984; East et al. 1993;
Holekamp et al. 1997). In both taxa, high-ranking animals are preferred over
lower-ranking individuals as social companions (Seyfarth 1980, 1981; Holekamp
et al. 1997). Furthermore, patterns of greeting behavior in Crocuta follow primate
patterns of social grooming (East et al. 1993), in which individuals prefer to direct
afﬁliative behavior towards high-ranking non-kin (Seyfarth & Cheney 1984).
Finally, triadic and more complex interactions (e.g., coalitions) appear to play
important roles in the acquisition and maintenance of social rank in Crocuta
(Mills 1990; Zabel et al. 1992; Holekamp & Smale 1991; 1993; Smale et al. 1993;
Jenks et al. 1995), as they do in many cercopithecine primates (Cheney 1977;
Walters 1980; Datta 1983, 1986; Harcourt 1988, 1992; Pereira 1989; Chapais et
al. 1991; Chapais 1992).

A number of studies have attempted to determine how young primates
“inherit’ their mothers’ ranks, and several possible mechanisms have been
proposed. The simplest hypothesis suggests that genetic inheritance of rank-
related behavioral tendencies or physical characteristics account for the

‘inheritance’ of maternal rank. Unprovoked aggression directed preferentially by

15

adult females toward offspring of lower-ranking females has also been proposed
as a mechanism of maternal rank ‘inheritance’ (Berman 1980; Horrocks & Hunte
1983). Another mechanism known to function importantly in primates involves
defensive interventions by mothers on behalf of their offspring when the latter are
engaged in disputes with conspeciﬁcs (Cheney 1977; Berman 1980; Horrocks 8
Hunte 1983; Netto & Van Hooff 1986). Finally third-party support, or coalition
formation, during aggressive interactions appears to play an important role in
acquisition of maternal rank in primates (Cheney 1977; Walters 1980; Datta
1983; Bernstein & Ehardt 1985; Ehardt & Bernstein 1992).

‘lnheritance’ of maternal rank occurs in both wild and captive groups of
spotted hyenas, including groups that vary widely in size and composition
(Holekamp & Smale 1993; Smale et al. 1993; Jenks et al. 1995). Free-living
Crocuta cubs attain their ranks within their peer cohorts within the ﬁrst 8 months
of life (Holekamp & Smale 1993), and they attain their ranks relative to older clan
members somewhat later (Smale et al. 1993). In light of the striking similarities
between the social lives of spotted hyenas and cercopithecine primates, we set
out to determine here whether young hyenas ‘inherit’ maternal rank via the same
processes as do young monkeys. Working with one large group of free-living
hyenas in Kenya, we tested predictions of each of the four mechanistic

hypotheses enumerated above.

METHODS

We monitored a single clan of spotted hyenas from June 1988 to June

16

1999 in their 65 km2 core home range in the Talek area of the Masai Mara
National Reserve, Kenya. We made daily behavioral observations of the 67
offspring of 22 adult female clan members present in the clan between June
1988 and April 1991. Throughout this period, observers were present in the
study area 23 to 31 days per month. Individual hyenas were identiﬁed by their
spot patterns and other unique characteristics, such as ear notches. Mother-
offspring relationships were determined on the basis of nursing associations.

Behavioral observations took place near dawn (0600-0900) and dusk
(1630-1930), when hyenas were most active. Data were recorded throughout
the clan’s home range and during all activities. Behavior was recorded during
weekly 30-minute focal animal surveys of individual cubs, and continuous critical
incident (‘all-occurrence’ Altmann 1974) sampling of agonistic behavior emitted
by all Talek hyenas. Aggressive behaviors included threat, bite, chase, and
approach in attack posture, whereas submissive behaviors included avoidance or
withdrawal from interaction, submissive posture, head bob, and carpal crawl (see
Kruuk 1972 for deﬁnitions). All aggressive behaviors by adult females toward
cubs were classiﬁed as ‘provoked’ if they occurred during nursing or feeding, or if
they immediately followed pesky or aggressive behavior by the targeted cub.
Aggressive behaviors not meeting any of these criteria were classiﬁed as
‘unprovoked.’ Coalitions were deﬁned as two or more hyenas concurrently
behaving aggressively toward a third. When coalitions formed, the identities of
all aggressors were recorded, as well as the order in which they joined the

interaction. The ﬁrst aggressor involved in a coalition was called the initiator, any

17

individuals joining the initiator were called supporters, and the recipient of the
aggression was termed the victim.

Maternal interventions were identiﬁed when an adult female interrupted an
interaction between her cub and another individual. Interventions were classiﬁed
as either ‘aggressive’ or ‘non-aggressive.’ Aggressive interventions involved
approach followed by threat, attack, or displacement behavior by the mother
toward the individual interacting with her cub. Non-aggressive interventions
involved approach followed by distraction (e.g., initiation of a greeting ceremony)
or blocking (e.g., the mother positioning herself between her cub and its
opponent). Additionally, the context of each intervention was classiﬁed as either
food-related or non-food related. A food-related intervention took place during a
dispute over access to or possession of a food item. A maternal intervention was
considered effective when the hyena interacting with the cub in question desisted
from interacting with that cub during or immediately after the intervention.

The social rank of each clan member was determined based on its
position in a matrix of appeasement behaviors exhibited during dyadic agonistic
interactions (Martin & Bateson 1988). The ‘winner’ was the animal being
appeased, while the ‘Ioser’ was the animal exhibiting appeasement behavior. By
convention, the most dominant animal in the hierarchy was assigned a rank of
one. A juvenile’s ‘maternal rank’ referred to its mother's position within the adult
female hierarchy. Low-bom individuals were those of lower maternal rank than
the focal animal, while high-bom animals had a higher maternal rank than the

focal animal (e.g., Datta 1983). A cub was considered high-ranking if its mother's

18

rank was 10 or higher and low-ranking if her rank was below 10. There was no
signiﬁcant sex bias within the sample set (Table 2.1; X22 = 0.75, NS) and rank
was unrelated to cub sex in all of the sample subgroups examined (Table 1;
N=67, t65=0.78, P=0.44; N=38, t36=0.57, P=0.58; N=20, t18=-0.28, P=0.79)

Ages of all juvenile hyenas were known to within one week, as described
by Holekamp et al. (1996). In all analyses below, hyenas were called cubs until
18 months of age, subadults from 18 to 24 months, and adults over 24 months
(Matthews 1939; Smale et al. 1993). Spotted hyenas acquire their adult ranks by
18 months of age, and generally maintain these ranks as long as they remain in
the natal clan (Smale et al. 1993). However, as in primates, rank acquisition
occurs in stages in Crocuta: social rank relative to peers is acquired from 0-6
months of age, and this rank relative to peers becomes perfectly aligned with
maternal rank by 6 to 8 months of age (Holekamp & Smale 1993). Cubs acquire
their social ranks relative to adults between 8 and 18 months of age (Smale et al..
1993). Therefore, for analyses below of behavioral changes over time, separate
rate values were calculated for each cub during each of the following age
intervals: 0-6 months, 7-8 months, 9-18 months, and 19-24 months. Only
individuals who survived at least 24 months (N=20) were included in these
longitudinal analyses. Siblings, offspring, and parents were considered to be kin
in the analyses below, and all other animals were considered to be non-kin.

Rate data were compared in most analyses. Hourly rates of maternal
intervention, aggressions received, and coalition formation were calculated for

each cub by dividing the total number of interventions, aggressions received, or

19

Table 2.1. Relationships among the sample sets of hyena cubs.

 

Sample Group All cubs Cubs surviving Cubs surviving
participating to 18 months to 24 months
in coalitions of age of age

 

Figures using

sample group 2.4, 2.5 2.1, 2.2, 2.3 2.6

N 67 38 20

Female 30 15 7

Male 37 23 13

Mean Female Rank (SD) 11.27 (5.85) 10.47 (5.38) 8.71 (5.47)
Mean Male Rank (SD) 10.11 (6.22) 9.39 (5.94) 9.54 (6.79)
# Subjects with 21 Sib 64 31 12

# Littermate Pairs 23 10 5

 

20

coalitions involving that cub by the total number of hours during which the cub
was observed. We calculated rates of female aggression by dividing the number
of unprovoked aggressions directed by each adult female toward each cub by the
number of hours during which each female-cub dyad was observed within 200m
of each other. Mean rates of aggression directed against high-bom and low-born
cubs were then calculated for each adult female. Only cubs that survived for at
least 18 months (N=38) were considered in analyses of female aggression and
maternal intervention.

Directionality in coalitionary interactions was assessed for each cub by
calculating the proportion of all coalitions that reinforced the adult hierarchy in
which that cub participated. Thus, if a cub joined ﬁve coalitions, three of which
were directed against Iow-bom victims, then that cub was assigned a value of
0.60 for supporting coalitions. We then calculated 95% conﬁdence intervals (Cls)
for these proportions averaged over all cubs. If 95% Cls did not include the 0.50
predicted if cubs were equally likely to participate in coalitions directed at high
and low-bom animals, then results were considered to be signiﬁcantly different
from chance expectations. All cubs participating in at least one coalition were
included in these analyses.

Spearrnan’s correlation coefﬁcients were calculated in all analyses using
social rank as an independent variable, and the signiﬁcance of each relationship
was assessed via Kruskal-Wallis tests. In all cases in which data were not
normally distributed, standard non-parametric tests, such as Wilcoxon and

Freidman tests, were used for analysis.

21

RESULTS
Genetic heritability

Although rank relationships among adult females were generally
extremely stable throughout the 11-year study period, changes in rank
relationships associated with a single clan ﬁssion event early in the study
indicated that genetic heritability cannot account for ‘inheritance’ of maternal
rank in spotted hyenas. That is, when members of three low-ranking matrilines
permanently emigrated from the Talek clan between November 1989 and June
1990, they were initially accompanied by two adult females (sisters) and their
offspring from a middle-ranking matriline, as described previously (Holekamp et
al. 1993). Although no members of the low-ranking matrilines ever returned to
Talek, the two middle-ranking females returned after an absence of one year. As
they attempted to rejoin the Talek clan, both of these females were the targets of
severe attacks by all other adult females, including those which they had
previously outranked (Holekamp et al. 1993). Both absentee females were
ultimately able to rejoin the Talek clan in 1991, but their new ranks were at the
bottom of the adult female hierarchy, where they have remained ever since.
Offspring born to these females before clan ﬁssion fell in rank along with their
mothers when they returned to Talek, and all of the subsequent offspring
produced by these two females ‘inherited’ their mothers' new ranks, rather than
their mothers’ initial ranks in the clan’s hierarchy. The fact that the ranks of these

females and their offspring changed so dramatically during the study period

22

strongly suggested that genetic heritability is unlikely to function as an important
mechanism mediating rank acquisition in this species.
Umrovoked aggression

If unprovoked female aggression contributes to maternal rank
‘inheritance’, then rates of unprovoked aggression directed by adult females
toward offspring of lower-ranking females should be higher than rates at which
they harass offspring of higher-ranking females. Furthermore, this hypothesis
predicts that rates of unprovoked aggression received by cubs from adult
females should be correlated with the cubs’ maternal ranks. We observed no
difference between the hourly rates of unprovoked aggression directed by adult
females at Iow-bom and high-bom cubs (Wilcoxon 221:0.601, P=0.548). In
addition, hourly rates at which cubs received unprovoked aggression from adult
females were not correlated with cubs’ maternal ranks (Rs=0.14, K-W18=16.16
P=0.51; Figure 2.1 ). Thus unprovoked female aggression does not appear to
function importantly in acquisition of maternal rank in Crocuta.
Maternal intervention

If maternal interventions function to promote maternal rank ‘inheritance’ in
spotted hyenas, then intervention rates should be inversely correlated with
maternal rank, and high-ranking females should intervene on behalf of their
offspring more effectively than should low-ranking females. The rates at which
female hyenas intervened on behalf of their cubs were indeed signiﬁcantly
correlated with maternal rank (Rs=-0.41, K-W13=31.08, P=0.03; Figure 2.2). Only

27/818 (3.3%) of maternal interventions occurred in response to a direct

23

 

 

o o
524-
X
1':
E
“O
.23“
o
8
n: o
22-
.9
m
o
g o
c: o
o: 0
$17 . o
. O. Q . O
o . o
o o o
0" o oo o. 0
o i o .
0 1o 20

Maternal Rank

Figure 2.1. Hourly rates at which adult females directed unprovoked aggressive
behaviors at cubs, ordered on the basis of the victim’s maternal rank.

24

aggressive attack. Most interventions occurred before the “intruder" had a
chance to behave aggressively toward the cub. For example, mothers frequently
intervened against any individuals who approached their cubs while the cubs
were feeding on small scraps. This supports Bennan’s (1980) suggestion that the
mere presence of the mother may curb rates of aggression directed toward high-
ranking cubs. Indeed, there was a tendency for high-ranking cubs to receive
lower overall rates of aggression than did low-ranking cubs (Rs=0.64, K-
W13=24.27, P=0.15). Thus low-ranking mothers had more opportunities to
intervene on behalf of their cubs, yet they intervened at far lower hourly rates
than did high-ranking females.

Intervention strategies of adult females did not depend on either the
relative ranks of the contestants (U1=133.5, P=0.375) or the context in which
interventions occurred (U1=58.00, P=0.172). Aggressive interventions were
common regardless of whether or not the situation involved food. In food-related
contexts, mothers often stood next to cubs with food, and threatened any
approaching animal. Overall, high-born females were found to be more effective
when intervening on behalf of their offspring than were low-born females
(U1=829.5, P=0.03). In particular, high-born females were more effective than
Iow-bom females in food-related contexts (U1=127.5, P=0.01; Figure 2.3), where
there was an obvious beneﬁt to winning. Thus, our data conﬁrmed that high-
ranking mothers intervened on behalf of their offspring both more frequently and

more effectively than did low-ranking mothers.

Coalitionary supggrt

25

40-

30s

20 ." .

(Interventions/Hour) x 100

 

 

 

 

 

O 5 10 15 20 25

Maternal Rank

Figure 2.2. Hourly rates of at which adult female hyenas intervened on behalf of
their offspring, ordered on the basis of their social rank.

26

100 ~

 

so)
on
c
.9.
ii
a 60-
£3
E
a)
.2
‘3 40%
a:
LU
o\°
20~
0

23

 

\\\\\\\\\\\\\\\w a

 

 

- Intervention against lower-ranking animals
m Intervention against higher-ranking animals

16

.5
o

 

\\\\\\\\\\\k\\\\~r

 

 

 

 

 

Nonfood

i

Context

Figure 2.3. Mean proportion (i SEM) of maternal interventions against high-born
and low-born opponents that were effective in food and non-food contexts.
Sample sizes represent numbers of mother-cub pairs observed.

27

Acquisition of maternal rank in Crocuta might be facilitated either by
coalitionary support of young or coalitionary targeting of young. First, if a cub is
supported by conspeciﬁcs when it attacks another animal, then this supportive
behavior may function in both militaristic and operant conditioning senses to
reinforce the cub’s behavior toward that target animal. Second, if a cub engaged
in a dispute with another animal is attacked by a coalition of conspeciﬁcs, then its
aggressive behavior toward the original target may be rapidly extinguished. This
hypothesis predicts that, when hyena A joins hyena B in an aggressive
interaction against hyena C, hyena A will join the individual who should win
according to the ‘rules’ of maternal rank inheritance. This hypothesis also
predicts that hourly rates of coalition formation in support of cubs, or rates of
coalition formation against cubs, should vary with maternal rank. Besides
receiving higher rates of coalitionary support, the offspring of high-ranking
females should have more allies, and higher-ranking allies, to support them than
should the offspring of low-ranking females. Additionally, youngsters may
facilitate acquisition or maintenance of maternal rank by providing support in
coalitions against low-born individuals. Such joining behavior should permit cubs
to reinforce their dominant status over that of target animals.

When adult hyenas supported cubs in coalitions against other cubs or
adults, they were more likely to do so against low-born than high-born
opponents. This was true both for conﬂicts between cubs and their peers (95%
Cl =0.73 - 1.00; Figure 2.4), and for conﬂicts between cubs and adults (95% Cl

=0.70 - 0.99; Figure 2.4). Cubs also supported other cubs more frequently

28

against low-born than high-born adults (95% Cl =0.87 - 1.00) and cubs (95%
Cl=0.60 - 0.89). Adult aggression against cubs was rarely supported by others,
and when this occurred, there was no tendency for the adult initiator of
aggression to be higher-ranking than the cub’s mother (cub supporter: 95% CI =
0.46 - 0.89; adult supporter: 95% CI = 0.15 - 0.95).

When supporters were divided into kin and non-kin categories, youngsters
of high maternal rank were found to have signiﬁcantly more kin providing support
than were youngsters of low maternal rank (t = 3.48, P=0.004), but we found no
signiﬁcant difference in the number of non-kin providing support (1‘ = 1.73,
P=0.102). Additionally, the ranks of supporters were correlated with the maternal
ranks of the youngsters receiving support (Rs=0.600, K-W15=26.86, P=0.04).
Despite having more and higher ranking allies, high-ranking cubs received
coalitionary support at hourly rates that did not differ signiﬁcantly from rates at
which low-ranking youngsters received coalitionary support (Rs=-0.06, K-
W18=22.74, P=0.20). Thus, in comparison to cercopithecine primates, spotted
hyenas receive relatively little coalitionary support from non-relatives during rank
acquisition. Hourly rates of coalitionary attacks directed against cubs were not
signiﬁcantly related to maternal rank (Rs=0.49, K-W13=21.70, P=0.25), but in
most cases in which a cub was the victim of coalitionary attack, the initiator of
aggression was of higher maternal rank than was the victim (Figure 2.4).

When cubs joined coalitions against other animals, they tended to do so
against low-born individuals (Figure 2.5). Cubs joined coalitions with both adults

(95% CI = 0.86 -1.00) and cubs (95% CI =0.81 - 0.99) against Iow-bom adults

29

 

 

 

 

 

 

V///////////////f

 

 

..://////////////////.

 

 

 

...7/////////////////%.-

 

 

 

 

 

 

.i7//////////////////mﬂ.

 

 

 

 

. >555: 052853. mcozzmoo o\o

wmm

wmm

dmm

sting

nd supporter of the

indicated above bars.

ierarchy organized in terms of victim, initiator, a

Figure 2.4. Mean proportion (i SEM) of coalitions that reinforced the exi

d m ce h
attack. Numbers of cub are

30

much more commonly than against high-born adults. They also joined coalitions
initiated by cubs against low-born cubs more frequently than against high-born
cubs (95% CI = 0.60 - 0.89). Although cubs also tended to join adults attacking
low-born cubs more than high-born cubs, this tendency was not signiﬁcant (95%
CI = 0.46 - 0.89).
Age-related trends

Spotted hyena cubs attain their ranks within their peer cohorts within the
first 8 months of life (Holekamp & Smale 1993), and they attain their ranks
relative to older clan members when they are 8 to 18 months of age (Smale et al.
1993). If some mechanisms are important for attainment of rank in relation to
juvenile peers, but not for attainment of rank in relation to adults, then the rates of
the behaviors associated with those mechanisms should drop after the cub has
attained its rank in relation to its peers. Thus, behaviors which facilitate
attainment of rank in relation to peers should occur at high rates during the ﬁrst 8
months of life, behaviors which facilitate attainment of rank in relation to adults
should occur frequently between 8 and 18 months of age, and behaviors
functioning to maintain rank should be common after 18 months of age.

Hourly rates of maternal intervention on behalf of cubs dropped
signiﬁcantly between 8 and 18 months (F1=5.92, P=0.015; open triangles in
Figure 2.6) and again between 18 and 24 months (F1=5.0, P=0.025). Thus
maternal intervention appears to play a strong role in attainment of rank within
the peer cohort and possibly also within the entire clan, but probably plays a

relatively small role in rank maintenance. Hourly rates of third-party support

31

100 -

 

 

80‘

 

 

60-

40-

% Cub Maternal Rank > Victim

20‘

§t\\\\\\\\\\\\V—»

 

 

 

 

 

Ww

 

 

 

 

0
VICTIM cub adult adult
INITIATOR cub adult cub adult

SUPPORTER cub cub cub cub

 

Figure 2.5. Mean proportion (i SEM) of coalitions joined by cubs in which the
joining cub was higher-ranking than the victim of attack. Coalitions are grouped
according to the age of the victim and initiator of the attack. Numbers of
supporting cubs are indicated above bars.

