K .‘
{:21 .

3%.”: .-

mam _
. ‘
a; :7 vary! ‘7'
«H.555
.3.» .v.‘
3P9! .5329
9.: ..
I
.

n "a“

.: a.
s..¥d.,...¥
3“}? {f

p

I
‘. v a!
2 : «9:1...

113..
:3?!»
:10:

15‘. st 12?.
. 'AO‘K‘JH b ’l‘& v}
u. I: ‘tX.V...(.=.:l.-l

 

”NW

gem-m! I”?'

an."

.‘hmfdf: J13.
ﬁx .2. .4

 

THESIS

, 1- LIBRARY
-. Michigan State
University

This is to certify that the
dissertation entitled

SCENT MARKING IN A HIGHLY SOCIAL MAMMALIAN
SPECIES, THE SPOTTED HYENA, CROCUTA CROCUTA

presented by
KEVIN ROBERT THEIS

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Zoology and Ecology,
Evolutionary Biology and
Behavior

 

 

ﬂu Professor’s Signéﬁre
Axle/(2f

Date

MSU is an Afﬁrmative Action/Equal Opportunity Employer

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IPrqlAcc&Pres/CIRCIDateDue,indd

SCENT MARKING IN A HIGHLY SOCIAL MAMMALIAN SPECIES, THE
SPOTTED HYENA, CROCUTA CROCUTA

By

Kevin Robert Theis

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Zoology and Ecology, Evolutionary Biology and Behavior

2008

ABSTRACT

SCENT MARKING IN A HIGHLY SOCIAL MAMMALIAN SPECIES, THE
SPOTTED HYENA, CROC U TA CROC UT A

By
Kevin Robert Theis

Previous research on the functions of mammalian scent marking has often focused
on marking by adult males of solitary species, in territorial contexts. However, it has
become clear that the functions of scent marking vary not only among species, but also
among different age, sex, reproductive and social classes within a single species. Here I
evaluated multiple functional hypotheses for scent marking ('pasting') in the highly social
spotted hyena, C rocuta crocuta. I used naturalistic observational data to ascertain the
social contexts in which pasting occurred, and conducted scent discrimination
experiments with free-living hyenas to determine the information content of paste.

In demonstrating that hyenas discriminate between the scents of clanmates and
others, I reaffirmed a territorial function for pasting. Pasting also appeared to function in
advertising dominance status for adult males, but not females. Rates of pasting among
males were dependent upon social rank, and males pasted most often when their social
partners were primarily other males. For adult females, pasting appeared to function in
maintaining social cohesion within clans. Female pasting rates were highest at dens; the
den is the social center of each hyena clan. Additionally, adult females investigated
female more than male scents, and females that engaged in extended absences from their
clans pasted more often upon returning to their clans than they did prior to their
departing. Adult females did not paste to advertise their sexual receptivity to males, as

their pasting rates were highest during pregnancy, not late lactation, when females are

most likely to be sexually receptive. Although information about female reproductive
state is clearly available in paste, its ftmctional signiﬁcance, if any, remains unknown.
Among adult males, pasting did not appear to advertise availability for reproductive
opportunities, as male pasting rates did not predict their relative reproductive success.
Lastly, an ontogenetic investigation of scent marking revealed that hyena cubs 'pasted' as
often as older conspeciﬁcs, yet they likely did not deposit scents because hyenas did not
consistently produce paste until their third year of life. Instead, by 'pasting,’ cubs may be
acquiring group-speciﬁc odors and bacteria that facilitate their being recognized as group
members before they are widely known as individuals to conspeciﬁcs.

Symbiotic bacteria in the scent glands of mammals are believed to be responsible
for generating many of the semiochemicals their hosts use in communication. Members
of mammalian social groups might, via cross-infection, come to share similar bacterial
communities in their scent glands. Consequently, they might emit group-speciﬁc odors
that, at the functional level, promote social cohesion. Here I tested an hypothesis
suggesting a bacterial mechanism for group-speciﬁc odors using free-living spotted
hyenas and culture-independent sampling methods (l6S rRNA gene surveys). The scent
glands of hyenas contain many previously uncultured anaerobes, mostly species of
Peptostreptococcus, Anaerococcus, Propionibacterium and C orynebacterium. Members
of these genera produce compounds believed to ﬁgure prominently in hyena scent
marking. Also, graphical and statistical analyses illustrated that, although not universal,
group-speciﬁc differences in the structure of symbiotic bacterial communities did exist
within the scent glands of hyenas, thus supporting the hypothesis that bacteria are

responsible for the production of group-speciﬁc odors in this species.

ACKNOWLEDGMENTS

As a young adult, I followed a meandering academic and professional path. Along
the way I beneﬁted from the generous guidance of several mentors, but one deserves
special mention. The Reverend Harold Ridley was a towering, sophisticated, masterfully
intellectual man, who ﬁrst showed me that genuine concern and generosity distinguish
great teachers from ordinary ones. I cherished Father Ridley’s aura, his warmth, his
dedication to his students and, selﬁshly, his belief in the capacity of my mind. I wish
dearly I could share this accomplishment with him. As I cannot, I dedicate it to him.

The meandering path I followed eventually led me to behavioral biology, and to
Bill Shields. I have lauded Bill previously when endorsing him for a Distinguished
Teaching Professorship, so will refrain from doing so now. Sufﬁce it to say that Bill is
the most innovative and effective teacher I have ever observed. I am deeply indebted to
him for providing me an ideal toward which to strive. Additionally, were it not for Bill,
I'd never have met Kay Holekamp.

There is no evident reason why Kay accepted me into her prestigious lab.
Regardless, I am eternally grateful she did. Through Kay's graciousness I have lived
beside and studied large carnivores in the African bush, and through her guidance I have
begun developing into an independent behavioral biologist. Kay is tireless in her attempts
at furthering the careers of her graduate students. She secured incredibly valuable
teaching opportunities for me when I wasn't yet deserving, and she has repeatedly
provided me with ﬁnancial support when under no obligation to do so. Kay routinely
answers emails in minutes, not days, and returns meticulously edited manuscript drafts in

days, not weeks. If I eventually possess a lab of my own, the respect and treatment I

iv

afford my students will be due in large part to the example Kay has set for me, and any
success I achieve will be a direct result of having had Kay in my comer.

My intention when starting at MSU was to study vocal communication in hyenas.
However, after I attended a few meetings of the ODOR group, I shifted my focus to
chemical communication. The ODOR group contributed greatly to the development of
this dissertation. John Stoltzfus was the impetus behind the microbial work, and two
ODOR members were so broadly inﬂuential in the development of my research that I
requested they serve on my advisory committee. Heather Eisthen has been strongly
supportive throughout my graduate career. She has repeatedly made time for discussion
with me and always provided detailed edits of my writing. She has also often been a
welcome source of perspective. Jim Miller was instrumental in selling me on the
importance of chemical stimuli in behavioral biology, in my developing an understanding
of the scientiﬁc method, and in my recognizing the importance of focusing on
hypothesis-driven problems. Tom Getty is the other member of my advisory committee
and, although not an ODOR member, he was instrumental during the formative stages of
my dissertation as well. Tom is very adept at broad-level, conceptual thinking, and his
contributions to my understanding of the ﬁelds of animal communication and behavioral
ecology are greatly appreciated.

Barb Lundrigan and Tom Schmidt also substantially enhanced my professional
development. Barb offered insightful discussion on mammalian scent marking and
inspired me to continually improve as a teacher. Tom is an incredibly approachable and,
thankfully, patient man. Similar to Kay, Tom took me into his lab when reason must

surely have been screaming at him not to. Tom has already invested much time, thought,

and many resources in me and my research. The microbial work in this dissertation
simply would not have been possible without him. I am deeply indebted to Tom for his
continued generosity and intellectual curiosity. Each of his students, but especially Brad
Stevenson, Stephanie Eichorst, and Zarraz Lee, graciously provided technical assistance.

My dissertation work is a small part of a long term research project addressing the
behavioral ecology of spotted hyenas in the Masai Mara National Reserve (MMNR),
Kenya. The Mara Hyena Project has had numerous contributors, many of whom collected
data analyzed in this dissertation. Laura Smale, co-founder of the project, deserves
speciﬁc mention, as does Russ Van Horn, who provided the paternity analyses. My
enjoyment of the Mara was heightened by my relationships with John Keshe, James
Kerembe, Moses Sairowa, Stephen and Lesingo Nairon'. I am indebted to all members of
the Holekamp lab. Kay is such a proliﬁc mentor that I cannot name them all. However,
three must be acknowledged directly. Joe Kolowski offered humor, an ear upon which to
vent ﬁ'ustration and, occasionally, an excuse for a beer and an NFL game. Terri
McElhinny provided frequent moral support. Her boundless optimism proved a
comforting compliment to my chronic cynicism. Most deserving of my gratitude,
however, is Jaime Tanner. She was critical to my accomplishing my research objectives
in the Mara and at home, and she has become a dear friend. She has frequently been
quick to forgive my shortcomings as both a colleague and a friend. I simply cannot thank
her enough.

I have beneﬁted from the assistance of Pat Bills, our resident computer guru and a
welcome source of conversation. Also, I have had numerous undergraduate assistants:

Anna .Heckla, Joey Verge, Jessi Smashey, Elizabeth Brown, Kelly Pyle, Amy Kuczynski

vi

and Laura Cisneros. It was a true delight to watch Anna, Joey and Jessi progress from
collecting data to asking biological questions and formulating hypotheses of their own.
Of course, science requires money, and I have beneﬁted from the generosity of the
following funding sources: National Science Foundation (awards to K.E. Holekamp),
Ofﬁce of the Graduate School, College of Natural Science, Department of Zoology,
Program in EEBB, Shaver Fellowship Committee, American Society for Microbiology,
and Sigma Xi. I must also thank the Ofﬁce of the President of Kenya and the Senior
Warden of the MMNR for permission to conduct this research.

Lastly, I thank my family, which has always offered me unconditional support.
They remained patient as I meandered my way through life, seemingly delaying
adulthood. I cannot remember a single instance of them telling me, or even hinting at the
possibility, that I could not accomplish something. I have been blessed with an
unbelievably supportive family by marriage as well. My mother-in-law even sheltered
and fed me while I returned to school for a second bachelor's degree. Of course, this
dissertation would not have been possible without the full support of my wife, Lisa. She
has had to assume many of my parental responsibilities while I have been in graduate
school. I am immensely grateful to her, and to our beautiful daughters Juliana Rose and

Karissa Ann, for their understanding and patience.

vii

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... ix
LIST OF FIGURES ....................................................................................................... x
GENERAL INTRODUCTION ..................................................................................... 1

Synopses of chapters ......................................................................................... 12
CHAPTER 1

THE ONTOGENY OF PASTING BEHAVIOR IN FREE-LIV ING SPOTTED HYENAS
(CROCUTA CROCUTA)

Introduction ....................................................................................................... 23

Methods ............................................................................................................. 25

Results ............................................................................................................... 27

Discussion ......................................................................................................... 34
CHAPTER 2

INTRAGROUP FUNCTIONS OF SCENT MARKING BY ADULT SPOTTED
HYENAS (CROCUTA CROCUTA)

Introduction ....................................................................................................... 44

Methods ............................................................................................................. 5 1

Results ............................................................................................................... 59

Discussion ......................................................................................................... 78
CHAPTER 3

DISCRIMINATION OF CONSPECIFIC SCENT BY F REE—LIVING SPOTTED
HYENAS (CROCUTA CROCUTA)

Introduction ...................................................................................................... 86

Methods ............................................................................................................ 90

Results .............................................................................................................. 1 01

Discussion ........................................................................................................ 1 14
CHAPTER 4

A BACTERIAL MECHANISM FOR GROUP-SPECIFIC ODORS IN THE SPOTTED
HYENA (CROCUTA CROCUTA)

Introduction ..................................................................................................... 127
Methods ........................................................................................................... 13 1
Results ............................................................................................................. 13 8
Discussion ....................................................................................................... 1 50
LITERATURE CITED ............................................................................................... 156

viii

LIST OF TABLES

Table 2.1. Intragroup functions of pasting by adult spotted hyenas: hypotheses and
predictions. Italicized letters in parentheses indicate the data set used to test each
prediction. The composition of each data set is explained in the Methods section....47

Table 2.2. Socio-ecological contexts of pasting behavior considered in the current study.
If more than one of these contexts preceded a paste event, we assigned the context that
most immediately preceded the paste event ............................................................... 55

Table 2.3. Time for which each hyena contributing to data set A was observed, as well as
the number of paste events each hyena exhibited during this study .......................... 56

Table 2.4. Results of Wilcoxon matched pairs tests comparing the hourly pasting rates of
adult female hyenas across periods of high, medium and low rainfall, as well as at dens,
ungulate kills and elsewhere (N = 12) ........................................................................ 61

Table 2.5. Results of Wilcoxon matched pairs tests comparing hourly rates of pasting by
adult male Spotted hyenas in observation sessions characterized by high, medium and low
ratios of adult males to adult females (N = 16) .......................................................... 69

Table 2.6. Results of Wilcoxon matched pairs tests comparing the hourly pasting rates of
adult female hyenas across various reproductive states (N = 18) .............................. 74

Table 3.1. Results of Spearman’s rank tests assessing the monotonic relationship between
the age of paste samples and the amount of time they were investigated. Investigation
times were calculated by averaging the responses of all participating hyenas toward each
paste sample for a given trial. Therefore, sample sizes indicate the number of paste
samples per donor class used in each experiment. Within each experiment, samples from
the two scent donor classes were analyzed separately .............................................. 104

Table 4.1. OTUS in the current study that shared at least 97% sequence similarities with
previously cultured or well-categorized (i.e. Candidatus) bacterial species ............. 139

Table 4.2. The number of sequences analyzed, and the number of OTUS found, in each
clone library. Richness estimates and diversity indices of the bacterial communities
within the anal pouches of the individual hyenas have been included, along with their
95% conﬁdence intervals ........................................................................................... 145

ix

LIST OF FIGURES

Figure 1.1. Effect of age on paste production in spotted hyenas. Numbers above the bars
indicate the number of hyenas sampled in each age class. ................................ 28

Figure 1.2. Hourly pasting rates of juvenile male and female spotted hyenas by age. The
asterisk indicates a statistically signiﬁcant difference (P < 0.05). Numbers in parentheses
indicate samples sizes. ......................................................................... 30

Figure 1.3. Proportion of spotted hyena pasting events that were overmarks of conspeciﬁc
scent marks, by age and sex. Asterisks indicate statistically signiﬁcant differences (P <
0.05). Numbers in parentheses indicate sample sizes. ..................................... 31

Figure 1.4. Behavioral contexts of pasting by A) cubs and B) subadults. Approximately
one percent of paste events were assigned an ‘unknown’ context. These events were
excluded from all analyses. Asterisks indicate statistically signiﬁcant differences after
Holm’s sequentially-rejective Bonferroni method was employed. Numbers in parentheses
indicate sample sizes. ........................................................................... 33

Figure 2.1. Mean (1- SE) hourly pasting rates of adult spotted hyenas across periods of
high, medium and low A) prey abundance, and B) rainfall within their home range.
Letters above the error bars indicate statistically signiﬁcant differences (P < 0.05)....60

Figure 2.2. Mean (i SE) hourly pasting rates of adult male and female spotted hyenas at
dens, ungulate kills, and elsewhere. Letters above the error bars indicate statistically
signiﬁcant differences (P < 0.05) .................................................................................. 62

Figure 2.3. Mean (i SE) hourly pasting rates of adult male and female spotted hyenas
during observation sessions at which lions were absent or present. Letters above the error
bars indicate statistically signiﬁcant differences (P < 0.05) ........................................ 63

Figure 2.4. Socio-ecological contexts of pasting by adult male and female spotted hyenas.
Data represent individuals observed to paste at least 10 times. The x-axis labels refer to
spontaneous, overrnarking, adjacent marking, agonism, greeting and other contexts,
respectively. Approximately six percent of paste events were assigned an ‘unknown’
context. These events were excluded from all analyses. Asterisks indicate statistically
signiﬁcant differences (P < 0.05) ................................................................................. 64

Figure 2.5. Mean (:5 SE) hourly pasting rates of adult male and female spotted hyenas
residing concurrently in the same clan. The asterisk indicates a statistically signiﬁcant
difference (P < 0.05). Numbers above bars indicate individuals sampled .................. 66

Figure 2.6. Effect of intrasexual social rank on rates of pasting behavior among adult A)
male, and B) female spotted hyenas. By convention the hi ghest-ranking member in each
intrasexual hierarchy was assigned a rank of 1 ........................................................... 68

Figure 2.7. Effect of the sex composition among conspeciﬁcs in the immediate social
environment on the frequency of pasting behavior by adult male and female spotted
hyenas. Asterisks indicate statistically signiﬁcant differences (P < 0.05) ................. 70

Figure 2.8. Mean (i SE) hourly pasting rates of adult female hyenas before and after
absences from the Talek clan’s territory lasting longer than one month (N = 9). The
asterisk indicates a statistically signiﬁcant difference (P < 0.05) .............................. 72

Figure 2.9. Mean (3: SE) hourly pasting rates of female Spotted hyenas during pregnancy,
early (11), mid (l2), and late (13) lactation (N = 18). Letters above the error bars indicate
statistically signiﬁcant differences (P < 0.05 ............................................................. 75

Figure 2.10. Mean (i SE) hourly pasting rates of female spotted hyenas during late
lactation during observation sessions in which 1 — 2, 3 -— 4, or more than 4 immigrant
males were present (N = 10). Differences were not statistically signiﬁcant ............. 76

Figure 2.11. Relative reproductive success of immigrant male spotted hyenas in relation
to pasting rate during the ﬁrst two years of residency and during the remainder of each
male’s tenure in their new clans (N = 16) ................................................................. 77

Figure 3.1. Map illustrating the locations in the Reserve of the three clans used in the
current study. The territory sizes of the Mara River, West Talek, and East Talek clans

were 31 kmz, 28 kmz, and 19 kmz, respectively (Kolowski et a1. 2007). During the study
period there were 30 hyenas present in the Mara River clan, 51 in the West Talek clan,
and 32 in the East Talek clan. This map was adapted, with permission, from Kolowski et
a1. (2007) .................................................................................................................... 91

Figure 3.2. The presence and participation of different age / sex classes of spotted hyena
across all 44 scent discrimination trials conducted in this study .............................. 102

Figure 3.3. The relative amount of time cub, subadult and adult female Spotted hyenas
spent snifﬁng paste samples from non-clanmates, clanmates, and control wires. “n.s.”
indicates a statistically non-Signiﬁcant difference between time spent snifﬁng clanmate
and non-clanmate odors, following a signiﬁcant overall treatment effect for that age class
(P < 0.05). Sample sizes for cubs, subadults and adult females are 14, 13 and 9,
respectively ............................................................................................................... 1 05

Figure 3.4. The relative amount of time cub and subadult Spotted hyenas spent snifﬁng

paste samples from female non-clanmate and female clamnate donors. Letters above the
error bars indicate statistically signiﬁcant differences (P < 0.05 (one-sided)) ......... 107

xi

Figure 3.5. The relative amount of time cub, subadult and adult female spotted hyenas
spent snifﬁng paste samples from members of neighboring clans, distant clans, and
control wires. “n.s.” indicates a statistically non-signiﬁcant difference between time spent
snifﬁng odors from neighboring and distant clan members, following a signiﬁcant overall
treatment effect for that age class (P < 0.05). Sample sizes for cubs, subadults and adult
females are 12, 10 and 5, respectively ...................................................................... 108

Figure 3.6. The relative amount of time spotted hyenas spent snifﬁng paste samples from
members of the West Talek clan, East Talek clan, and control wires. Both the West and
East Talek clans' territories were separated from the test clan's territory by at least one
other clan. “n.s.” indicates a statistically non-signiﬁcant effect of clan membership,
following a signiﬁcant overall treatment effect (P < 0.05, N = 8) ............................ 110

Figure 3.7. The relative amount of time cub, subadult and adult female spotted hyenas
spent snifﬁng paste samples from adult male and female donors. The asterisk indicates a
statistically signiﬁcant difference (P < 0.05). Sample sizes for cubs, subadults and adult
females are 12, 11 and 7, respectively ...................................................................... 111

Figure 3.8. The relative amount of time natal spotted hyenas spent snifﬁng paste samples
from pregnant and late lactating female donors. The asterisk indicates a statistically
signiﬁcant difference (P < 0.05, N = 10) .................................................................. 113

Figure 4.1. Map illustrating the relative locations within the Reserve of the four hyena
clans sampled in this study. The dashed lines indicate the estimated territorial boundaries
of hyena clans in this area of the Reserve from 1999 - 2000. This map was adapted with
permission from Van Horn et al. (2004) and Kolowski et a]. (2007) ....................... 133

Figure 4.2. Rarefaction curve illustrating the degree of sample coverage in this study for
all sampled hyenas considered collectively. The dotted curves represent the 95%
conﬁdence interval for the plot. To provide perspective when gauging the degree to
which the rarefaction curve is leveling off, the function y = x is also illustrated....140

Figure 4.3. Rank abundance curve showing the relative abundance of the 75 OTUS found
in this study .............................................................................................................. 141

Figure 4.4. Figure 4.4. Phylogenetic analysis of representatives of the nine OTUS (in bold
face) that each represented more than 1% of all sequences analyzed in the current study.
These nine representatives were incorporated into the phylogenetic tree with their ten
nearest neighbors from the Silva reference library, as determined by SINA. With each
representative sequence, the total number of sequences constituting that OTU is indicated
in brackets. The numbers within branches indicate the number of nearest neighbor
sequences that have been grouped together (i.e. condensed) on that branch. For nearest
neighbor sequences obtained from uncultured bacteria, we have included information
from the Silva database about the environmental source from which they were obtained.
.................................................................................................................................. 143

xii

Figure 4.5. Rarefaction curves illustrating the degree of sample coverage for each hyena
sampled in each clan. The clone libraries have been separated by clan of origin. The
function y = x (dotted lines) appears on each graph for perspective ......................... 144

Figure 4.6. Rank abundance curve Showing the relative abundance of OTUS within the
anal pouch of each sampled hyena. The sample sizes indicate the number of libraries that
contained at least that number of OTUS. The error bars indicate standard errors ..... 146

Figure 4.7. Dendrograrn illustrating similarities in the structure of bacterial communities
inhabiting the anal pouches of 16 spotted hyenas representing four distinct clans: Mara
River (circles), Southern Comfort (squares), Emarti Hill (diamonds) and Fig Tree
(triangles) ................................................................................................................... 148

Figure 4.8. Multidimensional scaling (MDS) plot comparing similarities in the structure
of bacterial communities inhabiting the anal pouches of 16 Spotted hyenas representing
four distinct clans: Mara River (circles), Southern Comfort (squares), Emarti Hill
(diamonds) and Fig Tree (triangles). The plot was based on Bray-Curtis dissimilarity
indices. In MDS plots, sampled communities that are dissimilar are placed far apart
whereas those that are similar are placed close together (Quinn & Keough 2002; Gotelli
& Ellison 2004). The stress of the plot is 0.123 ....................................................... 149

xiii

GENERAL INTRODUCTION
The objectives of this introductory essay are to provide a working deﬁnition of
animal communication, to argue the continued importance of animal communication
studies, to emphasize the particular value of chemical communication studies, and to
explain why free-living spotted hyenas are intriguing subjects with which to expand our
understanding of the mechanistic, developmental, and functional explanations of
mammalian scent marking. This essay concludes with a discussion of my research

objectives, including synopses of the data chapters constituting my thesis.

What is animal communication?

Since a widely accepted deﬁnition of animal communication has not yet been
developed (Maynard Smith & Harper 2003), it is important for individual researchers to
provide clear, working deﬁnitions of communication early in their writings. Animal
communication is here deﬁned as the transmission of information from a sender to a
receiver that effects a change in the receiver’s behavior and/or physiology, thereby
beneﬁting, on average, both the sender and receiver (Dusenbery 1992; Bradbury &
Vehrencamp 1998). Information is transmitted via ‘Signals’ and effected changes in
receivers are labeled ‘responses.’ Given that signalers and receivers seldom have
identical interests, each party often has a substantially different view of what constitutes
an ideal response (Krebs & Dawkins 1984). Consequently, signalers are often perceived
as attempting to ‘manipulate’ receivers, while receivers are seen as discriminating ‘mind-
readers,’ selectively responding only to honest signals (Krebs & Dawkins 1984). If a

signal consistently conveys dishonest information, then receivers who indiscriminately

respond to the signal do so with consequent detriment to their ﬁtness. Therefore, there is
strong selective pressure on receivers to ensure that they respond only to signals that
transmit honest, reliable information (Krebs & Dawkins 1984; Bradbury & Vehrencamp
2000). Consequently, signaling systems based on unreliable signals are not evolutionarily
stable (Johnstone 1997; Bradbury & Vehrencamp 1998, Maynard Smith & Harper 2003),
and signals typically provide honest information about senders and/or their social and
physical environments (e.g. Gosling et al. 2000; Blas et a1. 2006; F orsman & Hagman
2006; Lopez et a1. 2006; Velando et a1. 2006; Vanpe et a1. 2007; Villasenor & Drummond

2007; but see Elwood et al. 2006; Setchell et a1. 2006).