32

+ Initiator
+ Supporter

14 —v— Maternal Intervention

121

 

(Rate/Hour) x 100
on

 

 

6 - \-
4 _ :\
2 - e a
. V
O l l I T T T r l l T I l
0 2 4 6 8 10 12 14 16 18 20 22 24
A e in Months Peer Rank Adult Rank
9 Established Established

Figure 2.6. Age-related changes during the ﬁrst two years of life in the hourly
rates (t_SEM) at which a mother intervened on behalf of her own cub, a cub
joined another hyena attacking a third animal, a cub was the victim of coalitionary
aggression, and a cub initiated an attack in which one or more conspeciﬁcs
joined. Means for 20 cubs are represented by each data point.

33

dropped signiﬁcantly between 18 and 24 months of age (F1=5.56, P=0.018; ﬁlled
circles in Figure 2.6). This indicates that coalitionary support may ﬁgure
prominently in rank attainment both within the peer cohort and in relation to the
rest of the clan, but also that coalitionary support may be less critical for rank
maintenance than for its establishment. Nevertheless, the high rates of
coalitionary support given to youngsters older than 18 months of age, suggest
that coalitions may function importantly at all stages of a hyena’s life. Hourly
rates of coalition formation against cubs were relatively low, and did not vary
signiﬁcantly with age over the ﬁrst 24 months of life (F3=6.17, P=0.104; open
circles in Figure 2.6). Hourly rates at which cubs joined coalitions also did not
vary signiﬁcantly among age intervals (F3=2.85, P=0.415; ﬁlled triangles in Figure
2.6), but were relatively high throughout the ﬁrst two years of life. Thus joining
behavior may function importantly in both acquisition and maintenance of

maternal rank.

DISCUSSION

During the process of socialization, rank relations of juvenile spotted
hyenas rapidly come to conform to their clan’s existing social structure as
expressed in the adult female hierarchy (Holekamp & Smale 1993; Smale et al.
1993; Jenks et al. 1995). Although youngsters are initially equally likely to direct
aggression towards high-born and low-born conspeciﬁcs, their behavior rapidly
changes such that they direct aggression exclusively at targets of lower maternal

rank, and youngsters soon assume rank positions immediately below those of

their mothers in the clan’s dominance hierarchy (Holekamp & Smale 1993;
Smale et al. 1993). In the current study we tested predictions of four hypotheses,
each suggesting a different mechanism by which the process of maternal rank
‘inheritance’ might be mediated in this species. Additionally, a ﬁfth hypothesis
that we were not able to address in the present study suggests that young
hyenas may ‘inherit’ their ranks by observing their mothers interact and thereby
Ieaming their relative ranks (Kawai 1965; Sade 1972; Gouzoules 1975; Altmann
1980; Walters 1980).

Various lines of indirect evidence from free-living hyenas in our study
population, and in those of other workers, fail to support the genetic heritability
hypothesis. For example, the social rank of a juvenile hyena changes when its
mother rises or falls in the clan’s dominance hierarchy (Holekamp & Smale 1993;
Holekamp et al. 1993), suggesting that social rank is far too labile to be directly
determined by behavioral or morphological traits that are strongly heritable. In
addition to our own observations “of rank reversals associated with Talek clan
ﬁssion, observers watching hyena populations elsewhere in Africa have observed
revolutionary coalitions formed by members of Iow-bom matrilines overthrowing
the alpha matrilines in their respective study clans (Mills 1990; H. Hofer & M. L.
East personal communication). It thus seems no more likely that genetic
heritability directly mediates acquisition of maternal rank in spotted hyenas than it
does in cercopithecine primates. This should perhaps be expected since, if
genetic differences led directly to differences in social rank, then the genes

associated with high rank would become ﬁxed within only a few generations, and

35

hence the genetic basis for differences in social rank would quickly be eliminated.

Although unprovoked female aggression (‘harassment’) may play a strong
role in rank acquisition in some primates (Gouzoules 1975; Lee 8 Oliver 1979;
Berman 1980; Silk et al. 1981a; Horrocks 8 Hunte 1983), it does not appear to
inﬂuence rank ‘inheritance’ in hyenas. Low-ranking juveniles in some monkey
species receive more unprovoked aggression than do high-ranking juveniles, and
rates at which such aggressive acts are received by youngsters are inversely
correlated with maternal rank (e.g., Horrocks 8 Hunte 1983). However, our data
failed to reveal these trends during the process of rank acquisition in Crocuta.
Adult female hyenas did not direct unprovoked aggression more frequently
toward low-born than toward high-born cubs, and we found no relationship
between the rates of unprovoked aggression received by cubs and their maternal
ranks.

As occurs in many primates (e.g., Cheney 1977; Berman 1980; Walters,
1980; Horrocks 8 Hunte 1983), interventions by mothers on behalf of their
offspring appear to play a critical role in hyena rank ‘inheritance’. High-ranking
female hyenas intervened on behalf of their offspring more frequently, and more
effectively, than did low-ranking females. When low-ranking females intervened,
they tended to use physical blocking or distraction tactics, whereas high-ranking
females used high levels of aggression against their cubs’ opponents. In a study
of maternal rank acquisition among captive Crocuta, Jenks et al. (1995) reported
that only high-born females intervened on behalf of their cubs, whereas we often

saw both low-born and high-born females intervene. We suspect this discrepancy

36

can be explained by the fact that Jenks et al. (1995) scored only aggressive
behaviors as interventions whereas we recorded both aggressive and non-
aggressive interventions. In any case, Jenks et al. (1995) concluded as we did
that maternal interventions were extremely important mechanisms of maternal
rank acquisition. Most maternal interventions occurred in response to a potential
threat, rather than in response to aggression directed toward cubs. This type of
intervention is likely to be very important in a species such as the hyena where
the potential is high for cubs to sustain injury when engaged in altercations with
larger, stronger adults. Low-ranking individuals who behave aggressively to
high-born youngsters when the latter are present with their mothers risk severe
retaliation for their actions. Subsequently, even when mothers are not present
with their offspring, the threat of maternal intervention may inﬂuence the behavior
of other group members (Cheney 1977; Berman 1980), thereby promoting both
acquisition and maintenance of maternal rank.

Finally, coalitionary support appears to play an important role in rank
‘inheritance’ in hyenas. Although the rate at which cubs receive coalitionary
support is not related to their maternal ranks, the vast majority of all support for
cubs is given when cubs behave aggressively toward low-born opponents
(Figures 2.4 8 2.5). Conspeciﬁc support of cubs during their aggressive
interactions with Iow-bom individuals may positively reinforce both aggressive
behavior by the attacking cubs and appeasement behavior by their victims. The
bulk of coalitionary attacks on juveniles are formed against low-born individuals,

and these attacks may serve to punish aggressive acts by cubs that are directed

37

toward inappropriate (i.e. higher-ranking) targets. High-born cubs have more
allies, and more powerful allies, than do Iow-bom cubs, and the threat of
coalitionary aggression from these allies may represent a ubiquitous and
powerful reminder of the status quo.

Our ﬁndings germane to coalition formation in wild hyenas were similar to
those obtained for captive prepubertal hyenas by Zabel et al. (1992). Although
hourly rates of coalition formation in the wild were roughly an order of magnitude
lower than those observed in captivity, in both situations high-born hyenas were
more likely than low-born hyenas to initiate and support attacks, and low-born
animals were more likely to be targets of group attacks. Among both captive and
wild hyenas, coalitionary attacks appeared to reinforce the existing dominance
hierarchy. However, ontogenetic trends described in the current study (Figure
2.6) indicate that coalition formation may play a far more important role in
challenging the existing hierarchy, and thus in rank acquisition, than was
suggested by captive data.

Cubs may play an active role in the acquisition and maintenance of
maternal rank by joining coalitions, particularly those directed against low-born
individuals, since cubs preferentially join hyenas attacking Iower-bom animals.
By joining in aggression against Iow-bom individuals, cubs can reinforce their
own dominant status relative to the victim. Although maternal interventions may
be critical for cubs to gain their peer ranks and align these with maternal ranks,
coalitions may play a larger role for older cubs who are attaining and later

maintaining their ranks in relation to hyenas outside their peer group.

38

When Crocuta cubs are ﬁrst brought to the communal den at
approximately one month of age, they exhibit appeasement behavior to all other
hyenas they meet there, and adults generally only direct mild aggression toward
very young cubs when their play interrupts adults’ naps by the den. During the
course of rank acquisition, cub submissive behavior appears to be extinguished
in interactions with some conspeciﬁcs, such that appeasement persists only
toward a subset of conspeciﬁcs. Cubs begin initiating attacks on non-sibling
conspeciﬁcs during their third month of life (Holekamp 8 Smale 1993), and enter
a stage during which they direct both aggressive and appeasement behaviors
simultaneously at larger-bodied hyenas. Aggressive behavior directed by cubs
toward high-born conspeciﬁcs appears to be rapidly extinguished via punishing
counterattack or coalitionary aggression, and perhaps also by relatively rare and
ineffective maternal interventions. By contrast, cub attacks on low-born
conspeciﬁcs are likely to be reinforced by effective protection from the mother,
coalitionary support from other clan members, and perhaps also by immediate
improved access to contested resources.

Our analysis of age-related trends observed during maternal rank
inheritance suggests that the process of rank acquisition begins with defensive
maternal interventions in which the cub does not participate. If play turns too
rough or another hyena gets nasty, cubs may solicit maternal aid with
vocalizations (East 8 Hofer 1991; Holekamp et al. 19993), but mothers often
intervene without any apparent cub solicitation of aid. Later in development, cubs

start joining mothers when mothers intervene on their behalf. Later still, cubs

39

begin initiating attacks on conspeciﬁcs, and cubs often enjoy third party support
from their mothers or other allies when these attacks are directed at low-born
animals. This sequence of events appears to lead to rank reversals, ﬁrst with
peers and then with older juveniles and adults. Low-bom opponents ﬁrst cease
behaving aggressively toward high-born cubs, and later begin appeasing them.
Once these rank reversals have occurred, cubs appear to maintain their ranks,
not only by winning in dyadic encounters with Iow-bom conspeciﬁcs (Holekamp 8
Smale 1993; Smale et al. 1993), but also by obtaining third party support against
Iow-bom animals, and perhaps also by joining attacks on low-born animals.

It appears that the proximate mechanisms of maternal rank ‘inheritance’ in
spotted hyenas are strikingly similar to those found in cercopithecine primates.
In both groups, maternal intervention in disputes plays a key role in rank
attainment, with the mother asserting her dependent offspring’s status just below
her own. Coalitions also seem to play a strong role in rank ‘inheritance’ in both
primates and hyenas. In hyenas, coalitionary support by kin and non-kin appears
to reinforce aggression against ‘appropriate’ targets, whereas aggressive
coalitions serve to punish ‘inappropriately’ directed aggressive behavior by cubs.
Additionally, cubs may join in coalitions against Iow-bom individuals to reinforce
their own status. Genetic heritability is not an adequate explanation for maternal
rank ‘inheritance’ in either primates or hyenas. The key mechanistic difference in
rank ‘inheritance' between these taxa is that low-ranking cercopithecine juveniles
may experience a high degree of unprovoked aggression from high-born adult

females (eg., Horrocks 8 Hunte 1983), whereas low-ranking juvenile hyenas do

40

not receive such harassment. However, the frequency with which unprovoked
aggression occurs among adult female hyenas suggests that harassment might
. function in the maintenance of rank relationships in animals older than those
examined here. In any case, the fact that the mechanisms of maternal rank
‘inheritance' are so strikingly similar in Crocuta and cercopithecine primates
suggests that rank acquisition in both taxa might be mediated by common

cognitive processes.

41

Chapter 3
SOCIAL COMPLEXITY AND THE EVOLUTION OF INTELLIGENCE IN THE
SPOTTED HYENA

INTRODUCTION

Intelligence, deﬁned broadly by Kamil (1987) as “those processes by
which animals obtain and maintain information about their environments, and use
that information to make behavioral decisions,” has been of the subject of many
recent primate studies (reviewed in Byme 8 Whitten 1988; Tomasello 8 Call
1997; Matsuzawa 2001). Primates appear to be endowed with cognitive abilities
that are superior to, and qualitatively different from, those of most other
mammals. Two different types of selection pressures have been hypothesized to
favor the evolution of large brains and great intelligence in primates. The ﬁrst
hypothesis suggests that intelligence has been favored in primates by selection
pressures associated with complexity in the physical environment, particularly
that confronted when navigating through a three-dimensional arboreal world
(e.g., Povinelli 8 Cant 1995; Povinelli 8 Preuss 1995) or when ﬁnding and
obtaining food (e.g., Jerison 1973; CIutton-Brock 8 Harvey 1980; Milton 1981).
The second hypothesis suggests instead that key selection pressures have been
imposed by complexity associated with the ﬂexible behavior of conspeciﬁc group
members (Jolly 1966; Humphrey 1976; Byme 8 Whitten 1988). This social
complexity hypothesis is supported by interspeciﬁc comparisons suggesting that

the evolution of intelligence in primates has been more strongly inﬂuenced by

42

social pressures than by nonsocial aspects of the environment (reviewed in
Byrne 1994; Tomasello 8 Call 1997). In particular, brain size, which is roughly
correlated with tactical deception (Byme 1993), is strongly correlated with group
size, and other variables, such as territory size, explain any additional variance
(Dunbar 1992; Barton 1993).

One key area in which primates are thought to differ qualitatively from
other gregarious mammals is in their ability to form mental representations of
tertiary, or third-party relationships among conspeciﬁc group members
(Tomasello 8 Call 1997). These involve interactions and relationships in which
the observer is not directly involved. For example, female vervet monkeys
(Chlorocebus aethiops) respond to the distress call of an infant by orienting
toward the infant’s mother, indicating that they perceive an association between
the mother and infant (Cheney 8 Seyfarth 1980). Several primate species have
been shown to utilize information about the social relationships of conspeciﬁcs in
activities such as recruiting useful allies, challenging competitors, and reconciling
after ﬁghts (Bach'mann 8 Kummer1980; Cheney 8 Seyfarth 1989; Silk 1999).
Tomasello and Call (1997) hypothesize that the ability to recognize third-party
relationships is unique to primates, and furthermore, that this distinguishes their
mental abilities from those of all other animals. The ability to recognize third-party
relationships has not been demonstrated in non-primate mammals, even in
species living in complex, stable societies, such as elephants (Loxodonta
africana: Moss 1988), dolphins (Tursiops truncatus: Conner et al. 1992), or lions

(Panthera Ieo: Packer 1994). However, few published studies of these other

43

gregarious mammals have focused on cognitive abilities, and none to date has
inquired speciﬁcally whether these animals can recognize third-party
relationships. Therefore, it remains unclear whether the absence of evidence that
any of these mammals can recognize tertiary relationships means that this ability
is truly absent.

The current dearth of information about social cognition in non-primates
severely limits the generality of the social complexity hypothesis (Kamil 1987;
Harcourt 8 de Waal 1992). The social complexity hypothesis predicts that non-
primate animals that share with cercopithecine primates most salient features of
their social life should possess many or all of the same features of social
intelligence as do these primates, assuming that the same proximate
mechanisms are operating. Here we test this prediction, using behavioral data
from a group of free-living spotted hyenas (Crocuta crocuta), gregarious
carnivores that live in remarkably primate-like societies. Evidence for the
existence of shared cognitive abilities would suggest convergent evolution in
these two distantly related taxa and would lend support to the social complexity
hypothesis. In contrast, the failure to ﬁnd such evidence would suggest that the
social complexity hypothesis should either be rejected or restricted to primate
species.

The social lives of spotted hyenas are very similar to those of vervet
monkeys, baboons (Papio spp.), and other cercopithecine primates (Frank 1986;
Holekamp et al. 1999a). Like these primates, hyenas live in large, permanent

social groups, and group members cooperatively defend a communal territory

44

(Harcourt 8 DeWaaI 1992; Boydston et al. 2001). The composition of hyena
clans is much like that of most cercopithecine primate troops, in that both contain
multiple adult males and multiple matrilines of adult female kin and their
offspring. Within a group, or clan, of hyenas, as in many cercopithecine societies,
adults can be ranked in a linear dominance hierarchy, and an individual’s position
within the hierarchy determines its priority of access to resources (Wrangham 8
Waterman 1981; Tilson 8 Hamilton 1984; Frank 1986; East 8 Hofer 2001). In
both taxa, females are philopatric, whereas males disperse from their natal
territories (Cheney 8 Seyfarth 1983; Henschel 8 Skinner 1987; Pusey 8 Packer
1987; Smale et al. 1997). Juvenile hyenas acquire ranks immediately below
those of their mothers (Holekamp 8 Smale 1991), and they do so via the same
associative Ieaming mechanisms as those documented in cercopithecine
primates (Horrocks 8 Hunte 1983; Engh et al. 2000). Both hyenas and numerous
cercopithecine primates have been shown to reconcile after ﬁghts (Aureli 8 De
Waal 2000; Wahaj et al. 2001), and hyenas direct greeting behavior toward
higher-ranking individuals (East et al. 1993) just as vervet monkeys direct
grooming behavior toward higher-ranking individuals (Seyfarth 1980). Like vervet
monkeys, hyena kin associate more closely than non-kin, and individuals direct
afﬁliative behavior towards kin more frequently than towards non-kin (Seyfarth
1980; East et al. 1993; Holekamp et al. 1997). In addition, members of both taxa
prefer higher-ranking individuals over lower-ranking individuals as social
companions (Seyfarth 1980; Holekamp et al. 1997). This suggests that hyenas,

like monkeys, may recognize that some group members are more valuable social

45

partners than others. Finally, triadic and more complex interactions appear to
play an important role in the acquisition and maintenance of social rank in
hyenas (Zabel et al. 1992; Smale et al. 1993; Engh et al. 2000), as they do in
many cercopithecine primates (Cheney 1977; Datta 1986; Chapais 1992).

If the social complexity hypothesis is correct, then the similarities between
the social lives of hyenas and monkeys suggest that hyenas should demonstrate
abilities in the domain of social cognition that are similar to those of
cercopithecine primates. In particular, we wanted to determine here whether or
not hyenas can recognize third-party relationships. A previous attempt to inquire
whether hyenas understand tertiary relationships (Holekamp et al. 1999a) failed
to produce any evidence that they could do so; however, it was unclear in that
study whether the lack of evidence meant that the hyenas lacked the ability to
recognize third party relationships, or rather that they lacked the motivation to
demonstrate this ability in that particular test situation. In the experiments
presented here, we attempted to test hyenas in situations in which actions based
on their social knowledge could improve the ﬁtness outcomes of the actors.
Since coalitions play an important role in the acquisition and maintenance of
social rank (Zabel et al. 1992; Engh et al. 2000), we reasoned that triadic
interactions might reveal a great deal about the cognitive abilities of hyenas. We
expected that, if indeed hyenas can recognize third-party relationships based on
the social ranks of other hyenas, then they would be able to utilize this
knowledge in two ways. First, we predicted that hyenas would be able to exploit

changes in sub—group composition, such that they would increase rates of

46

aggression toward lower-ranking animals when potential allies arrived on the
scene. Second, we predicted that hyenas would be able to discriminate between
the ranks of two individuals engaged in a ﬁght, and that they would aid the
higher-ranking combatant, regardless of their own social ranks in relation to

those of the ﬁghters.

METHODS

We collected data from a single large clan of spotted hyenas in the Masai Mara
National Reserve, in southwestern Kenya. The study area is an open, rolling
grassland interrupted by seasonal creek beds, and grazed year-round by
resident populations of several species of ungulates. All data used in this study
were collected in over 39,000 observation sessions between June 1988 and
October 1998. Each observation session involved recording the behavior of all
hyenas visible within 200m of one another. Duration of individual observation
sessions ranged from 5 minutes to several hours. On average during the study
period, the clan contained 63.2 :I: 2.4 resident animals, including 18.5 :I: 1.1 adult
females, 14.4 :I: 1.4 adult immigrant males, and 31.3 :I: 1.1 subadults. Each hyena
(was identiﬁed by its unique pattern of spots and other natural marks (e.g., ear
notches) and sexed based on the morphology of the erect phallus (Frank et al.
1990). Ages of all hyenas born in the study clan since 1988 were known to :I: 7
days, as described previously (Holekamp 8 Smale 1993), and mother-infant
relationships were determined on the basis of nursing associations. For all

analyses reported here, we used only interactions between animals over six

47

menths old who were not close relatives. Hyenas less than 6 months old are
unlikely to have attained a stable social rank, and their knowledge of other
Individuals’ social ranks is likely incomplete (Smale et al. 1993). In addition, close
relatives, deﬁned here as mothers, offspring, and maternal siblings, often support
each other in disputes, regardless of the relative social ranks of their opponents
(Engh et al. 2000).

During each observation session, all aggressive and appeasement
behaviors (Kruuk 1972) were recorded as critical incidents (the all-occurrence
sampling of Altmann 1974). We determined the social rank of each clan member
based on its position in a matrix of appeasement behaviors displayed during
dyadic agonistic interactions (Martin 8 Bateson 1988). By convention, the most
dominant animal in the hierarchy was assigned a rank of 1.

Hyenas live in a ﬁssion-fusion society, in which sub-group composition is
constantly shifting (Kruuk 1972), and individual hyenas are frequently arriving to,
and departing from, subgroups. Here, we tested whether hyenas are able to
judge the relative ranks of pairs of conspeciﬁcs, as predicted by the social
complexity hypothesis. We ﬁrst used data collected before and after the arrival of
a new individual at a behavioral observation session to analyze hyenas’ strategic
use of potential allies. In a second analysis, we examined hyenas’ tendencies to
assist individuals engaged in on-going ﬁghts by joining with one combatant

against the other.

48

Initiation of ﬁghts after t_he arrival of potential allies

Our ﬁrst method of assessing hyenas’ knowledge of third-party rank
relationships involved determining whether hyenas were able to take advantage
of the arrival of potential allies. When aggression between two hyenas escalates,
one or more additional attackers often join the dyadic skirmish by forming a
coalition with one contestant against the other. Typically, animals joining
coalitions are all dominant to the victim of an attack (Zabel et al. 1992). Thus
when attempting to displace a larger subordinate animal from food, a hyena
might beneﬁt, for example, by delaying its attack until the arrival of a potential
coalitionary ally who is higher ranking than the target. We reasoned that, if
hyenas can recognize third-party rank relationships, then when a new individual
arrives at a scene, rates of aggressive behavior among hyenas already present
there should change according to the rank of the arriving individual relative to the
target animal. Since hyenas frequently join coalitions against individuals lower-
ranking than themselves, any arriving hyena whose rank is higher than that of the
target of the attack is a potential ally, regardless of whether the arriving hyena
ranks above or below the animal who initiates the aggression. If hyenas can
recognize third-party rank relationships, then either when an arriving hyena ranks
higher than both the initiator and its target or when it ranks intermediate to them,
the higher-ranking (initiator) of the original pair should increase its rate of
aggression against the lower-ranking (target) animal in order to take advantage
of the presence of a potential ally. Alternatively, if hyenas do not recognize the

relative ranks of their peers and use a simple rule of thumb when assessing

49

potential allies, such as “utilize individuals higher-ranking than yourself as allies”,
then the attack rate should increase when a higher-ranking individual arrives, but
not when an individual arrives whose rank position was intermediate between the
initiator and its target. Because animals rarely join coalitions against individuals
higher-ranking than themselves, we made no predictions about the behavior of
pairs of hyenas when lower-ranking conspeciﬁcs arrives.

All aggressive behavior observed during 10-minute intervals prior to, and
immediately after, the arrival of a new hyena was recorded during an observation
session, as was the rank of the individual arriving relative to all pairs of animals
already present in the session. For this analysis, we only used sessions in which
a single animal arrived. If additional individuals arrived or left the scene during
the 20 minutes surrounding the arrival of the focal animal, we excluded the
session from analyses. We carried this analysis out ﬁrst on a large dataset, which
included all observation sessions, regardless of the location and context (e.g.,
whether food was present or not). Additionally, to focus on those animals whose
motivation to make use of potential allies was greatest, we carried out the same
analysis on a restricted dataset that included only aggressive behaviors
preformed by individuals between 6 months and 2 years of age in the presence
of a food source. In this situation, the presence of a desired resource should
theoretically maximize the willingness of relatively small but high-ranking
aggressors to take advantage of newly-arrived help against larger, lower-ranking

hyenas.

50

We evaluated the initiation of ﬁghts in two contexts: 1) when the aggressor
ranked intermediate between its target and a potential ally (potential ally >
aggressor > target), and 2) when the aggressor was dominant to both its target
and a potential ally (aggressor > potential ally > target). For each context, we
compared the rates of aggression of individuals toward lower-ranking targets in
10-minute intervals before and after the arrival of a potential ally. Each individual
scored a point for the 10-minute interval (before or after the potential ally arrived)
during which it initiated higher rates of aggression associated with each observed
arrival, and scored a zero for the interval with the lower rate of aggression. If
there was no change in an individual’s rate of aggression between the two 10-
minute intervals, that individual received a zero score for both intervals. We then
compared before and after scores of individuals summed across all observation
sessions using a Wilcoxon matched-pairs test. The Wilcoxon test allowed us to
utilize repeated observations of individuals without the problems associated with
pseudoreplication.

Joining on-goingﬁgﬁ

Next, we tested whether individuals demonstrated knowledge of the
relative ranks of conspeciﬁcs when they joined on-going dyadic ﬁghts. We
expected that, if a hyena recognizes relative rank relationships, then it should join
ﬁghts in support of the higher-ranking individual, even when both contestants are
lower-ranking than itself. Alternatively, if hyenas don’t recognize relative rank
relationships, then high-ranking individuals should not necessarily support the

dominant individual involved in a ﬁght when both opponents are lower-ranking

51

than the joining hyena. Instead, they might simply join whichever member of the
pair initiates the aggression. This would be expected, for example, if hyenas
were merely being socially facilitated (sensu Zajonc 1965) to join a ﬁght based on
a rule of thumb such as “join whichever hyena is behaving aggressively” or “join
whichever hyena is winning.” Because hyenas rarely join coalitions against
animals dominant to themselves, we did not expect to detect a pattern among
joiners subordinate to both members of a ﬁghting pair.

In interactions in which a third party joined in an on-going ﬁght, we used
binomial tests to determine whether higher-ranking contestants received more
support than lower-ranking contestants. When a hyena joined in a ﬁght between
two others, we recorded whether it joined in support of the dominant or
subordinate individual. We only collected data from the ﬁrst individual to join a
ﬁght, even when other hyenas joined the ﬁght subsequently. Because we had

clear, directional predictions for each test, all statistics reported are 1-tailed.

RESULTS
Mon of ﬁghts aﬂerthejrrivgl of potentialgllies

We recorded data from 231 sessions in which a stable group of hyenas
was present 10 minutes before and 10minutes after the arrival of a new hyena.
Eighty-eight individuals in 114 of these sessions behaved aggressively during the
observation period. Forty-nine sessions involved an animal arriving at a scene

where there was at least one animal between 6 months and 2 years of age and

52

food was present. Of these 49 sessions, 26 included aggressive behavior by 25
different subadult hyenas.

When data were analyzed together from all animals in contexts both
where food was absent and where it was present, we observed a pattern that
appeared to support the “rule of thumb” hypothesis rather than suggesting that
hyenas can recognize third party relationships (Figure 3.1 ). That is, when the
individual arriving was higher-ranking than both members of a pair of opponents,
the dominant opponent increased its aggression rate against the subordinate
signiﬁcantly more often after than before the arrival (N=44, Z=1.627, P=0.050).
However, we observed no difference when the arriving individual ranked between
the two opponents (N=32, Z=0.219, P=0.414).

When we restricted our analysis to only those situations in which
exploiting knowledge about third party rank relationships would unambiguously
improve ﬁtness of aggressors, we observed a pattern of behavior consistent with
the hypothesis that spotted hyenas can recognize third-party rank relationships
(Figure 3.2). In situations where food was present and aggressors had not yet
reached adult size (<2 years old), we reasoned that hyenas would gain clear
beneﬁts by utilizing potential allies against larger, lower-ranking opponents.
Although this greatly reduced sample sizes, since youngsters receive relatively
little coalitionary support from non-relatives (Engh et al. 2000), the results of the
restricted analysis suggest that hyenas can and do strategically exploit
knowledge of third-party rank relationships. As in the initial analysis, the

dominant contestant tended to increase its aggression rate more often after the

53

1.25 -

 

 

 

 

 

 

 

 

_ before
after
'6
E 1.00 - *
'E |
\
8 Fl—
0 0.75 -
c
a
10.0
in
.E
a. 0.50 -
o
at
5
a: 0.25 -
E
0.00 _ . . ‘ '
'Domrnant to both Intermediate in rank
(arrrver>aggressor>target) (aggressoparriveptarget)

Rank of arriving hyena relative to contestants

Figure 3.1. Initiation of ﬁghts by hyenas in all contexts. Bars represent the mean
number of instances per hyena in which an individual’s rate of aggression was
higher either before (black) or after (gray) the arrival of a hyena that ranked
dominant to, or intermediate between, each pair of opponents. An asterisk
indicates a signiﬁcant difference (P<0.05).

 

 

 

 

 

 

 

 

   

 

 

1'25 ‘ - before
I: after
2 1.00 - p < 0'08 .,
an) I 7" '1
o 0.75 - '
c VIII
a ‘7'
m ,
o ,
=lll=
:
8 o 25
E .
0.00 " r i I 1
Dominant to both lnterrnediate in rank
(arrrver>aggressor>target) (aggressoparriveptarget)

Rank of arriving hyena relative to contestants

Figure 3.2. Initiation of ﬁghts by subadult hyenas in food-related contexts. Bars
represent the mean number of instances per animal in which an individual’s rate
of aggression was higher either before (black) or after (gray) the arrival of a
hyena that ranked dominant to, or intermediate between, each pair of opponents.
An asterisk indicates a signiﬁcant difference (P<0.05).

55

arrival of a higher-ranking individual than before (N=8, Z=1.414, P=0.078). In
contrast to the initial analysis, however, in ﬁtness-enhancing situations the
dominant contestant also increased its aggression rate signiﬁcantly more often
after the arrival of an individual ranking below itself but above its opponent (N=9,
Z=1.732, P=0.042). The increase in aggression rates was not due simply to the
arrival of an additional animal, because there was not a signiﬁcant increase in
aggression after the arrival of an individual lower-ranking than both contestants
(N=12, Z=-0.277, P=0.391).
Joining on-going ﬁghts

We used data from 114 coalitions to test whether individuals
demonstrated knowledge of the relative ranks of conspeciﬁcs when they joined
on-going dyadic ﬁghts. Hyenas supported the dominant individual in a ﬁght, even
when they outranked both contestants (Figure 3.3), suggesting that they
recognize third-party rank relationships. When an individual joining a skirmish
was intermediate in rank between the two contestants, it inevitably supported the
dominant individual (N=28, Y=28, P<0.001). Similarly, when an individual lending
support was dominant to both ﬁghting animals, it almost always supported the
dominant animal (N=81, Y=75, P<0.001). Animals lower-ranking than both
members of a ﬁghting pair rarely supported either party, and when they did lend
support, they did not differentiate between the dominant and the subordinate
contestants (N=5, Y=3, P=0.50).

To assess whether hyenas were simply joining ﬁghts in support of

attacking animals regardless of their relative ranks, we restricted our next

56

 

 

   

 

 

 

 

 

80 ' __ Rank of supported contestant
- Dominant

7° ' l::r Subordinate
t 60 -
O
8
:l 50 ‘
on
I.-
O 40 -
rn
H t
5 30 -
IE
0
.E 20 -

10 n

0 r L W: ‘2 1
Dominant to both Intermediate in rank ‘
'oiner>ﬁ htin air dominant >. subordinate
0 g g p ) ( contestant 10in") contestant

Rank of joining hyena relative to ﬁghting pair

Figure 3.3. Coalitionary support in on-going disputes. Bars represent the number
of times that hyenas ranking dominant to both members of a ﬁghting pair, or
ranking intermediate between the members of the pair, joined in ﬁghts. Bar color
indicates whether supporters joined on the side of the dominant (black) or
subordinate (gray) contestant. An asterisk indicates a signiﬁcant difference
(P<0.05).

57

analysis to reversals, deﬁned here as situations in which a subordinate initiates
an aggressive attack against a dominant animal, and we found a pattern
suggesting that hyenas base their decisions to join skirmishes on the relative
ranks of the contestants (Figure 3.4). Reversals are quite rare; of 7,990 dyadic
aggressive interactions recorded over more than 10 years, only 177 (2.2%) were
reversals. Of these 177 reversals, 19 involved a third animal joining in support of
one contestant. When the animal joining the ﬁght was higher ranking than both
contestants, it joined in support of the dominant contestant, even though that

animal was losing the ﬁght at that time (N=18, Y=3, P<0.01).

DISCUSSION

Results from earlier attempts to assess hyenas’ cognitive abilities have
been equivocal. When hyenas move N cubs from one den location to another,
they make N + 1 trips, suggesting that they operate according to a rule of thumb
that says “revisit your old den until you ﬁnd no more of your cubs there”
(Holekamp et al. 2000). In addition, the cooperative hunting behavior of hyenas
and other carnivores, which often appears to involve intelligent coordination and
division of labor among hunters (e.g., Guggisberg 1962; Stander 1992), can also
be explained most parsimoniously by simple rules, such as “always keep the
selected prey animal between you and another hunter" (Holekamp et al. 2000). A
previous experiment in which hyenas' knowledge of third-party relationships was
directly investigated failed to provide any support for the social complexity

hypothesis. That is, in an experiment modeled after that performed by Cheney

58

2° ‘ Rank of supported contestant

- Dominant
. [:1 Subordinate

Incidents of support

 

o , l -_,—_
Dominant to both Intermediate in. rank
(joiner>target>initiator) (target>jomer>lnltlator)

 

Rank of joining hyena relative to fighting pair

Figure 3.4. Coalitionary support in attacks initiated by the subordinate member of
a ﬁghting dyad (reversals). Bars represent the number of times that hyenas
ranking dominant to, or intermediate to, ﬁghting pairs joined in disputes. Bar color
indicates whether supporters joined on the side of the dominant (black) or
subordinate (gray) contestant. An asterisk indicates a signiﬁcant difference
(P<0.05).

59

and Seyfarth (1980), Holekamp et al. (1999) played recordings of cub whoops to
groups of female hyenas and monitored reactions of the mother and other adult
(control) females. Mothers and relatives of the whooping cub responded strongly,
indicating that hyenas can recognize the whoops of individuals. However, control
females were no more likely to look toward the mother of the whooping cub after
the playback than before. One interpretation of these results is that hyenas do
not recognize tertiary relationships. An alternative explanation, however, is that
hyenas do recognize third party relationships, but that the orientation responses
of control females measured in that study were poor indicators of this ability.
Several lines of indirect evidence suggest that hyenas might, in fact,
possess some intellectual abilities similar to those of monkeys. First, neocortex
sizes of hyenas and other carnivores are correlated with group size and lie on the
same grade as do neocortex sizes of primates (Dunbar & Bever 1998). In other
words, hyenas have neocortices comparable in size to those of group-living
primates. Second, like many cercopithecine primates, the ranks of very young
hyenas are dependent on the presence or absence of their mothers (Holekamp &
Smale 1991). When the mother is absent, animals lower-ranking than the mother
sometimes behave aggressively toward the cub, but when the mother is nearby,
lower-ranking animals are rarely aggressive toward her cub. Because hyenas
treat these youngsters differently in the presence of their mothers than in the
presence of other higher-ranking adults, it may be that they are mentally
categorizing the association between the mother-cub pair. On the other hand, it

may be that the hyenas are simply responding to a discriminative stimulus. If they

60

distress the cub when its mother is present, they are likely to be counterattacked,
whereas bothering the cub in the absence of the mother results in no
punishment. Studies of long-tailed macaques (Macaca fascicularis: Dasser 1988)
and vervet monkeys (Cheney & Seyfarth 1980) indicate that these primates
understand something about the relationship between mothers and their
offspring. At present, we cannot discriminate between the hypothesis that hyenas
also understand mother-infant relationships and the hypothesis that dependent
rank is a phenomenon based upon simple punishment.

Finally, male preferences for high-ranking females suggest that male
hyenas are able to discern relative rank relationships among their prospective
mates. Even when female reproductive state was controlled for, immigrant male
hyenas, who always rank lower than adult females, associated more with high-
ranking females than with low-ranking females (Szykman et al. 2001). Further
evidence that males can distinguish female ranks comes from the fact that
immigrant males who are relatively recent arrivals to the clan do not distinguish
between high and low-ranking females, indicating that it may take them some
time to learn the females' relative ranks. Although this evidence indicates that
males can distinguish between high and low-ranking females, the males’
preferences for high-ranking females may be a result of using simple direct cues,
such as body condition or odors.

The results of the current study show that hyenas can and do recognize
third party rank relationships. In the ﬁrst test of the hypothesis, we found

evidence that young hyenas altered their rates of aggressive behavior in food-

61

related contexts when potential allies arrived who ranked between themselves
and the victims of their attacks. These young hyenas are usually smaller than
most of their lower-ranking competitors and may be taking advantage of the
arrival of larger conspeciﬁcs to maintain better access to food.

The behavior of hyenas who join on-going disputes also supports the
assertion that hyenas recognize tertiary relationships. When hyenas joined ﬁghts
between other hyenas, they almost always joined on the side of the dominant
animal, even when that animal was lower-ranking then they were. Zabel et al.
(1995) suggest that hyenas have a strong tendency to do what other hyenas are
doing and therefore that hyenas often join coalitions as a result of social
facilitation (Zajonc 1965) rather than based on an assessment of relative ranks.
Since most aggression in hyena society is directed toward lower-ranking
individuals, simply joining an aggressor is likely to result in the pattern observed
here, in which the dominant animal is aided farmore frequently than the
subordinate animal. However, when we looked at rare instances of rank
reversals, situations in which the initiator of aggression was lower-ranking than
the victim, animals that intervened in the ﬁght overwhelmingly came to the aid of
the dominant animal. This demonstrates that hyenas do indeed use knowledge of
rank relationships. They are not just following simple rules, such as “join in
support of aggressors” or “join whichever animal is winning.” Clearly, hyenas will
aid the dominant animal even when that individual is losing the ﬁght.

The evidence presented here shows that hyenas, like many primates,

can recognize the relative ranks of the members of their social groups, but these

62

data do not tell us how they mentally represent information about dominance
relationships. Tomasello and Call (1997) argue that non-primates may not
understand transitive rank relationships and instead memorize relationships
between all possible pairs of conspeciﬁcs. While we cannot rule out this
hypothesis, it seems unlikely that each hyena in our study clan has memorized
the outcomes of conﬂicts between 1891 potential pairwise combinations of clan
members (number of clanmates!/[(number of clanmates - 2)!(2!)]), and that it
accesses this large mental database when joining or recruiting allies. In over
30,000 hours of observation, we have not yet observed agonistic interactions
between all possible pairs of adult hyenas. It therefore seems reasonable to
expect that hyenas may not have observed all pairs interacting and may have to
make inferences about rank based upon patterns of relatedness, association, or
rank relationships with other individuals. Although there is no evidence that they
can recognize 3'd party relationships, laboratory rats (Rattus norvegicus) appear
to organize information about food sources hierarchically (Macuda & Roberts
1995), and it seems likely that hyenas may be utilizing similar mental capabilities
to store relevant information about social relationships.