Why is the study of animal communication important?

Although the study of animal communication was heralded in Darwin’s The
Expression of the Emotions in Man and the Animals (1872), it was ﬁrst formally pursued
in the early twentieth century by the classical ethologists (reviewed by Burkhardt 2005),
and continued by their students (e.g. Marler 1967; Kruuk 1972). It remains a vibrant and
robust ﬁeld of study for four primary reasons (Bradbury & Vehrencamp 1998). First,
communication plays a central role in animal societies. Many animals participate in
societies because group-living affords improved defense of resources, reduced predation
risk, increased foraging efﬁciency, and enhanced reproductive output (Alcock 2005).
However, social animals must also deal with the substantial costs associated with a
gregarious lifestyle. These costs include increased competition for resources, greater
investment and risk of injury in establishing and maintaining dominance hierarchies,

higher prevalence of disease and parasite transmission, and an increased likelihood of

mating with close kin (Alcock 2005). Communication has been lauded as the “glue that
holds animal societies together” (Bradbury & Vehrencamp 1998, p. 5), because it
facilitates the beneﬁts, and modulates the costs, of living in groups. Although
communication is often most conspicuous among highly social animals, even solitary
ones require it to maintain spacing between competitors and to bring potential mates
together during reproductive periods (Bradbury & Vehrencamp 1998). In fact,
communication is ubiquitous throughout the animal kingdom, leading one preeminent
scholar to proclaim that “nothing would work in the absence of communication” (Hauser
1996,p.1)

The second reason researchers study animal communication is to ascertain the full
extent of information that animals convey to each other, and to elucidate the mechanisms
underlying the production, reception and interpretation of signals (Dusenbery 1992;
Bradbury & Vehrencamp 1998). Numerous studies have demonstrated that animals
communicate information about their species, group, sex and individual identity, their
reproductive, social, and motivational status, and their immediate social and physical
environment (Bradbury & Vehrencamp 1998; Maynard Smith & Harper 2003).
Ascertaining the information that animals communicate to others allows behavioral
ecologists to formulate hypotheses about the adaptive signiﬁcance, and optimization, of
signaling for different animal species, age groups and sex classes. Furthermore, in
addition to increasing our understanding of animal physiology and sensory ecology,
attempts at elucidating the mechanisms underlying animal signaling encourage
collaborations among researchers with disparate scientiﬁc backgrounds, most notably

engineering, physics, chemistry, microbiology, neuroscience, behavioral ecology and

 

psychology (see proceedings from the Sackler Colloquium of the National Academy of
Sciences, Irvine, CA, 2003; Chemical Signals in Vertebrates XI, Chester, England, 2006).
Collaborations among researchers with different backgrounds likely promote theoretical,
methodological, statistical, and technological innovation.

A third reason the study of animal communication remains productive is that it
can serve as a tool for exploring general evolutionary principles (Bradbury &
Vehrencamp 1998). One of the primary interests of the classical ethologists was the
evolution of Signals (Kruuk 2003; Burkhardt 2005). Over the last half-century their
pioneering efforts have mushroomed into a myriad of studies on evolutionary topics in
communication, such as signal optimization in response to background noise and predator
eavesdropping (e.g. Ryan et al. 1982; Bayly & Evans 2003; Aubin 2004; Brumm et al.
2004; Remage-Healey et al. 2006; 0rd et al. 2007; Poesel et al. 2007), and the role of
sexually selected signals in reproductive isolation and speciation (e. g. Sorenson et al.
2003; Kingston & Rossiter 2004; Boul et al. 2007; Bradbury & Vehrencamp 1998).

The fourth reason animal communication studies are important is that they have
practical applications (Bradbury & Vehrencamp 1998). Studies of insect pheromones
have lead to the development of pheromone bait-traps that help control agricultural and
forest pests (e.g. Subchev et al. 2004; Jeans-Williams & Borden 2006; Ibeas et a1. 2007).
Recently, behavioral ecologists have begun investigating whether similar tactics can be
employed to minimize carnivore-livestock conﬂicts around nature reserves (e. g. Parker
2006). Conservationists have also utilized audio playback experiments and animal
signaling activity to estimate the population size of endangered Species and to determine

species diversity in vulnerable areas (e.g. Apollonio et a1. 2004; Barbraud & DeLord

2006; Geissmann & Nijman 2006; Ryan et al. 2006; Barea-Azcon et al. 2007). Lastly,
knowledge gained about how animals communicate with conspeciﬁcs has lead to the

improved welfare of captive animals (e.g. Olsson et al. 2003; Schulte et al. 2007), and
enhanced the success of captive breeding programs (e. g. Fisher et al. 2003; Roberts &

Gosling 2004; Zhang et al. 2004).

Why study chemical communication, in particular?

If the value of animal communication studies is recognized, then the value of
chemical communication studies, in particular, Should be acknowledged as well, given
that chemical Signaling is the most ancient and widely utilized form of communication in
the animal kingdom (Agosta 1992; Dusenbery 1992; Doty 1995; Wyatt 2003). One
challenge that has historically hindered the investigation of chemical communication is
that human sensory systems, although well-adapted for perceiving many of the vocal and
visual signals generated by animals, are relatively insensitive to the chemical messages
that ﬁgure so prominently in the umwelten of most animal species, including the majority
of non-human mammals (Blaustein 1981; Penn & Potts 1998). For example, consider the
despondent reaction of George Schaller when he realized, after months of studying wild
giant pandas, that he and they occupied disparate sensory worlds. “How would I ever
understand pandas? They moved from odor to odor, the air ﬁlled with important
messages where I detected nothing” (1993, p. 99). While Schaller’s observation
poignantly illustrates the challenge associated with studies of chemical communication, it

also serves to intrigue many scientists.

If the presumption is correct that we fail to perceive the vast majority of
communications that occur around us, then the prospect of utilizing technological
innovations and the systematic observation of animals to discover how they communicate
via scent is, for myself and others, a highly attractive endeavor. Over the past few
decades, chemical and behavioral ecologists have sought to document the full extent of
information communicated via chemicals among animals, to determine how these odors
are produced, sensed and interpreted, and to demonstrate how chemical signaling is
evolutionarily adaptive for a wide range of species (Bradbury & Vehrencamp 1998;
Wyatt 2003). One form of chemical signaling that has received a great deal of
investigation is mammalian scent marking (see Johnston 2003; Hurst & Beynon 2004).

Scent marking is the deliberate deposition of urine, feces and/or the products of
secretory glands in the environment, and is one of the primary forms of signaling
employed by mammals (Johnson 1973). A unique characteristic of scent marks is that
they persist as viable messages for considerable periods of time where they were
deposited, even after their sender has departed (Wynne-Edwards 1962; Marler 1967;
Eisenberg & Kleiman 1972). Consequently, although communication is traditionally
described as occurring within a dyad (Bradbury & Vehrencamp 1998), scent marking is
most appropriately viewed as an act of communication conducted within a network of
individuals (Hurst 2005). Communication netWorks exist whenever signals travel farther
than the average Spacing among individuals in the signaling system (McGregor 1993;
McGregor 2005). Therefore, the likelihood of network formation is dependent upon both
population density and signal properties, speciﬁcally broadcast distance and fade-out

time. Due to limited broadcast distance, visual, tactile and electrical signaling are

generally restricted to individuals in immediate proximity of one another (Dusenbery
1992; Bradbury & Vehrencamp 1998). A dyadic representation of communication in
these modalities is often, therefore, sufﬁcient. Audible signals, however, can travel
distances far in excess of the average spacing between individuals and thus potentially
provide information to multiple receivers over a large area (e. g. Grafe 2005; McComb &
Reby 2005). Some chemical signals also travel considerable distances on air or water
currents and may be perceived by numerous receivers Spaced far apart (Wilson & Bossert
1963; Elkinton & Carde 1984). Scent marks, however, do not travel far, yet their
persistence makes them available to multiple receivers over time, and, therefore, they are
best considered as occurring within a network (Hurst 2005). The distinction here is not
trivial, as scent marks cannot be directed toward speciﬁc recipients. Rather, once
deposited, they are available to all individuals who happen upon them (Hurst 2005).
Therefore, there has likely been strong selection pressure to ensure that the information
contained in scent marks is appropriate for communication with all animals in the area
rather than one speciﬁc individual (Hurst 2005).

Mammalian scent marks can convey information about the donor’s species,
group, sex, age, genetic and individual identity, as well as its reproductive, social,
nutritional and health status (reviewed by Penn & Potts 1998; Gosling & Roberts 2001;
Wyatt 2003; Hurst & Beynon 2004; Hurst 2005). Communicating this information to
conspeciﬁcs functions to establish and maintain territories, advertise dominance and
reproductive status to competitors and potential mates, facilitate social cohesion, and
increase foraging efﬁciency (reviewed by Ralls 1971; Eisenberg & Kleiman 1972;

Johnson 1973; Thiessen & Rice 1976; Gorman & Trowbridge 1989; Wyatt 2003; Hurst

2005; Lewis 2006). Scent marking is therefore an intriguing form of communication to
study, as scent marks potentially contain a wealth of information about their donors and
the use of scent marks likely serves different adaptive functions for each age, sex and
social class that deposits and/or accesses the information available in them (Yahr 1983;
Hurst 2005).

However, historically, the adaptive function of mammalian scent marking that has
received the vast majority of attention is territorial maintenance (see Gosling 1982;
Gorman & Trowbridge 1989; Gosling & Roberts 2001). This over-emphasis on territorial
scent marking has occurred despite recommendations that research emphases be more
comprehensive (e.g. Ralls 1971; Eisenberg & Kleiman 1972; Johnson 1973; Bearder &
Randall 1978). The argument of these authors was not that scent marking is afforded too
prominent a role in territorial maintenance, but rather that emphasizing a single function
to the exclusion of others potentially underestimates the complexity of mammalian
chemical Signaling systems and the number of functions they serve. Recently,
mammalian scent marking studies have become more comprehensive, and as a result, we
now have evidence that the functions of scent marking vary not only among mammalian
species, but also among different age, sex, reproductive and social classes within the
same species (e.g. Drea et al. 2002; White et a1. 2003; Palagi et al. 2004; Hurst 2005;
Lewis 2006; Mateo 2006; Scordato & Drea 2007). Furthermore, it has become apparent
that we have a particularly poor understanding of chemical communication in highly
social species (Scordato & Drea 2007). The broad objective of my dissertation research is

therefore to provide a comprehensive understanding of the scent marking behavior of the

gregarious spotted hyena, Crocuta crocuta, a large, keystone predator found throughout

much of sub-Saharan Africa (Estes 1992).

Why study scent marking in the spotted hyena?

Spotted hyenas live in complex social groups, called clans, that typically contain
40 — 80 individuals (Kruuk 1972; Trinkel et al. 2007). Members of a clan cooperatively
defend territories from neighboring hyenas, and they defend ungulate kills from both
other hyenas and African lions, Panthera leo (Kruuk 1972; Mills 1990; Henschel &
Skinner 1991; Boydston et a1. 2001). Similar to the complex societies of many
cercopithecine primates, hyena clans contain multiple breeding males and multiple
overlapping generations of females (Holekamp et a1. 2007). Another characteristic hyena
clans share with some primate societies is that they are structured by linear dominance
hierarchies (Frank 1986b; Holekamp et al. 2007). In contrast to most mammalian social
systems, a hyena’s position within its clan’s hierarchy is not determined by its ﬁghting
ability. Instead, natal hyenas inherit their mother’s rank (Frank 1986b; Holekamp &
Smale 1991, 1993; Smale et al. 1993; Engh et al. 2000). Additionally, adult immigrant
males, all of whom are lower-ranking socially than natal animals, queue for social status
with other immigrants (Smale et al. 1997; East & Hofer 2001). Interestingly, the
dominance queue among immigrant males is maintained despite males exhibiting very
low rates of aggression towards one another (Frank 1986b; Smale et al. 1997; East &
Hofer 2001). The low rates of aggression among males are interesting because, as with
female hyenas (Holekamp et al. 1996), a male’s position within the clan’s hierarchy has a

substantial effect on his reproductive success (Engh et al. 2002; Hofer & East 2003).

Therefore, a substitute for aggression in maintaining the stability of male dominance
queues likely exists. Communication is a likely candidate.

Given the complexity of spotted hyena societies, it is not surprising that hyenas
emit a rich array of vocal, visual and chemical signals (Blumstein & Armitage 1997;
Freeberg 2006; Kruuk 1972; Mills 1990). Crocuta’s vocal repertoire includes long-
distance whooping, aggression-elicited giggling, groaning over cubs, and alarm rumbling
(Kruuk 1972; East & Hofer 1991a). Spotted hyenas also exhibit numerous facial, postural
and genital displays that can indicate agonistic motivation and intent, sexual interest,
and/or general arousal (Kruuk 1972; East et al. 1993). The vocal and visual signals of
spotted hyenas are routinely utilized to communicate information among members of the
same clan (Kruuk 1972; Mills 1990; East & Hofer 1991b; East et al. 1993; Holekamp et
al. 1999; Theis et al. 2007).

The potential chemical signaling behaviors of spotted hyenas include rolling,
scent-rubbing, forepaw scraping, defecating at latrines and anal gland ‘pasting’ (Kruuk
1972; Bearder & Randall 1978; Mills 1990; Henschel & Skinner 1991). Pasting is the
most common and conspicuous of these behaviors. It is a form of scent marking exhibited
by all four extent members of Family Hyaenidae, during which hyenas drag their
extruded anal pouches over grass stalks or other objects in the environment (Kruuk 1972;
Mills 1990; Sliwa & Richardson 1998). As hyena anal pouches contain a pasty mass,
consisting of degenerated epidermal cells bound by a waxy substratum, pasting results in
hyenas depositing a thin layer of anal gland secretion near the tops of grass stalks
(Matthews 1939; Kruuk 1972; Mills 1990). Among Spotted hyenas, pasting is sometimes

repeated multiple times on the same stalk (Kruuk 1972). Occasionally other hyenas join

10

in as well, adding their secretion to the same substrate, a phenomenon called
‘overmarking’ (Kruuk 1972; Mills 1990).

Interestingly, although there is a robust, positive correlation between dominance
status and scent marking behavior in most mammals (Ralls 1971), in Crocuta societies it
is the immigrant males, the lowest-ranking members of the clan, who scent mark most
frequently (Mills & Gorman 1987). Given this anomaly, and the fact that vocal and visual
communication ﬁgure prominently in modulating hyena social interactions within clans,
it has been suggested that spotted hyena pasting behavior is utilized exclusively for inter-
group communication, i.e. maintaining territories (Gorman & Mills 1984). However, this
restricted view of pasting is unwarranted. As will be Shown in this dissertation, pasting is
performed by all spotted hyenas, regardless of age or sex. Additionally, it is exhibited in a
variety of social and behavioral contexts (Kruuk 1972). Therefore, while pasting certainly
has a territorial function (Kruuk 1972; Mills & Gorman 1987; Henschel & Skinner 1991;
Boydston et al. 2001), as does most mammalian scent marking, pasting likely also has
additional intra-clan functions that have yet to be discovered (Hofer et al. 2001; Drea et
a1. 2002). The ﬁrst aim of my dissertation research is to ascertain the functions of pasting
for all age, sex, social and reproductive classes of spotted hyenas by systematically
observing the frequency and context of pasting behavior among free-living subjects, and
by conducting scent discrimination experiments with these wild hyenas as well, to
determine the potential information communicated via paste. The second aim of my
dissertation research is to examine the role of bacteria in generating the information
contained in spotted hyena anal gland secretions. The justiﬁcations and approaches for

achieving these aims are discussed in the chapter synopses that follow.

11

Throughout the remainder of this dissertation I have employed the pronoun “we”
in favor of “I,” because, although I assume ownership of the original ideas and analyses
included in this document, data collection has truly been a collective effort. Many of the
data presented here were extracted from a long-term (currently 19 years), longitudinal
study of spotted hyena behavioral ecology that has been conducted by Dr. Kay Holekamp
and her many colleagues, students, and research assistants. Other data were generated
through collaboration with experts in other ﬁelds of study, without whom I would have
been unable to achieve my research objectives. Furthermore, each of my dissertation

chapters has been, or will be, submitted as a multiple-author manuscript.

Synopses of dissertation chapters
Chapter One

Among the hundreds of investigations of mammalian scent marking, only a few
have substantively addressed the development of the behavior in juveniles (e. g. French &
Cleveland 1984; Woodmansee et al. 1991; Sliwa 1996). It is possible that the lack of
information on the ontogeny of scent marking is due to the behavior not being observed
to an appreciable degree until sexual maturity in many mammalian species (Yahr 1983).
However, we cannot be sure, as very few studies have indicated that juvenile scent
marking was considered, and not observed (e. g. Lewis 2006). With greater emphasis
recently being placed on understanding the operation of natural selection during juvenile
life stages in mammals (Holekamp & Smale 1998; Altmann & Altmann 2003),

investigating the ontogeny of scent marking in mammalian species in which the behavior

12

is commonly observed among juveniles will provide us with a more comprehensive
understanding of chemical communication in general.

All spotted hyenas frequently paste, regardless of age, although juveniles may do
so ineffectively as they reportedly do not produce secretions in their anal glands until
they are 12-18 months old (Kruuk 1972). If Kruuk’s observation is accurate, then pasting
by juvenile hyenas presents a Darwinian puzzle (Alcock 2005), in that performing
ineffectual behaviors typically reduces one’s ﬁtness. Although unable to further our
understanding of pasting by cubs (<1 yr.), a preliminary investigation by Woodmansee et
al. (1991) of the ontogeny of pasting in captive hyenas revealed an increase in scent
marking as hyenas reached puberty (~24 mos), and a modest correlation between social
status and marking rate at this time as well. These results suggest that the scent marking
behavior of older juvenile spotted hyenas adheres to mammalian norms, and potentially
functions in communicating dominance status and reproductive condition to competitors
and potential mates (Ralls, 1971; Johnson 1973; Yahr 1983). In this chapter, we study the
development of pasting behavior in free-living spotted hyenas in the Masai Mara
National Reserve, Kenya (MMNR). This study builds upon the preliminary ontogenetic
investigation of pasting in captive hyenas (Woodmansee et al. 1991), by including larger
sample sizes in natural (as opposed to highly artiﬁcial) rearing conditions, and by more
strongly emphasizing the scent marking behavior of cubs.

The speciﬁc objectives of this chapter are to: 1) accurately determine the age at
which spotted hyenas begin consistently producing secretions in their anal glands, 2)
describe how rates of juvenile pasting and overmarking vary during development, 3)

evaluate the effect of social rank on pasting behavior among juveniles, 4) identify the

13

behavioral contexts in which pasting occurs among juvenile Spotted hyenas, and 5)
consider whether juvenile pasting is likely an adaptive behavior, and if so, what functions
it potentially serves.

These objectives will be achieved by two means. First, we will examine the anal
pouches of Spotted hyenas anaesthetized in the MMNR, for the presence of paste
secretions. These hyenas will be either known-age animals or individuals in which we
can accurately estimate ages via patterns of tooth wear. Second, we will systematically
observe the pasting behavior of numerous individually-known hyenas from birth through
reproductive maturity to obtain information on scent marking rates and the behavioral

and social contexts that elicit pasting.

Chapter Two

Mammalian scent marking can function to establish and maintain territories,
facilitate social cohesion, increase foraging efﬁciency, and advertise dominance and
reproductive status to competitors and potential mates (reviewed by Ralls 1971;
Eisenberg & Kleiman 1972; Johnson 1973; Thiessen & Rice 1976; Gorman &
Trowbridge 1989; Wyatt 2003; Hurst 2005; Lewis 2006). Historically, territorial marking
has received the vast majority of attention (see Gosling & Roberts 2001). In spotted
hyenas, territorial maintenance has been the function of pasting behavior most
emphasized as well (Gorman & Mills 1984; Henschel & Skinner 1991; Boydston et al.
2001). However, given that different age/sex classes of spotted hyenas paste at different
rates (Mills & Gorman 1987), that hyenas frequently paste at locations other than

territorial boundaries, including areas where intruders seldom appear (Hofer et a1. 2001),

14

and that hyenas preferentially investigate the paste secretions of some donors more
intensively than others (Drea et al. 2002), it is probable that pasting has intra-clan
functions in addition to that of facilitating territorial maintenance. The functions served
by scent marking can be best ascertained by determining the speciﬁc stimulus situations
that elicit marking behavior, and the responses of animals to conspeciﬁc scents (Ralls
1971; Yahr 1983). In this chapter, we describe the general and speciﬁc contexts in which
pasting is exhibited, and apply these data to predictions derived from competitive and
reproductive functional hypotheses for Spotted hyena pasting behavior.

First, we will determine whether pasting communicates dominance to conspeciﬁc
competitors by systematically observing the pasting behavior of each adult hyena in a
single clan over a two-year period, during which the hierarchical relationships among all
adults remained stable. Previous research has indicated that adult immigrant males paste
at higher rates than other members of hyena clans (Mills & Gorman 1987). Given that
immigrant males are the lowest-ranking members of a clan, Mills & Gorman (1987)
opined that the elevated pasting rates of males indicate that, contrary to most mammalian
scent marking, pasting does not convey dominance status to competitors. However,
before this assertion can properly be made, intra-sex hierarchical relationships need to be
considered (Kappeler 1990; Woodmansee et al. 1991). Among immigrant males, high-
ranking individuals may paste more frequently than low-ranking individuals, indicating
that pasting likely is associated with social status.

Pasting by immigrant males may also serve as a form of sexual advertisement. A
male’s reproductive success is strongly dependent upon his length of tenure in the clan

(Engh et a1. 2002; Honer et al. 2007). Males do not typically sire cubs until they have

15

resided in a clan for at least two years (Engh et al. 2002; East et al. 2003). Since male
hyenas are frequently targets of female aggression, pasting may allow newly tenured
males to advertise their presence to females without suffering the harassment incurred by
being in close proximity to females (Mills & Gorman 1987). This hypothesis does not
require that newly arrived males paste at higher rates than males who have resided in a
clan for a long time, but it does predict that males who invest in pasting early in their
tenure should ultimately experience greater reproductive success than males who do not
paste as often. We will test this hypothesis by determining whether a correlation exists
between a male’s pasting rate early in residency and the proportion of cubs he sires in the
clan during his entire tenure.

Pasting by females could potentially function to advertise their reproductive state
to males. In many mammals, females emit chemical cues that provide males with
information about their reproductive condition (reviewed by Johnson 1973; Gorman &
Trowbridge 1989). The beneﬁts of advertising their impending sexual receptivity are
likely greatest for females of solitary species wherein mates are difﬁcult to locate and/or
species in which females are dominant to, and aggressive toward, males, and
consequently males are hesitant to approach females (e. g. Petrulis & Johnston 1997;
Palagi et a1. 2004). Additionally, in female-dominated species, there is little beneﬁt
associated with concealing ovulation, as creating paternity confusion is unnecessary for
preventing male infanticide (Palagi et al. 2004). Female spotted hyenas are frequently
aggressive towards adult male clanmates, and infanticide by male hyenas is a very rare
event (East et al. 2003). Therefore, it is possible that female hyenas could beneﬁt from

advertising their impending ovulation to males. If female spotted hyenas do advertise

l6

their sexual receptivity via scent marking than females should paste most often during
periods when they are most likely to be receptive (i.e. late lactation period). This will be
determined by systematically observing the scent marking behavior of adult female

spotted hyenas across multiple reproductive cycles.

Chapter Three

As discussed above, the functions of scent marking are best ascertained by
determining the stimulus conditions that elicit marking, and observing responses to marks
by conspeciﬁcs (Yahr 1983). In the previous chapter we considered the contexts in which
pasting occurred. In this chapter we consider the responses of spotted hyenas to
conspeciﬁc scent. Deﬁnitive functional explanations for animal signals require knowing
speciﬁcally what information is available in them (Drea et al. 2002). One of the primary
means by which to determine the information content of scent marks is by conducting
scent discrimination experiments. During these bioassays, subjects are presented with
scents from donors that differ in only a single characteristic. If over multiple trials,
utilizing multiple donors and subjects, one scent is consistently responded to in a
different manner than other scents, then the scents are interpreted as being distinctive in
some way and, therefore, capable of communicating information about the single
characteristic to receivers. In an effort to determine the information content of paste, Drea
et a1. (2002) performed a series of scent bioassays with captive spotted hyenas, through
which they demonstrated that paste contains information about the individual identity of

the donor, as well as its sex. The authors suggest that these results indicate a potential

17

reproductive function for pasting behavior, a function previously believed to be
inapplicable to spotted hyenas (Matthews 1939; Gorman & Mills 1984).

Muller-Schwarze (2001) has recommended that, if the logistic and experimental
obstacles associated with ﬁeld studies can be overcome, scent discrimination experiments
should be conducted on free-ranging animals in natural environments. His reasoning was
that the context for proper responses are likely to be encountered only in the ﬁeld.
Recently, others have echoed his sentiment (Drea et al. 2002; Begg et al. 2003). In this
chapter, we will conduct scent discrimination experiments on three different clans of
free-living spotted hyenas in the MMNR. The experiments will be staged primarily at
communal den sites, where a majority of clan members congregate on a nightly basis. As
hyenas are primarily nocturnal (Kruuk 1972; Kolowski et al. 2007), the bioassays will be
conducted in the evening with the aid of night-vision binoculars. The paste samples that
will be used in the experiments were obtained previously from the anal pouches of
hyenas anaesthetized in the MMNR.

We have three speciﬁc objectives in performing these experiments. First, to
discern whether hyenas discriminate between paste secretions from clanmates and non-
clanmates, and if so, whether discrimination is due strictly to subjects’ familiarity with
donors, or whether donors from different clans have group-speciﬁc odors, as has
previously been reported (Hofer et al. 2001; Burgener et al. 2006). Second, to verify that
Spotted hyenas are able to discriminate between paste samples based on the sex of
donors. Third, to further pursue the suggestion by Drea et al. (2002) that pasting has a
reproductive function, by determining whether paste from female donors contains

information about their reproductive state. In concert with Chapter Two, these scent

l8

discrimination experiments will provide us with a more comprehensive understanding of

the adaptive signiﬁcance of pasting in the spotted hyena.

Chapter Four

Parasitic and commensal organisms that reside on animal bodies have likely
inﬂuenced behavioral evolution to a large extent (Jog & Watve 2005). Among these
organisms are the bacteria associated with mammalian Skin surfaces, which are widely
believed to generate many of the volatile semiochemicals used by mammals during
chemical communication (Albone et al. 1974; Leyden et al. 1981; Rennie et al. 1990;
Parekh 2002; Alexy et al. 2003; Buesching et al. 2003; James et al. 2004; Voigt et al.
2005). One place these bacteria ﬂourish is in mammalian scent organs (Albone 1984;
Wyatt 2003). Most terrestrial members of the order Carnivora possess anal scent glands
that provide warm, moist, anaerobic environments, ideal for the proliferation and growth
of bacteria (Albone et al. 1978; Gorman & Trowbridge 1989). These scent glands also
provide a substrate of lipids and proteins that are metabolized by the bacteria, generating
volatile carboxylic acids that ﬁgure prominently in communication among carnivores
(Gorman et al. 1974; Albone et al. 1978; Gorman & Trowbridge 1989). Through the
production of carboxylic acids, commensal bacteria may provide carnivores with
individual and/or group-speciﬁc odors (Gorman 1976). This explanation for the
derivation of identity-speciﬁc odors in mammals is called the ‘fermentation hypothesis’
(Gorman 1976). All else being equal, if individual animals in a group harbor individually
distinctive microbial populations in their scent glands, then their scent proﬁles will each

be unique, and their scent marks will provide information to receivers about their

19

individual identity. Additionally, through frequent bodily contact and overmarking of one
another’s scent marks, members of the same social group can potentially come to share a
common microbial population. Consequently, the scent proﬁles of members of the same
social group will be similar and their scent marks will provide information to receivers
about their group membership (Albone et al. 1974; Gorman 19.76; Buesching et al. 2003).
Group-speciﬁc odors potentially facilitate group cohesion and make cooperative defense
of territories more efﬁcient (Rasa 1973; Kruuk et a1. 1984; Buesching et al. 2003).

In this chapter, we test the hypothesis that commensal bacteria are responsible for
the production of group-speciﬁc odors in the Spotted hyena (Burgener et al. 2008).
Hyenas are highly gregarious, cooperatively territorial, large carnivores that have clan-
distinctive scent proﬁles (Hofer et a1. 2001; Burgener et al. 2008). If group-speciﬁc odors
are the products of bacterial fermentation, then members of the same social group should
have more similar bacterial populations in their scent glands than do members of
different social groups. During a previous test of this prediction in the red fox, Vulpes
vulpes, the prediction was not substantiated. However, the results in that study were
likely biased by the differential recoverability of anaerobic bacterial forms in culture
(Albone et al. 1978; Schloss & Handelsman 2004). We will avoid this potential bias by
extracting bacterial 16S rDNA from paste secretions obtained from anaesthetized spotted
hyenas, and performing recombinant cloning to generate a ‘snapshot’ of the microbial
community of each individual hyena’s anal pouch.

If group-speciﬁc odors in Crocuta are due to commensal bacteria, then after
controlling for sex and reproductive state, clanmates Should harbor more similar

microbiota in their anal pouches than do hyenas from different clans. If we do ﬁnd

20

differences between clans, further analysis of the generated clone libraries will allow us
to describe speciﬁcally how bacterial populations differ among clans. If microbial
differences between clans are not found, then we will have convincingly demonstrated
that the bacterial mechanism for group-speciﬁc odors hypothesis is not applicable to
spotted hyenas, as they come into frequent contact with conspeciﬁcs, and regularly

overrnark one another’s scent marks.

21

CHAPTER ONE

Theis, K. R., A. L. Heckla, J. R. Verge, and K. E. Holekamp (2008). The ontogeny of
pasting behavior in free-living spotted hyenas, Crocuta crocuta. In: Chemical Signals in
Vertebrates 11 (Hurst, J .L., R.J. Beynon, S.C. Roberts, T.D. Wyatt, Eds.). Springer, New
York. pp. 179-188.

CHAPTER ONE
THE ONTOGENY OF PASTING BEHAVIOR IN FREE-LIVING SPOTTED HYENAS
(CROCUTA CROCUTA)
INTRODUCTION

An efﬁcient communication system is a critical component of any animal society.
Communication among members of a society can function to coordinate group activities,
modulate intragroup aggression, and facilitate complex social interactions. In mammals,
communication is often achieved via chemical signals that may convey information about
the caller’s reproductive state, social status, or individual, species and group identity
(Wyatt 2003). One particularly common form of chemical communication among
mammals is scent marking, which is the deliberate deposition of glandular secretions,
urine or feces in the environment (Marler 1967; Ralls 1971; Eisenberg & Kleiman 1972;
Johnson 1973). For many mammals, the temporal and spatial patterns of scent marking
facilitate territorial maintenance and mate attraction via honest advertisement of resource
holding potential (Rich & Hurst 1998; Gosling & Roberts 2001). However, it is
becoming increasingly clear that the function of scent marking varies not only among
species, but also among different age/sex classes within the same species (e.g. Drea et al.
2002; White et al. 2003). To date, few studies have addressed the ontogeny of scent
marking in mammals (but see Rasa 1973; French & Cleveland 1984; Woodmansee et al.
1991; Palagi et al. 2002). The current study describes the ontogeny of anal gland scent
marking (‘pasting’) in free-living spotted hyenas. Although some earlier studies have

considered pasting in spotted hyenas (Kruuk 1972; Mills 1990; Woodmansee et al. 1991;

23

Boydston et al. 2001; Drea et al. 2002), we still do not have a complete understanding of
its potential adaptive functions, particularly among juveniles.

Spotted hyenas are large, gregarious carnivores found throughout sub-Saharan
Africa, living in clans of up to 90 individuals (Kruuk 1972). They are unique in that their
societies are female-dominated, with adult immigrant males being lower-ranking socially
than all natal animals. Natal animals inherit their mothers’ rank positions within the
clan’s linear dominance hierarchy (Holekamp & Smale 1993; Engh et al. 2000), while
immigrant males queue for social status amongst themselves (East & Hofer 2001). In

addition to reasons associated with their unique biology, spotted hyenas are also

 

intriguing subjects for communication studies because they emit a rich array of visual,
vocal and chemical signals. Their most conspicuous chemical Signaling behavior is
pasting, wherein a hyena deposits anal gland secretions on grass stalks or, occasionally,
on other objects in the environment (Kruuk 1972; Mills 1990). In the Spotted hyena, all
age/sex classes frequently ‘paste’ (Mills & Gorman 1987), although the pasting-like
behaviors of many juveniles are likely ineffective as juvenile hyenas reportedly do not
produce paste within their anal glands until they are 12-18 months old (Kruuk 1972).
Juvenile pasting therefore presents a Darwinian puzzle (Alcock 2005), in that performing
ineffectual behaviors typically reduces one’s ﬁtness.

The objectives of this study were: 1) to systematically determine the age at which
spotted hyenas begin producing paste in their anal glands, 2) to determine how rates of
juvenile pasting and overmarking vary during development, 3) to evaluate the effect of

social rank on pasting behavior among juveniles, 4) to identify the immediate behavioral

24

contexts in which pasting occurs among juvenile Spotted hyenas, and 5) to consider the

potential functions of pasting by juvenile hyenas in light of this study’s other ﬁndings.

METHODS

This study was conducted in the Talek region of the Masai Mara National
Reserve, Kenya. The subject population was a single large Crocut'a clan that has been
monitored continuously since 1988. Its members were identiﬁed by their unique spot
patterns and other conspicuous characteristics, such as ear notches. Sex was determined
from the dimorphic glans morphology of the erect phallus (Frank et al. 1990). Cubs were
assigned birthdates by estimating their ages (:1: 7d) when ﬁrst observed, based on their
pelage and size. In this study, hyenas were considered cubs until they were one year old,
and subadults from 13 to 30 months old. We further deﬁned early, middle and late
subadulthood as 13-18, 19-24 and 25-30 months respectively, in effort to draw
comparisons with a previous study of captive juvenile hyenas (Woodmansee et al. 1991).
Since juvenile hyenas ‘inherit’ their social ranks directly from their mothers (Engh et al.
2000), each subject was assigned its mother’s relative rank within the clan at the time of
the subject’s birth.

Between 1994 and 2003, 113 different Talek hyenas were anaesthetized with
Telazol (W.A. Butler Co.; 6.5 mg/kg) delivered from a COz-powered riﬂe (Telinject
Inc.). To determine the age at which hyenas begin to produce paste, we examined the anal
pouch of each hyena for the presence/absence of secretions. If a hyena was darted
multiple times during this period, we only used its datum from the ﬁrst darting. The ages

of immigrant males were estimated from patterns of tooth wear using methods described

25

 

by Van Horn et al. (2003). To evaluate the effects of age and sex on the likelihood of
paste production by the anal glands we employed multiple logistic regression (SAS v8.2,
2001). Throughout this study, results were considered statistically Signiﬁcant when P <
0.05 (two-sided). All descriptive statistics have been presented as mean i standard error.

In the current study we also utilized archived behavioral observation data
gathered between 1988 and 1996. We collected repeated behavioral measures from 14
male and 12 female hyenas, from birth through 30 months of age. Behavioral
observations were made at dens, kills and elsewhere. During observations, scent marking,
agonistic interactions, and greeting behaviors were recorded via all occurrence sampling
(Altmann 1974). Rates of pasting and overmarking were calculated as the number of
events per individual per hour observed (Woodmansee et al. 1991). Since scent marking
rate data were not normally distributed, comparisons were made from these data using
nonparametric Friedman ANOVA, Mann-Whitney U (Statistica v6.1, 2002), and
Wilcoxon signed-rank tests (SAS).

Each pasting event was assigned a context based upon the behavioral and social
events occurring in the two minutes preceding the pasting behavior. If a hyena pasted on
grass stalks, etc. that had been previously scent marked by others in the same observation
session, the context was ‘overmarking’ (om). If a hyena pasted in a spot not previously
scent marked by others in the observation session, yet just after another hyena had pasted
elsewhere, the context was ‘ adjacent marking’ (am). If a hyena pasted following a
submissive and/or aggressive interaction with a social partner, or following a signiﬁcant,
high-intensity aggressive interaction between group-mates in the same observation

session, the context was ‘agonism’ (agon). If a hyena pasted after greeting with another

26

hyena (i.e. mutual snifﬁng of the anogenital region), or after offering or receiving an
unrequited greeting invitation (i.e. one hyena lifts its hindleg thereby conspicuously
exposing its anogenital region to a second hyena that does not subsequently sniff the
region), the context was ‘greeting’ (grt). If pasting was preceded by more rare stimuli,
such as participation in border patrols, inter-clan conﬂicts or allomarking of hyena cubs,
the context was categorized as ‘other’ (0th). If more than one of these contexts preceded a
paste event, we assigned the context that most immediately preceded the paste event.
Lastly, if a pasting event was not preceded within two minutes by evident stimuli, the
context was ‘spontaneous’ (spon). Paste events that were not preceded by two complete
minutes of all-occurrence sampling were labeled ‘unknown.’ For each hyena, we
calculated the proportion of pasting events that occurred in each of these contexts.
Proportion data were arcsin square root transformed and comparisons were made using
paired and unpaired t-tests (Statistica). We employed the Holm’s sequentially-rejective
Bonferroni method to evaluate the statistical signiﬁcance of multiple comparisons

(Shaffer 1995).

RESULTS

Anaesthetized hyenas ranged from 7 to 143 months of age. Investigation of their

anal pouch contents revealed a highly signiﬁcant effect of age (Wald x2 = 16.85, P <

0.0001), but not of sex (Wald x2 = 0.007, P = 0.93), on the likelihood of paste production.

It appears that spotted hyenas do not consistently produce paste until their third year of

life (Figure 1.1). The youngest age at which secretions were observed within an anal

27

 

16

 

 

0.6 -

20

 

0.4 -

46

 

0.2 -

proportion anal pouches with paste

 

 

 

 

 

 

 

 

 

 

 

 

 

0.0 I L l 1 I
1-12 13-24 25-36 37-48 49+

age (months)

Figure 1.1. Effect of age on paste production in Spotted hyenas. Numbers above the bars

indicate the number of hyenas sampled in each age class.

28

pouch was 9 months, although paste production at this age is uncommon, as only three of
the 12 cubs darted in their ninth month had appreciable secretions in their anal pouch.
Eight cubs were darted prior to nine months of age, and all their pouches appeared devoid
of secretions. In this study, the sex, social rank and body mass of cubs did not appear to
affect paste production (unpublished data), however, these factors should be considered
in a more controlled manner, with larger sample sizes, before ﬁrm conclusions are drawn.
For the behavioral portion of this study, we observed each of the 26 juvenile

hyenas for 98 :t 7 hrs (range: 37 — 165). Collectively, they exhibited 1491 paste events,

for an average of 57 :t 7 pastings per hyena (range: 8 — 138). Among both males and

females, cub and subadult pasting rates were similar (Wilcoxon signed-rank test; N“: 14,

= -16.5, P = 0.33; Nf: 12, S = 14, P = 0.30). Additionally, neither males nor females
varied their pasting rates among early, middle and late subadult periods (Friedman
ANOVA; Nm= 8, x2 = 1.75, P = 0.42; Nf: 12, x2 = 3.87, P = 0.14; analyses of individuals
observed > 3 hr in each period). Although there was no effect of age, males pasted more

frequently than females as cubs (Mann-Whitney U test, U = 42, P = 0.03; Figure 1.2), but

not as subadults (U = 76, P = 0.71). Pasting rates were not inﬂuenced by social rank in

any sex/age class examined here (male cub: R2 = 0.133, F1512 = 1.85, P = 0.2; female cub:
R2 < 0.001, F1,10 < 0.01, P = 0.98; male subadult: R2 = 0.082, F1712 = 1.07, P = 0.32;
female subadult: R2 = 0.073, F1,10 = 0.79, P = 0.4).

Males overmarked more when they were subadults than cubs (Wilcoxon signed —
rank test, S = 32.5, P = 0.04; Figure 1.3). Males did not, however, vary the proportional

abundance of overmarking across early, middle and late subadult periods (Friedman

29

1-0 r - male(14)
* [:l female (12)

 

0.8 r

.9
ox

 

 

.3
4:.

 

pastes per hour

 

0.2

 

 

 

 

 

 

0.0

 

subadult

Figure 1.2. Hourly pasting rates of juvenile male and female spotted hyenas by age. The
asterisk indicates a statistically signiﬁcant difference (P < 0.05). Numbers in parentheses

indicate samples sizes.

30

0.4 -

- male (14)
l:| female (12)

 

.9
C)

.O
N

 

 

proportion overmarks

.0
p—i

 

 

0.0

 

 

 

 

 

 

 

 

 

subadult

Figure 1.3. Proportion of Spotted hyena pasting events that were overmarks of conspeciﬁc

scent marks, by age and sex. Asterisks indicate statistically signiﬁcant differences (P <

0.05). Numbers in parentheses indicate sample sizes.

31

ANOVA, N = 6, x2 = 3.0, P = 0.22). Female overmarking did not vary with age (cub vs.

subadult: S = -15.5, P = 0.24; across subadult periods: N = 6, Friedman ANOVA x2 = 3.5,

P = 0.17). Although cubs did not overmark disproportionately by sex (Mann-Whitney U
test, U = 75, P = 0.67), subadult males overmarked more than females (U = 37, P = 0.02).

Social rank did not inﬂuence the extent to which hyenas overmarked in any age/sex class
(male cub: R2 = 0.21, Fm: 3.14, P = 0.1; female cub: R2 < 0.001, PL10 < 0.01, P = 0.97;
male subadult: R2 < 0.001, F1,12 < 0.01, P = 1.0; female subadult: R2 = 007,131,10 = 0.71,
P = 0.42).

In general, the primary contexts ofpasting for all age/sex classes in this study
were spontaneous, overmarking, adjacent marking, and following agonistic interactions
(Figure 1.4A,B). Among males, the proportion of pasting events occurring spontaneously
and in the context of overmarking varied between cub and subadult periods (paired t-test,
N = 9; Spon: t = 3.49, P = 0.01; om: t = -2.4, P = P = 0.04; males pasting > 10 times each
category; Figure 1.4A,B), with male subadults exhibiting decreased spontaneous pasting
behavior and increased overmarking. Males did not vary their propensity for pasting in
other contexts between cub and subadult age periods (am: t = -0.09, P = 0.93; agon: t = -
1.46, P = 0.18; grt: t = -0.4, P = 0.7). Female hyenas did not vary the proportion of
pasting events occurring in each behavioral context between cub and subadult periods (N
= 6; Spon: t = -0.49, P = 0.64; om: t = 0.71, P = 0.51; am: t = 0.61, P = 0.57; agon: t = -
1.41, P = 0.22; grt: t = 0.09, P = 0.94; Figure 1.4A,B).

Among cubs, there were no signiﬁcant differences between males and females in

the behavioral contexts preceding pasting behavior (t-test; Spon: t = 0.39, P = 0.7; om: t =

0‘5 l -maIe(ll) A

 

 

 

 

 

 

 

 

 

 

1:] female (8)

In
*5 0.4 .
o
5
§ 0.3 ~
a
Q.
5-:
c
g 0.2 -
p
E
8. 01
e .
G.

0.0 -

spon om am agon grt oth
0.5 - * B

 

 

 

proportion of paste events

 

 

 

 

 

if

spon om am agon grt 0th

 

 

 

Figure 1.4. Behavioral contexts of pasting by A) cubs and B) subadults. Approximately
one percent of paste events were assigned an ‘unknown’ context. These events were
excluded from all analyses. Asterisks indicate statistically signiﬁcant differences after
Holm’s sequentially-rejective Bonferroni method was employed. Numbers in parentheses

indicate sample sizes.

33

-0.24, P = 0.81; am: t = 0.77, P = 0.45; agon: t = 0.84, P = 0.41; grt: t = -0.73, P = 0.47;
Figure 1.4A). However, sex did signiﬁcantly affect pasting context among subadults,
with females pasting spontaneously more often than males (t = -2.89, P = 0.01), and
males overmarking more often than females (t = 2.39, P = 0.03; Figure 1.4B). Male and
female subadults did not vary in the degree to which they pasted in other contexts (am: t

= 1.26, P = 0.23; agon: t = -1.03, P = 0.32; grt: t = -0.28, P = 0.78).

DISCUSSION

Although a few individuals in our study population produced paste as early as
nine months, it appears that spotted hyenas do not consistently generate anal gland
secretions until their third year of life. This is when both males and females mature
sexually (Matthews 1939), with males beginning to disperse from their natal clans (Frank
et al. 1995 ; Smale et al. 1997), and many females conceiving their ﬁrst litters (Holekamp
et al. 1996). It is also during this period that juveniles in our study began to participate in
territorial defense (see also Boydston et al. 2001). Therefore, it appears that paste
production in spotted hyenas is similar to exocrine scent gland activity in many
mammals, in that it seemingly coincides with the appearance of adult typical behaviors
(Yahr 1983). Interestingly, although consistent paste production appeared to coincide
with reproductive maturity, this did not lead to an increased frequency of scent marking
activity. Our ﬁnding contrasts with that of Woodmansee et al. (1.991), who found that
captive spotted hyenas reared in peer-only groups greatly increased their frequency of

pasting late in their second year of life. The difference between their study and ours

34

reafﬁrms the inﬂuence of social and environmental contexts on mammalian scent
marking behavior (French & Cleveland 1984; Muller-Schwarze 2001).

In this longitudinal study of 26 individuals, the frequency of ‘pasting’ by juvenile
spotted hyenas was consistently high throughout early development. Furthermore, we
have conﬁrmed that cubs engage in this behavior despite their not producing anal gland
secretions (Kruuk 1972). With greater emphasis being placed in recent studies on
understanding selection during juvenile life stages (Holekamp & Smale 1998; Altmann &
Altmann 2003), this seemingly ineffectual behavior presents an intriguing Darwinian
puzzle (Alcock 2005). Although pasting by juvenile hyenas does not appear to bear
substantial costs (e.g. time, energy, social repercussion), and althoughjuvenile pasting
need not necessarily be an adaptive trait (Gould & Lewontin 1979), given the high
frequency at which young hyenas pasted in this study, functional hypotheses should be

considered.

Results in relation to possible adaptive functions of pasting by juvenile hyenas
One adaptive hypothesis suggests that, rather than depositing scents in the
environment, cubs are acquiring group-speciﬁc odors that facilitate conspeciﬁc
recognition of them as members of the group before they are widely known as
individuals. Rasa (1973) proposed that mature African dwarf mongooses (Helogale
undulate rufula) communally allomark newborns to ensure that they are recognized by
conspeciﬁcs as mongooses rather than prey. Interestingly, in the current study, we also
observed multiple instances of mature animals allomarking young cubs (unpublished

data). For spotted hyena cubs, overmarking conspeciﬁc paste may be a mechanism for

35

actively donning group-speciﬁc odors. Among social animals, being effectively
recognized as a member of a group can signiﬁcantly reduce the amount of aggression an
individual receives from its social partners (e. g. Hurst et a1. 1993). Being recognized as a
group member may be especially important for young carnivores because they have not
yet developed the morphological weapons or behavioral repertoire necessary for
defending themselves against conspeciﬁcs, particularly in species that exhibit conspeciﬁc
infanticide, such as spotted hyenas (Hershel & Skinner 1991; Frank et al. 1995 ; personal
observation). There have been numerous reports of other carnivores rubbing their cheeks,
necks and backs against conspeciﬁc scent marks (Reiger 1979). For example, European
badger (Meles meles) cubs occasionally rub themselves against the extruded scent glands
of adult conspeciﬁcs that are engaged in scent marking (Buesching et al. 2003), and
meerkats (Suricata suricatta) frequently rub their bodies against communal scent posts
(Moran & Sorensen 1986). In both species, the rubbing behavior is believed to result in
the actors picking up, rather than depositing, group-speciﬁc scents (Buesching et al.
2003; Moran & Sorensen 1986).

A second functional hypothesis to explain pasting by hyena cubs is that it exposes
them to symbiotic bacteria that are potentially necessary for paste production in general,
and for the generation of group-speciﬁc odors in particular. In bobtail squid (Euprymna
scolopes), bacterial colonization is a prerequisite for normal development of the
luminescent organ in juveniles (McFall—Ngai & Ruby 1991). The rudimentary organs of
juvenile squid have ciliated projections that facilitate bacterial inoculation. Once
inoculation has occurred, the ciliated structures regress (McFall-Ngai & Ruby 1991). We

see a rough analogy between these ciliated structures and overmarking of conspeciﬁc

36

 

paste by spotted hyena cubs. In mammals, bacteria are likely responsible for generating
many of the semiochemicals emanating from scent glands (Albone 1984). One
mechanistic hypothesis to explain the generation and maintenance of mammalian group-
speciﬁc odors is that members of a social group, through bodily contact and overmarking,
come to Share a common microbial population in their scent glands (Albone et al. 1974;
Gorman 1976). Provided they exist on Similar organic substrates, shared microbial
communities should theoretically produce shared volatile odorants as a by-product of
their metabolism. Pasting therefore potentially provides both short and long-term group-
recognition beneﬁts to hyena cubs, by facilitating their acquisition of both group-speciﬁc
odors and bacteria.

We found that male cubs pasted more frequently than their female peers,
suggesting that this sexually dimorphic behavior might be mediated by differential
exposure of males and females to speciﬁc gonadal steroid hormones during development
(Yahr 1983). Given that most cubs did not produce paste secretions, and that male and
female cubs did not signiﬁcantly differ with respect to the behavioral contexts in which
they pasted, it is doubtful that the functional explanations for pasting behavior differ
between male and female cubs. Although male and female subadults did not paste at
different rates (see also Mills & Gorman 1987; Woodmansee et al. 1991), they did paste
under different behavioral and social circumstances. Subadult females pasted
spontaneously twice as often as males did, whereas subadult males overmarked
conspeciﬁc scent marks far more frequently than did females. These sex differences in
context suggest that the functions of pasting likely differ between subadult females and

males. Spontaneous scent marking is typically viewed as a form of indiscriminate self-

37

 

advertisement wherein the signaler communicates information about its identity and
presence to members of its social group as well as its neighbors (Wolff et al. 2002). Self-
advertisement may be particularly beneﬁcial for subadult female Spotted hyenas as
females in this species are philopatric (Smale et al. 1997). For female Crocuta,
establishing and reinforcing their individual identity, group membership, and position
within the clan’s linear dominance hierarchy may facilitate their access to contested
resources and ultimately their reproductive success (Holekamp & Smale 1993; Smale et
a1. 1993; Holekamp et al. 1996; Holekamp & Smale 1998). It is important to note that
self-advertisement would beneﬁt all subadult females regardless of social rank and that,
in the current study, female scent marking activity was not inﬂuenced by social rank.