Carnivore and primate lineages diverged over 90 million years ago and
are not sister taxa (Lillegraven et al. 1979, Waddell et al. 1999), yet in both taxa,
similar social environments are associated with similar cognitive abilities. This
convergent evolution provides support for the generality of the hypothesis that
social complexity led to the evolution of intelligence in primates and other

mammalian taxa. As with the data from primates, our hyena data are

63

correlational. Instead of social complexity promoting the evolution of intelligence,
it may be that intelligence is a prerequisite of social complexity, or that both

social complexity and social intelligence are correlated with some third variable of
major importance. Nonetheless, the results of this study raise the question of
whether the neural mechanisms underlying the abilities of hyenas and
cercopithecine primates to recognize 3rd party relationships are also convergent.
Additionally, these results suggest that we might expect to ﬁnd similar cognitive
abilities among elephants, cetaceans, and other mammals living in large,
complex societies in which individual ﬁtness is strongly inﬂuenced by social

dexterity.

64

Chapter 4
REPRODUCTIVE SKEW AMONG MALES IN A FEMALE-DOMINATED

MAMMALIAN SOCIETY

INTRODUCTION

Mammalian societies vary in the extent to which reproduction is skewed
toward one or a few socially dominant individuals. In societies organized by
hierarchical rank relationships, reproductive success (RS) among males is often
positively correlated with social rank, which in turn is largely determined by
ﬁghting ability. Rank-related reproductive skew among males has now been
documented with behavioral and molecular data in many group-living mammals
(ungulates: CIutton-Brock 1988; Pemberton et al. 1992; rodents: Sherman et al.
1991; primates: Altmann et al. 1996; Bercovitch & Numberg 1997; De Ruiter et
al. 1994; Smith et al. 1993; carnivores: Creel et al. 1997; Girrnan et al. 1997;
Grifﬁn 1999; Keane et al. 1994). In all of these species, and in most other
mammals as well, sexual selection has favored males that are larger, stronger,
and more pugnacious than females (Darwin 1871 ). Like societies of many other
gregarious mammals, social groups of spotted hyenas (Crocuta crocuta) are
structured by linear dominance hierarchies (Frank 1986; Holekamp & Smale
1990, 1993; Kruuk 1972; Mills 1990). However, in contrast to most other
mammals, spotted hyenas exhibit some striking sex-role reversals. That is,
among adult hyenas, females are larger and more aggressive than males, and

females dominate males in all dyadic contexts (Kruuk 1972; Frank 1986).

65

Furthermore, ﬁghting among adult male hyenas is less frequent and less intense
than is that observed among adult females (Frank 1986; Kruuk 1972; Mills 1990),
suggesting that male-male contest competition over mates may not function as
importantly in sexual selection in hyenas as it does in many other mammals.

Reproductive success among female Crocuta is strongly correlated with
social rank (Frank et al. 1995; Holekamp et al. 1996), but little is known about R8
in males. Limited observations of wild spotted hyenas have suggested that only
the highest-ranking immigrant male fathers offspring in this species (Frank 1986;
Mills 1990). However, copulations are rarely observed, so the extent of
reproductive skew among male hyenas remains unknown. Here we combine ﬁeld
observations with the use of genetic markers as indicators of paternity to
examine patterns of reproductive skew among male hyenas, as well as to test
predictions of competing models suggesting alternative relationships between
male rank and R8 in this species. The ﬁrst, null, hypothesis suggests that males
of all social ranks should be equally likely to father cubs. Second, a dominance-
based priority of access model predicts that the number of offspring sired by a
given male should depend on both his social rank and the number of females
simultaneously in estrus. In a less restrictive test of the priority of access model,
we inquire whether the highest-ranking male observed with a fertile female is the
male most likely to sire her cubs.

Spotted hyenas are gregarious carnivores that live in social groups, called
clans, composed of multiple matrilines of adult females and their offspring, as

well as one or more adult immigrant males. An individual's social rank determines

66

its priority of access to food (Frank 1986; Kruuk 1972; Tilson and Hamilton 1984).
Both male and female cubs “inherit” their mothers’ ranks; that is, they attain ranks
in the dominance hierarchy immediately below those of their mothers (Engh et al.
2000; Holekamp & Smale 1993; Smale et al. 1993). Females are usually
philopatric, whereas males invariably emigrate from their natal clans at 25 to 62
months of age (Frank 1986; Henschel & Skinner 1987; Holekamp & Smale 1998;
Smale et al. 1997). Adult natal males dominate adult females ranked lower than
their own mothers in the clan’s dominance hierarchy for as long as they remain in
their natal clans. However, when males disperse, they behave submissively to all
new hyenas encountered outside the natal area (Holekamp & Smale 1998;
Smale et al. 1993, 1997). Thus, by joining a new clan, each immigrant assumes
the lowest rank in that clan’s dominance hierarchy, and his status only improves
through attrition by death or secondary dispersal of the immigrant males that
arrived before him (Smale et al. 1997).

Because males typically remain in their natal clans for several months or
years after puberty, there are two classes of reproductively mature males in
every Crocuta clan: adult natal males born in the clan and adult immigrant males
born elsewhere. Since all natal animals outrank all immigrants, a male’s rank-
determined priority of access to food is inevitably worse after he emigrates from
his natal clan than before dispersal (Holekamp 8. Smale 1998; Smale et al. 1993,
1997). Nevertheless, virtually all males eventually disperse. If dispersal is
adaptive, then the ﬁtness beneﬁts of dispersing must exceed costs associated

with reduced access to food. The ultimate cause of male dispersal in this species

67

remains unknown. The most plausible hypothesis to account for a behavior
pattern with such dire nutritional consequences is that males may only be able to
mate outside of their natal clans. Therefore, in addition to using our genetic data
to examine patterns of reproductive skew among males, we also use them to test

this hypothesis.

METHODS

Study population
We conducted this study in the Talek area (1° 40's, 35° 50’E) of the

Masai Mara National Reserve. This is an area of open grasslands grazed year
round by large concentrations of several ungulate species. The subject
population was one large Crocuta clan inhabiting a home range of approximately
65 km2. We monitored Talek hyenas from June 1988 to January 1999.

We identiﬁed all hyenas in the Talek clan by their unique spot patterns,
and we determined their sex from the dimorphic glans morphology of the erect
phallus (Frank et al. 1990). All hyenas initially appearing in the Talek home range
as adults were photographed when ﬁrst sighted, and the day an individual was
ﬁrst seen in Talek was considered its date of arrival. Immigrant males remaining
in Talek for at least six months were called ‘resident immigrant males,’ whereas
males remaining for shorter periods were called ‘transients.’ The duration of
residence of any given immigrant male in the Talek clan was called his ‘tenure’ in
the clan. Between 1988 and 1999, the Talek study clan varied in size from 50 to
78 individuals, and usually contained 20 to 22 adult females (range: 16-28) and

theirjuvenile offspring. On average during the study period, 13.8 1 2.1 resident

68

immigrant males and 4.4 i 0.6 adult natal males were present in the clan. In
addition, 2 or 3 transient immigrant males appeared in the clan’s territory each
month and remained for shorter periods, usually only a few days (Smale et al.
1997)

Mother-offspring relationships were established on the basis of regular
nursing associations, which last up to 21 months in this species (Hofer & East
1995; Holekamp et al. 1996). We assigned birth dates to litters by estimating cub
ages when they were ﬁrst observed above ground at natal or communal dens.
Cub ages could be estimated to :l: 7 days based on pelage, size, and other
aspects of cub appearance and behavior (Holekamp et al. 1996). Male Crocuta
have viable sperm in their testes by 24 months of age (Matthews 1939), and we
therefore considered natal males older than 24 months to be adults. We
estimated the ages of immigrant males based upon the degree of tooth wear
(Van Horn et al. in review).

We conducted behavioral observations from vehicles between 0530 and
0900 h and between 1700 and 2000 h. During the study period, we observed
hyenas for over 30,000 person hours during 35,040 observation sessions. We
deﬁned an observation session to include all of the time spent watching hyenas
located within 200 m of one another. We identiﬁed all individual hyenas present
in these sessions, and we excluded from analyses any sessions in which one or
more unidentiﬁed hyenas were present. Throughout the study period, all
mounting behavior and complete copulatory sequences (those involving

intromission and ejaculation) were recbrded during critical incident sampling (the

69

‘all occurrence sampling’ of Altmann 1974).

We determined social ranks of individuals based on outcomes of several
thousand dyadic agonistic interactions during which one individual exhibited
appeasement behavior to the other, as described previously (Holekamp & Smale
1990; Smale et al. 1993). All resident immigrants present in the clan concurrently
were organized into a matrix based on direction, not number, of interactions
(Martin & Bateson 1988). The resident immigrant matrix was composed of 997
dyadic agonistic interactions in which 38 interactions (3.8%) were inconsistent
with assigned ranks. By convention, the highest-ranking immigrant, who was able
to win against all other immigrants in agonistic interactions, was called the ‘alpha
male,’ and was assigned a rank of 1. The period during which any male held
alpha status was referred to as his ‘reign.’ Due to male attrition and the arrival of
new immigrants, not all resident males present in Talek during a given reign were
necessarily there concurrently. For tenure-based analyses, only resident
immigrant males whose tenure began after the start of the study period were

used.

Sample collectjon and patemity analysis
Most DNA samples originated from venous blood collected while hyenas

 

were anaesthetized with Telazol (2.5 mg/kg) delivered from a COz-powered
darting riﬂe (T elinject Inc, Saugus, CA). Most cubs were darted between 8 and
12 months of age, and immigrant males were darted as early in their tenure as
possible. We collected body and tooth measurements from all immobilized
hyenas. We used a Puregene (Gentra Systems, Minneapolis, MN) kit to extract

DNA from blood shortly after collection, and then stored the DNA in liquid

70

nitrogen. We also collected tissue samples from dead hyenas whenever we
found them. DNA was isolated from tissues using either standard phenol-
chloroforrn extraction techniques or Qiagen kits (Qiagen, Inc, Bothel, WA). We
genotyped sampled individuals using eleven autosomal microsatellite loci and
one X-linked microsatellite locus (Funk & Engh in review; Libants et al. 2000;
Tables A1, A2 8 A3 in Appendix A of this dissertation). Across the 12 loci, on
average, there were 5.1 alleles/locus (range: 2-8), and the expected
heterozygosity was 0.62. We estimated typing error (1.0%) from the frequency of
mother-cub mismatches. Overall exclusion probability was 0.998.

All resident immigrant, transient, and unrelated adult natal males present
in the clan at conception of a particular litter were considered as potential fathers.
Males were considered unrelated to particular females when their maternal
coefﬁcients of relatedness, calculated from genealogies, were less than 0.125.
We employed a maximum likelihood-based approach (Thompson 1975; Meagher
1986) to assign paternity to cubs with known, genotyped mothers using the
program CERVUS (Marshall et al. 1998). We used the following parameters in
CERVUS: 85 candidate males (total number of resident immigrant males + adult
natal males + sampled transient males), 92% of candidate males sampled, 91%
of loci typed, and a 1% error rate. A male was considered the father of a cub
when CERVUS assigned him at 95% conﬁdence. Conﬁdence levels were derived
based upon population allele frequencies at each locus and represent levels of
tolerance of false paternity that would be assigned to males whose genotypes

are consistent with those of cubs by chance alone (see Marshall et al. 1998 for

71

details).

Modeling the relamshjp between male rank and male R§
We used results of our paternity analysis to examine predictions of two

mutually exclusive models, each suggesting a different relationship between
male rank and R8 in Crocuta. The ﬁrst hypothesis suggests that males of all
social ranks should be equally likely to father cubs. A second, dominance-based
priority of access model predicts that the number of offspring fathered by a given
male should depend on both his dominance status and the number of females
simultaneously in estrus (Altmann et al. 1996). Since a male can monopolize only
one fertile female at a time, the priority of access model predicts that the alpha
immigrant male will father all offspring conceived when only one female in the
group is in estrus, the alpha and beta males will equally share paternity of cubs
sired when two females are simultaneously in estrus, and so on (Altmann et al.
1996). Conception dates of hyena litters were calculated by subtracting the
gestation period of 110 days (Schneider 1926) from birth dates. The 2-week
interval surrounding each conception date was considered the fertile period
during each breeding cycle of each female hyena. In addition, we determined
whether the highest-ranking male present with a particular female during her
fertile period was the male most likely to father her cubs. To do so, we recorded
the identities and social ranks of all males found with that female during her
fertile period. Finally, to evaluate whether access to females was affected by
male rank, we calculated the percent of all adult Talek females that were
encountered by each immigrant male per observation session during 26 two-

week intervals in 1994 as [(number of females encountered/number of females

72

present in the clan)/number of sessions each male observed]*100.

Statistical analysis
Correlations between behavioral variables or measures of male RS and

male social rank were evaluated with Spearrnan’s correlation coefﬁcients using
SYSTAT software. Other correlations were assessed using Pearson’s R.
Probabilities of fathering cubs were compared for adult natal and adult immigrant
males using Chi-square tests, as were observed and predicted values obtained
in testing the priority of access model. We compared reproductive rates of age-
matched immigrant and adult natal males with a Mann-Whitney U test, and we
used Analysis of Variance to evaluate differences in measures of RS among
males based upon total tenure and highest rank achieved. For analyses in which
the dependent variable was a proportion (percent of assigned cubs sired), we
used an arcsine-square root transformation to normalize the distribution of the
data. Differences between groups were considered signiﬁcant when P<0.05.
Please note that the data presented in ﬁgures 4.1 and 4.6 represent the
same male at more than one rank or in more than one year. Almost all resident
immigrant males were present for several years (range=0.67-8.08) and occupied
several ranks (range=1-12) during their tenure in the clan, and we believe that
using data from these same individuals over multiple years and ranks presents a
more complete view of long term male reproductive success than simply
presenting a “snapshot” of RS at only one point in a male’s life. However, to
avoid pseudoreplication, we include additional analyses in which individual males
are represented only once or repeated measures where each male is ‘

represented over the same time/rank span.

73

RESULTS

Paternity assignment and multiple patemig
DNA samples from 199 hyenas were collected between December, 1990

and August, 2000, including 85 natal females, 74 natal males, 5 transient males,
and 35 resident immigrant males. Between December, 1990 and January, 1999,
each of eight different resident immigrant males held alpha status in Talek for 1
to 34 months. DNA was sampled from at least 92% of all resident immigrant and
adult natal males during this period, and paternity was veriﬁed in 75 (68.2%) out
of 110 cubs sampled (Table 4.1). Twelve of the 75 cubs assigned fathers had
mismatches with their putative fathers at one locus. The rate of cub-father
mismatches is 0.97%, which is similar to the 1.0% mismatch rate for known
mother-cub pairs. No male assigned paternity by CERVUS at 95% conﬁdence
had more than a single mismatch with his putative offspring.

Seventeen cubs for which paternity could be assigned were from singleton
litters, and 58 were from 31 twin litters. Of 14 twin litters in which sires could be
assigned for both littermates, three (21.4%) were fathered by two different males,
and 11 (78.6%) were fathered by single males. In 5 (38.5%) of an additional 13
twin litters in which only one cub could be assigned a father, the assigned father
could be excluded from siring the other cub at 2 or more loci. Because litter size
varied from 1 to 2 cubs and the proportion of cubs in a litter sired by a single
male differed in this population, all analyses below use individual cubs rather

than litters as the units of analysis unless otherwise indicated.

74

@cNemm §8N @034 me o: 2: 89.58. A538 NetwomENF j<

 

 

Agog "N 0 cs; F .N 8 5 av 082 ANNV $8 3:535 z“.
308: N o o N .5 8 a: 039 :Nv $8 385-42%? 5
cit m @0va N o N Na N E $8 av $2: ELFNLEQNNE EN
258: N o o N v o 6v $8? as $8? 8:89853 <20
A058: m o o N v m 6V $02 E $8? BENENQNN .5: ,
3.58 N o €33 N w m o E #8 a: OBS BENNNQNB 2m.
cs8: N o o N m w E #3 a: so? N22955:: mas.
Asma m o 3.; 5 F a 2 2 AN: #3 a: $8? 553825? 22
mm_mE
2995:: 835 g g
2869 ER: 29: 2053 20:80:00 8359... vanmm
2995: :26 3 22m >9 cccgmmm noEEmm $505. 3.9: Em: mEEQEé 22m mm o_mE

3 25m s new a 8.5 u 38 a 8.6 u 5 u :32 as 628”. .x. :92 so 923 22¢.

 

...mcm_w... m_mE 29m 590 9:3 £593 do mEmzmn. 6v can...

75

Male dispersal status
Of the 75 cubs for which paternity could be assigned, 73 (97.3%) were

fathered by immigrant males, and two (2.7%) by adult natal males (Table 1). No
cubs were found to be sired by transient immigrant males; however, only ﬁve
transients were sampled. Of 31 sampled cubs conceived during the 24 months
when all 5 natal and 22 resident immigrant males were sampled (August, 1995 —
September, 1997), 22 (71.0%) were fathered by immigrants and none were sired
by natal males. Twenty-nine of these 31 cubs (93.5%) could be assigned to
Talek males using CERVUS at 80% or greater conﬁdence, and the remaining
two cubs (6.5%) could not be assigned with conﬁdence to any Talek males.
Uncertain paternity of these two cubs could be due to their having been sired
either by unsampled transient males in the Talek area or by males encountered
during female excursions into surrounding territories. We observed several
transient males in the Talek home range each month (Smale et al. 1997), and
females occasionally made forays into surrounding territories. On average, 94.4
i 1.9% of all locations at which adult Talek females are found fall within the
clan’s territorial boundaries (Boydston et al. 2001).

Mean annual RS among immigrant males was 0.66 i 0.12 cubs/year
(range: 0 - 2.33; N=35), and was thus over ten times higher than mean annual
RS among adult natal males (mean=0.06 i 0.06 cub/year; range: 0-2.15; N=35).
Natal males were less likely to father any cubs at all than were immigrants (1 of
35 sampled adult natal males vs. 22 of 35 immigrants; X21=28.6, P < 0.001 ).
When we restricted our analysis of RS to immigrant males estimated to be less

than 3 years old, the mean annual RS of immigrant males (mean=0.33 :t 0.19

76

cubs/year; N=12) was still higher than that of adult natal males (Mann-Whitney
U1=255.0; P=0.02). It is therefore unlikely that adult natal males are simply too
young to reproduce. These data indicate that adult natal males generally
experience very low RS compared to that of males managing to join a new clan
after emigration from their natal groups.

Male rank and the priority of access model
Most immigrant males that eventually attained alpha status sired at least

one cub (range: 0-13 cubs; N=8 males), but these males sired an average of
87.5 i 11.0% of their sampled offspring before they attained alpha status.
Clearly, alpha males were not monopolizing reproductive opportunities (Table
4.1, Figure 4.1).

Social rank of resident immigrant males was positively correlated with their
annual reproductive rate (Figure 4.1; Rs=-0.31; P=0.041; N=35 males). In a
conservative analysis in which each male’s rate of yearly cub production was
calculated for a single, randomly assigned rank, we still found a positive
correlation between male rank and RS (Rs=-0.30; P<0.042; N=35). These data
allowed us to mle out the null hypothesis, which suggested that males of all
ranks were equally likely to sire cubs. However, although the correlation we
found between male rank and R8 was signiﬁcant, rank failed to account for even
half of the variance in male RS, and the pattern of RS observed among Talek
males failed to match that predicted by the priority of access model (Figure 4.2;
x21=17o.2, P<0.001).

One potential problem with the application of the priority of access model

to the ﬁssion-fusion society of the spotted hyena is that it assumes that males

77

N
l

10

 

 

14

13

.3
l
m

 

14
10

14

Number of cubs sired / year

12
, 15
12

0 5 10 15
Immigrant male rank

 

Figure 4.1. Mean (i SEM) reproductive rate of immigrant males at each rank
position. Numbers above bars represent the number of males at each rank
position.

78

50 -
0 Predicted by priority of access model

0 Observed

Number of cubs sired

 

 

 

Immigrant male rank

Figure 4.2. Observed paternity of 75 cubs in relation to male social rank
compared with the pattem of paternity predicted by a dominance-based priority of
access model (Altmann et al. 1996).