It is less apparent why, in an adaptive sense, subadult males overmark clanmates’
scents more often than do subadult females. Scent marking directly on top of, or
immediately beside, the previously deposited scent marks of conspeciﬁcs is typically
perceived as an honest and effective means by which to advertise one’s competitive
ability to both competitors and prospective mates (reviewed by Gosling & Roberts 2001;
Hurst & Beynon 2004; Johnston 2005 ; Ferkin & Pierce 2007). However, at least among
subadult male hyenas this does not seem to be the case. In the current study, subadult
males directed three-fourths of their overmarks toward other subadult male and adult
male clanmates (unpublished data), suggesting that subadult males may be targeting their
competitors for overmarking. We also found that males overmarked conspeciﬁcs
Signiﬁcantly more as subadults than cubs, and that subadult males exhibited a tendency to
increase overmarking activity as they matured (unpublished data). However, we did not

ﬁnd any correlation between overmarking activity and social rank. Therefore, a

38

competitive function for subadult male overmarking behavior appears unlikely. Of
course, a confounding factor here is that we cannot be certain which individuals were
producing secretions in their anal glands and, therefore, were capable of actually
depositing scent marks upon others’ marks when overmarking. Ideally, an analysis
assessing the degree of correlation between overmarking and social rank would include
only individuals that were already producing paste. However, even for subadult males
that have begun to produce both paste in their anal glands and sperm in their testes,
overmarking is unlikely to make them more competitive for reproductive opportunities
because male hyenas seldom sire offspring in their natal clans (Engh et al. 2002; Van
Horn et al. 2008).

Although overmarking by subadult male hyenas is unlikely to function in
advertising dominance or attracting mates, given the high degree of sociality in Crocuta,
overmarking by subadult males may function to reinforce their group membership,
thereby facilitating social cohesion. Spotted hyena clans are ﬁssion-fusion societies in
which subgroups change frequently in size and composition (Kruuk 1972; Smith et al.
2007). Frequently, when reuniting after being separated, members of hyena clans engage
in conspicuous greeting behaviors, suggesting that the maintenance of social bonds is
very important in this Species (East et al. 1993). Additionally, extrusion of the anal pouch
during greeting is most common among individuals who meet only infrequently (East et
al. 1993; Hofer et al. 2001). As subadult males approach their third year of life they begin
the process of natal dispersal (Smale et al. 1997). The dispersal process entails multiple
exploratory forays into foreign territories, resulting in subadult males becomingly

increasingly absent from their natal clanmates for extended periods of time (Smale et al.

39

1997). It is conceivable that the heightened overmarking activity of subadult males
effectively compensates for their infrequent social interaction with clanmates as a result
of their dispersal forays, and facilitates their being non-aggressively received upon their
return (see Hurst et al. 1993; Rostain et al. 2004). A problem that must be resolved here,
however, is why males increase overmarking activity, but not self-advertising
Spontaneous pasting behavior, when they become subadults. Resolution may lie in the

mechanisms underlying scent deposition and information transfer in this species.

Results in relation to mechanistic considerations

When animals deposit scent marks upon those of conspeciﬁcs, three mechanistic
results are possible (Johnston et al. 1994). First, the t0p mark may entirely mask the
bottom mark. Second, the top and bottom marks may remain distinct and individually
identiﬁable. Third, the top and bottom marks may blend, forming a new, composite scent.
Scent-masking is typically attributed to solitary species and is believed to function in
concealing the scent marks of competitors and/or in mate-guarding (Johnston et al. 1994;
F erkin & Pierce 2007). In the current study of juvenile hyenas, overmarking behavior
was most common among subadult males. The heightened frequency of overmarking by
subadult male hyenas is not consistent with the scent-masking hypothesis because
overmarking by subadult males was not correlated with social status, male hyenas seldom
sire cubs in their natal territories (Engh et a1. 2002), and mate-guarding is not an effective
reproductive strategy for males in this species (East et a1. 2003; Szykman et al. 2007).

If the top and bottom scent marks remain distinct and individually identiﬁable

following overmarking, then a chemical bulletin board is generated (Johnston et al.

40

1994). Chemical bulletin boards are assumed to occur among wide-ranging mammalian
Species, as they permit each participant to advertise their identity and presence to all
receivers that happen by (Johnston et al. 1994; F erkin & Pierce 2007). Consequently,
contributing to bulletin boards is an effective means of self-advertisement, perhaps even
more so than scent marking spontaneously. Given the aforementioned value of self-
advertisement for subadult females, if overmarking by spotted hyenas generates a
chemical bulletin board, then subadult females should overmark clanmates as frequently,
if not more often, than subadult males. We have shown this is not the case.
Scent-blending through overmarking is believed to occur among colonial species,
as the consequent composite, group-speciﬁc odor permits rapid and efﬁcient recognition
of group-mates (Johnston et al. 1994; F erkin & Pierce 2007). We have already
hypothesized that spotted hyena cubs overmark clanmates to acquire group-speciﬁc odors
that facilitate their being recognized as members of their social group, and perhaps also to
acquire group-speciﬁc bacterial communities responsible for the production of clan-
speciﬁc odors. We further postulate that subadult male hyenas, by frequently contributing
their scent to the composite odor of the clan (see Mills & Gorman 1987), simultaneously
reinforce their group membership and ensure their possession of group-speciﬁc bacterial
communities. Either subadult females can accomplish these same objectives by
overmarking less often, perhaps as a result of their continued presence in the clan’s
territory, or the importance of advertising individual identity supercedes communicating

group membership for subadult female hyenas.

41

Conclusions

In this study, we have proposed multiple hypotheses for pasting behavior in male
and female juvenile hyenas. Although most spotted hyena cubs do not produce
appreciable quantities of paste in their anal scent glands, they still frequently engage in
pasting behavior. We here suggest that hyena cubs are picking up, rather than depositing,
scents. AS spotted hyena paste secretions bear clan-Speciﬁc chemical Signatures (Hofer et
al. 2001), cubs may be donning clan-Speciﬁc odors that facilitate their being recognized
as members of the group before they are widely known as individuals. Furthermore, by
engaging in pasting behavior, cubs may be acquiring symbiotic bacteria that potentially
ﬁgure prominently in producing the odorants associated with paste secretions.

Among subadults, paste production is much more common, as approximately
two-thirds of all subadult spotted hyenas investigated in this study had perceptible
amounts of paste in their anal pouches. Given the high incidence of spontaneous pasting
behavior among subadult female hyenas, the current study suggests that pasting may
serve as a form of self-advertisement for them. Among subadult males, however,
Spontaneous pasting was relatively infrequent, while overmarking of clanmates’ scent
marks was common. Whereas in most mammals, overmarking is believed to serve
competitive and reproductive functions (Johnston 2005; Ferkin & Pierce 2007), our data
suggest that, at least among subadult male spotted hyenas, overmarking may reinforce the
signaler’s group membership.

We emphasize that each of these hypotheses is speculative and that collectively
they do not represent an exhaustive list of the potential adaptive functions of scent

marking by juvenile spotted hyenas. A line of further inquiry that appears most necessary

42

is an investigation of the bacterial communities of Spotted hyena anal pouches. A
preliminary study revealed clan-speciﬁc differences in the volatile odorants emanating
from paste secretions (Hofer et al. 2001). Although a bacterial mechanism for these
differences has been proposed (Burgener et al. 2006), this hypothesis has not yet been
tested. In the future, we will determine the degree of variability in the bacterial
communities among Spotted hyena anal pouches, and, if variability exists, whether it is
clan-Speciﬁc. If it is established that the bacterial communities in Crocuta anal pouches
are clan—Speciﬁc, then the hypotheses introduced in the current study will have received

considerable support.

43

CHAPTER TWO
INTRAGROUP FUNCTIONS OF SCENT MARKING BY ADULT SPOTTED
HYENAS (CROCUTA CROCUTA)
INTRODUCTION

Scent marking, deﬁned as the deliberate deposition of urine, feces, and/or the
products of exocrine scent glands in the environment, is a common form of
communication among mammals (Ralls 1971; Eisenberg & Kleiman 1972; Johnson
1973). The information potentially communicated via mammalian scent marking includes
the donor’s species, group, sex, age and individual identity, as well as its reproductive,
social, nutritional and health status (Ferkin et al. 1997; Gosling & Roberts 2001; Zala et
al. 2004; Hurst 2005). In general, mammals signal this information to achieve three broad
functional objectives: 1) territory and resource defense, 2) maintenance of dominance and
social relationships, and 3) mate attraction and reproductive synchrony (Ralls 1971;
Eisenberg & Kleiman 1972; Johnson 1973; Gosling 1982; Rich & Hurst 1998; Heymann
2000)

Although historically, emphasis has been placed on the ﬁtness beneﬁts accrued by
adult males scent marking in territorial contexts (Gosling & Roberts 2001; Hurst &
Beynon 2004), recent studies have demonstrated that the functions of scent marking vary
widely, not only among mammalian species, but also among different age, sex,
reproductive and social classes within a single Species (e.g. Begg et al. 2003; Miller et al.
2003; Swaisgood et a1. 2004; Hurst 2005; Lewis 2006; Jordan 2007). Additionally, it has
become increasingly clear that we need to develop a more comprehensive understanding

of the functions of scent marking in highly gregarious species (Scordato & Drea 2007;

44

Muller & Manser 2008), especially given that social complexity potentially drives
signaling complexity (e. g. Freeberg 2006). In this study, we evaluated potential functions
for scent marking in a highly gregarious carnivore, the Spotted hyena, Crocuta crocuta.

Spotted hyenas live in complex social groups, called clans, that typically contain
40 - 80 individuals (Kruuk 1972; Trinkel et al. 2007). Members of each clan
cooperatively defend their group’s territory from neighboring hyenas, and they also
defend ungulate kills from both other hyenas and African lions, Panthera leo (Kruuk
1972; Mills 1990; Henschel & Skinner 1991; Boydston et al. 2001). Hyena clans contain
multiple breeding males and multiple overlapping generations of females. Clans are
structured by rigid linear dominance hierarchies (Frank 1986a,b). In contrast to most
mammalian social systems, a hyena’s position within its clan’s dominance hierarchy is
not determined by its ﬁghting ability. Instead, natal hyenas ‘inherit’ the social ranks of
their mothers (Frank 1986b; Holekamp & Smale 1991, 1993; Smale et al. 1993; Engh et
al. 2000), and adult immigrant males, all of whom are subordinate to natal animals, queue
for dominance status with other immigrant males (Smale et al. 1997; East & Hofer 2001).

In addition to possessing a rich repertoire of vocal and visual signals, spotted
hyenas communicate chemically through greeting ceremonies, scent-rubbing, forepaw
scratching, defecating at latrines, and anal gland “pasting” (Kruuk 1972; Bearder &
Randall 1978; Mills 1990; Henschel & Skinner 1991; East et al. 1993). Anal gland
pasting is a common and conspicuous form of scent marking wherein hyenas drag their
extruded anal pouches over grass stalks and, in doing so, typically deposit a thin layer of
anal gland secretion near the tops of the stalks (Matthews 1939; Kruuk 1972; Mills

1990). Previous investigations of pasting behavior among spotted hyenas have shown that

45

it ﬁgures prominently in territorial maintenance (Kruuk 1972; Mills & Gorman 1987;
Henschel & Skinner 1991; Boydston et al. 2001). Indeed, some have suggested that
pasting functions exclusively in territorial maintenance for C rocuta (Gorman & Mills
1984). However, this seems unlikely given that all age/sex classes of spotted hyenas paste
(Mills & Gorman 1987; Woodmansee et al. 1991; Theis et al. 2008), and that they often
do so, even in areas where hyena population density is high, at locations other than
territorial boundaries, such as communal dens where intruders are typically not seen
(Hofer et al. 2001; Theis et al. 2008). Furthermore, spotted hyenas have been observed
pasting in a wide variety of social and behavioral contexts (Kruuk 1972; Mills & Gorman
1987; Woodmansee et al. 1991; Theis et al. 2008). Although others have suggested that
pasting has intragroup functions for adult spotted hyenas (Mills & Gorman 1987;
Woodmansee et al. 1991; East et al. 1993; Hofer et al. 2001; Drea et al. 2002), no one has
yet systematically evaluated these hypotheses. Therefore, the objectives of our study
were to document the circumstances under which pasting occurs, and to evaluate four
hypotheses suggesting intragroup functions for pasting behavior by adult spotted hyenas
(Table 2.1). This was achieved by evaluating pasting rates and determining the broad and
Speciﬁc socio-ecological contexts in which pasting occurs among free-living hyenas.
Determining the contexts in which scent marking occurs among free-living animals has
been deemed critical in ascertaining its functional signiﬁcance (Muller-Schwarze 2001;

Thomas & Kaczmarek 2002; Begg et al. 2003).
The ﬁrst hypothesis we evaluated (H1 in Table 2.1) suggests that pasting

functions to advertise dominance status to competitors (Woodmansee et al. 1991). Scent

46

Table 2.1 Intragroup functions of pasting by adult Spotted hyenas: hypotheses and
predictions. Hypotheses are neither exhaustive nor mutually exclusive. Italicized letters in
parentheses indicate the data set used to test each prediction. The composition of each

data set is explained in the Methods section.

Hypothesis Prediction Data Set

H1: Advertise dominance status
P1: positive correlation between social rank and pasting rate (B)
P2: pasting rates highest in vicinity of competitors (A)

H2: Facilitate social cohesion
P1: pasting rates higher after absence from clan than before (C)

H3: Advertise female sexual receptivity
P1: pasting rates highest when likelihood of fertility is greatest (D)

P2: during fertile periods, pasting rates highest in vicinity of males (D)

H4: Advertise male availability for reproductive opportunities
P1: positive correlation between pasting rate early in tenure and relative
reproductive success (A)

 

47

marking has been Shown to be an honest indicator of competitive ability in many
mammals (Ralls 1971; Gosling & Roberts 2001; Rich & Hurst 1998, 1999). If pasting has
a competitive function among spotted hyenas, then we expected to observe a strong
positive correlation between social status and pasting frequency (Ralls 1971). It was
previously suggested that pasting does not communicate dominance status to competitors
because adult immigrant male hyenas seemingly engage in the behavior more frequently
than other members of hyena clans, and they are also the lowest-ranking members of
their respective clans (Mills & Gorman 1987). However, given that all adult female
hyenas are higher-ranking socially than all adult immigrant males, dominance should be
considered separately for each sex when determining the effect of social status on scent
marking behavior (Kappeler 1990; Woodmansee et al. 1991). Therefore, in this study, we
evaluated the dominance hypothesis separately for males and females. We additionally
tested the prediction that, if pasting advertises dominance status, then rates of pasting
should be highest when hyenas are in the vicinity of their intragroup rivals (Macdonald
1985; Mech et al. 2003).

The second hypothesis we assessed suggests that pasting functions to facilitate
social cohesion within clans by reafﬁrming group membership (East et al. 1993). In
house mice, it has been shown that scent marking plays a critical role in maintaining
familiarity and tolerance among group members (Hurst et al. 1993). If pasting serves this
purpose in adult spotted hyenas, then individuals who are absent from their clans for
extended periods of time should paste more frequently following their return than they

did prior to their departure (East et al. 1993).

48

The third hypothesis suggests that pasting functions to advertise female sexual
receptivity. In many mammals, females emit chemical cues that provide males with
information about their reproductive condition and receptivity (Johnson 1973; Johansson
& Jones 2007). The beneﬁts of advertising sexual receptivity are likely greatest for
females of asocial species in which mates are difﬁcult to locate, and in Species in which
females are dominant to, and aggressive toward, males such that males are hesitant to
approach females (Petrulis & Johnston 1997; Palagi et al. 2004). Female spotted hyenas
are indeed frequently aggressive toward male conspeciﬁcs (Frank 1986b; Szykman et al.
2003; East et al. 2003), so female hyenas could potentially beneﬁt from advertising their
impending ovulation to males. Furthermore, male Spotted hyenas associate most closely
with females when they are fertile (Szykman et al. 2001), so information about ovulation
and female sexual receptivity appears to exist in this Species. This information is
potentially communicated via paste. Paste secretions do contain information on donor
sex, and it has been suggested that they contain information on female reproductive state
as well (Drea et al. 2002). If female spotted hyenas advertise their sexual receptivity via
pasting, then they should paste most frequently during periods when they are most likely
to be fertile: during late lactation. This is the period during which females typically wean
their latest litters and become pregnant again (Szykman et al. 2003). Additionally, during
fertile periods, female hyenas should paste most frequently when in the presence of
prospective mates.

The fourth hypothesis suggests that pasting by adult males functions to advertise
their presence in the clan, and thereby their availability to females for reproductive

Opportunities (Mills & Gorman 1987). Male hyenas seldom sire offspring in their natal

49

clans and most males disperse (Smale et al. 1997; Van Horn et al. 2008). Among adult
immigrant males, reproductive success is largely dependent upon a male’s length of
tenure in his current clan (Engh et al. 2002; East et al. 2003). Most immigrant males do
not successfully sire cubs until they have resided in a clan for at least two years, and most
progeny appear to be sired by males with at least four years of tenure (Engh et al. 2002;
East et al. 2003). As females exercise an unusual degree of mate choice in this species
(Honer et al. 2007), and as females appear to be selecting mates that have been
immigrants in their clans for several years (Engh et al. 2002; East et al. 2003), males
could seemingly beneﬁt from advertising their presence in their new clans early and
often. Since male hyenas are frequent targets of female aggression (Frank 1986b;
Szykman et al. 2003; East et a1. 2003), pasting may afford newly tenured males the
opportunity to advertise their presence to females without suffering the harassment
incurred by being in close physical proximity to them (Mills & Gorman 1987). This
hypothesis does not require that newly arrived males paste at higher rates than males who
have resided in a clan for a long time, but it does predict that males who invest in pasting
early in their tenures should ultimately experience greater reproductive success than
males who do not do so. Speciﬁcally, among immigrant males, there should be a positive
correlation between males’ pasting rates early in their residencies and the proportional

abundances of cubs they sire during their tenures.

50

METHODS
Study area

We have been intensively Studying a single clan of spotted hyenas in the Talek
region of the Masai Mara National Reserve, Kenya, since May, 1988. This region
consists of rolling grassland, scattered brushland, and narrow stretches of riparian forest.
The Reserve supports sizable concentrations of resident ungulates (Thomson’s gazelle,
Gazella thomsonii; topi, Damiscilus korrigum; impala, A epyceros melampus), as well as
large migratory herds of wildebeest (Connochaetes taurinus) and zebra (Equus burchelli)
that are present with the resident antelope from approximately June through September
each year (Frank 1986a; Holekamp et al. 1999; Cooper et al. 1999). As part of our long-
term study of the Talek clan, we have conducted bi-monthly prey censuses within the
clan’s home range, as described in Holekamp et al. (1999). In the current study, we
averaged the total prey counts of censuses occurring in each month of the study, and
characterized each month as having high (474 — 3068 head; upper third of monthly
values), medium (264 - 473, middle third), or low (34 — 263, lower third) prey
abundance.

There are typically two rainy seasons in the Reserve, one from March — May, and
another from November —— December, although inter-annual variation exists (Darling
1960; Holekamp et al. 1999). As part of our long-term monitoring efforts we recorded
rainfall levels (mm) daily from a rain gauge located within the Talek clan’s home range.
In the current study, we summed the daily rainfall measures within each month of the

study, and characterized each month as having high (105 — 336 mm, upper third of

51

monthly values), medium (47.1 — 104.9, middle third), or low (3 — 47, lower third)

rainfall volumes.

Study subjects
The current study uses systematic behavioral data collected daily via direct

observations of Talek hyenas between September, 1988 and April, 2000. During that time
the Talek clan maintained a territory of approximately 65 kmz, and was comprised of 20

— 23 breeding females, 10 — 12 adult immigrant males, and 30 — 40 offspring (Holekamp
et al. 1999). Members of the clan were identiﬁed by their unique spot patterns and other
conspicuous physical characteristics. Sex was determined from the dimorphic glans
morphology of the erect phallus (Frank et al. 1990). Birth dates were assigned to natal
hyenas by estimating their ages (i 7 days) when ﬁrst observed above ground, based
primarily on their pelage and size. We restricted our investigation to natal females that
were at least 30 months of age, and immigrant males. Maternity assignments were based
on genotyping and nursing associations, and paternity assignments were determined
through genotyping alone, as described in Engh et al. (2002) and Van Horn et al. (2004).
We assigned individuals social ranks based on the observed outcomes of dyadic agonistic
interactions (Holekamp & Smale 1991; Smale et al. 1993). By convention, we assigned

the highest-ranking individual in each intrasexual dominance hierarchy a rank of one.
Observational and quantitative methods

Throughout the study, behavioral observations were made twice daily, around

dawn and dusk, when hyenas were socially most active (Kruuk 1972; Kolowski et al.

52

2007). Observations were made at dens, ungulate kills, and elsewhere within the clan’s
territory. At each observation session we recorded the identity of each hyena present, the
time in minutes during which each hyena was observed, and all occurrences (Altmann
1974) of pasting, agonistic interactions, and greeting behaviors. Greeting is a common
afﬁliative behavior among hyenas, wherein two individuals stand head-to-tail, lift their
rear legs closest to the other, and sniff one another’s anogenital region (Kruuk 1972; East
et al. 1993).

For each session, we generated a male to female ratio by dividing the number of
immigrant males present by the number of adult females present. Ratios were then
categorized as high (> 1.3), medium (0.67 — 1.3), or low (< 0.67). Rates of pasting were
calculated as the number of events per individual per hour observed (Woodmansee et al.
1991). Pasting rate data were not normally distributed and for many analyses could not be
successfully transformed. Therefore, comparisons were made using nonparametric
Friedman ANOVA, Mann-Whitney U, and Wilcoxon matched pairs tests (Statistica v6.1,
2002). As suggested by Mundry & Fischer (1998), we reported exact values (T) for
Wilcoxon matched pairs tests with sample sizes less than 16, and asymptotic values (Z)
for tests with sample sizes of 16 or greater. Holm’s sequentially-rejective Bonferroni
method was used to evaluate the statistical signiﬁcance of multiple comparisons (Shaffer

1995). To assess the monotonic relationships between continuous variables, we utilized
Spearman’s rank correlation coefﬁcients (rs) (Statistica; Quinn & Keough 2002). For all

analyses in which pasting rate was the dependent variable, individual hyenas were

excluded if they were not observed for at least three hours in each pertinent independent

53

variable category. Descriptive statistics have been presented as mean i standard error
throughout.

Each pasting event was assigned a context based upon the social and behavioral
events that occurred in the observation session during the preceding two minutes (Table
2.2). For each hyena, we calculated the proportion of its pasting events that occurred in
each context. Proportion data were arcsin square root transformed and comparisons were
made using paired and unpaired t-tests (Statistica). Hyenas were excluded from

contextual analyses if they did not exhibit at least ten paste events.

Data sets used in study (A, B, C and D)

Given the broad scope of functional hypotheses addressed in this study, and the
fact that it encompassed 12 years of behavioral observation data, multiple data sets were
used to address the hypotheses (Table 2.1). The composition and value of each data set is

explained below.

Data set A

To obtain general descriptive information on pasting behavior, including the
socio-ecological contexts in which it occurs among free-living adult hyenas, we studied
its occurrence among 16 immigrant male and 12 natal female hyenas (Table 2.3). The
immigrant males were studied from the time they were ﬁrst seen interacting with
members of the Talek clan until they either died or secondarily dispersed to another clan.
Each of these males resided in the clan for at least two years, but died or secondarily

dispersed before the end of the study. These criteria ensured that we obtained an accurate

54

Table 2.2. Socio-ecological contexts of pasting behavior considered in the current study.

If more than one of these contexts preceded a paste event, we assigned the context that

most immediately preceded the paste event.

Context

abbn

Description

 

spontaneous

overmarking

adjacent marking

agonism

greeting

other

unknown

spon

om

am

agon

grt

oth

unk

paste event was not preceded by evident stimuli

paste placed on grass stalk scent marked previously
by others in the same observation session

paste event followed within two minutes of pasting
by another, but paste was not placed on a grass stalk
scent marked previously by others in the same
observation session

paste event followed a submissive and/or aggressive
interaction with a social partner, or following a
high-intensity conﬂict between clanmates in the
same observation session

paste event followed reciprocal leg-lifting and
anogenital investigation with a social partner, or
after offering or receiving a greeting invitation
(i.e. leg-lifting)

paste event preceded by rare stimuli, such as
arriving in the observation session, participating in
border patrols, inter-clan conﬂicts, lion — hyena
conﬂicts, or allomarking of hyena cubs

paste event was not preceded by two minutes of all—
occurrence sampling

 

55

Table 2.3. Time for which each hyena contributing to data set A was observed, as well as

the number of paste events each hyena exhibited during this study.