79

obtain perfect information about the whereabouts and reproductive status of
breeding females. In fact, however, higher-ranking immigrants typically
encountered a larger proportion of clan females per time interval than did lower-
ranking immigrants (Figure 4.3; R8 = -0.66; P <0.005; N=20 males). Thus high-
ranking immigrant males may be more efﬁcient than their lower-ranking peers
with respect to moving around the Talek territory or meeting up with multiple
females. On average, males are only observed with 22.7% of females during
each two week period, suggesting that even high-ranking males may not have
accurate information about the reproductive status of most clan females.

We next assessed the possibility that the sire of a female’s litter is likely to
be the highest-ranking male meeting up with her during her fertile period. We
recorded the behavior of 29 females during 53 fertile periods in which we
assigned paternity to the litter conceived. On average, each fertile female was
observed for 260 i 32 minutes in 11.4 i 1.4 sessions. In 10 (18.9%) of the 53
fertile periods, the highest-ranking male observed with the mother fathered at
least half of the litter she produced 110 days later. However, in 43 (81.1%) of
these fertile periods, the highest-ranking male observed with the female failed to
sire any of her cubs. In fact, the highest-ranking male observed with a fertile
female was no more likely to sire her cubs than any other immigrant male
observed with her (X21=1.37, P<0.25, N=64 cubs). On average during these
fertile periods, females were found with 2.51 d: 0.33 immigrants that were higher-
ranking, and with 1.17 i 0.19 unrelated adult natal males that were higher-

ranking, than the sires of the cubs conceived.

80

P
01
l

.0
.5

.o
co

.9
N

.0
u—L

 

Proportion of clan females encountered

 

 

.9
o

10 15
Immigrant male rank

C
(II

Figure 4.3. Mean proportion of all 21 adult females present in the Talek clan in
1994 that were encountered per observation session during each of 209 2-wk
long intervals by the immigrant males present in the clan during this period,
ordered on the basis of male rank in the immigrant hierarchy.

81

Copplatorv behavior
We watched Talek hyenas for over 30,000 person hours during the study

period, but we observed only 39 mounting bouts and 20 bouts of complete
copulation (Figure 4.4). Each bout typically included several mounts or complete
copulations over a period of hours. Five of 39 mounting bouts (12.8%) were
performed by ﬁve different adult natal males, and the remainder (87.2%) by 16
different immigrant males. Adult natal males were never observed to achieve
intromission, so all complete copulations involved immigrants. Neither the
number of mounting bouts performed (R8 = -0.38; NS) nor the number of
copulatiOns achieved (Rs=-.14; NS) was signiﬁcantly correlated with immigrant
male social rank (N=25 ranks), although this may have been due to the small
sample size of observed copulations. Alpha males achieved only 1 (5.0%) of 20
complete copulations observed, and over 50% of all observed copulations were
achieved by animals ranked fourth through sixth in the immigrant male hierarchy.
No observed mounts or copulations involved transient males. The overall trends
of the behavioral data are similar to those observed in the molecular paternity
data. During a single estrus period, Talek females were mounted by one to three
different males (mean =1.18 i- 0.08 males/estrus; N=33 estrus periods). On
average, each resident immigrant male seen copulating mated with 1.42 i 0.29
different females. Thus mating appears to be highly promiscuous in both sexes.
Ma_I_e rank and tenure

Immigrant male social rank in this species is highly correlated with arrival
order in the new clan (Pusey and Packer 1997; Smale et al. 1997), so we next

attempted to evaluate effects of immigrant male tenure on RS. Among immigrant

82

25-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I Copulations (N=20)

.0 D Mounts (N=34)

“2’ 20 —

(D

(D

8

a) 15 ‘-

.§

>

2

g 10 -

To

*5

t; 5 -

o\

O -_1 L
0 4 8 12 16

Immigrant male rank

Figure 4.4. Distribution of mounting and copulatory behavior across immigrant
male social rank.

83

males who fathered any cubs, the mean period between arriving in Talek and
fathering a ﬁrst cub was 20.8 i 4.0 months (range: 0 - 55.2; N=24 males), and
long term male R8 was positively correlated with total time in the clan (Figure
4.5; Rp = 0.82; P<0.001; N=24 males). One possible explanation for this pattern
is that males failing to reproduce soon after immigrating to Talek engaged in
secondary dispersal whereas successful males remained there longer. However,
this interpretation seems unlikely since immigrant males that remained in Talek
for several years were no more likely to have fathered cubs during their ﬁrst two
years in Talek than were immigrants that disappeared during their third year (2 of
6 vs. 3 of 8 males: X12 with Yates’ correction for small samples = 0.18; NS).
To disentangle the effects of rank and tenure in Talek on immigrant male

RS, we simultaneously regressed the transformed dependent variable of percent
of assigned cubs sired against total male tenure in the clan and highest social
rank achieved during that tenure. Whereas both of these independent variables
were signiﬁcantly related to percent cubs sired by immigrant males, total tenure
(P=0.02) explained 60% of the total variance in this model whereas highest rank .
achieved (P=0.45) explained only 18% of the variance (N=24 males). The rate at
which males produce cubs is low early in their tenure, then it increases after they
have resided in the clan for several years (Figure 4.6; F7,128=2.457; P=0.021).
Repeated measures analysis of the 3 males who were present in the clan for a
minimum of 5 years and 8 social ranks detected a similar relationship between
tenure and reproductive rate despite the small sample size and low power

(F4,8=4.25; P=0.04), but found no relationship between rank and reproductive

84

10

Number of cubs sired

 

 

 

I I l
0 1 2 3 4 5 6 7 8 9
Total years in clan

Figure 4.5. Total number of cubs sired by 24 immigrant males shown in relation
to their tenure in the Talek clan.

85

N
01
I

 

 

 

 

 

 

 

{3.
8 2.0—
>
\ 2'9
'0 ..
93
'5 1.5-
(D
.o
3
0
*5 1.0-
8
is”
3 0.5- 32
z 31
2 1
0.0 ' '
0 2 4 6 8 10

Year of immigrant tenure

Figure 4.6. Mean reproductive rate (iSEM) for immigrant males during each
year of their tenure. Numbers above bars represent the number of males
present during each year of tenure.

86

rate (F7,14=0.58; NS).

Tenure was not signiﬁcantly correlated with any measure of male body
size, including body length (Rp = 0.277, NS; N=35), skull length (Rp = 0.289, NS;
N=35) and shoulder height (Rp = 0.201; NS; N=35). Thus immigrant males who
had been present in the Talek clan for an extended period were not any larger
than those males that had arrived more recently, and the observed relationship
between male tenure and RS did not appear to be due to an effect of increasing

body size.

DISCUSSION

The spotted hyenas in our Kenyan study clan exhibit a promiscuous
mating system. Approximately one fifth of all twin litters are sired by multiple
males. Genetic paternity tests have revealed that multiple paternity similarly
occurs in other carnivore species (Eurasian badgers: Da Silva et al. 1994;
Ethiopian wolves: Gottelli et al. 1994; dwarf mongooses: Keane et al. 1994; lions:
Packer et al. 1991). The most important determinants of R8 in male hyenas
appear to be an individual’s dispersal status and the duration of his residence in
his new clan after dispersal. Adult natal male hyenas win all ﬁghts against
immigrants and show sexual interest in clan females (Holekamp & Smale 1998),
yet they seldom father any cubs. It seems unlikely that natal males are too young
to father cubs, since reproductive rates of natal males were signiﬁcantly lower
than reproductive rates of age-matched immigrant males. This suggests a

compelling ultimate explanation for natal (primary) dispersal behavior in Crocuta,

87

a behavior that inevitably leads to a substantial drop in social rank and hence in
priority of access to food. The current data suggest that dispersing males may be
trading superior access to food in their natal areas for superior access to mates
elsewhere. Male-biased dispersal is common in mammals (Chepko-Sade &
Halpin 1987; Dobson 1982; Greenwood 1980; Pusey and Packer 1987; Stenseth
& Lidicker 1992), and this behavior may have evolved in many species as a
mechanism to promote outbreeding, either directly via selection on males or
indirectly via female preferences for immigrant males as mates.

Among resident immigrant males, social rank is correlated with male RS,
but tenure is a far better predictor of RS than is rank. Our data do not conform to
predictions of a dominance-based priority of access model (Altmann et al. 1996),
and the data documenting associations between males and fertile females are
inconsistent with the hypothesis that the highest-ranking male present with a
female during her fertile period is most likely to sire some or all of her offspring.
Our results thus stand in striking contrast to those obtained from other
mammalian carnivores in which male RS increases with social dominance and
ability to win ﬁghts with other males (eg., sea lions: Campagna & LeBoeuf 1988;
Cape hunting dogs: Creel et al. 1997; raccoons: Gehrt & Fritzelll 1999; meerkats:
Grifﬁn 1999; kinkajous: Kays et al. 2000; wolves: Packard et al. 1985; polar
bears: Ramsay & Stirling 1986). The social lives of spotted hyenas are similar in
many respects to those of most cercopithecine primates (Frank 1986; Holekamp
et al. 1999a). However, most studies of paternity in wild cercopithecine primates

generally ﬁnd that, as in most other mammals (Ellis 1995), male social rank is

88

highly correlated with reproductive success (Altmann et al. 1996; de Ruiter et al.
1994; Melnick 1987).

Several different hypotheses might explain why high-ranking male hyenas
are not monopolizing reproduction. Social rank of male hyenas is not correlated
with body size, so unlike high-ranking males in other polygynous mammals, high-
ranking male hyenas cannot necessarily physically exclude lower-ranking males
from mating. In fact, ﬁghting among male hyenas is rare, and when it does occur,
it is usually restricted to low-level threats (Frank 1986; Kruuk 1972). The only
stage in the life history of this species in which males confront intense aggression
from other males is’ during the process of immigrating into a new clan, and male-
male ﬁghting is rare even under these circumstances. Whereas all males
disperse, few successfully immigrate. From 1988 to 1997, although 49% of 291
prospective immigrants remained in Talek for more than 1 month, only 15%
remained there for longer than 2 years (Smale et al. 1997). Aggression directed
toward potential immigrants might permit only the best ﬁghters to gain entry into
the clan, and successful immigrants might then be too well matched to risk
escalated ﬁghts. This scenario, however, seems unlikely, since most new
immigrants behave extremely submissively to both male and female residents
when trying to enter a new clan (Smale et al. 1997), and because substantial
variation in body size, and thus presumably in ﬁghting ability, exists within the
immigrant male hierarchy.

High-ranking male hyenas may not be able to monopolize reproduction

because they cannot adequately keep track of female location and reproductive

89

state. Hyenas live in ﬁssion-fusion societies in which subgroup composition
changes from day to day, or even hour to hour (East et al. 1993; Holekamp et al.
1997), so tracking multiple reproductive females might be difﬁcult. Our analysis
of male rank in relation to the proportion of clan females encountered shows that
high-ranking males have a slight advantage over low-ranking males with respect
to encountering females, so high-ranking males may fare better in scramble
competition over mates. However, unlike lions, hyenas do not appear to respect
“ownership” of estrous females by the ﬁrst male to mate (Packer & Pusey 1983).
Indeed, we have observed two males alternately mounting a single estrous
female (Szykman 2001). Furthermore, our behavioral data show that even when
high-ranking males are observed with fertile females, they are unlikely to father
cubs. This is unlikely to be the result of sperm depletion (Preston et al. 2001),
since estrus is asynchronous in female hyenas, and 60% of fertile periods do not
overlap at all with those of any other females (Szykman et al. 2001). Thus,
neither difﬁculty in ﬁnding receptive females nor inability to fertilize them can
satisfactorily explain the patterns of reproductive skew observed among male
hyenas.

Prohibitive energetic costs might keep high-ranking males from
monopolizing reproduction. We frequently observed males following particular
females for days or even weeks (Szykman 2001). These ‘consorting’ males may
undergo serious energetic stress unless they periodically leave the females to
feed. Female hyenas always have higher priority of access to food than

immigrant males, so it may be difﬁcult for a male to guard a female continuously

90

for extended periods of time without starving. Male hyenas with the highest
priority of access to food at ungulate kills may be best able to fortify themselves
for extended attendance of females, and thereby obtain an advantage during
male-male scramble competition (see also Bercovitch & Numberg 1996, 1997;
Bercovitch 1997). If this were true, however, we would expect to see a stronger
relationship between male rank and RS than between tenure and RS.

Finally, high-ranking males may not be able to monopolize reproduction if
females choose not to mate with them. Female hyenas are dominant to
immigrant males, and their unusual genital morphology makes forced copulation
impossible. Several lines of evidence suggest that female choice has important
effects on patterns of paternity in Crocuta. First, despite the fact that adult natal
males can always win in ﬁghts with immigrants, and although they express
sexual interest in Talek females, natal males rarely father any cubs. These
patterns suggest that females strongly prefer immigrants over natal males,
perhaps to avoid the deleterious consequences of inbreeding. Second, patterns
of association between Talek females and immigrant males show that the
highest-ranking male present with a female during her fertile period fathers her
cubs in only 19% of cases, and mid-ranking males do better than expected
based on their ability to win in either agonistic interactions or scramble
competition with other males. Third, RS varies substantially within most male
rank positions, and also within any given year of immigrant tenure in the Talek
clan, suggesting that some males present during each alpha reign are more

attractive to females than others. These data suggest that female choice of

91

mates may limit control of reproduction by high-ranking male hyenas.

Optimal skew models (e.g., Vehrencamp 1983; Reeve et al. 1998)
suggest that dominant individuals have full control of subordinate reproduction
and allow subordinates to breed only when necessary to entice them to remain in
the group and cooperate. Although Optimal skew models may be applicable to
some animal societies, they fail to account for observed partitioning of
reproduction within many vertebrate social groups, including those of other
mammalian carnivores (e.g., Clutton-Brock et al. 2001). Instead, these recent
empirical data are more consistent with predictions of limited control models of
reproductive skew, which predict that subordinates will breed when dominants’
capacity for control of reproduction is reduced. Similarly, the data presented here
for male spotted hyenas are more compatible with limited control than optimal
skew models, but control in this species is clearly limited by variables other than
those associated with male-male ﬁghting. Indeed, as suggested recently by
Koenig and Haydock (2001), interactions between males and females, including
incest avoidance and mate choice, may limit dominants’ control of reproduction.

In contrast to what occurs in most polygynous mammals, male-male
contest competition appears to have little inﬂuence over male reproductive
success in the spotted hyena. The highest-ranking (natal) males have very low
reproductive success, and high-ranking immigrant males cannot prevent lower-
ranking males from copulating with females. Rather than favoring big, ﬁerce
males, sexual selection in hyenas has apparently favored males who compete by

successfully immigrating, persisting in their new clans for several years, and

92

overcoming the energetic handicaps imposed by low social rank.

93

CHAPTER 5
SOCIAL AND DEMOGRAPHIC VARIABLES INFLUENCE PARASITISM.

IN A WILD CARNIVORE

INTRODUCTION

Various factors are known to inﬂuence the susceptibility of hosts to
parasitic infections, including age, sex, genotype, behavior, and habitat features
(Rau 1983; Wakelin 1985; Alexander & Stimson 1988; Wakelin & Blackwell 1988;
Keymer & Read 1991; Poulin 1996; Muller-Graf et al. 1999; MacDonald et al.
1999). Little is known, however, about how these variables interact to inﬂuence
the distribution of parasites in populations of wild vertebrates. For example,
although high-ranking captive house mice (Mus domesticus) are known to be
more susceptible to Babesia infections than are low-ranking mice (Barnard et al.
1993), social status and other behavioral variables are seldom considered in
relation to parasite prevalence among animals living in their natural habitats (but
see Hausfater & Watson 1976; Miiller-Graf et al. 1996). Because detailed
information‘about host populations is rarely available, few studies have
investigated the combined impact of social and demographic variables on
parasitism. Because both social and demographic variables may inﬂuence an
individual’s exposure and resistance to parasites, studies incorporating both
types of variables should yield a more complete understanding of parasite.

epidemiology in wild populations than is currently available.

94

Studies conducted to date with free-living animals have revealed that even
seemingly benign levels of infection can lead to dramatic decreases in
reproductive success and survivorship (Edwards & Barnard 1987; Moller1990;
Hudson et al. 1992). For example, in a study of snowshoe hare (Lepus
amen'canus) predation, Murray et al. (1997) found that hares treated with an
antihelminthic drug experienced a 17% reduction in annual mortality due to
predation compared to control animals. Because natural levels of parasitism may
have substantial impacts upon host survival, understanding patterns and
processes of parasiticinfections in wild animals is critical to our understanding of
their population dynamics. Here we use data collected during a 12-year study of
the behavioral ecology of free-living spotted hyenas (Crocuta crocuta) to examine
the effects of age, sex, season, social rank, and dispersal status on the

prevalence and intensity of three common species of hyena endoparasites.

METHODS
Study Animals

Spotted hyenas are gregarious carnivores that live in large groups, called
clans, structured by rigid dominance hierarchies. Although often thought of as
scavengers, hyenas are in fact capable hunters, and kill up to 96% of their prey
themselves (Kruuk 1972). Their diets include a wide variety of prey species,
ranging from invertebrates and passerine birds to wildebeest and zebras
(Holekamp et al. 1997; Cooper et al. 1999) (Table 5.1.). Hyenas frequently

compete with lions (Panthera Ieo), jackals (Canis mesomelas), and other

95

Table 5.1. Foods consumed by spotted hyenas in the Masai Mara National
Reserve, Kenya (data from Cooper et al. 1999).

Common Name

Scientiﬁc Name

 

Ungulates
Wildebeest
Thompson’s gazelle
Topi

Zebra

Impala

Grant’s gazelle
Hartebeest
Giraffe

Buffalo
Warthog
Hippopotamus
Dik dik

Other
Elephant
Springhare
l-Iare

Hyena
Jackal
Bat-eared fox
Mongoose
Amphibian
Bird
Invertebrate

Connochaetes taun'nus
Gaze/Ia thompsoni
Damaliscus Iunatus
Equus burchelli
Aepyceros melampus
Gazelle granti
AIcelaphus buselaphus
Giraffa camelopardalis
Syncerus caffer
Phacochoerus africanus
Hippopotamus amphibius
Madoqua kirkii

Loxodonta Africana
Pedetes capensis
Lepus capensis
Crocuta crocuta ’
Canis mesomelas
Otocyon mega/otis
Helogale parvula
Various

Various

Various

96

carnivores at kill sites (Kruuk 1972). As a result, they may be exposed to a
variety of parasites and pathogens from both their prey species and their
carnivore competitors.

Between 1998 and 1999, there were 26 adult females, 20 adult immigrant
males, 11 adult natal males, and 29 subadults in the clan. Individual hyenas were
identiﬁed based upon unique patterns of spots and natural markings (e.g., ear
notches). Although hyena females have a peniforrn clitoris, we could distinguish
males from females by differences in the morphology of the erect glans (Frank
Glickman & Powch 1990). Ages of all hyenas born in the clan were known to
within i 7 days, based upon characteristic developmental changes in juvenile
pelage and other features.

Social ranks were assigned to all clan members by organizing the
outcomes of dyadic agonistic interactions into a matrix based upon the direction
of interactions (as described in Smale et al. 1993). Individuals who exhibited
appeasement behavior during dyadic interactions were considered lower-ranking
than their partners. By convention, the highest-ranking animal was assigned a
rank of 1.