 

Months Hours Pastes Alive, in clan, at study’s end?

Name Sex Observed Observed Observed (If yes. age in years)
BAIL f 70 204 46 yes, 8.3
BERN f 59 95 126 yes, 7.4
CR f 51 61 20 yes, 6.8
HOB f 65 97 42 no

KIP f 48 131 65 yes, 6.5
MP f 40 77 18 yes, 5.8
PAR f 42 56 30 yes, 6.8
PT f 33 96 19 no
SCY f 25 55 26 no

SD f 34 87 50 no
SEIN f 31 99 30 yes, 5.1
WR f 41 83 70 yes, 5.9
98 m 32 37 6 no

FA m 78 181 48 no

FN m 104 218 103 no
HOL m 55 118 34 no
MARK m 31 28 2 no

POS m 38 32 5 no
QUAI m 58 118 67 no
QUO m 28 42 19 no
RCN m 76 121 23 no

RS m 49 36 7 no

SIM m 47 53 23 no
STG m 56 36 11 no

SY m 90 258 135 no

TZ m 72 86 35 no

VD m 26 24 23 no

ZIP m 54 1 17 28 no

 

56

measure of each male’s reproductive success within the Talek clan. We calculated each
male’s reproductive success as the number of cubs it sired divided by the total number of
cubs born throughout its tenure. Cubs that had not been genotyped were excluded from
these analyses. The 12 natal females were studied from the time they reached 30 months
of age until either their death or the conclusion of the study. The initial selection criterion
for these females was that they lived to be at least 4.5 years old. When multiple females

from the same matriline met the initial criterion, we randomly selected one for study.

Data set B

To best determine the effect of social status on pasting behavior we studied all the
adult male (N = 15) and female (N = 20) hyenas present in the Talek clan from
September, 1997 through June, 1999. During this time, both intrasexual dominance
hierarchies remained stable, and there was minimal emigration or immigration by adult

males. Pasting rates were calculated for each of the 35 adult hyenas during this period.

Data set C

Although each female spotted hyena resides in a particular territory, She may
occasionally be absent from her clan’s territory for periods of days or weeks, presumably
while on foraging or exploratory visits to other areas (Hofer & East 1993a,b,c; Holekamp
et al. 1993). This data set allowed us to ascertain whether being separated from clanmates
by absence from the clan’s territory for extended periods of time affected scent marking
activity. We studied all adult females that went unseen by observers in the Talek territory

for periods of at least one month between 1989 and 1999. For each female studied, we

57

determined pasting rates during the two week periods preceding extended absences, and
the two week periods following absences. Immigrant males were not studied because,
once they had successfully immigrated into the clan, they were seldom absent for

extended periods of time.

Data set D

To evaluate the effect of reproductive state on female pasting behavior, we
studied all females that successfully raised at least one offspring through weaning,
between 1996 and 2000. Pasting rates were calculated for both pregnancy and lactation,
deﬁned as follows (Szykman et al. 2003):
Pregnancy (p): The gestation period in Crocuta lasts 110 days (Matthews 1939; Kruuk
1972). Therefore, females were considered pregnant for the 110 days preceding the births
of their litters.
Lactation (ll, 12, 13): The length of the lactation period was determined from observations
of nursing behavior and weaning conﬂicts (Szykman et al. 2003). As the length of the
lactation period varies among female spotted hyenas (Holekamp et al. 1996; Holekamp &
Smale 2000), we divided the overall lactation interval into three periods of equal length
for each female (Szykman et al. 2003). Females often became pregnant while still nursing
offspring from previous litters. When this occurred, females were categorized as

pregnant, rather than lactating.

58

RESULTS

Pasting rates among adult hyenas did not vary signiﬁcantly with local prey

abundance (data set A; Friedman ANOVA, Nmalc = 16, X2163 = 2.00, P = 0.368; Nfemalc =
12, X2”; = 3.50, P = 0.174; Figure 2.1A). Additionally, pasting rates among adult males

did not vary with rainfall (Friedman ANOVA, 1216,; = 0.50, P = 0.779; Figure 2.1B).
Adult females, however, increased their scent marking activity during periods of high

rainfall (x2122 = 13.17, P = 0.001; Table 2.4). Although location did not signiﬁcantly
affect male pasting activity (Friedman ANOVA, X2162 = 2.63, P = 0.269), adult females

pasted more often at dens than at ungulate kills or elsewhere in their territory (x2123 =

12.67, P = 0.002; Table 2.4, Figure 2.2). Given the elevated rates of pasting by females at
dens, and the additional observation that lions were seldom seen at hyena dens during this
study, we excluded observation sessions conducted at dens from an analysis evaluating
the effect of lion presence on pasting behavior. The presence of lions in observation
sessions inhibited pasting behavior in both males (Wilcoxon matched pairs, N = 16, Z =
3.15, P = 0.002), and females (N = 12, T: 0.00, P < 0.001; Figure 2.3).

Adult hyenas were observed pasting in a variety of socio-ecological contexts
(Figure 2.4). Approximately one-third of the observed paste events were spontaneous.
Overmarking and adjacent marking of conspeciﬁc scent marks were also relatively
common, as was pasting immediately after agonistic interactions with clanmates. Within
the context of agonism, neither males nor females pasted more frequently after

interactions in which they were the socially dominant partner than after those in which

they were subordinate (paired t-test, NmalC = 12, t = -1.88, P = 0.086; Nfemale = 12, t =

59

1.0

0.8

 

0.6 ‘ a

 

 

 

 

 

 

 

 

 

0.4 ~

0.2 -

pastes per hour
I
|

 

 

 

 

 

 

 

 

 

0.0

 

high medium

prey abundance in home range

1.0 -

0.8 -

 

 

0.6 - —— ——

 

 

 

0.4 -

 

 

0.2 ~

pastes per hour

 

 

 

 

 

0.0

 

 

 

   

medium

rainfall levels in home range

Figure 2.1. Mean (i SE) hourly pasting rates of adult Spotted hyenas across periods of
high, medium and low A) prey abundance, and B) rainfall within their home range.

Letters above the error bars indicate statistically signiﬁcant differences (P < 0.05).

60

Table 2.4. Results of Wilcoxon matched pairs tests comparing the hourly pasting rates of

adult female hyenas across periods of high, medium and low rainfall, as well as at dens,

ungulate kills and elsewhere (N = 12).

 

Comparison T P — value
Rainfall levels
high vs. medium 9.00 0.016
high vs. low 0.00 0.001
medium vs. low 28.00 0.424
General location
den vs. kill 1.00 0.001
den vs. other 8.00 0.012
kill vs. other 33.00 0.677

 

61

0.9 .

 

- male(N=16)
0.8 - ——b——— D female (N =12)
0.7 -
0.6 -

 

 

 

 

 

pastes per hour

 

 

 

 

 

 

 

 

 

den kill other

general location

Figure 2.2. Mean (i SE) hourly pasting rates of adult male and female spotted hyenas at
dens, ungulate kills, and elsewhere. Letters above the error bars indicate statistically

signiﬁcant differences (P < 0.05).

62

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0-6 F - male (N = 16)
[:1 female (N = 12)
a
l-
:3
O
.1:
I-
o
G.
g
V)
a:
D.
_b_
lions absent lions present

Figure 2.3. Mean (i SE) hourly pasting rates of adult male and female spotted hyenas
during observation sessions at which lions were absent or present. Letters above the error

bars indicate statistically signiﬁcant differences ( P < 0.05).

63

0°" .male(N=12)

[j female(N=12)

 

 

 

 

 

 

 

 

 

proportion of paste events

 

 

 

 

 

 

 

 

 

 

 

 

 

spon om am agon

Figure 2.4. Socio-ecological contexts of pasting by adult male and female spotted hyenas.
Data represent individuals observed to paste at least 10 times. The x-axis labels refer to
spontaneous, overmarking, adjacent marking, agonism, greeting and other contexts,
respectively. Approximately six percent of paste events were assigned an ‘unknown’
context. These events were excluded from all analyses. Asterisks indicate statistically

signiﬁcant differences (P < 0.05).

64

0.46, P = 0.652). Despite the substantial variability in the circumstances preceding
pasting behavior, two sex differences in pasting context were evident. First, a greater
proportion of female paste events followed participation in greeting behaviors (t-test, t =
3.62, P = 0.002). Among both males and females, a majority of paste events in this
context (75/ 109, 69%) occurred after individuals participated in full greetings (i.e.
reciprocal leg-lifting and investigation of the anogenital region). The remainder of paste
events in this context were equally represented by Situations in which the pasting
individual lifted its leg in invitation to another but the invitation was not reciprocated
(18/109, 17%), and in which the pasting individual had received a greeting invitation that
it did not reciprocate (16/109, 14%). The second evident sex difference in pasting context
was that a greater proportion of male paste events were overmarks of conspeciﬁc scent
marks (t = -2.10, P = 0.047). Additionally, adult males and females overmarked the
scents of different groups of conspeciﬁcs. Of the 113 observed instances of overmarking
by adult males, 86 (76%) involved pasting over scent marks deposited by other adult
males. Adult males did occasionally overmark scents deposited by juveniles (23, 20%),
but they seldom overmarked adult females (4, 4%). Of the 54 observed instances of
overmarking by adult females, 25 (46%) involved pasting over scents deposited by other

adult females, and 29 (54%) involved scents deposited by juveniles.

Testing predictions of H1: Advertise dominance status

Immigrant male hyenas pasted more frequently than did adult females (data set B;
Mann-Whitney U, Nmale = 15, Nfemalc = 20, U: 60.00, P = 0.002; Figure 2.5). Among

males, we observed a strong positive correlation between social rank and pasting

65

0.6 -

 

0.5 - 15

 

 

.9
is

 

 

20

 

pastes per hour
O
DJ

.0
to

 

 

 

0.1 -

 

 

 

 

 

0.0

 

male female

Figure 2.5. Mean (i SE) hourly pasting rates of adult male and female spotted hyenas
residing concurrently in the same clan. The asterisk indicates a statistically signiﬁcant

difference (P < 0.05). Numbers above bars indicate individuals sampled.

66

rate (Spearman rank R, r, = -0.60, P = 0.018; Figure 2.6A), but no such correlation was

evident among females (rs = -0.12, P = 0.616; Figure 2.68). The proportion of paste
events that were overmarks of conspeciﬁc scent marks was not correlated with

intrasexual rank in either sex (male: rs = -0.33, P = 0.225; female: rs = -0.02, P = 0.946).

The ratio of adult males to adult females present in the immediate social
environment had a marked effect on the frequency of scent marking by immigrant males.

Males pasted most frequently during observation sessions in which their social partners
were primarily other males (data set A; Friedman ANOVA, N = 16, X2162 = 8.00, P =
0.018; Table 2.5, Figure 2.7). For females, the effect of the ratio of males to females in
sessions was not statistically signiﬁcant (N = 7, 787; = 5.43, P = 0.066), but this result

appears to be a product of low sample size: 5 females were excluded from this analysis
because they were too seldom present in high ratio sessions. Regardless, signiﬁcant
differences between males and females were evident. During observation sessions in
which at least two-thirds of the hyenas present were male, males pasted at signiﬁcantly
higher rates than females (Mann-Whitney U, U = 17.00, P = 0.008). During sessions in
which at least two-thirds of the individuals present were female, females pasted more
frequently than males (U = 17.00, P = 0.008).

Interestingly, although the presence of adult male clanmates clearly elicited
pasting behavior by immigrant males, the appearance in the Talek territory of new
potential immigrant males did not appear to affect the frequency of pasting by established
immigrant males. The hourly pasting rates of established immigrant males (data set A, but
excluding data from ﬁrst year of tenure), did not differ signiﬁcantly between the two

weeks after the appearance of new potential immigrant males in the territory and periods

67

1.2 .

1.0 .

pastes per hour

0.0 ~

1.2

1.0 -

pastes per hour

0.2 -

0.0 ~

 

 

 

0.8 -

0.6 -

0.4 ~

 

m . o - . . . a
0 2 4 6 8 10 12 14 16
male intrasexual social rank
’ B

O
O
O O
O
O O O O o O
l o o o o O
9 . a . l 9 a

 

10 12 14 16 18 20

female intrasexual social rank

Figure 2.6. Effect of intrasexual social rank on rates of pasting behavior among adult A)

male, and B) female spotted hyenas. By convention the highest-ranking member in each

intrasexual hierarchy was assigned a rank of l.

68

Table 2.5. Results of Wilcoxon matched pairs tests comparing hourly rates of pasting by

adult male spotted hyenas in observation sessions characterized by high, medium and low

ratios of adult males to adult females (N = 16).

Comparison Z

P — value

 

Ratio of males to females

high vs. medium 2.28
high vs. low 2.79
medium vs. low 1.66

0.023
0.005
0.098

 

69

-male(N=16)
1.0 -CIfemale(N=7)

0.8 ~

 

 

_o
ox

 

 

.9
4:.

 

pastes per hour

.9
N

 

 

 

   

 

0.0

 

 

 

 

 

 

 

 

 

 

 

high medium

low

ratio of males to females in session

Figure 2.7. Effect of the sex composition among conspeciﬁcs in the immediate social

environment on the frequency of pasting behavior by adult male and female spotted

hyenas. Asterisks indicate statistically signiﬁcant differences (P < 0.05).

70

in which no potential immigrants were observed (N = 15, Wilcoxon matched pairs, T =

51, P = 0.639; new males: 0.42 i 0.06, no new males: 0.43 i 0.14).

Testing predictions of H2: Facilitate social cohesion

From 1989 — 1999, nine resident adult females were absent from the Talek clan
for periods of at least one month and were observed often enough to be included in the
social cohesion analysis (7.6 i 1.1 extended absences per female; 67.6 i 15.3 days per
absence). Eight of these females were very low-ranking socially, and one was mid-
ranking. Among them, prolonged absence from the clan had a signiﬁcant, positive effect
on pasting rate (Figure 2.8). Females pasted approximately three times more often upon
their return after an absence from the clan’s territory of at least one month than they did

prior to their departure (data set C; Wilcoxon matched pairs, N = 9, T: 5.00, P = 0.039).

Testing predictions of H3: Advertise female sexual receptivity

Although presumably physically capable (Matthews 1939; Holekamp et al. 1996),
most of the females in data set A were not reproductively active at thirty months of age.
From thirty months of age until onset of their ﬁrst conﬁrmed pregnancy, females pasted
nearly three times more often than after they started breeding (Wilcoxon matched pairs,
N = 10, T: 1.00, P = 0.004; 0.92 i 0.17 pastes / hr vs. 0.30 i 0.05 pastes / hr).

Among breeding females, pasting rates were not highest during late lactation, as
predicted by this hypothesis; rather, females pasted considerably more often during

pregnancy than during early, mid, or late lactation (data set D; Friedman ANOVA, N =

71

2.5 -

 

 

2.0 ~

p—A
LII

 

p—A
O

 

pastes per hour

 

0.5 -

 

l
l

 

 

 

 

 

 

 

0.0 ‘
2 weeks prior 2 weeks following

timing with respect to prolonged absence from the clan

Figure 2.8. Mean (i SE) hourly pasting rates of adult female hyenas before and after
absences from the Talek clan’s territory lasting longer than one month (N = 9). The

asterisk indicates a statistically signiﬁcant difference (P < 0.05).

72

18, x2183 = 15.80, P = 0.001; Table 2.6, Figure 2.9). During late lactation, females were
seldom observed in sessions where no adult males were present, so we were unable to
determine whether the presence of at least one male increased the likelihood of female
scent marking behavior. However, females in late lactation did not signiﬁcantly vary their
pasting rates among sessions in which 1 — 2, 3 — 4, and 5 or more immigrant males were

present (Friedman ANOVA, N = 10, 78102 = 3.07, P = 0.215; Figure 2.10). Thus none of

the predictions of this hypothesis were conﬁrmed.

Testing predictions of H4: Advertise male availability for reproductive opportunities
The rates at which immigrant males pasted during their ﬁrst two years in their
new clans was not Signiﬁcantly correlated with their relative reproductive success
throughout their tenures (data set A; Spearman rank R, N = 16, r, = 0.02, P = 0.956;
Figure 2.11). Additionally, the rates at which immigrant males pasted in their third year
of tenure and beyond were not signiﬁcant predictors of relative reproductive success (r, =
0.20, P = 0.451). Lastly, the proportion of male pasting events that were overmarks of
conspeciﬁc scent marks, in both their ﬁrst two years in new clans and throughout the
remainder of their tenures, did not appear to affect their reproductive success (ﬁrst two

years: rs = -0.21, P = 0.444; remainder of tenure: N = 14, r, = 0.10, P = 0.742). Thus, as

with H3, none of the predictions of this hypothesis were conﬁrmed by our data.

73

Table 2.6. Results of Wilcoxon matched pairs tests comparing the hourly pasting rates of

adult female hyenas across various reproductive states (N = 18).

Comparison Z

Reproductive state

pregnancy vs. early lactation 3.42
pregnancy vs. mid lactation 3.59
pregnancy vs. late lactation 2.24
early vs. mid lactation 0.97
early vs. late lactation 0.37
mid vs. late lactation 1.14

P — value

< 0.001
< 0.001
0.025
0.332
0.711
0.256

 

74

0.6 r

0.5 -

pastes per hour

0.1 -

0.0

 

0.4 ~

0.3 -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l.

 

p 11 12 B

female reproductive state

Figure 2.9. Mean (t SE) hourly pasting rates of female spotted hyenas during pregnancy,

early (11), mid (12), and late (13) lactation (N = 18). Letters above the error bars indicate

statistically Signiﬁcant differences (P < 0.05).

75

0.8 r

.0
ox

 

pastes per hour
0
4:.

 

.0
N

 

 

 

 

 

 

 

 

 

 

 

0,, . i 1_._1
1-2 3-4 5+

number of immigra nt males in session

Figure 2.10. Mean (i SE) hourly pasting rates of female spotted hyenas during late
lactation during observation sessions in which 1 — 2, 3 — 4, or more than 4 immigrant

males were present (N = 10). Differences were not statistically signiﬁcant.

76

 

 

 

 

 

 

0 O ﬁrst two years incbn

I; O remainder of tenure inclan

:

3 0.20 - 9 0

or)

CE

*5 O O

'5 0.15 ~

8

,_ o o

'17:

3 0.10» 9 0

3 o o

H-

c t O

i:

,c 0.05 ~ o

g o c

a. O O

E

c.000loe c... .0) .o . ,o L
0.0 0.2 0.4 0.6 0.8 1.0 1.2

hourly pasting rate

Figure 2.1 1. Relative reproductive success of immigrant male Spotted hyenas in relation
to pasting rate during the ﬁrst two years of residency and during the remainder of each

male’s tenure in their new clans (N = 16).

77

DISCUSSION

Scent marking ﬁgures prominently in territorial maintenance and defense for
many mammalian species (Gosling 1982; Gorman & Trowbridge 1989). However,
placing exclusive emphasis on the territorial function of scent marking in gregarious
mammals potentially overshadows important intragroup functions of scent marking, such
as social integration or reproductive advertisement (Eisenberg & Kleiman 1972; Hofer et
al. 2001). It has been suggested that pasting behavior in the spotted hyena functions
exclusively in territorial maintenance and defense (Gorman & Mills 1984). Here we have
demonstrated that, in addition to functioning in territorial defense, pasting has multiple
intragroup functions and that at least some of these functions are sexually dimorphic. Our
analyses support the hypothesis that adult immigrant males, but not adult females,
advertise their dominance status via pasting. They further suggest that females utilize
pasting to facilitate social cohesion within clans. Unfortunately, our data were
insufﬁcient to permit us to test this hypothesis among adult males. Lastly, our data fail to
support hypotheses suggesting that pasting advertises female sexual receptivity or that it

serves as a form of male sexual advertisement.

Assessment of H1: Advertising dominance status

Scent marking advertises dominance status to competitors and potential mates in
many mammals (Ralls 1971; Gosling & Roberts 2001; Hurst & Beynon 2004). However,
noting that immigrant males pasted more frequently than other age/sex classes of spotted
hyena, and that immigrant males are the lowest-ranking members of hyena societies,

previous researchers concluded that pasting does not function to advertise dominance

78

status among adult spotted hyenas (Mills & Gorman 1987). In this study, we have
conﬁrmed that, as is typical among mammals (Johnson 1973), adult male spotted hyenas
scent mark more often than their female peers. However, we have further demonstrated a
positive correlation between intrasexual social rank and the frequency of scent marking
among males. We did not ﬁnd a correlation between rank and scent marking frequency
among females, however. Although a deﬁnitive explanation for this dimorphism will
require further investigation, a preliminary supposition is that using scent marking to
advertise dominance status is less important for maintaining hierarchies formed among
philopatric individuals than it is for maintaining those formed among immigrants, as the
latter need to routinely accommodate the addition of strangers. Interestingly, the same
sexually dimorphic patterns in scent marking frequency have been observed in the ring-
tailed lemur (Lemur catta), another gregarious mammal with male-biased dispersal and
female-dominated societies (Kappeler 1990; Sussman 1992; Drea & Scordato 2008).

A common method used to advertise dominance status among male mammals is
to engage in competitive scent marking or ‘scent wars’ (Gosling & Roberts 2001; Hurst
& Beynon 2004), in which an individual ensures that the preponderance of recently
deposited scent marks in an area are its own rather than those of its competitors (Rich &
Hurst 1998, 1999). A high quality individual can ensure that its marks predominate by
physically excluding competitors from the area, discouraging competitors who are in the
area from engaging in scent marking themselves, and/or by effectively countermarking
scent marks deposited in the area by competitors (Rich & Hurst 1998, 1999). Although
male spotted hyenas attempting to immigrate into new clans can be viciously attacked by

males who have already successfully immigrated into those clans, this is uncommon

79

(Smale et al. 1997; East & Hofer 2001). Furthermore, hyena clans typically contain
multiple breeding males that exhibit very low rates of aggression amongst themselves
(Kruuk 1972; Frank 1986b; Mills 1990; Smale et al. 1997; East & Hofer 2001).
Additionally, we have not observed low-ranking immigrant males being attacked by
higher-ranking males subsequent to scent marking by the former. Given that socially
dominant male hyenas mark at the highest rates, that male pasting rates are highest in the
vicinity of male competitors, and that one-half of all pasting events exhibited by males in
this study were counterrnarks (overmarks + adjacent marks) of clanmates’ scent marks, it
appears that males are advertising their dominance status through the sheer volume of
their marking and potentially through their effectiveness in countermarking male
competitors as well. In the current study, the vast majority of overmarking by immigrant
males appeared to be directed toward the scent marks of other adult males. However,
before it can be deﬁnitively asserted that immigrant males are targeting the scent marks
of other males for overmarking, the opportunities available for males to overmark scents

laid down by different age/sex classes of hyena need to be controlled.

Assessment of H2: Facilitate social cohesion

The social cohesion hypothesis for scent marking suggests that it facilitates the
beneﬁts, and modulates the costs, of group-living by reafﬁrming individual presence and
membership within groups, thereby strengthening the afﬁliative ties that exist among
group members (modiﬁed from Willis & Brigham 2004; Sharpe 2005 ; McCowan et al.
2008). Persistent reafﬁrmation of presence is believed to be crucial in maintaining

familiarity and social tolerance among group members in gregarious species (East et al.

80

1993; Hurst 2005; McCowan et al. 2008). For example, when subordinate male house
mice (Mus musculus) are experimentally prevented from contributing their odors to the
substrates shared by members of their groups, they are attacked more quickly and harshly
than control males (Hurst et al. 1993). Similarly, naked mole-rats (Heterocephalus
glaber) react aggressively to colony members that have been removed from their burrow
systems for only twelve hours, suggesting that odor familiarity needs to be continually
reinforced within mole-rat societies (O’Riain & Jarvis 1997).

Maintaining social cohesion may be particularly challenging for large groups of
animals, particularly those living in ﬁssion-fusion societies wherein long-term
associations must be maintained despite frequent, and sometimes lengthy, physical
separation of group members (Willis & Brigham 2004; Lehmann et a1. 2007). Spotted
hyena clans are large ﬁssion-fusion societies in which members travel, rest and forage in
subgroups that change in composition multiple times each day (Smith et al. 2008). In
fact, it is exceedingly rare for all members of medium and large size hyena clans to be
concurrently present in the same area (Holekamp et al. 1997; Smith et al. 2008).

As predicted by East et al. (1993), in the current study we found that females that
had engaged in extended absences from their clans pasted far more often upon returning
than they did prior to departing. This suggests that pasting may facilitate re-assimilation
into clans and enhance social cohesion. Among spotted hyenas, greeting ceremonies are
particularly common during subgroup reunions, and extrusion of the anal pouch during
spotted hyena greeting ceremonies is most common among clanmates who meet least

frequently (East et al. 1993; Hofer et al. 2001). One-ﬁfth of female pasting events in the

81

current study occurred just after participation in greeting, further suggesting that pasting
facilitates social cohesion in hyenas clans, at least among adult females.