Male hyenas reach sexual maturity at 2 years of age (Matthews 1939),
and we considered all animals over 2 years old to be adults. Females generally
spend their entire lives in their natal clans, whereas all males emigrate from their
natal clans between the ages of 2 and 5 years (Smale et al. 1997). When males
immigrate into new clans, they join at the bottom of the rank hierarchy. Thus, all

natal animals (adult females and their offspring) are socially dominant to all

97

immigrant males (Smale et al. 1993).
Study Site & Sampling

Fecal samples were collected from a single, large clan of hyenas in the
Talek area of the Masai Mara National Reserve, Kenya. The Mara is an area
characterized by open, rolling grasslands grazed year-round by large
concentrations of ungulates. Between June 1999 and July 2000, we collected
205 fecal samples from 70 individuals (19 adult females, 19 adult immigrant
males, 9 adult natal males, and 23 subadults). All samples were collected from
known individuals observed defecating. In order to avoid pseudo-replication, we
used only one randomly selected sample from each individual for our analyses.
Parasite prevalence and intensity of infection were calculated using a modiﬁed
McMaster technique (Engh et al. in review; Appendix B of this dissertation).
Throughout the study, we recorded daily precipitation (in mm) and conducted
censuses of local ungulate populations twice each month, as described
previously (Holekamp et al. 1999b).
Meal analyses

Patterns of prevalence and intensity of infection were analyzed for three
common parasites within the population: Ancylostoma sp., Spirometra sp., and
Isospora sp. (Engh et al. in review; Table B1 in Appendix B). Prevalence, the
proportion of animals infected with a parasite, was analyzed using logistic
regression (SAS Proc Logistic, SAS Institute, 1988). Independent variables were
age, sex, social rank, month, and dispersal status (natal or immigrant). An initial

logistic regression was run using forward stepwise selection (p < 0.05) to

98

determine which individual variables should be included in the model. If more
than one variable was included, we entered appropriate 2nd order interactions
and tested for signiﬁcance (Hosmer and Lemeshow, 2000).

We used general linear models (SAS Proc GLM, SAS Institute, 1988) to
analyze intensity, the number of eggs of a particular parasite species present in
infected hosts, as well as total eggs counted and the total number of species
present in each sample (range=1-5). All egg counts were log (x + 1) transformed
to correct for overdispersion. We included all independent variables used in the
logistic regression, plus appropriate second order interaction terms. Interaction
terms that were not biologically feasible were not included in the model. For
example, we did not include dispersal status by sex or dispersal status by age
interactions, because all immigrants were adult males.

We calculated Pearson correlation coefﬁcients (rp) between parasite
prevalence and intensity and monthly rainfall total and mean ungulate counts. To
evaluate the relationship between social rank and intensity cf parasite infections,
we used Spearrnan correlations (rs). Analyses of dichotomous variables were

carried out using Student’s t-tests.

RESULTS
Egg copnts and species richness

The number of parasite species with which hyenas were infected was
signiﬁcantly related to the month of the year in which the sample was collected

(Table 5.2). Sample sizes were small in some months and may have resulted in

99

Table 5.2. Results of general linear modeling analysis on parasite intensity data.
An asterisk indicates a signiﬁcant relationship.

Variable All Eggs

F D
Social Rank 4.88 0.03*
Month 0.01 0.92
Age 0.28 0.59
Sex 0.00 0.97

Dispersal Status 0.06 0.81

Month* Rank 0.02 0.89
Month*Age 0.00 0.99
Month*Sex 3.93 0.05*
Month*Dispersal 0.12 0.73
Age*Sex 0.00 0.95

# 399

F p
2.37 0.12

4.67 0.03*

0.71 0.40
0.23 0.63
1.61 0.20
0.14 0.71
0.00 0.94
1.31 0.25
0.09 0.75
0.02 0.89

‘ Species
Ancylostoma Spirometra Isospora

F P

1.58 0.21
0.46 0.49
0.95 0.33
0.02 0.89
1.74 0.19
0.09 0.76
0.01 0.91
0.10 0.75
1.09 0.30
0.19 0.66

100

F p

F

p
5.26 0.02* 2.52 0.15

2.83 0.10
1.49 0.22
0.22 0.63
0.67 0.41
0.26 0.61
0.52 0.47
1.74 0.19
0.20 0.65
3.27 0.07

0.38 0.55
1.47 0.25
5.34 0.04*
0.04 0.85
1.32 0.28
1.85 0.21
0.74 0.41
0.99 0.34
0.01 0.99

a spurious relationship between month and species richness. There was no
correlation between mean monthly parasite richness and monthly rainfall (RP: -
0.023, d.f. = 11, p = 0.95) or mean monthly prey counts (Rp= -0.161, d.f. = 11, p=
0.62).

Total egg counts were higher in higher-ranking individuals (Table 5.2), but
this result was apparently strongly inﬂuenced by high Spirometra counts. On
average, the total egg count was 9601 i 1892 eggs per gram of feces, with 7037
i 1770 of those eggs identiﬁed as Spirometra. In addition to the effect of social
rank on egg burden, there was a marginally signiﬁcant effect of month and sex
on egg burden.

Ancylostoma

None of the variables tested appeared to affect the intensity of
Ancylostoma infections (Table 5.2). However, we found a signiﬁcant relationship
between hyena age and prevalence of infection with Ancylostoma (Table 5.3).
Ninety percent of all individuals and 100% of juveniles were infected.

Spirometra

The prevalence of Spirometra infection was strongly affected by dispersal
status (Table 5.3), with natal animals being over 4 times more likely to be
infected than immigrants (Odds Ratio = 4.20). Social rank was an important
determinant of the intensity of Spirometra infection (Table 5.2). Higher-ranking
individuals had higher intensities than did lower-ranking animals (Rs= -0.36, N =
52, p = 0.01) (Figure 5.1). In addition, there was a nearly signiﬁcant (p = 0.07)

age by sex interaction affecting intensity. The interaction was mainly a result of

101

Table 5.3. Results of logistic regression on parasite prevalence data. An asterisk
indicates a signiﬁcant relationship.

Species
Variable Ancylostoma Spirometra Isospora
x2 p x p x2 D
Social Rank 1.04 0.30 2.12 0.14 0.75 0.38
Month 1.35 0.24 0.00 0.96 1.81 0.17
Age 3.80 0.05* 0.28 0.59 0.28 0.59
Sex 0.00 0.93 1.17 0.27 4.91 0.02*

Dispersal Status 0.97 0.32 6.40 0.01* 0.29 0.58

102

Log (Spirometra count + 1)

 

 

2 a r r .. r I I

 

 

0 10 20 30 40 50

Social Rank

Figure 5.1. Log(x + 1) transformed Spirometra intensity in relation to hyena social
rank.

103

the high intensity of infections in sub-adult females (mean : s.e. = 3.99 :t 0.29)
compared to all other age/sex classes (sub-adult males: mean = 3.33 i 0.24;
adult females: mean = 3.17 i 0.22; adult males: mean = 3.29 i 0.16); however,
post-hoc pairwise comparisons revealed no signiﬁcant differences between
age/sex groups.
lsos ora

Both the intensity and prevalence of Isospora. infection were higher in
females than in males (Tables 5.2 & 5.3). Female hyenas were almost 3.5 times
(Odds Ratio = 3.47) more likely than males to be infected, and the mean intensity
of infection was signiﬁcantly higher in females than in males (female mean = 2.60

i 0.14; male mean = 2.05 i 0.05; t15= 2.67, p = 0.02) (Figure 5.2).

DISCUSSION

No single independent variable was able to predict differences in
susceptibility of hosts to three common parasite species. In fact, for each
parasite species, different variables had high predictive value. Certainly, no
single factor can be expected to account for the distribution of all parasites in the
ﬁeld (Price 1980). Instead, complicated interactions between host physiology and
behavior, in combination with the transmission biology of the parasite, are likely
to shape patterns of parasite prevalence and intensity.
Egg Coppts ang Species Richness

Although month of the year was related to the number of species present

in samples, there was no apparent relationship between monthly egg counts and

104

 

 

 

 

 

 

 

 

: T
+
*0
c
8
T
O 2 -
t‘.’
o
o.
(I)
o
2
v
8’
._r 1‘
0 I 1
female male

Sex

Figure 5.2. Log(x + 1)transformed Isospora intensity in male and female hyenas.

105

either rainfall or ungulate abundance. While both of these variables should
theoretically be expected to strongly affect exposure and resistance to parasites,
there may be a time delay between changes in environmental variables and
changes in gut fauna. In addition, the relationship between environmental
variables and parasite richness may be extremely complex, and thus may not be
evident in simple statistical tests.

The number of parasite eggs present in a sample was related to social
rank. This result, however, was strongly inﬂuenced by the Spirometra data.
Spirometra had the highest mean intensity of infection. Almost 75% of all eggs
recovered from fecal samples were Spirometra, so the fact that social rank
predicts egg intensity is largely due to the underlying relationship between
Spirometra and social rank. Although there was a statistically signiﬁcant
relationship between the month by sex interaction and total egg count, there was
no clear-cut pattern to the relationship.

Ancylostoma

Ancylostoma is a hookworm parasite that is extremely prevalent (90%)
among Talek hyenas (Engh et al. in review). Eggs are passed in feces, and
larvae usually infect their hosts by penetrating the skin and migrating to the
digestive tract. Young animals are frequently infected with Ancylostoma when
they ingest eggs passed in their mothers’ milk (Bowman 1995). Severe
hookworm infections can cause blood loss anemia, dermatitis, and pneumonia
(Roberts & Janovy 1996). We found no good predictors of intensity of

Ancylostoma infection. Most animals were infected, but the vast majority of

106

individuals maintained a low level of infection. However, prevalence of
Ancylostoma was higher in subadults than in adults. In fact, all subadults
sampled were infected. Hyena cubs spend the ﬁrst 6-10 months of life living in an
underground communal den and thus may be exposed to more larvae than
adults; however, all hyenas frequently sleep on bare soil, and hyenas rarely
defecate near the den site (pers. obs.). It therefore seems unlikely that young
hyenas are exposed to the parasite more frequently than adults. Instead, sub-
adult hyenas are probably more likely to be infected because they have not
developed acquired immunity to the parasite. Numerous studies have attributed
higher parasite prevalence in young animals to their naive immune systems (e.g.,
MacDonald et al. 1999; Miiller-Graf et al. 1999).
Spirometra

Spirometra, a pseudophyllidean tapeworm, was responsible for the bulk of
the total egg burden carried by hyenas. Spirometra eggs are passed in the feces
and hatch in water. The coracidium is eaten by copepod intermediates, which are
later eaten by additional intermediate hosts (Bowman 1995). In the Serengeti
ecosystem, Spirometra plerocercoids (spargana) are found in the muscles and
tendons of a variety of ungulates that serve as intermediate hosts (Sachs 8
Sachs 1968). Hyenas presumably become infected when they ingest the
muscles and viscera of infected prey animals. Although Spirometra is not
particularly pathogenic, severe infections can cause vitamin B12 deﬁciency

anemia (Bowman 1995).

107

To our surprise, we found that natal animals were far more likely than
immigrants to be infected with Spirometra. Immigrant males necessarily move
through multiple hyena territories during dispersal, whereas females spend the
vast majority of their time within their natal territory (Boydston et al. 2001);
therefore we expected that immigrant males would be more likely to be exposed
each parasite species. Additionally, immigrant males invariably have low social
status and thus experience poor priority of access to food (T ilson 8 Hamilton
1984). Presumably, poor access to food should inhibit immune function and
result in a higher likelihood of becoming infected when exposed to parasites. Due
to this presumed combination of greater exposure and lower resistance, we
expected that if any relationship between dispersal status and prevalence of a
parasite were to exist, infection would be more prevalent in immigrants. Instead,
infection was.less likely in immigrants. The explanation here may be derived from
concurrent consideration of both hyena behavioral ecology and the transmission
biology of Spirometra. Social rank determines a hyena’s priority of access to food
(Tilson 8 Hamilton 1984). Since immigrant males hold low ranks, they tend to
spend less time feeding on meat, and more time feeding on skin and bones than
do females (Heschel 8 Skinner 1987, 1990). As a result, immigrant males may
be exposed to fewer Spirometra spargana than are higher-ranking natal animals.

Alternatively, it may be that Spirometra is locally adapted to particular
clans of hyenas (e.g., Lively 8 Dybdahl 2000). If this were true, then immigrant
males who do not share the most common genotypes in their new clans might be

less susceptible to parasitic infection. This, however, seems unlikely, because

108

natal hyenas within a single clan have low average relatedness values, and there
appears to be little genetic distinction between clans (RC Van Horn, unpublished
data). Another possible explanation for this pattern is that the males that
successfully immigrate have superior abilities to fend off parasitic infection. Since
female hyenas rarely mate with natal males (Engh et al. 2002), if this second
hypothesis were correct, then sexual selection could be acting to favor
immunologically-superior males. It is possible that vitamin B12 deﬁciency anemia
from Spirometra combined with a lack of fresh meat in the diet of dispersing
males forces infected males to abandon attempts at dispersal.

The idea that natal animals have a higher Spirometra prevalence as a
result of greater exposure at ungulate kills ﬁts nicely with the intensity data. We
observed a correlation between intensity of Spirometra infection and social rank,
with higher-ranking individuals tending to have worse infections. This makes
intuitive sense because higher-ranking animals eat larger quantities of fresh meat
than lower-ranking conspeciﬁcs, and they also tend to have better access to
viscera and muscle tissue. Hausfater and Watson (1976) found a similar
relationship between high rank and high parasite load in male baboons (but see
Mtiller-Graf et al. 1996). Although Hausfater and Watson (1976) could not identify
the source of variation among baboons, they speculated that it was due to rank-
based differences in feeding behavior.

Sub-adult female hyenas tended to have more intense Spirometra
infections than any other age/sex class. Sub-adults, in general, may be less

resistant to Spirometra than are adults. MiJlIer-Graf et al. (1999) examined

109

Spirometra infections in lions and found the highest intensity of infection in
unweaned cubs, suggesting that sub-adults may be more susceptible to infection
by this parasite than are adults. A key physiological difference between sub-adult
male and female hyenas is that young females have higher levels of circulating
androstenedione, a weak androgen, than do sub-adult males (Lindeque et al.
1986; Glickman et al. 1992). This contrasts sharply with the pattern observed in
most mammals, in which males have higher androgen levels than do females.
Androgens have been implicated as immunosuppressants, and in numerous
studies, the higher parasite burdens observed in male animals have been linked
to higher androgen levels (Alexander 8 Stimson 1988; Poulin 1996). Thus, a
combination of poor resistance and androgen-mediated immunosuppression may
lead to exceptionally intense Spirometra infections in young female hyenas.
Isospora

Isospora is a coccidian protozoan that can denude gut mucosa, cause
bloody diarrhea, and be life-threatening, particularly when exacerbated by stress
(Roberts 8 Janovy 1996). The parasite is usually directly transmitted from fecal
contamination, but some forms of Isospora have paratenic hosts (Bowman
1995)

Female hyenas were both more likely to have Isospora, and have higher
loads of Isospora, than were males. Again, differences in feeding ecology may
explain this pattern. Hyenas rarely feed on fecal matter, but while they are
feeding on viscera, particularly the nutrient-rich liver and spleen, they are likely to

ingest some intestinal contents of their prey. As a result, they may be infected

110

from direct exposure to oocysts in their prey’s feces or by ingesting sporozoites
encysted in paratenic hosts. Because female hyenas are higher-ranking than
males, they tend to have access to the muscles and viscera of their prey more
often than do males (Henschel 8 Skinner 1987). Alternatively, physiological
differences between males and females may result in differential resistance to
Isospora. The energetic stress of maintaining pregnancy and lactation, and the
relatively high androstenedione levels of female hyenas compared to males
(Glickman et al. 1992) may make female hyenas particularly susceptible to
parasite infections. Certainly, high androgen levels among males have been
linked to higher levels of parasitic infection in males than in females of numerous
species (Klein 2000), and future studies combining the hormone levels and
parasite loads of individual hyenas may provide further evidence of androgen-
induced immunosuppression.
Implications

There is some debate about how well fecal egg counts represent actual
levels of parasitic infection (Keymer 8 Hioms 1986; Uhlinger 1993). The number
of eggs shed by parasites can be inﬂuenced by hormonal cycles (Hausfater 8
Watson 1976), parasite density (Anderson 8 Schad 1985), or rates of feeding
(Keymer 8 Hioms 1986). We were unable to verify whether the egg counts we
recorded here were accurate indicators of actual parasite loads in hyenas, but
evidence from other species suggests that fecal egg counts are correlated with

intestinal parasite load (Roberts 8 Swan 1981; Sithithawom et al. 1991).

111

The results of this study suggest a strong role for social behavior in
understanding the epidemiology of parasitic infections. The prevalence and
intensity of Spirometra infections were overwhelmingly inﬂuenced by dispersal
status and social rank, two social variables that have a huge impact on the diet of
an individual hyena. In addition, sex- and rank-related differences in feeding
behavior may also account for differences in Isospora infections between males
and females. While many studies examine parasitism in the wild, they rarely take
into account the impact of social behavior on individual susceptibility. This study
suggests that there is a strong link between social behavior and parasite

epidemiology.

112

APPENDICES

113

Table A1. Characteristics of microsatellite repeats used to assess paternity in

spotted hyenas.

Locus
CCR01
CCR04
CCR05
CCR07
CCR11
CCR12
CCR13
CCR14
CCR16
CCR17
CCRA3*

No. Alleles

ONNOD-k-thUIV-ﬁ

*CCRA3 is sex-linked

Allele Range Repeat Size Heterozygosity

191 -1 99
123-149
145-153
179-195
142-146
143-149
95-101

179-189
176-178
140-156
151 -1 66

APPENDIX A

wNNNNN-hNNNN

114

0.365
0.660
0.675
0.480
0.505
0.727
0.702
0.796
0.429
0.789
0.919

NV
mm

mm

mv
mm

mes
vm

me.
mm

mv
vm

0N?
vm

N?
mm

mm

mv
om

mv
vm

mv
om

mv
vm

our
vm

N_.

cm
00
we

om
mm

om
vm

cm P
V@

N?
mm

mm

mv
om

mv
vm

mv
om

mv
.3

our
vm

N?
mm

mm

mv
om

mv
wm

mv
om

mv

om?
vm

NF
mm

mm

mv
mm

mv
vm

mv
mm

new
vm

om _.
om

NF
mm

mm

mv

mv
vm

9.

no
em

om _.
vm

N?

on
00
we

cm
on

co
Va

, _.
our
vm

N?

on
om
mo

cm
mm

co
.3

cm?
va

9.
mm

mm

me
No

mv
vm

mv
N0

3»
vm

ON _.
#0

N?

on
00
mo

om
om

om
vm

on F
vm

N?

on
00
mm

cm
on

om
vm

owe
vm

m<m00 tmoo 9100 9850 #560 9.100 NFKOO :moo nomoo momoo vomOO vomoo

ease 2e:
“mecca

mEc.
9:8.
.852.
95...
aEck
gage.
oE_._.
aEm...
23950
N 226
“modem
m:5...
an...
_mccc<
oEc.
an...
.c__ee<
oE_._.
9:8.
93950
P 280
“modem
95...
an...
23950 .53...

9.98922

.m2___2mmo._o_E mac? .8 255980 p.98 cozomo. mun. .m.< can...

115

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
v.N.. FOP ...O.. ON? ON_. ONN.: ...O_. ON? FNF Om.m.. OO.N.. P.N..
.VO vO v.0 .NO vO YO vO vO .V.O vO YO YO

0.0 ONO ONO ON ON Oé ONO ON ON ON ON ON

ON ON... ON... ON ON ONO... ON... ON ON ON... ON... ON

O O O O O O O O O O .. O

O O O O O O O O O O ON O

ON ON ON ON ON ON ON ON ON ON O ON
méoo tmoo meoo OPKQO v5.00 mwmoo Nwmoo EEOC NOmOO OOKOO VOEOO wOmOO

<20 39.. ON
ON...
399922. .58.
ESE

3.25

92

8.25 ms. 26..
5:3
3.93225.

.2283. mun. .1 ON .0. 203.950 cozomom .N.< p.98.

116

APPENDIX B

A_n_alvsis of fecal samples

From June, 1999 to July, 2000, we collected 207 fresh fecal samples from
70 hyenas residing in a single social group in the Masai Mara National Reserve,
Kenya. Individual hyenas were recognized from their unique patterns of spots
and other physical characteristics, such as ear notches. All samples were
collected from individuals observed defecating. We used liquid displacement to
measure 1/2 ml of fecal matter into 7 ml of Sheather’s sugar water solution (Zajac
1994). The mixture was homogenized and centrifuged for 5 minutes at 7200 rpm,
then supernatant was transferred to a McMaster slide. All eggs and oocytes
observed within the gridlines were identiﬁed and the number of eggs per gram
was recorded. Because we analyzed eggs and oocytes rather than whole worms,
we were able to identify most parasites only to genus. In order to avoid pseudo-
replication in analyses of prevalence and intensity of infection, only a single,
randomly chosen sample from each individual was used.