Two additional ﬁndings of this study lend support to the social cohesion
hypothesis for pasting among adult female hyenas. First, female pasting rates were
highest in the vicinity of dens (see also Hofer et al. 2001). Communal dens are the social
centers of hyena societies and are frequently attended by most clan members (Boydston
et al. 2006; White 2007). They are therefore ideal places at which to reafﬁrm one’s
presence within the group and strengthen afﬁliative ties. Interestingly, in the current
study, the observed positive effect of rainfall on female scent marking behavior remains
when we considered only observation sessions at dens (unpublished data). Given that
rainfall degrades scent marks (Alberts 1992), the increased pasting activity by females at
dens during periods of high rainfall further suggests the importance to females of
effectively communicating their presence in the area to clanmates.

The second additional ﬁnding that lends support to the social cohesion hypothesis
is that females in the current study pasted particularly often during pregnancy. This is
important to the current argument for two reasons. First, pregnant females are more apt
than females in other reproductive conditions to take extended absences from their clans’
home ranges (Holekamp et al. 1993). Second, given that communal dens are the social
centers of hyena clans, they pose signiﬁcant risks of aggression to mid- and low-ranking
females and, of greatest consequence to the argument, their offspring (Boydston et al.
2006; White 2007). Investment in self-advertisement and reafﬁrmation of group
membership prior to giving birth may be of particular value. In the future, it should be

determined whether, controlled for social rank, investment in scent marking while

82

pregnant reaps rewards in terms of reduced rates of aggression directed towards oneself

and one’s offspring at communal dens.

Assessment of H3: Advertise female sexual receptivity

Chemical advertisement of estrus status has been widely observed among female
mammals (Johnson 1973; Johansson & Jones 2007). Advertising estrus may be
particularly important for females in Species with female-dominated societies, as
advertising sexual receptivity may encourage the approach of apprehensive males,
thereby increasing the likelihood of fertilization (Petrulis & Johnston 1997; Palagi et al.
2004). Chemical cues to estrus do appear to exist in the Spotted hyena (Szykman et al.
2001); however, our analyses suggest they may not be available in paste. In the current
study, females did not paste most often during late lactation, the reproductive period
during which female hyenas are most likely to be in estrus. Instead, female hyenas pasted
most often when they were pregnant. Additionally, during late lactation, female pasting
activity did not vary with the degree of male presence in observation sessions, further
suggesting that they were not signaling sexual receptivity to immigrant males.

Our results are consistent with the idea that scent marking is likely an inefﬁcient
means through which to communicate information about estrus in highly social
mammals, given that prospective mates are oftentimes in the immediate vicinity of
females (Wolff et al. 2002; F erkin et al. 2004). Although male hyenas likely cannot
perfectly monitor the reproductive status of all females in their clans (Engh et al. 2002),
males have frequent contact with most female clanmates (Szykman et al. 2001), and

males are often observed snifﬁng females directly and snifﬁng areas in which females

83

have recently urinated or lain down (Frank 1986b). Although we currently have a
relatively poor understanding of the female reproductive cycle in Crocuta, estrus in
female hyenas appears to last for 24 — 48 hrs, and seems to reoccur every two weeks,
barring fertilization (East et al. 2003). As paste readily persists in the environment for a
month or more (Gorman & Mills 1984), pasting would not be an ideal means by which to
chemically advertise estrus in this species. Urine is a much more likely candidate, given
that mammalian urine contains numerous steroid metabolites (Hauser et al. 2008), and
that the biological signiﬁcance of estrus-speciﬁc urinary products has been demonstrated
in a wide variety of mammals (e.g. Michael 1975; Swaisgood et al. 2002; Slade et al.

2003; Rajanarayanan & Archunan 2004; Achiraman & Archunan 2006).

Assessment of H4: Advertise male availability for reproductive opportunities

Among immigrant male Spotted hyenas, reproductive success is dependent to a
large degree upon a male’s length of tenure in his current clan (Engh et al. 2002; East et
al. 2003). If female spotted hyenas are considering the length of time males have been
members of their clans when making mate choice decisions (Honer et al. 2007), then
males would seemingly beneﬁt from advertising their presence in clans early and often,
as doing so would potentially increase females’ familiarity with their scents and increase
the likelihood that they will be considered for mating opportunities in the future (Mills &
Gorman 1987). In the current study, we have, to our knowledge, provided the ﬁrst
evaluation of the potential effect of scent marking frequency on the long-term
reproductive success of male mammals in a wild population. The frequency at which

male hyenas pasted early, as well as late, in their tenures was unrelated to their relative

84

reproductive success. Moreover, we found that the degree to which males invested in
overmarking conspeciﬁc paste also did not affect their reproductive success.

Only persistently motivated, high-quality males can saturate their environments
with scent marks (Fisher et al. 2003). Therefore, females that base mate choice decisions
on their familiarity with individual male odors, the frequencies at which they encounter
the odors of individual males, or how consistently they observe individual males placing
scent marks atop those of other males, presumably select highly competitive, genetically
sound mates (Johnston et al. 1997; Fisher et al. 2003; F erkin et a1. 2005; Cheetham et a1.
2008). However, even females in species that do not use these criteria when selecting
mates can obtain valuable information about the suitability of potential mates from male
scent marks (Mech et al. 2003). Information about male quality can be obtained directly
from a single scent mark if the marks of dominant, high-quality males include
compounds, or levels of compounds, unique to their condition (e. g. Hayes et al. 2001).
Furthermore, male scent marks potentially provide females with opportunities to assess
the genetic compatibility of potential mates (reviewed by Thom et al. 2008). Given that
genetic compatibility is potentially driving mate choice decisionsiin the spotted hyena
(East et al. 2003), the role pasting by immigrant males plays in allowing females to assess

their genetic compatibility with prospective mates needs to be determined.

85

CHAPTER THREE
DISCRIMINATION OF CONSPECIFIC SCENT BY FREE-LIVING SPOTTED
HYENAS (CROCUTA CROCUTA)
INTRODUCTION
An effective communication system is a critical component of the behavioral

repertoire of nearly every animal species. Among mammals, a particularly common form
of communication is scent marking, which is the deliberate deposition of urine, feces
and/or the products of exocrine scent glands in the environment for the purpose of
signaling conspeciﬁcs (Ralls 1971; Eisenberg & Kleiman 1972; Johnson 1973). '
Mammalian scent marking can serve to facilitate the defense of resources, maintain
dominance and social relationships, attract mates, or promote reproductive synchrony
(Ralls 1971; Eisenberg & Kleiman 1972; Johnson 1973; Gosling & Roberts 2001; Rich &
Hurst 1998). However, the functions of scent marking vary widely, not only among
mammalian species, but also among different age, sex, reproductive and social classes
within a single species (e.g. Begg et al. 2003; Miller et al. 2003; Swaisgood et al. 2004;
Hurst 2005; Lewis 2006; Jordan 2007). Additionally, it has become increasingly clear
that we need to develop a more comprehensive understanding of the functions of scent
marking in highly gregarious species (Scordato & Drea 2007; Muller & Manser 2008),
especially given that social complexity potentially drives signaling complexity (e. g.
Freeberg 2006). Therefore, the purpose of the current study was to further our
understanding of the functions of scent marking in a highly gregarious large carnivore,

the spotted hyena, Crocuta crocuta.

86

The functions of animal signals are best ascertained by identifying the broad and
speciﬁc socioecological contexts in which the signals are emitted, and by determining the
types of information the signals transmit (Yahr 1983; Drea et al. 2002). In a previous
study, we described the contexts in which spotted hyenas exhibit scent marking behavior
(Theis et al. submitted). In the current study, we extended our previous ﬁndings by
conducting scent discrimination experiments with free-living spotted hyenas to determine
the information available in hyena scent marks.

Spotted hyenas live in complex social groups, called clans, that typically contain
40 — 80 individuals (Kruuk 1972; Trinkel et al. 2007). Hyena clans are ﬁssion-fusion
societies, wherein members are seldom all present in the same place at the same time;
rather members ﬁssion into subgroups that change in composition and size several times
each day (Holekamp et al. 1997; Smith et al. 2008). Hyena clans contain multiple
breeding males and multiple overlapping generations of females. In contrast to most
mammalian social systems, hyena societies are female-dominated, and social status
within a clan’s linear dominance hierarchy is not determined by size or ﬁghting ability
(Frank 1986a,b). Instead, natal hyenas ‘inherit’ the social ranks of their mothers (Frank
1986b; Holekamp & Smale 1991, 1993; Smale et a1. 1993; Engh et al. 2000), and adult
immigrant males, all of which are subordinate to natal individuals, queue for dominance
status with other immigrant males (Smale et al. 1997; East & Hofer 2001).

To mediate the complex social relationships between and within clans, spotted
hyenas utilize a rich array of vocal, visual and chemical signaling behaviors (Kruuk 1972;
Mills 1990). A common and conspicuous form of chemical signaling among spotted

hyenas is ‘pasting’ behavior (Matthews 1939). When pasting, hyenas extrude their anal

87

scent pouches and drag them over grass stalks. In doing so, hyenas typically deposit anal
gland secretions near the tops of the grass stalks (Matthews 1939; Kruuk 1972; Mills
1990). These secretions potentially remain as viable signals for more than a month
(Gorman & Mills 1984).

Previous investigations of pasting among spotted hyenas have Shown that it
ﬁgures prominently in territorial maintenance. In high-density populations, spotted
hyenas demarcate the boundaries of their territories with pastings, often engaging in
border patrols wherein several hyenas collectively engage in high rates of scent marking
along their clan’s territorial boundaries (Kruuk 1972; Henschel & Skinner 1991;
Boydston et al. 2001). In low-density hyena populations, members of clans are likely
unable to effectively replenish scent marks along their territorial borders and,
consequently, instead engage in hinterland marking, concentrating their pasting activity
beside key resources within their territories rather than along territorial boundaries (Mills
& Gorman 1987). Additionally, it has been shown that captive spotted hyenas investigate
paste from donors with which they are unfamiliar more often than they do paste from
familiar donors (Drea et al. 2002), and it has been reported anecdotally that Crocuta in
natural p0pulations react strongly to scent marks from non-clanmates found in their
territories. However, no one has yet systematically demonstrated that free-living spotted
hyenas discriminate between the scents of clanmates and others. Therefore, the ﬁrst
objective of the current study was to determine whether hyenas discriminate between
paste samples obtained from clanmate and non-clanmate donors, and to afﬁrm that they

Spend more time investigating scents from non-clanmates.

88

The second objective of the current study was to discern whether the ability of
free-living hyenas to discriminate between the scents of clanmates and others is based on
the degree of familiarity between respondents and donors, or is instead based on chemical
badges of group membership available in paste. Again, captive hyenas discriminated
between paste samples from donors with which they were familiar and samples from
donors with which they were unfamiliar (Drea et al. 2002). However, previous
investigations have also revealed that the chemical compositions of paste samples from
members of the same clans are more similar than the chemical compositions of samples
from members of different clans (Hofer et al. 2001; Burgener et al. 2008). Thus, spotted
hyena paste potentially contains information about the donor's group membership. If
discrimination between paste samples from clanmates and non-clanmates is based on
familiarity, then hyenas should additionally discriminate between samples obtained from
members of neighboring clans and those obtained from members of distant clans (Leger
1993). If discrimination is based on the presence of group-speciﬁc odors in paste, then
hyenas should discriminate between paste samples obtained from members of two distant
clans with which the respondents are presumably equally unfamiliar. Note that these
mechanistic hypotheses are not mutually exclusive.

In addition to facilitating territorial maintenance, several intragroup functions
have been proposed for spotted hyena pasting as well. Speciﬁcally, it has been suggested
that pasting serves to advertise reproductive condition (Mills 1990; Drea et al. 2002), to
advertise dominance status (Woodmansee et a1. 1991; Theis et al. submitted), and to
facilitate social cohesion within clans (East et al. 1993; Theis et al. submitted). Given that

these hypotheses primarily involve signaling potential mates or same-sex competitors,

89

they imply that paste should convey information on donor sex. Drea et al. (2002) showed
this to be the case, as captive adult spotted hyenas discriminated between paste samples
from male and female donors. The third objective of the current study was to verify this
ﬁnding with free-living hyenas. A reproductive function for pasting implies that, at least
among females, paste contains information about reproductive state. This is not
necessarily so for male hyenas, as they are in reproduction condition year round once
they reach adulthood (Matthews 1939). In many mammals, the scent marks of females
provide males with information about their reproductive condition and receptivity
(Johnson 1973; Johansson & Jones 2007). The fourth objective of this study was
therefore to ascertain whether such information is available in the paste produced by
females, by determining whether hyenas discriminate between paste samples from
pregnant and lactating female donors. A ﬁnal objective of this study was to determine,
when possible, whether these discriminative abilities or tendencies are age- or sex-

dependent.

METHODS
Study site and subjects
This study was conducted in the Masai Mara National Reserve, Kenya. The
Reserve consists of rolling grassland, scattered brushland, and narrow stretches of
riparian forest, and supports sizable concentrations of resident and migratory ungulates
(Frank 1986a; Holekamp et a1. 1999; Cooper et al. 1999). From June — December, 2004,
we conducted scent discrimination experiments on members of three distinct hyena clans

in the Reserve: West Talek, East Talek, and Mara River clans (Figure 3.1). Prior to 2000,

90

{#y’"“
hIArhAIlAA01A1; ,»
NATIONAL RESERVE g

.2

Mara River f

clan

            

, n
‘ "“p ”(1’ -

West Talek.

clan \\

R noun-nu: a. of a." a,
K “‘\ .ﬂ. ...- o .....
__\ East Talek

K/._\\/clan

Figure 3.1. Map illustrating the locations in the Reserve of the three clans used in the

current study. The territory sizes of the Mara River, West Talek, and East Talek clans
were 31 kmz, 28 kmz, and 19 kmz, respectively (Kolowski et al. 2007). During the study
period there were 30 hyenas present in the Mara River clan, 51 in the West Talek clan,

and 32 in the East Talek clan. This map was adapted, with permission, from Kolowski et

aL(2007)

there was a single clan in the Talek region of the Reserve. By 2002, however, the hyenas
comprising the Talek clan had completely ﬁssioned into two separate, yet contiguous
clans, which we now refer to as the West and East Talek clans (J .E. Smith, Michigan
State University, unpublished data). The Mara River clan resides approximately 10 km to
the west of the West Talek clan.

The hyenas in these three clans had been previously habituated to the presence of
our research vehicles. Clan members were individually identiﬁed by their unique spot
patterns and other conspicuous physical characteristics. Sex was determined from the
dimorphic glans morphology of the erect phallus, as described in Frank et al. (1990).
Birth dates were assigned to natal hyenas by estimating their ages (i 7 days) when ﬁrst
observed above ground at dens, based primarily on their pelage and size. In this study,
hyenas less than 12 months of age were considered to be cubs, and those between 13 and
30 months old were categorized as subadults. Together, cubs and subadults were called
"juveniles." All natal females greater than 30 months of age were considered to be adults,

as were all immigrant males.

Sample collection and storage
Paste samples were obtained directly from the anal scent pouches of anaesthetized
adult spotted hyenas sampled throughout much of the Reserve between 1994 and 2004.

Scent donors were immobilized with Telazol (W.A. Butler; 6.5 mg/kg), delivered from a
COz-powered darting riﬂe (Telinject), and paste samples were scraped directly from their

anal pouches with a sterilized scalpel handle. Upon collection, paste samples were placed

in sterile cryogenic vials (Coming). The vials were placed in liquid nitrogen and

remained continuously frozen until being used in experiments. The age of samples in this
study refers to the time between date of sample collection from an immobilized hyena
and date of use in a scent discrimination experiment. We utilized 63 samples from 60
different hyenas. We made aliquots of the frozen samples (25/63 samples) that were to be
used multiple times during the study.

Since the chemical composition of scent marks can change over time (Gorman
1976; Buesching et al. 2002; Cavaggioni et al. 2006), it is ideal for frozen samples to be
thawed shortly before their use in scent discrimination experiments. Given the logistics of
conducting scent discrimination experiments here with free-living large carnivores, there
were occasional instances in which experiments were postponed at the last moment, after
aliquots had already been thawed. Delays occurred as a result of rainfall, car trouble,
human and wildlife interference, and hyenas having moved their den sites without our
knowledge. When delays occurred, we used the aliquots in experiments as soon as
conditions permitted; usually the next evening. Overall, most aliquots were used in
experiments within one hour of being removed from the liquid nitrogen tank (54/88,
61%), and the vast majority (82/88, 93%) were used by the next day. Six aliquots were
used in experiments four days after being thawed. Although this was not ideal, hyena
paste is a very long-lasting signal that remains potent in the environment for a month or
more (Gorman & Mills 1984; Apps et al. 1989). Most importantly, in every trial during
this study, the paired paste aliquots had been thawed for exactly the same amount of time.

On average, aliquots were set out 2.2 i 0.3 hrs before hyenas arrived at the test Sites.

93

Location and setup of experiments

The scent discrimination experiments were conducted primarily at communal den
sites within the territories of the three study clans (39/44 trials, 89%). Communal dens are
the social centers of hyena societies and are attended by most clan members each evening
(Boydston et al. 2006; White 2007). They are therefore ideal locations for maximizing
participation in experiments. There was a period of approximately one month during the
study in which the Mara River clan did not have an active communal den. To conduct
experiments with this clan during that time, we located its members before dawn and,
based on their direction of travel, anticipated where they would rest during the day. We
then drove ahead of the traveling hyenas and set up the experiments beside the
anticipated resting sites (5/44 trials, 11%).

In this study, we conducted both three- and two-choice scent discrimination
experiments. In the three-choice discrimination experiments (e. g. Fisher et al. 2003), we
presented subjects with three new 70-cm segments of ﬂexible, plastic-covered electrical
wire (no. 7 AWG) positioned vertically in the ground with approximately 55 cm of each
wire remaining exposed. The wires were placed one meter apart in a straight line, and
were positioned perpendicular to the current wind direction, and as upwind as possible, of
the den or resting site. Each wire was ﬁrst thoroughly wiped with 70% isopropyl alcohol.
We then used clean wooden applicators to apply paste samples (~0.05 g), 5 cm from the
tops of two wires. The height was chosen because Crocuta typically place paste marks 40
— 50 cm high on grass stalks (Kruuk 1972). The top of the third wire was rubbed with a
clean wooden applicator and served as a control. The positioning of samples among the

three wires was randomized.

94

Two-choice discrimination experiments (e.g. Smith et al. 1997; Swaisgood et al.
2004), were conducted toward the end of this study for two reasons. First, as noted above,
there was a period of time when the Mara River clan did not have an active den site and
we had to rely on the behavioral cues of subjects to predict appropriate locations for
testing sites. Conserving time during setup of the experiment was a prime concern during
this period. Second, by this time it had become clear that the scent samples themselves,
not the experimental apparatus, were primarily driving olfactory investigation of the
wires, so including control wires in experiments was no longer informative (see Results).

There was only a single instance of a sample being used twice within the same
experiment. The second time it was used, it was paired with a different sample than the
one with which it had been paired originally, and the second trial was also conducted
with a different clan. There was also a single instance of a sample being presented twice
in testing of the same clan. The second presentation was for a different experiment and
occurred more than two months after the sample had ﬁrst been presented. Only two of the
eight participants in the latter trial had participated in the ﬁrst trial. Excluding the
responses of these two individuals in the latter trial would not have changed the statistical

signiﬁcance of any result in that experiment.

Descriptions of discrimination experiments

Clanmate vs. non-clanmate: In the ﬁrst experiment we tested whether spotted hyenas
preferentially investigate the scents of non-clanmates over those of clanmates. We
conducted 11 trials in which we presented participants with paste samples from a

clanmate, a non-clanmate, and a control simultaneously. In seven of the trials the paste

95

donors were female, and in four they were male. We deﬁned a "neighboring" clan as one
with a territory abutting that of the tested clan, and a "distant" clan as one separated from
the territory of the tested clan by at least one other territory. The non-clanmate paste
donor was a member of a neighboring clan in four trials, and was from a distant clan in
the other seven trials. Eight trials were conducted with West Talek, two with East Talek,
and one with the Mara River clan.

Neighboring vs. distant clan: In the second experiment we sought to determine whether
hyenas discriminate between scents from neighbors and donors from distant clans. We
conducted 10 trials in which we presented participants with paste samples from a
neighbor, a member of a distant clan, and a control simultaneously. All donors from
neighboring clans were conﬁrmed members of their respective clans within the previous
year. In six of the trials the paste donors were female, and in four they were male. Seven
trials were conducted with the East Talek clan and three were conducted with the West
Talek clan.

Distant vs. distant clan: In the third experiment we attempted to ascertain whether
hyenas discriminate between the scents of two distant clans. We assumed that the
experimental participants had had no prior contact with scent donors from either distant
clan. We conducted six trials with the Mara River clan in which we presented participants
with paste from a member of the West Talek clan, paste from a same-sex member of the
East Talek clan, and a control simultaneously. In four of the trials the paste donors were
female, and in two they were male.

Male vs. female: In the fourth experiment we determined whether free-living hyenas

discriminate between the scents of males and females. We conducted 10 trials in which

96

we simultaneously presented participants with paste samples from an adult male and
female that were both members of the same distant clan. Each of the female donors were
lactating when sampled. Five trials were conducted with the West Talek clan and ﬁve
were conducted with the Mara River clan.

Pregnant vs. late lactating: In the ﬁnal experiment we sought to ﬁnd out whether
hyenas discriminate between paste samples from pregnant females and those sampled
during late lactation. We conducted 7 trials in which we simultaneously presented
participants with samples from two female clanmates, one pregnant, the other in late
lactation. Pregnancy was deﬁned as the 110 days preceding parturition (Matthews 1939;
Kruuk 1972). Since the mean weaning age in Crocuta is 14 months (Holekamp et al.
1996), late lactation was deﬁned as the period between seven months post-parturition and
weaning. Five trials were conducted with the Mara River clan and two were conducted

with the East Talek clan.

Data collection

AS most of the discrimination trials were conducted at night (3 7/44, 84%), we
dictated observations into a digital voice recorder and transcribed our observations the
following morning. When transcribing, we determined the lengths of snifﬁng (nose
within 10 cm of stimulus) bouts by participants by timing our vocalized observations
with a stopwatch. All nighttime observations were made through night vision goggles
(AN/PVS-7 Generation 3), supplemented with an infrared light source (SONY DCR-

TRV530 digital handycam with a SONY HVL-IRH2 video infrared light).

97

We recorded the presence and identity of every hyena that approached within 5 m
downwind of the experimental stimuli. This represented the vast majority of animals in
the general vicinity of the experiments; only a few hyenas remained at the peripheries of
the sites and departed before their identities could be conﬁrmed. Given that the fur of
newborn hyena cubs is typically uniformly black, they are generally difﬁcult to
individually identify using night vision goggles. Therefore, unless they bore some feature
that facilitated our readily identifying them as individuals, newborn cubs were not
included in analyses.

When a hyena approached the stimuli, we recorded its direction of approach with
respect to the wind. Although exceptionally rare, if a hyena approached the experimental
apparatus from upwind, its responses were excluded from analyses for that experiment,
because hyenas approaching from upwind may be responding primarily to visual, rather
than chemical, stimuli. For each hyena, we recorded the total time spent snifﬁng each of
the stimuli, as well as any incidences of overmarking the stimuli (pasting on an
experimental wire). Occasionally, two hyenas concurrently investigated the stimuli.
When this occurred, to avoid potential bias due to social facilitation, we only recorded the
response of the ﬁrst hyena to approach the stimuli. If the other hyena returned to the

stimuli at some point later in the trial, we then recorded data from it.

Conclusion of experiments
Trials concluded following 1) prolonged periods of inactivity at Sites after
substantial participation (1 i 0.1 hrs; 17/44, 39%), 2) participants pulled an experimental

wire from the ground (16/44, 36%), 3) substantial rainfall (9/44, 20 %), and, 4) on two

98

occasions, equipment failure (5%). Note that, experimental trials did not conclude as
soon as any overmarking of the stimuli occurred (see also Palphramand & White 2007).
Although the application of additional scents on experimental stimuli can potentially
inﬂuence subsequent responses by other participants (Bel et al. 1999), given the logistic
challenges of the current study, we found it unfeasible to conclude trials as soon as
overmarking occurred.

As most trials were conducted at communal den Sites, the earliest participants in
them were usually cubs, who tended to emerge from den holes Shortly before their
mothers and older siblings arrived. After investigating the stimuli, cubs commonly
overmarked them. Therefore, had we ended trials after the ﬁrst incidence of overmarking,
we could not have obtained response data from subadults or adults. Through pilot trials,
we had learned that young hyenas even attempted to overmark samples on inﬂexible
wooden dowels. The wooden dowels inevitably snapped and, not only did we then lose
exceptionally rare and valuable scent samples, but the splintered dowels also posed an
injury risk to the hyenas. Therefore, we chose instead to present the samples on electrical
wires, as they re-assume their original vertical position after being overmarked and can
subsequently be investigated further by other participants. Also, it should be noted that
most spotted hyena cubs and subadults do not consistently produce paste in their anal
glands (Theis et al. 2008). Therefore, it is unlikely that they were actually depositing
paste when overmarking experimental stimuli. Only once during the entire study did an
adult hyena overmark a stimulus. Nevertheless, for each discriminative ability that

hyenas exhibited in this study, we attempted to verify that overmarking had not been a

99

factor by determining whether participants exhibited the discriminative ability prior to
any overmarking of stimuli having occurred.