Nine parasite taxa were identiﬁed in the 70 fecal samples (Table B.1),
including 1 spurious species (Nematodirus sp.). In 135 additional samples from
the same individuals, 2 more species, Tn'churis sp. and Moniezia sp. were
identiﬁed. Like Nematodirus sp., Moniezia sp. was probably spurious; that is, it
was a parasite of the hyenas’ prey rather than of the hyenas. Neither spurious
species was included in measures of parasite richness. All 70 individuals were

positive for at least one species of parasite. On average, each individual was

117

infected with 2.60 1 0.13 species (range: 1-5). Intensity of infection was
overdispersed (variance > mean) for most parasite taxa, so median levels of
infection are displayed in Table B.1.

Hookworms (Ancylostoma sp.) were the most common parasite, with 90%
of individuals infected. Graber and Blanc (1979) reported infections of
Ancylostoma duodenale in Ethiopian hyenas, and the eggs found in the current
study were consistent with their description. The spirurid eggs measured
approximately 36 x 12 pm and were very likely Spirocerca Iupi. S. Iupi is
frequently found in Kenyan dogs, but it was not observed in any of 3 hyenas
dissected by Brodey et al. (1977). We also found eggs of Toxocara sp. Although
Toxocara canis has been identiﬁed in hyenas by previous researchers (Baylis
1937), the dimensions of the eggs observed in the current study (62 x 60 pm)
more closely resembled those of Toxocara cati.

Spirometra sp. eggs were extremely common (74.3% of individuals
infected) and had the highest median intensity of infection. These eggs were
probably Spirometra preton‘ensis, which Nelson et al. (1965) and Graber and
Blanc (1979) found in hyenas. Pleurocercoids (spargana) of Spirometra are
relatively common in the hock joints and ﬂesh of several ungulate species that
hyenas feed upon (Sachs 8 Sachs 1968), and Spirometra sp. was the most
prevalent parasite in a survey of lion feces in the Tanzania (Millier-Graf 1995).
Several other types of cestode eggs were also identiﬁed. We found numerous
taeniid eggs, which could be Taenia crocutae, T. hyaenae, or Echinococcus

granulosus, all of which have been reported in hyenas (Mettrick 8 Beverley-

118

Burton 1961; Nelson et al. 1965). In addition, we found eggs of Mesocestoides
sp., which have not previously been recorded in hyenas but are known to infect
other African carnivores, including lions (Panthera Ieo), servals (Fe/is serval), and
carcals (Fe/is caracal) (Round 1968). In some samples, we found eggs of
Dipylidium sp., a common parasite of Kenyan dogs (Canis familiaris) and jackals
(Canis mesomelas) (Nelson et al. 1965), which was previously reported in
hyenas (Round 1968).

A single protozoan, Isospora sp. was observed. The size of the oocysts
(35 x 26 um) was intermediate between that of Isospora felis and I. Ieonina, two
species that have been recorded in African lions but not in hyenas (Patnaik 8
Acharjyo 1970; Bjork et al. 2000). The paucity of protozoan species recovered in
this study may have been a result of the concentration method used. Protozoans
tend to become distorted and desiccated in saturated sugar and salt solutions

(Zajac 1994). Finally, 2.9% of samples contained unidentiﬁed parasites.

119

Table 8.1. Prevalence and intensity (eggs per gram of feces) of parasite eggs
and oocytes in the feces of 70 wild spotted hyenas.

Parasite Prevalence Median Maximum Variance/Mean
(%) Intensity Intensity Ratio of Abundance

(e-p-e-l (e-p-g-I

 

Ancylostoma sp. 90.0 1000 17600 0.44
Spirometra sp. 74.3 3000 67200 1.07
Isospora sp. 25.7 200 4000 1.91
Dipylidium sp. 21.4 300 1200 2.00
Spirurida 15.7 200 300 1.90
Taeniidae 12.9 200 3400 2.36
Mesocestoides sp. 11.4 200 500 2.11
Toxocara sp. 5.7 200 1600 2.44
Nematodirus sp. 4.3 200 300 2.30
Unknown 2.9 1 50 200 -

Trichuris sp.* 0 0 0 -

Moneizia sp.* 0 0 0 -

 

*Trichun's sp. and Moniezia sp. were observed in additional samples from the

same individuals.

120

LITERATURE CITED

Alexander J, Stimson WH (1988) Sex hormones and the course of parasitic
infection. Parasitology Today 4:189-193

Altmann J (1974) Observational study of behaviour: sampling methods.
Behaviour 492227-267

Altmann J (1980) Baboon Mothers and Infants. Harvard University Press,
Cambridge, MA -

Altmann J, Alberts SC, Haines SA, Dubach J, Muruthi P, Coote T, Gefen E,
Cheesman DJ, Mututua RS, Saiyalel SN, Wayne RK, Lacy RC, Bruford
MW (1996) Behavior predicts genetic structure in a wild primate group.
Proceedings of the National Academy of Sciences 93:5797-5801

Altmann J, Hausfater G, Altmann SA (1988) Determinants of reproductive
success in savannah baboons, Papio cynocephalus. In: Clutton-Brock T
(ed) Reproductive Success. University of Chicago Press, Chicago, pp
403-418

Andelman SJ (1985) Ecology and reproductive strategies of vervet monkeys
(Cercopithecus aethiops) in Amboseli National Park, Kenya (PhD
dissertation). University of Washington, Seattle, WA

Anderson RM, Schad GA (1985) Hookwonn burdens and faecal egg counts: An
analysis of the biological basis of variation. Transactions of the Royal
Society of Tropical Medicine and Hygiene 79:812-825

Aureli F, De Waal FBM (2000) Natural Conﬂict Resolution. University of
California Press, Berkeley

Bachmann C, Kummer H (1980) Male assessment of female choice in
hamadryas baboons. Behavioral Ecology and Sociobiology 6:315-321

Barnard CJ, Behnke JM, Sewell J (1993) Social behaviour, stress and
susceptibility to infection in house mice (Mus musculus): effects of
duration of grouping and aggressive behaviour prior to infection on
susceptibility to Babesia microti. Parasitology 107:183-192

Barton RA (1993) Independent contrasts analysis of neocortical size and
socioecology in primates. Behavioral and Brain Sciences 16:694-695

121

Baylis HA (1937) Records of some helminths from the spotted hyaena. Annals
and Magazine of Natural History 20:438-441

Bercovitch F8 (1997) Reproductive strategies of rhesus macaques. Primates
38:247-263.

Bercovitch FB, Numberg P (1996) Socioendocrine and morphological correlates
of paternity in rhesus macaques (Macaca mulatta). Journal of
Reproduction and Fertility 107259-68

Bercovitch FB, Numberg P (1997) Genetic determination of paternity and
variation in male reproductive success in two populations of rhesus
macaques. Electrophoresis 18:1701-1705

Berman CM (1980) Early agonistic experience and rank acquisition among free-
ranging infant rhesus monkeys. lntemational Journal of Primatology
1 :1 53-170

Bernstein IS (1969) The stability of the status hierarchy in a pigtail monkey group
(Macaca nemestn'na). Animal Behaviour 17:452-458

Bernstein IS, Ehardt CL (1985) Agonistic aiding: kinship, rank, age, and sex
inﬂuences. American Journal of Primatology 8:37-52

Bjork KE, Averbeck GA, Stromberg BE (2000) Parasites and parasite stages of
free-ranging wild lions (Panthera [90) of northern Tanzania. Journal of Zoo
and Wildlife Medicine 31:56-61

Bowman DD (1995) Georgis’ Parasitology for Veterinarians, 6th Edition. WB
Saunders Company, Philadelphia

Boydston EE, Morelli TL, Holekamp KE (2001) Sex differences in territorial
behavior exhibited by the spotted hyena (Hyaenidae, Crocuta crocuta).
Ethology 107:369-385

Brodey RS, Thompson RG, Sayer PD, Eugster B (1977) Spirocerca Iupi infection
in dogs in Kenya. Veterinary Parasitology 3:49-59

Byme RW (1993) Do larger brains mean greater intelligence? Behavioral and
Brain Sciences 162696-697

Byme RW (1994) The evolution of intelligence. In: Slater PJB, Halliday TR (eds)
Behaviour and Evolution. Cambridge University Press, Cambridge, pp
223-265

Byme RW, Whiten A (1988) Machiavellian Intelligence. Clarendon Press, Oxford

122

Campagna C, Le Boeuf BJ, (1988) Reproductive behaviour of southern sea lions.
Behaviour 104:233-260

Carl EA (1971) Population control in arctic ground squirrels. Ecology 52:395-413

Chapais B (1983) Dominance, relatedness and the structure of female social
relationships in rhesus monkeys. In: Hinde RA (ed) Primate Social
Relationships. Sinauer Associates, Sunderland, Mass, pp 208-219

Chapais B (1992) The role of alliances in social inheritance of rank among
female primates. In: Harcourt AH, De Waal FBM (eds) Coalitions and
Alliances in Humans and Other Animals. Oxford Science, Oxford, pp 29-
59 '

Chapais B, Girard M, Primi G (1991) Non-kin alliances and the stability of
matrilineal dominance relationships in Japanese macaques. Animal
Behaviour 41 :481-492

Cheney DL (1977) The acquisition of rank and the development of reciprocal
alliances among free-ranging immature baboons. Behavioral Ecology and
Sociobiology 22303-318

Cheney DL, Seyfarth RM (1980) Vocal recognition in free-ranging vervet
monkeys. Animal Behaviour 28: 362- 367

Cheney DL, Seyfarth RM (1983) Nonrandom dispersal in free-ranging vervet
monkeys: Social and genetic consequences. American Naturalist 122:392-
412

Cheney DL, Seyfarth RM (1989) Reconciliation and redirected aggression in
vervet monkeys. Behaviour 1 10:258-275

Chepko-Sade BD, Halpin ZT (1987) Mammalian Dispersal Patterns. University of
Chicago Press, Chicago

CIutton-Brock TH (1988) Reproductive Success. University of Chicago Press,
Chicago

CIutton-Brock TH, Brotherton PNM, Russell AF, O'Riain MJ, Gaynor D, Kansky
R, Grifﬁn R, Manser M, Sharpe L, Mcllrath GM, Small T, Moss A, Montfort

S (2001) Cooperation, control, and concession in meerkat groups.
Science 291:478-481

Clutton-Brock TH, Guiness FE, Albon SD (1982) Red Deer: Behavior and

123

Ecology of Two sexes. University of Chicago press, Chicago

Clutton-Brock TH, Harvey PH (1980) Primates, brains and behaviour. Journal of
Zoology 190:309-323

Cooper SM, Holekamp KE, Smale L (1999) A seasonal feast: long-term analysis
of feeding behaviour in the spotted hyaena (Crocuta crocuta). African
Journal of Ecology 37:149-160

Conner RC, Smolker RA, Richards AF (1992) Dolphin coalitions and alliances.
In: Harcourt AH, De Waal FBM (eds) Coalitions and Alliances in Humans
and Other Animals. Oxford University Press, Oxford, pp 415-443.

Cords M (1987) Forest guenons and patas monkeys: Male-male competition in
one-male groups. In: Smuts BB, Cheney DL, Seyfarth RM, Wrangham
RW, Struhsaker TT (eds) Primate Societies. University of Chicago Press,
Chicago, pp 98-111

Craig JV, Ortman LL, Guhl AM (1965) Genetic selection for social dominance
ability in chickens. Animal Behaviour 13:114-131

Creel S, Creel NM, Mills MGL, Montfort SL (1997) Rank and reproduction in
cooperatively breeding African wild dogs: behavioral and endocrine
correlates. Behavioral Ecology and Sociobiology 8:298-306

Da Silva J, Macdonald DW, Evans PGH (1994) Net costs of group living in a
solitary forager, the Eurasian badger. Behavioral Ecology 5:151-158

Darwin C (1871) The descent of man and selection in relation to sex. Murray,
London

Dasser V (1988) A social concept in Java monkeys. Animal Behaviour 36:225-
230

Datta SB (1983) Patterns of agonistic interference. In: Hinde RA (ed) Primate
Social Relationships: an Integrated Approach. Oxford University Press,
Oxford. 99 289-297

Datta SB (1986) The role of alliances in the acquisition of rank. In: Else JG, Lee
PC (eds) Primate Ontogeny, Cognition and Social Behavior. Cambridge
University Press, Cambridge, pp 219-225

De Ruiter JR, Van Hooff JARAM, Scheffrahn W (1994) Social and genetic

aspects of paternity in wild long-tailed macaques (Macaca fascicularis).
Behaviour 129:203-224

124

De Waal FBM (1977) The organisation of agonistic relations within two captive
troops of Java monkeys (Macaca fascicularis). Zeitschrift fur
Tierpsychologie 442225-282

De Waal FBM ( 1982) Chimpanzee politics. Jonathan Cape, London

Dobson SF (1982) Competition for mates and predominant juvenile dispersal in
mammals. Animal Behaviour 30:361-369

Drickhammer LC, Vessey S (1973) Group changing in free-ranging male rhesus
monkeys. Primates 14:359-368

Dunbar RIM (1980) Determinants and evolutionary consequences of dominance
among female gelada baboons. Behavioral Ecology and Sociobiology
72253-265

Dunbar RIM (1992) Neocortex size as a constraint on group size in primates.
Journal of Human Evolution 20:469-493

Dunbar RIM, Bever J (1998) Neocortex size predicts group size in carnivores and
some insectivores. Ethology 104:695-708

East ML, Hofer H (1991) Loud-calling in a female-dominated mammalian society:
II. Behavioural contexts and functions of whooping of spotted hyaenas.
Animal Behaviour 422651-670

East ML, Hofer H (2001) Male spotted hyenas (Crocuta crocuta) queue for status
in social groups dominated by females. Behavioral Ecology 12:558-568

East ML, Hofer H, Wickler W (1993) The erect ‘penis’ is a ﬂag of submission in a
female-dominated society: greetings in Serengeti spotted hyenas.
Behavioral Ecology and Sociobiology 332355-370

Edwards JC, Barnard CJ (1987) The effects of Trichinella infection on intersexual
interactions between mice. Animal Behavior 352353-540

Ehardt CL, Bernstein IS (1992) Conﬂict intervention behavior by adult male
macaques: structural and functional aspects. In: Harcourt AH, De Waal
FBM (eds) Coalitions and Alliances in Humans and Other Animals. Oxford
University Press, New York, pp 83-111

Ellis L (1995) Dominance and reproductive success among non-human animals:
A cross-species comparison. Ethology and Sociobiology 162257-333

Engh AL, Esch K, Smale L, Holekamp KE (2000) Mechanisms of maternal rank
‘inheritance’ in the spotted hyaena. Animal Behaviour 60:323-332

125

Engh AL, Funk SM, Van Horn RC, Scribner KT, Bruford MW, Libants S,
Szykman M, Smale L, Holekamp KE (2002) Reproductive skew among
males in a female-dominated mammalian society. Behavioral Ecology
13:193-200

Engh AL, Nelson KG, Peebles CR, Hernandez AD, Hubbard KK, Holekamp KE
(In review) A coprological survey of the parasites of spotted hyenas
(Crocuta crocuta) in the Masai Mara National Reserve, Kenya. Submitted
to Journal of Wildlife Diseases.

Estrada A (1978) Social relations in a free-ranging group of Macaca arctoides. In:
Chivers DJ, Herbert J (eds) Recent Advances in Primatology. Behavior.
Academic Press, New York, pp 55-58

Feldhammer GA, Drickamer LC, Vessey SH, Merritt JF (1999) Mammalogy:
Adaptation, Diversity, and Ecology. McGraw-Hill, Boston

Frank LG (1986) Social organisation of the spotted hyaena (Crocuta crocuta). ll.
Dominance and reproduction. Animal Behaviour 35:1510-1527

Frank LG, Glickman SE, Powch I, (1990) Sexual dimorphism in the spotted
hyena. Journal of Zoology London 221 :308-313

Frank LG, Holekamp KE, Smale L (1995) Dominance, demography, and
reproductive success of female spotted hyenas. In: Sinclair ARE and
Arcese P (eds) Serengeti II: Conservation, Research, and Management.
University of Chicago Press, Chicago, pp 364-384

Funk SM, Engh AL (In review) Identiﬁcation of polymorphic microsatellite loci in
Hyaenidae using loci conserved among Camivora. Submitted to
Conservation Genetetics

Gehrt SD, Fritzell EK (1999) Behavioural aspects of the raccoon mating system:
determinants of consortship success. Animal Behaviour 572593-601

Girman DJ, Mills MGL, Geffen E, Wayne RK (1997) A molecular genetic analysis
of social structure, dispersal, and interpack relationships of the African
wild dog (Lycaon pictus). Behavioral Ecology and Sociobiology 40:187-
1 98

Glickman SE, Frank LG, Pavgi S, Licht P (1992) Hormonal correlates of
‘masculinization’ in female spotted hyaenas (Crocuta crocuta). 1. Infancy
to sexual maturity. Journal of Reproduction and Fertility 952451 -462

Goodall J (1986) The Chimpanzees of Gombe. Belknap Press, Cambridge, MA

126

Gotelli D, Sillero-Zubiri C, Applebaum D, Girman D, Roy M, Garcia-Moreno J,
Ostrander E, Wayne RK (1995) Molecular genetics of the most
endangered canid: The Ethiopian wolf, Canis simensis. Molecular Ecology
3:249-257

Gouzoules H (1975) Maternal rank and early social interactions of infant
stumptail macaques (Macaca arctoides). Primates 162405-418

Graber M, Blanc JP (1979) Ancylostoma duodenale (Dubini, 1843) Creplin, 1843
(Nematoda: Ancylostomidae) parasite de I’hyene tachetee Crocuta
crocuta (Erxleben), en Ethiopie. Revue d’ Elevage et Medecine Veterinaire
des Pays Tropicaux 322155-160

Greenwood PJ (1980) Mating systems, philopatry, and dispersal in birds and
mammals. Animal Behaviour 28:1140-1 162

Grifﬁn AS (1999) A genetic analysis of cooperative breeding in meerkats (PhD
dissertation). University of Edinburgh, Edinburgh, UK

Gross MR (1996) Alternative reproductive strategies and tactics: diversity within
the sexes. Trends in Ecology and Evolution 11:92-98

Guggisberg CAW (1962) Simba. Bailey Brothers and Swinfen, London

Harcourt AH (1988) Alliances in contests and social intelligence. In: Byme RW,
Whiten A (eds) Machiavellian Intelligence. Clarendon Press, Oxford, pp
1 32-1 52

Harcourt AH (1992) Coalitions and alliances: are primates more complex than
non-primates? In: Harcourt AH, De Waal FBM (eds) Coalitions and
Alliances in Humans and Other Animals. Oxford Science, Oxford. DD 445-
471

Harcourt AH, Stewart KJ (1987) The inﬂuence of help in contests on dominance
rank in primates: hints from gorillas. Animal Behaviour 35:182-190

Hausfater G (1975) Dominance and reproduction in baboons (Papio
cynocephalus): A quantitative analysis. Contributions in Primatology 7: 1-
1 50

Hausfater G, Watson DP (1976) Social and reproductive correlates of parasite
ova emissions by baboons. Nature 262:688-689

Henschel JR, Skinner JD (1987) Social relationships and dispersal patterns in a
clan of spotted hyaenas Crocuta crocuta in the Kruger National Park.