In some trials, members of each age/sex class were not able to participate because
an early participant had pulled a wire from the ground, there was exceptionally poor
turnout at the test site, or rainfall occurred early during the trial. In cases with poor
turnout or rainfall, if no overmarking of the stimuli had yet occurred, we retrieved the
wires, secured the top of each with a Ziploc bag, stored the wires in a cool dry place
overnight, and used them again, at the same site, the following evening (4/44, 9%). In
instances where rainfall occurred, we quickly retrieved the wires and, in each case, paste
samples remained clearly visible on them. If a wire had been pulled from the ground or
overmarking of stimuli had occurred, and we had additional aliquots of the samples, new
wires and new samples were utilized the following night instead (5/44, 11%). Whenever
trials were replicated in these ways, we only used the data from individual hyenas that
were obtained during their ﬁrst evening interacting with the stimuli. Whenever it could be
done without unduly disturbing the hyenas, we removed the experimental wires from the
test site immediately following the conclusion of a trial. When this was not possible, the

wires were removed very early the next morning.

Quantitative analyses

An inherent challenge with stationary ﬁeld experiments on free-living animal
subjects is that researchers cannot control the arrivals and departures of individual
participants. Consequently, the lengths of time participants were present varies

substantially across trials. Additionally, the numbers of trials at which individual

100

participants were present within each experiment varies. As a result, the total time
individuals spent snifﬁng samples varied considerably. Therefore, in an effort to
standardize responses and to avoid pseudoreplication, we calculated the proportion of
time each individual hyena spent snifﬁng each stimulus type, out of the total time spent
snifﬁng all stimuli combined, across all trials within an experiment. Our sampling unit in
all experiments was the individual hyena.

Proportions could not be consistently normalized, so we employed nonparametric
statistical analyses (Statistica, v6.1, 2002) . We used Friedman ANOVA tests to assess
treatment effects for 3-choice scent bioassays. If treatment effects were statistically
signiﬁcant, we evaluated whether hyenas discriminated between non-control samples by
using Wilcoxon matched pairs tests (T -statistic; Mundry & Fischer 1998). Wilcoxon
matched pairs tests were also used to evaluate the outcomes of 2-choice bioassays. All
statistical tests were two-tailed except for Wilcoxon matched pairs tests in which we
assessed the differential investigation of samples from clanmates and non-clanmates
because a directional response had been predicted. When comparing two independent
sampling groups, we utilized Mann-Whitney U tests. Descriptive statistics have been

presented as mean i standard error throughout.

RESULTS
Cubs and subadults frequently participated in the scent discrimination
experiments (Figure 3.2). Participation by adults was less common, with adult females
participating at moderate levels and immigrant males doing so only rarely. The lack of

participation by immigrant males seemed due primarily to their low levels of attendance

101

4'0 1:] participated intrial

- present at trial

 

3.5

3.0 -

2.5 —

 

 

1.5 -

 

 

1.0 -

number of individuals

0.5 -

   

 

 

 

 

 

 

subadult adult female adult male

 

0.0

 

age / sex class

Figure 3.2. The presence and participation of different age / sex classes of spotted hyena

across all 44 scent discrimination trials conducted in this study.

102

at den sites, and their remaining at the peripheries of den areas when they were in
attendance (see also East & Hofer 1991, 1995; White 2007). Given that immigrant males
seldom participated in experiments, we excluded them from all statistical analyses,

although we have included descriptions of their responses when available.

Age of paste samples

In this study, the age of samples used in experiments ranged from one month to
ten years (3.3 i 0.2 yrs). Sample age did not affect the degree to which samples were
investigated by respondents. Analyses of samples from each class of scent donor revealed
that the time elapsed since sample collection did not affect the amount of time during

which samples were sniffed by participating hyenas (Table 3.1).

Clanmate vs. non-clanmate

Signiﬁcant treatment effects were evident for cubs (Friedman ANOVA, N = 14,

x2142 = 7.00, P = 0.03), and subadults (N = 13, x21 3,2 = 7.88, P = 0.019), but not adult
females (N = 9, x29; = 0.81, P = 0.666; Figure 3.3). Only a single immigrant male

participated in this experiment, and he exclusively investigated paste samples from non-
clanmates. The signiﬁcant treatment effects among juveniles were due primarily to their
lack of investigating control wires, as neither cubs (Wilcoxon matched pairs, T = 32.00,
P = 0.109 (one-sided)), nor subadults (T= 35.00, P = 0.377), investigated paste samples
from non-clanmates more than samples from clanmates. However, considering only trials
in which paste samples from female donors were used, both cubs (N = 1 1, T = 10.50, P =

0.023 (one-sided)) and subadults (N = 9, T= 8.00, P = 0.049) investigated paste samples

103

Table 3.1. Results of Spearrnan’s rank tests assessing the monotonic relationship between

the age of paste samples and the amount of time they were investigated. Investigation

times were calculated by averaging the responses of all participating hyenas toward each

paste sample for a given trial. Therefore, sample sizes indicate the number of paste

samples per donor class used in each experiment. Within each experiment, samples from

the two scent donor classes were analyzed separately.

 

Experiment Donor class N r5 P-value
clanmate vs. non-clanmate

clanmate 1 l -0.22 0.520

non-clanmate 1 1 -0.04 0.916
neighboring clan vs. distant clan

neighboring clan 10 -0.30 0.404

distant clan 10 0.19 0.604
distant clan vs. distant clan

West Talek 6 -0.03 0.958

East Talek 6 -0.37 0.469
male vs. female

male 10 -0.25 0.489

female 10 -0.20 0.580
pregnant vs. late lactating

pregnant 7 0.29 0.535

late lactating 7 0.54 0.216

 

104

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

0'7 ' - non-clanmate
ns. chnmate

0.6 ~ us. [:1 control

E 0.5 -

'5

m _ __

g 0.4~ f—

;- ﬂ

= T

.O 0.3 r

E / __

o __

D.

e 0.2 -

D- __
0.1 - —_
0,0 A %

 

 

 

 

 

 

 

 

 

 

 

cub subadult adult female

respondent age class

Figure 3.3. The relative amount of time cub, subadult and adult female spotted hyenas
spent snifﬁng paste samples from non-clanmates, clanmates, and control wires. “n.s.”
indicates a statistically non-signiﬁcant difference between time spent snifﬁng clanmate
and non-clanmate odors, following a signiﬁcant overall treatment effect for that age class
(P < 0.05). Sample sizes for cubs, subadults and adult females are 14, 13 and 9,

respectively.

105

from non-clanmates more than those from clanmates (Figure 3.4). Among juveniles
collectively, the effect of donor clan membership on relative investigation time was
evident before any overmarking of experimental stimuli occurred (N = 8, T= 0.00, P =
0.004 (one-sided)).

Ten juveniles (8 cubs, 2 subadults) overmarked an experimental stimulus at least

once. They did not differentially overmark paste samples from non-clanmates, clanmates,
or the control wires (Friedman ANOVA, x2103 = 0.94, P = 0.625). There was also no
signiﬁcant effect of treatment on overmarking activity when considering only trials using

female scent donors (x272 = 1.00, P = 0.607).

Neighboring clan vs. distant clan

Signiﬁcant treatment effects were again evident for cubs (Friedman ANOVA, N =

12, x2122 = 6.68, P = 0.035), and subadults (N = 10, x2103. = 7.19, P = 0.027), but not

adult females (N = 5, x253 = 3.60, P = 0.165; Figure 3.5). The signiﬁcant treatment effect

observed with cubs was again due to minimal investigation of the control wires, as they
did not differentially investigate paste samples obtained from neighboring and distant
clan members (Wilcoxon matched pairs, T = 18.00, P = 0.110). This lack of
discrimination remained when considering only trials using scent from female donors (N
= 12, T= 33.00, P = 0.677). Subadults also did not differentially investigate samples from
neighboring and distant clan members (all trials: T= 21.00, P = 0.557; female donor trials
only: N = 5, T = 7.00, P = 1.000). Three immigrant males participated in this experiment,
and each sniffed paste samples from neighboring hyenas more than samples from

members of distant clans. Six juveniles (3 cubs, 3 subadults) overmarked an experimental

106

0.9 -

 

 

 

 

 

 

 

 

 

 

 

 

- cub (N =11)
CI subadult (N = 9)
a

'8 __

2‘5

1:

v:

o

E b __b_

'5

:1

p

E

o ______.._._

Q.

E

:-

non-clanmate chnmate

clan membership of paste donor

Figure 3.4. The relative amount of time cub and subadult spotted hyenas spent snifﬁng
paste samples from female non-clanmate and female clanmate donors. Letters above the

error bars indicate statistically signiﬁcant differences (P < 0.05 (one-sided)).

107

0'8 ' - neighboring clan

 

 

 

 

  

 

 

 
   
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

n.s. distant chn
0'7 P ns. 1: 90111101
B __
E
=
m
o V
E
e y
5
'E y
o
a.
o
in
Q:
I __
4- //
subadult adult female

 

respondent age class

Figure 3.5. The relative amount of time cub, subadult and adult female spotted hyenas
Spent snifﬁng paste samples from members of neighboring clans, distant clans, and
control wires. “n.s.” indicates a statistically non-signiﬁcant difference between time spent
sniffing odors from neighboring and distant clan members, following a signiﬁcant overall
treatment effect for that age class (P < 0.05). Sample sizes for cubs, subadults and adult

females are 12, 10 and 5, respectively.

108

stimulus at least once. They did not differentially overmark the scent of neighboring and

distant clan members, or the control wires (Friedman ANOVA, X262 = 1.60, P = 0.449).

Distant clan vs. distant clan

Given that this experiment was conducted exclusively on a Single clan, we did not
obtain sufﬁcient sample sizes for examining differences in response among participants
based on their age / sex class. Therefore, for this experiment all participants, excluding

immigrant males, were analyzed collectively. There was a signiﬁcant effect of treatment
(Friedman ANOVA, N = 8, X282 = 7.16, P = 0.028; Figure 3.6), but participants did not

signiﬁcantly differ in the relative lengths of time they sniffed paste samples from West
Talek and East Talek hyenas (Wilcoxon matched pairs, T = 8.00, P = 0.195). The lone
immigrant male to participate in this experiment investigated the samples from West
Talek and East Talek hyenas for the same amount of time. Four juveniles (2 cubs, 2
subadults) overmarked an experimental stimulus at least once. There were no apparent

patterns in their overmarking responses.

Male vs. female

Neither cubs (Wilcoxon matched pairs, N = 12, T = 36.00, P = 0.850), nor
subadults (N = 11, T = 18.00, P = 0.206), differentially investigated paste samples from
male and female donors (Figure 3.7). Among subadult participants, males and females

did not signiﬁcantly differ in their responses to male and female odors (Mann-Whitney

U, Nmale = 5, Nfemale = 6, U = 7.00, P = 0.177). Unlike juveniles, adult females spent

109

0.6 ~ n.s.

 

0.5 ~

 

 

 

 

 

proportion time sniffed

 

 

 

 

 

 

 

 

0.0

 

West Talek East Talek control

clan membership of paste donor

Figure 3.6. The relative amount of time spotted hyenas spent snifﬁng paste samples from
members of the West Talek clan, East Talek clan, and control wires. Both the West and
East Talek clans' territories were separated from the test clan's territory by at least one
other clan. “n.S.” indicates a statistically non-signiﬁcant effect of clan membership,

following a signiﬁcant overall treatment effect (P < 0.05, N = 8).

110

1.0 P-male III
()_9 [:1 female

0.8 *

 

 

 

 

0.6 -

 

 

 

 

 

 

 

proportion time sniffed

 

 

 

 

   

 

 

 

 

subadult adult female

respondent age class

Figure 3.7. The relative amount of time cub, subadult and adult female spotted hyenas
spent snifﬁng paste samples from adult male and female donors. The asterisk indicates a
statistically signiﬁcant difference (P < 0.05). Sample sizes for cubs, subadults and adult

females are 12, 11 and 7, respectively.

111

substantially more time sniffing paste samples from female than male donors (N = 7, T =
1.00, P = 0.031; Figure 3.7). Due to low sample size, we could not statistically evaluate
whether adult females made this discrimination prior to any overmarking of experimental
stimuli. However, each of the four females that did investigate stimuli before any
overmarking occurred investigated paste from female donors more than paste from male
donors. A single immigrant male participated in this experiment and he spent three times
as much time snifﬁng paste from females as from males.

Eleven juveniles (6 cubs, 5 subadults) overmarked an experimental stimulus at
least once. They did not overmark paste samples from one sex more than the other (T =
17.50, P = 0.945). There was also a single instance of overmarking by an adult female.
This female overmarked a paste sample obtained from an adult male residing in a distant
clan. Interestingly, she was pregnant when she participated in this trial, as She gave birth

to a litter 77 days later.

Pregnant vs. late lactating

Sample sizes were insufﬁcient for us to evaluate effects of treatment among
different age classes separately (Ncub = 1, N subadult = 6, Naduh female = 3). Considered
collectively, however, natal spotted hyenas spent substantially more time snifﬁng paste
samples from pregnant females than females sampled in late lactation (N = 10, T = 0.00,
P = 0.002; Figure 3.8). The effect of donor reproductive state on relative investigation
time was apparent when limiting the analysis to responses that occurred before any
hyenas overmarked the experimental stimuli (N = 6, T = 0.00, P = 0.031). Three

immigrant males participated in this experiment. Two of them exclusively sniffed paste

112

0.9 ~ *

 

 

0.8 -

 

 

 

0.7 .

0.6 -

0.5 -

0.4 ~

0.3 .

 

 

proportion time sniffed

 

0.2 -

 

T

0.1

 

 

 

 

 

0.0

 

pregnant late lactating

reproductive state of paste donor

Figure 3.8. The relative amount of time natal spotted hyenas Spent snifﬁng paste samples
from pregnant and late lactating female donors. The asterisk indicates a statistically

signiﬁcant difference (P < 0.05, N = 10).

113

samples from late lactating females. The third investigated scent from a pregnant female
for a slightly longer period of time than it did scent from a late lactating female. A single
subadult male overmarked stimuli during this experiment, overmarking samples from

pregnant female donors twice.

DISCUSSION

This study was the ﬁrst systematic investigation Of the discrimination of
conspeciﬁc scent by free-living spotted hyenas. It was also the ﬁrst investigation of the
scent discrimination abilities of hyenas that included cub, subadult and adult female
respondents. Drea et al. (2002) showed that captive adult hyenas discriminated between
paste samples from unfamiliar and familiar individuals, as well as between samples from
male and female donors. Here we have extended these ﬁndings by Showing that free-
living hyenas discriminated between paste samples from female clanmates and non-

clanmates, males and females, and pregnant and late lactating female donors.

Clanmate vs. non-clanmate

The efﬁcient and accurate recognition of group members is a critical component
of territorial behavior in gregarious species (Brennan & Kendrick 2006). Several social
mammals have been shown to recognize group members on the basis of scent alone (e. g.
O'Riain & Jarvis 1997; Muller & Manser 2007; Palphramand & White 2007). Although
captive Spotted hyenas have been shown to discriminate between paste samples from
familiar and unfamiliar donors (Drea et al. 2002), and hyenas have been observed

reacting aversively to alien scent marks encountered in their territories (Kruuk 1972;

114

Mills 1990), prior to the current study, no one had yet systematically determined whether
free-living hyenas discriminate between clanmates and non-clanmates on the basis of
scent alone. Here we have demonstrated that juvenile hyenas clearly discriminated
between paste samples from female clanmates and non—clanmates. Interestingly, juveniles
did not discriminate between paste samples from male clanmates and non-clanmates.
This may be a reﬂection of the labile nature of adult male membership in hyena clans, as
compared to that of philopatric females (Frank 1986a).

Although juvenile hyenas differentially investigated paste samples based on clan
membership, adult females did not. Considering that adult female spotted hyenas assume
primary roles in territorial advertisement and defense (Boydston et al. 2001), we had
expected adult females to spend considerably more time snifﬁng the scents of non-
clanmates (Kruuk 1972; Mills 1990), especially those from female donors. Scent marks
from alien female donors represent potential threats to territorial security, whereas scent
marks from alien males likely do not. As is typical among mammals, spotted hyenas
exhibit male-biased dispersal (Smale et al. 1997) and, therefore, the appearance of
transient males within a clan's territory is a relatively common occurrence.

A challenge associated with scent discrimination studies is that, although positive
experimental results indicate discriminative ability, negative results do not necessarily
indicate inability to discriminate (Brown 1979). Instead, negative results may be a
product of insufﬁciently motivated participants, underlying contextual factors, or the
scent samples from different donor groups being equally informative and biologically
relevant (Brown 1979; White et al. 2004). In a previous study (Theis et al. submitted), we

suggested that, for adult female hyenas, pasting functions to facilitate social cohesion

115

within groups. It is therefore possible that, for adult female hyenas, monitoring the
presence and condition of same-sex competitors, whether they are clanmates or non-
clanmates, is equally important. Certainly, more data on the responses of adult female

hyenas to scents from clanmate and non-clanmate female donors are needed.

Neighboring clan vs. distant clan

If social animals discriminate between odors from members of their own social
group and others, as juvenile hyenas in this study have shown they do, their
discriminative ability may be based on the degree of familiarity they have with donors or
instead on their recognition of chemical badges of group membership. Among free-living
animals, individuals are more familiar with their neighbors than they are with strangers
that reside farther away (see Muller & Manser 2007). Therefore, if discrimination
between the scents of group members and others is based on degrees of familiarity, then
free-living animals should discriminate between odors from neighbors and strangers.
Indeed, it has been shown that members of multiple mammalian species make this
discrimination (e. g. Sliwa & Richardson 1998; Rosell & Bjorkoyli 2002; Muller &
Manser 2007; Palphramand & White 2007). In general, mammals that make this
discrimination spend more time investigating scent marks from strangers than neighbors
(e.g. Sliwa & Richardson 1998; Rosell & Bjorkoyli 2002; Palphramand & White 2007),
because unfamiliar individuals typically pose a greater threat to territorial maintenance
than do established neighbors ('dear enemy' effect: Fisher 1954; Temeles 1994).
However, among territorial social mammals, this is not necessarily the case (Muller &

Manser 2007). For highly gregarious mammals, the scent mark of a single transient

116

stranger within their territory likely represents less of a threat to a social group than does
the scent mark of a member of a neighboring social group placed beyond an established
border ('nasty neighbor' effect: Muller & Manser 2007). To date, very few studies have
considered whether highly social mammals discriminate between the scent marks of
neighbors and members of distant social groups and, if so, whether they exhibit greater
interest in scents from their neighbors (e. g. Muller & Manser 2007).

Drea et al. (2002) demonstrated earlier that captive adult Spotted hyenas
discriminated between pastes samples from familiar and unfamiliar donors. In the current
study, there was insufﬁcient participation by adult females for us to evaluate whether
they discriminated between paste samples from members of neighboring and distant
clans. However, four of the ﬁve adult female participants sniffed paste samples from
members of distant clans more than they did samples from members of neighboring clans
or the control wires. These data, while preliminary, suggest that adult hyenas discriminate
between the scents of neighbors and strangers, and that a dear enemy effect may be
operating even in this highly gregarious species.

Although few adult females participated in this experiment, several cubs and
subadults did participate. Despite suggestive trends in the data, neither cubs nor subadults
clearly discriminated between paste samples from neighboring and distant clans. It was
unsurprising that cubs did not differentially respond to scents from neighboring and
distant clans, as they should be equally unfamiliar with both. Hyena cubs are primarily
conﬁned to their clan's communal den (Holekamp & Smale 1998) and, Since alien hyenas
are hardly ever observed at communal dens (personal observation), cubs are unlikely to

be familiar with the scents of any hyena that is not a clanmate. The similarity in the

117

responses exhibited by cubs and subadults was moderately surprising, however, as
subadult hyenas do travel, feed and rest beyond the conﬁnes of the communal den within
their clan's territory (Holekamp & Smale 1998). Older subadults have even been
observed participating in border patrols (Theis et al. 2008). During a border patrol
members of a hyena clan cooperatively investigate and demarcate one of their territorial
boundaries with scent marks (Kruuk 1972; Henschel & Skinner 1991; Boydston et al.
2001). Participating in border patrols would likely increase subadult familiarity with the
scent marks of their neighbors. Additionally, male hyenas begin engaging in predispersal
forays from their natal clans at approximately 18 -24 months of age (Smale et al. 1997).
During their forays into neighboring territories, male hyenas investigate the odors they
encounter, presumably including the scent marks of the hyenas that reside there (Smale et
al. 1997). Consequently, predispersal exploratory forays would increase a male subadult's
familiarity with the odors of its neighbors, and potentially with the odors of members of
distant clans as well. Therefore, responses to conspeciﬁc scents by subadult hyenas may
be substantially inﬂuenced by experience, and the lack of discriminative ability exhibited
by younger (12 — 18 months of age; 5/10 subadult respondents) and older (18 - 30

months of age) subadult hyenas in this experiment should be interpreted with caution.

Distant clan vs. distant clan

Social animals that discriminate between odors from members of their own social
group and others, may do SO based on their degree of familiarity with donors, or instead
based on chemical badges of group membership. Chemical badges of group membership

are available when the odors associated with individuals in the same social group are

118

more similar than the odors associated with members of different social groups. Group-
speciﬁc odors have been identiﬁed in social insect species (e. g. vander Meer et al. 1989;
Ruther et al. 2002), as well as in several gregarious mammals (e.g. O'Riain & Jarvis
1997; Bloss et al. 2002; Buesching et al. 2003; Burgener et al. 2008). Group-speciﬁc
odors are believed to promote cohesion within social groups by facilitating accurate
recognition among group members (Ruther et al. 2002; Buesching et al. 2003). For
instance, among gregarious naked mole-rats (Heterocephalus glaber), being recognized
as a group member substantially reduces the amount of aggression individuals receive
from their social partners (O'Riain & Jarvis 1997).

Group-speciﬁc odors seemingly exist in the spotted hyena as well: the proﬁles of
volatile carboxylic acids emanating from paste samples are more similar within than
between hyena clans (Hofer et al. 2001; Burgener et al. 2008). Prior to the current study,
however, no one had attempted to determine whether hyenas discriminated between paste
samples obtained from members of two different clans. Our analyses failed to
demonstrate that they do, but two confounding issues should be discussed. First, there
was only modest participation in this experiment, with half (4/8) of the respondents being
cubs. It is possible that cubs lack the experience or the motivation to discriminate
between paste samples from members of two clans with which they are entirely
unfamiliar. We mention this because each of the subadult (2) and adult female (2)
participants sniffed paste samples from West Talek donors more than they did samples
from East Talek donors. A second potentially confounding issue is that, when this
experiment was conducted, we thought that the original Talek clan had ﬁssioned into the

East and West Talek clans by 2001. Therefore, we used paste samples collected in 2001

119

or later. However, recent analyses of the association patterns of hyenas in these clans
suggest that the two Talek clans did not completely split until 2002 (J .E. Smith, Michigan
State University, unpublished data). Since almost half (5/12) of the samples we used in
this experiment were collected in 2001, the samples may not truly reﬂect group-Speciﬁc

characteristics.

Male vs. female

Chemical communication plays a critical role in the reproductive biology of most
mammals (Eisenberg & Kleiman 1972; Johnson 1973; Johansson & Jones 2007). Even
among mammals that appear highly sexually monomorphic, the odors produced by the
two sexes are usually substantially dimorphic (Blaustein 1981). Given the importance of
opposite-sex odors in mammalian reproductive biology, it is surprising that few studies
have considered when during development mammals begin to discriminate between male
and female odors (White et al. 2004). Here we have demonstrated that adult female, but
not cub or subadult, spotted hyenas discriminate between paste samples from male and
female donors. As noted above, in discrimination experiments, negative results do not
necessarily mean that participants cannot discriminate between scent sources, but rather
only that they do not do so (Brown 1979). The lack of discrimination between male and
female odors exhibited by juvenile hyenas may reﬂect the absence of a functional role for
such a discrimination in the social lives of young hyenas (White et al. 2004). The primary
challenges for hyena cubs are being recognized as members of their social groups,
gathering information about their clanmates, and becoming assimilated into their clan's

dominance hierarchy (Holekamp & Smale 1998). Scent marks from unfamiliar

120

individuals may provide cubs with important information about their social
environments, regardless of whether the marks were deposited by male or female donors
(Holekamp & Smale 1993; Smale et al. 1993). For subadult hyenas, the primary
challenge is independently obtaining food following weaning (Holekamp et al. 1997;
Holekamp & Smale 1998). For subadult hyenas then, there is also no apparent functional
role for differentially investigating male and female odors.

Among adult hyenas, however, a primary challenge is obviously maximizing
one’s reproductive potential (Holekamp et a1. 1996; Holekamp & Smale 1998; Engh et al.
2002). In general, if adults investigate the chemical signals of Opposite-sex individuals
more than those of same-sex individuals, the signals under investigation are believed to
function more in reproduction than in intrasexual competition or COOperation (White et al.
2004). For example, upon attaining reproductive maturity, female mice (Mus musculus)
switch from avoiding lures scented by adult males to exhibiting a marked preference for
them, presumably thereby facilitating reproduction (Drickamer & Brown 1998). In their
study of captive spotted hyenas, Drea et al. (2002) found that adult females investigated
paste samples from male donors more than those from female donors. Accordingly, the
authors suggested that pasting likely has a reproductive function for spotted hyenas. In
contrast, in the current study, we found that free-living adult female hyenas spent far
more time investigating paste samples from female than male donors. Rather than
supporting a reproductive function for pasting by Spotted hyenas, our analysis suggests
that pasting functions in intrasexual competition among female hyenas. A potential
explanation for the difference between the results obtained in these two studies is that

Drea et al. (2002) used captive animals while we utilized free-living subjects. The

121

importance and potential inﬂuence of context on the outcome of discrimination studies
has been widely recognized (e. g. Eisenberg & Kleiman 1972; Brown 1979; Leger 1993),
and it has been recommended that, if the logistic challenges associated with ﬁeld
experiments can be overcome, investigators should conduct their scent discrimination
experiments in the ﬁeld with free-living participants (Muller-Schwarze 2001).