127

South African Journal of Zoology 22:18-24

Henschel JR, Skinner JD (1990) The diet of the spotted hyaenas Crocuta crocuta
in Kruger National Park. African Journal of Ecology 28269-82

Hofer H, East ML (1995) Population dynamics, population size, and the
commuting system of Serengeti spotted hyaenas. In: Sinclair ARE, Arcese
P (esd) Serengeti ll: Dynamics, Management, and Conservation of an
Ecosystem. University of Chicago Press, Chicago, pp 332-363

Holekamp KE, Boydston EE, Smale L (2000) Group movements in social
carnivores. In: Boinski S, Garber P (eds) On the Move: Group Travel in
Primates and Other Animals. University of Chicago Press, Chicago, pp
587-627

Holekamp KE, Boydston EE, Szykman M, Graham I. Nutt K, Piskiel A, Singh, M
(1999a) Vocal recognition in the spotted hyaena and its possible
implications regarding the evolution of intelligence. Animal Behaviour
582383-395

Holekamp KE, Cooper SM, Katona CI, Berry NA, Frank LG, Smale L (1997)
Patterns of association among female spotted hyenas (Crocuta crocuta).
Journal of Mammalogy 78:55-64

Holekamp KE, Ogutu JO, Dublin HT, Frank LG, Smale L (1993) Fission of a
spotted hyena clan: Consequences of prolonged female absenteeism and
causes of female emigration. Ethology 932285-299

Holekamp KE, Smale L (1990) Provisioning and food-sharing by lactating spotted
hyenas (Crocuta crocuta). Ethology 862191-202

Holekamp KE, Smale L (1991) Dominance acquisition during mammalian social
development: the ‘inheritance’ of maternal rank. American Zoologist
31 2306-317

Holekamp KE, Smale L (1993) Ontogeny of dominance in free-living spotted
hyaenas: juvenile rank relations with other immature individuals. Animal
Behaviour 462451 -466

Holekamp KE, Smale L (1998) Dispersal status inﬂuences hormones and
behavior in the male spotted hyena. Hormones and Behavior 332205-216

Holekamp KE, Smale L, Berg R, Cooper SM (1997) Hunting rates and hunting

success in the spotted hyaena (Crocuta crocuta). Journal of Zoology
(London) 24221-15

128

Holekamp KE, Smale L, Szykman M (1996) Rank and reproduction in the female
spotted hyaena. Journal of Reproduction and Fertility 108:229-237

Holekamp KE, Szykman M, Boydston EE, Smale L (1999b) Association of
seasonal reproductive patterns with changing food availability in an
equatorial carnivore, the spotted hyaena (Crocuta crocuta). Journal of
Reproduction and Fertility 116287-93

Hoogland JL (1979) Aggression, ectoparasitism, and other possible costs of
prairie dog (Cynomys spp.) coloniality. Behaviour 6921-35.

Horrocks J, Hunte W (1983) Maternal rank and offspring rank in vervet monkeys:
an appraisal of the mechanisms of rank acquisition. Animal Behaviour
312772-782

Hosmer DW, Lemeshow S (2000) Applied Logistic Regression, 2“d Edition. John
Wiley and Sons, Inc, New York

Hudson PJ, Dobson AP, Newborn D (1992) Do parasites make prey vulnerable
to predation? Red grouse and parasites. Journal of Animal Ecology
612681-692

Humphrey NK (1976) The social function of intellect. In: Bateson PPG, Hinde RA
(eds) Growing Points in Ethology. Cambridge University Press,
Cambridge. pp 303-317

Jenks S, Weldele M, Frank L, Glickman SE (1995) Acquisition of matrilineal rank
in captive spotted hyaenas: emergence of a natural social system in peer-
reared animals and their offspring. Animal Behaviour 502893-904

Jerison HJ (1973) Evolution of the Brain and Intelligence. Academic Press, New
York.

Johnson J (1987) Dominance rank in juvenile olive baboons, Papio anubis: the
inﬂuence of gender, size, maternal rank, and orphaning. Animal Behaviour
35:1694-1708

Jolly A (1966) Lemur social behavior and primate intelligence. Science 1522501 -
506

Kamil AC (1987) A synthetic approach to the study of animal intelligence.
Nebraska Symposium on Motivation 72257-308

Kawai M (1958) On the system of social ranks in a natural troop of Japanese
monkeys, I, II. Primates 12111-148

129

Kawai M (1965) On the system of social rank in a natural troop of Japanese
monkeys, I. Basic rank and dependent rank. In: Altmann SA (ed)
Japanese monkeys, A Collection of Translations. Privately published,
Atlanta

Kays RW, Gittleman JL, Wayne RK (2000) Microsatellite analysis of kinkajou
social organization. Molecular Ecology 92743-751

Keane B, Waser PM, Creel SR, Creel NM, Elliot LF, Minchella DJ (1994)
Subordinate reproduction in dwarf mongooses. Animal Behaviour 47:65-
75

Keymer AE, Hiorns RW (1986) Faecal egg counts and nematode fecundity:
Heligmosomoides polygyrus and laboratory mice. Parasitology 93:189-203

Keymer AE, Read AF (1991) Behavioral ecology: The impact of parasitism. In:
Toft CA, Aeschlimann A, Bolis L (eds) Parasite-Host Associations:
Coexistence or Conﬂict? Oxford University Press, Oxford. PP 37-62

Kikkawa J, Smith JNM, Prys-Jones R, Fisk P, Catterall C (1986) Determinants of
social dominance and inheritance of agonistic behavior in an island
population of silvereyes, Zosterops Iateralis. Behavioral ecology and
Sociobiology 192165-169

Kingdon J (1997) The Kingdon Field Guide to African Mammals. Academic
Press, San Diego

Klein SL (2000) The effects of hormones on sex differences in infection: from
genes to behavior. Neuroscience and Biochemical Reviews 51 :149-166

Koenig WD, Haydock J (2001) Dividing up the kids. Science 291:442-443

Kruuk H (1972) The spotted hyena: a study of predation and social behavior.
University of Chicago Press, Chicago

Lee PC, Oliver JI (1979) Competition, dominance, and the acquisition of rank in
juvenile yellow baboons (Papio cynocephalus). Animal Behaviour 27:576-
585

Libants S, Olle E, Scribner KT (2000) Microsatellite loci in the spotted hyena
Crocuta crocuta. Molecular Ecology 921443-1445 '

Lillegraven JA, Kielan-Jaworowska Z, Clemens WA (1979) Mesozoic Mammals:

The First Two-thirds of Mammalian History. University of California Press,
Berkeley

130

Lindeque M, Skinner JD, Millar RP (1986) Adrenal and gonadal contribution to
circulating androgens in spotted hyaenas (Crocuta crocuta) as revealed by
LHRH, hCG, and ACTH stimulation. Journal of Reproduction and Fertility
782211-217

Lively CM, Dybdahl MF (2000) Parasite adaptation to locally common host
genotypes. Nature 405:679-681

MacDonald DW, Anwar M, Newman C, Woodroffe R, Johnson PG (1999) Inter-
annual differences in the age-related prevalence of Babesia and
Trypanosoma parasites of European badgers (Me/es meles). Journal of
Zoology, London 247265-70

Macuda T, Roberts WA (1995) Further evidence for hierarchical chunking in rat
spatial memory. Journal of Comparative Psychology: Animal Behavior
21:20-32

Marshall TC, Slate J, Kruuk LEB, Pemberton JM (1998) Statistical conﬁdence for
likelihood-based paternity inference in natural populations. Molecular
Ecology 72639-655

Martin P, Bateson P (1988) Measuring Behavior: An Introductory Guide.
Cambridge University Press, Cambridge

Matsuzawa T (2001) Primate Origins of Human Cognition and Behavior.
Springer-Verlag, Tokyo

Matthews LH (1939) Reproduction in the spotted hyaena, Crocuta crocuta
(Erxleben). Philosophical Transactions of the Royal Society of London,
Series B 23021-78

Meagher TR (1986) Analysis of paternity within a natural population of
Chemaelin'um Iuteum. I. Identiﬁcation of most-likely male parents.
American Naturalist 1282199-215

Melnick DJ (1987) The genetic consequences of primate social organization: A
review of macaques, baboons, and vervet monkeys. Genetica 732117-135

Melnick DJ, Pearl MC (1987) Cercopithecines in multimale groups: Genetic
diversity and population structure. In: Smuts BB, Cheney DL, Seyfarth
RM, Wrangham RW, Struhsaker 'I'I' (eds) Primate Societies. University of
Chicago Press, Chicago, pp 121-134

Mettrick DF, Beverley-Burton M (1961) Some Cyclophyllidean cestodes from
carnivores in southern Rhodesia. Parasitology 512533-544

131

Mills MGL (1990) Kalahari Hyaenas: The comparative behavioral ecology of two
species. Chapman and Hall, London

Milton K (1981) Distribution patterns of tropical plant-food as an evolutionary
stimulus to primate mental development. American Anthropologist 832534-
548.

Missakian EA (1972) Genealogical and cross-genealogical dominance
relationships in a group of free-ranging monkeys (Macaca mulatta) on
Cayo Santiago. Primates 132169-180

Moller AP (1990) Effects of parasitism by the haematophagous mite on
reproduction in the barn swallow. Ecology 71:2345-2357

Moss CJ (1988) Elephant Memories. Houghton Mifﬂin, Boston, MA

Moss R, Watson A, Rothery P, Glennie W (1982) Inheritance of dominance and
aggressiveness in captive red grouse, Lagopus Iagopus scoticus.
Aggressive Behavior 821-18

Miiller-Graf CDM (1995) A coprological survey of intestinal parasites of wild lions
(Panthera Ieo) in the Serengeti and Ngorongoro Crater, Tanzania, East
Africa. Journal of Parasitology 812812-814

MiJller-Graf CDM, Collins DA, Woolhouse MEJ (1996) Intestinal parasite burden
in ﬁve troops of olive baboons (Papio cynocephalus anubis) in Gombe
Stream National Park, Tanzania. Parasitology 112:489-497

Miiller-Graf CDM, Woolhouse MEJ, Packer C (1999) Epidemiology of an
intestinal parasite (Spirometra spp.) in two populations of African lions
(Panthera Ieo). Parasitology 1 182407415

Murray DL, Cary JR, Keith LB (1997) Interactive effects of sublethal nematodes
and nutritional status on snowshoe hare vulnerability to predation. Journal
of Animal Ecology 662250-264

Nelson GS, Pester FRN, Rickman R (1965) The signiﬁcance of wild animals in
the transmission of cestodes of medical importance in Kenya.
Transactions of the Royal Society of Tropical Medicine and Hygiene
592507-524

Netto WJ, Van Hooff JARAM (1986) Conﬂict interference and the development of
dominance relationships in immature Macaca fascicularis. In: Else JG, Lee
PC (eds) Primate Ontogeny, Cognition, and Social Behaviour. Cambridge
University Press, Cambridge, pp 291-300

132

Packard JM, Seal US, Mech DL, Plotka ED (1985) Causes of reproductive failure
in two family groups of wolves (Canis lupus). Zeitscrift fur Tierpsychologie
69:24-40

Packer C (1979) Male dominance and reproductive activity in Papio anubis.
Animal Behaviour 27:37-45

Packer C (1994) Into Africa. University of Chicago Press, Chicago, IL

Packer C, Collins DA, Sindimwo A, Goodall J (1995) Reproductive constraints on
aggressive competition in female baboons. Nature 373260-63

Packer C, Gilbert DA, Pusey AE, O’Brien SJ (1991) A molecular genetic analysis
of kinship and cooperation in African lions. Nature 3512562-565

Packer C, Pusey AE (1983) Adaptations of female lions to infanticide by
incoming males. American Naturalist 1212716-728

Patnaik MM, Acharjyo LN (1970) Notes on the helminth parasites of vertebrates
in Baranga Zoo (Orissa). Indian Veterinary Medical Journal 472723-730

Paul A, Kuester J (1987) Dominance, kinship, and reproductive value in Barbary
macaques (Macaca sylvanus) at Affenburg Salem. Behavioral Ecology
and Sociobiology 212323-331

Pemberton JM, Albon SD, Guiness FE, Cluttonbrock TH, Dover GA (1992)
Behavioral estimates of male mating success tested by DNA ﬁngerprinting
in a polygynous mammal. Behavioral Ecology 3:66-75

Periera ME (1989) Agonistic interactions of juvenile savannah baboons ll.
Agonistic support and rank acquisition. Ethology 802152-171

Periera ME, Altmann J (1985) Development of social behavior in free-living non-
human primates. ln: Watts ES (ed) Non-human Primate Models for
Human Growth and Development. Alan R. Liss, New York

Poulin R (1996) Sexual inequalities in helminth infections: A cost of being male?
American Naturalist 147:287-295

Povinelli DJ, Cant JGH (1995) Arboreal clambering and the evolution of self-
conception. Quarterly Review of Biology 702393-421

Povinelli DJ, Preuss TM (1995) Theory of mind: evolutionary history of a
cognitive specialization. Trends in Neuroscience 182418-424

Preston BT, Stevenson IR, Pemberton JM, Wilson K (2001) Dominant rams lose

133

out by sperm depletion. Nature 4092681-682

Price P (1980) Evolutionary Biology of Parasites. Princeton University Press,
Princeton, NJ

Pusey AE, Packer C (1987) Dispersal and philopatry. In: Smuts BB, Cheney DL,
Seyfarth RM, Wrangham WR, Struhsaker TT (eds) Primate Societies.
University of Chicago Press, Chicago, pp 250-266

Pusey AE, Packer C (1997) The ecology of relationships. In: Krebs JR, Davies
NB (eds) Behavioural Ecology Blackwell Scientiﬁc, Oxford, pp 254-283

Ramsay MA, Stirling l (1986) On the mating system of polar bears. Canadian
Journal of Zoology 6422142-2151

Rau ME (1983) Establishment and maintenance of behavioural dominance in
male mice infected with Trichinella spiralis. Parasitology 862319-322

Reeve HK, Emlen ST, Keller L (1998) Reproductive sharing in animal societies:
reproductive incentives or incomplete control by dominant breeders?
Behavioral Ecology 92267-278

Roberts JL, Swan RA (1981) Quantitative studies of ovine haemonchosis. 1.
Relationship between faecal egg counts and total worm counts. Veterinary
Parasitology 82165-171

Roberts LS, Janovy J (1996) Gerald D. Schmidt 8 Lany S. Roberts’ Foundations
of Parasitology. 5th Edition. William C. Brown Publishers, Dubuque, Iowa

Rood JP (1980) Mating relationships and breeding suppression in the dwarf
mongoose. Animal Behaviour 282143-150

Round MC (1968) Checklist of the helminth parasites of African mammals.
Technical Communications of the Commonwealth Bureaux of
Helminthology no. 38, Commonwealth Agricultural Bureaux, Famham
Royal, UK

Sachs R, Sachs C (1968) A survey of the parasitic infestation of wild herbivores
in the Serengeti region in northern Tanzania and the Lake Rukwa region
in southern Tanzania. Bulletin of Epizootic Diseases of Africa 162455-472

Sade DS (1972) A longitudinal study of social behavior of rhesus monkeys. In:
Tuttle R (ed) The Functional and Evolutionary Biology of Primates. Aldine,
Chicago, pp. 378-398

Schaller GB (1972) The Serengeti Lion: A study of predator-prey relations.

134

University of Chicago Press, Chicago, IL

Seyfarth RM (1980) The distribution of grooming and related behaviors among
adult female vervet monkeys. Animal Behaviour 282798-813

Seyfarth RM (1981) Do monkeys rank each other? Behavioral Brain Science
42447-448

Seyfarth RM, Cheney DL (1984) Grooming, alliances, and reciprocal altruism in
vervet monkeys. Nature 308:541-543

Sherman PW, Jarvis JUM, Alexander RD (1991) The Biology of the Naked Mole-
Rat. Princeton University Press, Princeton

Silk JB (1987) Social behavior in an evolutionary perspective. In: Smuts BB,
Cheney DL, Seyfarth RM, Wrangham RW, Struhsaker TT (eds) Primate
Societies. University of Chicago Press, Chicago, pp 318-329

Silk JB (1999) Male bonnet macaques use information about third-party rank
relationships to recruit allies. Animal Behaviour 58:45-51

Silk JB, Samuels A, Rodman PS (1981a) Hierarchical organisation of female
Macaca radiate. Primates 22:84-95

Silk JB, Samuels A, Rodman PS (1981b) The inﬂuence of kinship, rank and sex
on afﬁliation and aggression between adult females and immature bonnet
macaques (Macaca radiate). Behaviour 782111-137

Sithithawom P, Tesana S, Pipigool V, Kaewkes S, Pairojkul C, Sripa B, Paupeiroj
A, Thaiklar K (1991) Relationship between fecal egg count and worm
burden of Opisthorchis-vivem‘ni in human autopsy cases. Parasitology
1 022277-281

Smale L, Frank LG, Holekamp KE (1993) Ontogeny of dominance in free-living
spotted hyaenas: juvenile relations with adult females and immigrant
males. Animal Behaviour 462467-477

Smale L, Nunes S, Holekamp KE (1997) Sexually dimorphic dispersal in
mammals: patterns, causes, and consequences. Advances in the Study of
Behavior 262180-250

Smith DG (1993) A 15-year study of the association between dominance rank
and reproductive success of male rhesus macaques. Primates 34:471-480

Smuts BB (1985) Sex and Friendship in Baboons. Harvard University Press,
Cambridge, UK

135

Stander PE ( 1992) Cooperative hunting in lions: the role of the individual.
Behavioral Ecology and Sociobiology 292445-454

Stenseth NC, Lidicker WZ (1992) Animal Dispersal: Smell Mammals as a Model.
Chapman and Hall, New York

Sugiyama Y (1976) Life history of male Japanese monkeys. Advances in the
Study of Behavior 72255-284

Szykman M (2001) Sexual behavior and male mate choice in the spotted hyena,
Crocuta crocuta (PhD dissertation). Michigan State University, East
Lansing, MI

Szykman M, Engh AL, Funk SM, Van Horn RC, Scribner KT, Holekamp KE
(2001) Association patterns among male and female spotted hyenas
(Crocuta crocuta) reﬂect male mate choice. Behavioral Ecology and
Sociobiology 502231 -238

Thompson EA (1975) The estimation of pairwise relationships. Annals of Human
Genetics 392173-188

Tilson RT, Hamilton WJ (1984) Social dominance and feeding patterns of
spotted hyaenas. Animal Behaviour 322715-724

Tomasello M, Call J (1997) Primate Cognition. Oxford University Press, Oxford

Uhlinger CA (1993) Uses of fecal egg count data in equine practice.
Compendium on Continuing Education for the Practicing Veterinarian
152742-749

Van Horn RC, McElhinny TL, Holekamp KE (In review) Age estimation and
dispersal in the spotted hyena (Crocuta crocuta). Submitted to Journal of
Mammalogy

Vehrencamp SL (1983) A model for the evolution of despotic versus egalitarian
societies. Animal Behaviour 312667-682

Weddell PJ, Okada N, Hasegawa M (1999) Towards resolving the interordinal
relationships of placental mammals. Systematic Biology 4821-5

Wahaj SA, Guse KR, Holekamp KE (2001) Reconciliation in the spotted hyena
(Crocuta crocuta). Ethology 1 0721 057-1 074

Wakelin D (1985) Genetic control of immunity to helminth infections. Parasitology
Today 1217-23

136

Wakelin D, Blackwell JM (1988) Genetics of Resistance to Bacterial and Parasitic
Infections. Taylor and Francis, London

Walters J (1980) Interventions and the development of dominance relationships
in female baboons. Folia Primatologia 34261-89

Walters JR, Seyfarth RM (1987) Conﬂict and cooperation. In: Smuts BB, Cheney
DL, Seyfarth RM, Wrangham RW, Struhsaker TT (eds) Primate Societies.
University of Chicago Press, Chicago, pp 306-317

Wrangham RW, Waterman PG (1981) Feeding behaviour of vervet monkeys on
A. tortilis and A. xanthophloea: with special reference to reproductive
strategies and tannin production. Journal of Animal Ecology 502715-731

Zabel CJ, Glickman SE, Frank LG, Woodmansee KB, Keppel G (1992) Coalition
formation in a colony of prepubertal spotted hyaenas. In: Harcourt AH, De
Waal FBM (eds) Coalitions and Alliances in Humans and Other Animals.
Oxford University Press, New York, pp 113-135

Zajac AM (1994) Fecal examination in the diagnosis of parasitism. In: Sloss MW,
Kemp RK, Zajac AM (eds) Veterinary Clinical Parasitology, 6th Edition,
Iowa State University Press, Ames, Iowa, pp. 3-88

Zajonc RB (1965) Social facilitation. Science 149:269-274

137

   

lllilillllillillllllll