Our ﬁnding that free-living adult female hyenas spent more time investigating
odors from alien females than from alien males is consistent with results obtained in a
study of female ringtailed lemurs (Lemur catta; Scordato & Drea 2007). Female
ringtailed lemurs consistently investigated odors from extra-group females more than
odors from extra-group males, suggesting that scent marking by female lemurs functions
in intrasexual competition and resource defense (Scordato & Drea 2007). The degree of
interest female lemurs showed toward the odors of other females may be due to their
living in female-dominated societies (Scordato & Drea 2007). The results of the current
study lend support to this hypothesis, as spotted hyenas also live in female-dominated
societies (Kruuk 1972; Frank 1986b; Mills 1990). However, the results of both studies
may be more broadly indicative of patterns that exist among highly social female
mammals in general, at least among those that participate in resource defense. In the
current study, female hyenas were responding to the presence, inside their clan's territory,
of scent marks from alien males and females. As noted above, hyenas exhibit male-biased
dispersal (Frank 1986a; Smale et al. 1997) and, therefore, the appearance of a transient
male inside a clan's territory is a common occurrence. However, since female hyenas are
philopatric (Frank 1986a), and highly territorial (Kruuk 1972; Mills 1990; Boydston et al.

2001), the appearance of an alien female in a clan's territory is much less common (but

122

see Hofer & East 1993b). Additionally, the presence of alien females or their scent marks
are potentially of great consequence to adult female hyenas because they may indicate
intrusion pressure. Among social species in which females participate in resource
defense, females likely beneﬁt from actively monitoring the presence and condition of
same-sex, extra-group competitors, as this information potentially serves to modulate

interactions between social groups (Wittemyer & Getz 2007; Meyer et al. 2008).

Pregnant vs. late lactating

In many mammals, including social species (e.g. Scordato & Drea 2007; Meyer et
al. 2008), females advertise their reproductive condition and sexual receptivity via scent
marking (Johnson 1973; Johansson & Jones 2007). In doing so, they can facilitate mating
and conception, and potentially reduce the amount of harassment they receive from male
conspeciﬁcs during periods when they are not receptive (Johansson & Jones 2007). Here
we simultaneously presented hyenas with paste samples from pregnant females and
females in late lactation, to determine whether pasting by female hyenas advertises their
sexual receptivity. As sexual receptivity among female hyenas typically occurs during
late lactation (Szykman et al. 2003), if females advertise their receptivity via pasting,
adult males would be expected to exhibit greater interest in scents from late lactating than
pregnant female donors. Unfortunately, only three adult males participated in this
experiment and their responses to the stimuli varied. Therefore, we could not evaluate
this prediction.

Although we could not characterize the responses of males collectively, each of

the other participants in this experiment investigated paste samples from pregnant

123

females more than those from late lactating donors. Therefore, paste from female donors
clearly contains information about their reproductive condition, and natal animals
exhibited a particular interest in the scents of pregnant females. Interestingly, in a
previous study (Theis et al. submitted), we found that adult female hyenas pasted
substantially more often during pregnancy than they did at other times. Although we do
not yet clearly understand how female hyenas beneﬁt from communicating information
about their pregnancies, studies in other mammals have provided clues about the types of
information they may be broadcasting. The urine of pregnant female mammals contains
information about the fetuses females carry, speciﬁcally their paternal haplotypes at the
major histocompatibility locus (Beauchamp et al. 1994; Beauchamp et al. 1995). In other
words, odorants in a pregnant female's urine provide receivers with the identity of her
offspring's sire. These ﬁndings are pertinent to the current study because this information
is not necessarily available only in urine, rather it may additionally be available in other
female secretions, including saliva, milk and sweat (Beauchamp et al 1995; Beauchamp
et al. 2005). It is possible that this information is available in the paste secretions of
pregnant female hyenas as well.

For female mice, communicating information about an offspring's paternal
haplotype may reduce the likelihood of that offspring becoming a victim of infanticide
(Beauchamp et al. 2005). Additionally, in being pre-exposed to the fetal odortypes of
their offspring, females may be more committed to providing maternal care after giving
birth (Beauchamp et al. 2005). We ﬁnd it intriguing to expand consideration of this idea
from the mother alone to the entire social group. A primary challenge for newborn social

carnivores is being effectively recognized as members of their social groups, thereby

124

reducing the amount of aggression they receive from their social partners (Rasa 1973;
Fell et al. 2006). If paste from pregnant female hyenas contains fetal odortypes, then
increased rates of pasting by females during pregnancy might serve to familiarize
clanmates with the odors of their offspring prior to their being born. This could

potentially facilitate the recognition and assimilation of their offspring into the clan.

Future directions

As appropriate samples become available, some of our experiments should be
replicated to bolster sample sizes and increase the resolution of response patterns among
different age / sex classes. The protocols used in the current study were effective and are
simple enough that the experiments can be readily replicated. However, a minor
adjustment to the protocols should be made to facilitate participation by adult hyenas,
especially immigrant males, in the experiments. The location of experiments should be
moved from the immediate vicinity of communal dens to alongside well-traveled trails
that hyenas use when arriving at these dens. Although this adjustment will be labor
intensive, it has two substantial beneﬁts. First, by moving the experimental stimuli away
from the immediate vicinity of dens, juvenile hyenas are less likely to engage the stimuli
before adults have the opportunity to do so. This will remedy the potentially confounding
issue of juveniles overmarking the stimuli. Second, since immigrant males in attendance
at communal dens remain at the peripheries of den areas (East & Hofer 1991, 1995;
White 2007), setting up experiments beside primary paths leading to dens will increase
the likelihood of adult males encountering and interacting with the stimuli. If sample

sizes of adult respondents can be bolstered in all experiments, then we will be better able

125

to consider how the reproductive condition (White et al. 2004; Rasmussen et al. 2005;
Scordato & Drea 2007; Palphramand & White 2007) and social status (Rostain et al.
2004) of experimental participants inﬂuences their responses to conspeciﬁc scents.
Although the current study has further developed our understanding of the
information content of paste, and consequently improved our understanding of the
functions of pasting behavior among spotted hyenas, there is likely additional information
available in paste as well. Speciﬁcally, paste may provide receivers with information
about the donor's age (White et al. 2003), health (Zala et al. 2004), diet (F erkin et al.
1997), or social status (Hayes et al. 2001). In the future, it would be beneﬁcial to conduct
additional scent discrimination experiments with free-living hyenas to discern whether

this information is available in paste as well.

126

CHAPTER FOUR
A BACTERIAL MECHANISM FOR GROUP-SPECIFIC ODORS IN THE SPOTTED
HYENA (CROCUTA CROCUTA)
INTRODUCTION

Animal bodies contain substantially more bacterial than animal cells (Savage
1977). Many of these symbiotic bacteria are not pathogenic to their hosts; rather they are
mutualists, some being absolutely necessary for the proper development and functioning
of animal bodies (Dethlefsen et al. 2007; Cogen et al. 2008). These realizations have
catalyzed numerous investigations of the effects of symbiotic bacteria on animal
development, physiology, and health (6. g. Rawls et al. 2004; Dethlefsen et al. 2006;
Haine 2008), but few studies have yet addressed the potential roles that bacteria might
play in animal behavior.

Animal behavior is a dynamic comprehensive ﬁeld of study that synthesizes
material from other ﬁelds of biological study, including evolution, genetics, development,
physiology, neurobiology, endocrinology and psychology (Dugatkin 2004). However,
one ﬁeld remains glaringly outside the current scope of animal behavior: microbial
ecology. Within two widely used contemporary textbooks of animal behavior, there
exists only a single discussion of bacteria in each (Dugatkin 2004; Alcock 2005). In both
cases, the discussion is of behavioral adaptations animals have evolved to avoid being
infected by pathogenic bacteria. A lack of consideration of bacteria in animal behavior

research is evident in the contemporary primary literature as well. A recent Web of
Science keyword search (bacteria*) of six peer-reviewed animal behavior research

journals yielded only ﬁfteen publications addressing bacteria Since the year 2000

127

(http://apps.isiknowledge.com; accessed 08/ 15/08; Animal Behaviour, Behavioral
Ecology, Behavioral Ecology and Sociobiology, Behaviour, Ethology, Journal of
Ethology). Only three of these ﬁfteen publications discussed potential beneﬁcial
associations between bacterial symbionts and their animal hosts (Buesching et al. 2003;
Lam et al. 2007; Lombardo 2008).

One aspect of animal behavior in which bacteria likely ﬁgure prominently is
communication (Albone 1984; Muller-Schwarze 2006). For most animals,
communication is achieved largely via chemical means (Wyatt 2003). The components of
chemical signals may be obtained directly from signalers' diets, synthesized by signalers
themselves, or generated as by-products of symbiotic bacterial metabolism (Gorman &
Trowbridge 1989). Many mammals communicate chemically by scent marking with
secretions from specialized scent glands (Albone 1984; Wyatt 2003; Muller-Schwarze
2006). Mammalian scent glands are warm, moist, anaerobic, organic-rich, and thus highly
conducive to the proliferation of symbiotic bacteria (Albone 1984; Gorman &
Trowbridge 1989). When symbiotic bacteria ferment the organic-rich sebum substrates
within these glands, they generate volatile fatty acid by-products (Albone et al. 1974;
Gorman et al. 1974), and these fatty acids seem to be prominent and active components
of mammalian scent marks (e. g. Gorman 1976). Consequently, it has been postulated that
symbiotic bacteria are critically involved in the production of social odors among
mammals that engage in scent marking (Albone et al. 1974; Gorman 1976). All else
being equal (age, sex, reproductive state, etc.), if conspeciﬁcs harbor unique communities
of bacteria within their scent glands, then they should emit individually distinct signature

odors via scent marking (Gorman 197 6). Alternatively, if they harbor similar microbial

128

communities in their scent glands, then they should emit similar odors (Albone et al.
1974; Macdonald 1985). Members of mammalian social groups may come to share
similar microbial communities in their scent glands, and thus emit group-speciﬁc odors,
as a result of occupying the same physical space, coming into frequent bodily contact
with one another, or repeatedly scent marking the same sites (Albone et al. 1974;
Macdonald 1985; Buesching et al. 2003). At the functional level, group-speciﬁc odors are
believed to facilitate recognition among group members, thereby reducing levels of
aggression among social partners and promoting social cohesion within groups (Albone
et al. 1974; O'riain & Jarvis 1997; Bloss et al. 2002; Saﬁ & Kerth 2003; Fell et al. 2006)
If the hypothesis were correct that symbiotic bacteria are responsible for group-
speciﬁc social odors in mammals, then the structure of bacterial communities in the scent
glands of members of the same social group should be more similar than the structure of
communities found in members of different groups. This prediction was previously tested
using captive red foxes (Vulpes vulpes), and the results did not support the hypothesis
(Albone et al. 1978). However, as the authors of the study themselves noted, their
culture-based approach to sampling bacterial communities was likely biased due to the
differential recoverability of anaerobic bacterial forms from culture (Albone et a1. 1978;
Schloss & Handelsman 2004). At present, most bacteria remain unculturable, so culture-
independent community sampling techniques, such as 16S rRNA gene surveys, provide
valuable alternatives to traditional culture-based approaches (Riesenfeld et al. 2004;
Dethlefsen et al. 2007). The 16S rRNA gene is critical in protein manufacture, and is
therefore a universal and highly conserved gene that allows researchers to make

taxonomic identiﬁcations of organisms based on their nucleotide sequences at this locus

129

(Dethlefsen et al. 2007). The objective of the current study was to re-evaluate the
hypothesis that bacteria are responsible for the production of group-speciﬁc social odors
in mammals, using free-living spotted hyenas (Crocuta crocuta) as subjects, and
employing culture-independent 16S rRNA gene survey sampling techniques.

Spotted hyenas are large carnivores found throughout sub-Saharan Africa. They
live in complex social groups, called clans, that typically consist of 40 - 80 individuals
(Kruuk 1972; Trinkel et al. 2007). Members of each clan cooperatively defend their
group’s territory from neighboring hyenas (Kruuk 1972; Mills 1990; Henschel & Skinner
1991; Boydston et al. 2001). Hyena clans are ﬁssion-fusion societies, in which members
are seldom all present in the same place at the same time; rather members ﬁssion into
subgroups that change in composition and size several times each day (Holekamp et al.
1997; Smith et al. 2008). Hyena clans contain multiple breeding males and multiple
overlapping generations of females. Genetic relatedness is not signiﬁcantly greater within
than among hyena clans, because of high levels of male—mediated gene ﬂow among clans
(Van Horn et al. 2004). In contrast to most mammalian social systems, hyena societies
are female-dominated, and social status within a clan’s linear dominance hierarchy is not
determined by size or ﬁghting ability (Frank 1986a,b). Instead, natal hyenas ‘inherit’ the
social ranks of their mothers (Frank 1986b; Holekamp & Smale 1991, 1993; Smale et al.
1993; Engh et al. 2000), and adult immigrant males, all of which are subordinate to natal
individuals, queue for dominance status with other immigrant males (Smale et al. 1997;
East & Hofer 2001).

To mediate the complex social relationships between and within clans, spotted

hyenas utilize a rich array of vocal, visual and chemical signaling behaviors (Kruuk 1972;

130

Mills 1990). A common and conspicuous form of chemical signaling among spotted
hyenas is ‘pasting.’ When pasting, hyenas extrude their anal scent pouch and drag it over
grass stalks. In doing so, hyenas typically deposit secretions near the tops of the grass
stalks (Matthews 1939; Kruuk 1972; Mills 1990). These secretions, referred to as 'paste,’
are produced in bi-lobed sebaceous glands that are separate from the gastrointestinal tract
and empty their products into the anal pouch (Matthews 193 9; Mills 1990). The paste of
spotted hyenas is rich in fatty acids (Buglass et al. 1991; Hofer et a1. 2001), and it has
been shown that the chemical proﬁles of paste are group-speciﬁc (Hofer et a1. 2001;
Burgener et al. 2008). Speciﬁcally, the relative abundances of volatile odorants, primarily
fatty acids, emanating from paste are more similar among clanmates than they are among
members of different clans (Burgener et al. 2008). It has been suggested that the group-
speciﬁc differences in the odor of paste may be due to the different metabolic activities of
different communities of symbiotic bacteria within the scent glands of Spotted hyenas
(Burgener et al. 2008; Theis et al. 2008). The primary objective of the current study was
to inquire whether group-speciﬁc bacteria may be responsible for the production of
group-Speciﬁc social odors in the spotted hyena. A secondary objective was to

preliminarily describe the microbiome of the spotted hyena anal pouch.

METHODS

Study area and sample collection
This study was conducted in the Masai Mara National Reserve, Kenya. From

1999 - 2000, we obtained paste samples directly from the anal pouches of sixteen spotted

hyenas anaesthetized with Telazol (W.A. Butler; 6.5 mg/kg) delivered from a C02-

131

powered darting riﬂe (Telinject). The sampled hyenas represented four distinct clans
within the north-central part of the Reserve: Emarti Hill, Mara River, Fig Tree and
Southern Comfort (Figure 4.1). We collected paste Samples from four adult lactating
females from each of these clans. Paste samples were placed in sterile cryogenic vials
(Corning), stored in liquid nitrogen, and transported to Michigan State University, where

they remained frozen (minimum -20 C) until being used in the current study.

Construction of 168 ribosomal DNA clone libraries from paste samples

We used a Mo Bio ultraclean fecal DNA kit to isolate DNA from each paste
sample (~ 0.1 g of paste per sample). Each extraction was then diluted (1:10), and the
16S rDNA in each was ampliﬁed by polymerase chain reaction (PCR) using two bacterial
primers (8F, 5' - AGA GTT TGA TCA TGG CTC AG - 3'; 1392R, 5' - ACG GGC GGT
GTG TAC - 3'). The PCR program was: 95 C for 3 min, followed by 30 cycles consisting
of 95 C for 30 sec, 55 C for 30 sec, 72 C for 45 sec, and a ﬁnal extension period of 72 C
for 10 min. Amplicons were then ligated into plasmid vectors (pCR 2.1-TOPO;
Invitrogen) that were used to transform Escherichia coli (TOP10) and generate a 16S
rDNA clone library for each paste sample. We randomly selected 47 clones from each
1 library for DNA sequencing. High throughput sequencing was performed at the Research
Technology Support Facility of Michigan State University, using the SF primer.

Eighty nine percent (670/ 752) of the sequencing reactions yielded quality
sequences (Geospiza Finch; Q>20) of typically 500 - 800 base pairs in length (at least
300 bp). These 670 sequences were aligned using the automated aligner in Arb software,

and then adjusted by hand to comply with secondary structure models of the 16S rRNA

132

MASAI MARA
NATIONAL RESERVE

 

 

’

 

Figure 4.1. Map illustrating the relative locations within the Reserve of the four hyena
clans sampled in this study. The dashed lines indicate the estimated territorial boundaries
of hyena clans in this area of the Reserve from 1999 - 2000. This map was adapted with

permission from Van Horn et al. (2004) and Kolowski et al. (2007).

133

gene (ssu_jan04 database release; Ludwig et al. 2004). The aligned sequences
(speciﬁcally the region between base positions 60 and 651 of the 16S rRNA gene; E. coli
numbering) were incorporated into a distance matrix, which was used to assess sequence
similarities and assign sequences to operational taxonomic units (OTU) via the program
DOTUR (average neighbor joining method; Schloss & Handelsman 2005). In the current
study, OTUS were deﬁned based upon a 97% sequence identity cutoff, and therefore

represent approximate species-level associations (Stackebrandt & Goebel 1994).

Describing the anal pouch microbiome

From the DOTUR output, we generated rarefaction and rank abundance curves
for all sequences considered collectively. Additionally, we recorded a Chaol richness
estimate and a Shannon diversity index (which considers evenness in addition to richness;
Magurran 2004) for the symbiotic bacterial community associated with the anal pouch.
Our purpose in constructing rarefaction and abundance curves and in calculating diversity

measures was to evaluate our sample coverage.

Species-level taxonomic identities of OTUS in the anal pouch microbiome

To determine the species-level identities of bacteria comprising the anal pouch
microbiome, we randomly selected a Single representative sequence from each OTU, and
submitted the representative sequences to the Silva incremental aligner (SINA;
http://www.arb-silva.de/aligner; Pruesse et al. 2007). SINA aligned the representative
sequences with their ten nearest neighbors in the Silva reference library. We then

imported the representative sequences and their ten nearest neighbors into Arb and

134

generated another distance matrix. We imported this matrix into DOTUR, and
determined whether the sequences representing the OTUS Shared 97% sequence
similarity with any of their nearest neighbors in the Silva reference library (Silva Release
95). If they did, and if the neighboring sequences were obtained from cultured or
Candidatus bacterial Specimens, then we assigned corresponding species-level taxonomic
identities to the representative sequences and their OTUS. One sequence shared 97%
sequence similarity with ten uncultured environmental samples. AS it also might have
shared 97% sequence similarity with a cultured bacterium that was not one of its ten
closest neighbors in the reference library, the sequence was resubmitted to SINA and the
process was duplicated using 40 nearest neighbors instead often. The sequence in
question shared 97% sequence similarity with all 40 of its nearest neighbors in the
reference library, and again the neighboring sequences were all obtained from uncultured
bacteria. We then used the sequence matching function (Seqmatch) available from the
Ribosomal Database Project (Release 10, update 4; Cole et al. 2007) to broadly identify
the sequence as belonging to the genus Cetobacterium. We used Arb to generate a
distance matrix from the sequence in question and several sequences obtained from
Cetobacterium and F usobacterium (sister genus to Cetobacterium) type specimens. The

percent similarities among these sequences were determined via DOTUR.

Approximate genus-level taxonomic identities of OT Us in the anal pouch microbiome
We selected a random representative sequence from each OTU that made up more
than 1% of all sequences analyzed in the current study. These sequences were aligned in

SINA using their ten nearest neighbors in the Silva reference library. The representative

135

sequences and their ten nearest neighbors were imported into Arb, and a neighbor-joining
phylogenetic tree was constructed illustrating the relationships among the sequences.
From this tree, we inferred genus-level taxonomic identities for the abundant members of

the anal pouch microbiome.

Alpha diversity: diversity within communities

To assess our sample coverage, we generated separate distance matrices in Arb
from sequences comprising each of the clone libraries. We also used DOTUR to
construct rarefaction and abundance curves, and to calculate richness and diversity

measures, for each of the clone libraries.

Beta diversity: diversity between communities

From the DOTUR output assessing degrees of Similarity among all sequences
considered collectively, we tabulated the number of sequences obtained per OTU per
clone library. We entered these sampling data into the program EstimateS (Colwell
2006), and calculated Bray-Curtis similarity indices among all 16 clone libraries. The
Bray-Curtis similarity index is widely used in ecological studies to assess the biotic
distinctness of sampling communities (Magurran 2004; Quinn & Keough 2002).
Moreover, the Bray-Curtis index was the most appropriate beta diversity measure for the
current study because it was the distance measure employed to demonstrate that hyena
paste has group-speciﬁc odorant proﬁles (Burgener et al. 2008).

We used the program Mega4 (Tamura et al. 2007) to perform a hierarchical

cluster analysis of the 16 clone libraries based upon Bray-Curtis indices. Cluster analyses

136

produce dendrograms that illustrate the degrees of difference among sampled
communities. We constructed our dendrograrn using the Unweighted Pair Group Method
with Arithmetic Mean (U PGMA; e. g. Pardo & Armitage 1997), based upon Bray-Curtis
dissimilarity indices (1 - Bray-Curtis). We used dissimilarity, as opposed to similarity,
measures so that the branch lengths on the dendrogram would correspond to degrees of
difference in community structure among the clone libraries. We additionally visualized
similarities in community structure among clone libraries via non-metric
multidimensional scaling (MDS) plots, based again on Bray-Curtis dissimilarity indices
(Statistica, v6.1, 2002). In MDS plots, sampled communities that are dissimilar are
placed far apart whereas those that are similar are placed close together (Quinn &
Keough 2002; Gotelli & Ellison 2004).

In addition to visually exploring our data, we also statistically evaluated
similarities in community structure among clone libraries. Analysis of similarity
(ANOSIM) is a conservative non-parametric analysis that evaluates whether signiﬁcant
differences exist between groups based on a distance measure (Clarke 1993; Quinn &
Keough 2002; Gotelli & Ellison 2004). We conducted an ANOSIM testing the prediction
that the structure of bacterial communities in the anal pouches of spotted hyenas would
be more similar within than between clans. We entered sampling data into the program
PAST (version 1.82b; Hammer et al. 2001), and logo (x + l) transformed them. The
ANOSIM included 10,000 permutations, and was based on Bray-Curtis indices with clan

membership as the independent variable.

137

RESULTS

Anal pouch microbiome

At a 97% sequence identity cutoff, we found 75 OTUS in the microbiome of the
Spotted hyena anal pouch (Chaol = 136.9 (103.7 - 208.3, 95% CI); Shannon = 2.50 (2.38
- 2.63)). Only eleven of the 75 OTUS shared 97% sequence identities with previously
cultured or well-categorized bacterial species (Table 4.1). Collectively, these eleven
OTUS represented only two percent (14 / 670) of the sequences analyzed in the current
study. Although our sample coverage did not appear complete (Figure 4.2), it was
representative, as the anal pouches of the 16 sampled spotted hyenas appeared to be
dominated by fewer than ten bacterial species (Figure 4.3). Nine OTUS each represented
more than one percent of all sequences analyzed. Collectively, these nine OTUS
represented 85 percent (569 / 670) of the sequences analyzed in this study, and appear to
be members of the genera Peptostreptococcus, C orynebacterium, Propionibacterium, and

Anaerococcus (Figure 4.4).

Alpha diversity

Within clone libraries, sample coverage was not complete (Figure 4.5). As alpha
diversity measures are affected by sampling effort, the Chaol and Shannon diversity
measures for these anal pouch communities should be interpreted cautiously (Table 4.2;
Magurran 2004). However, no differences in phylotype diversity among libraries from
different clans were apparent and, although sampling was not complete, it did appear to
be representative. A majority of libraries contained fewer than ten OTUS, and each

library was dominated by just a few phylotypes (Figure 4.6). Therefore, at a 97%

138

Table 4.1. OTUS in the current study that shared at least 97% sequence similarities with

previously cultured or well-categorized (i.e. Candidatus) bacterial species.

Representative
sequence for OTU

MaraRiver2_H04

MaraRiver3_C09

F i gTree4_E02

FigTree3_C10

EmartiHill4_F06

F i gTree4__E06

SouthemComfort2__G04

MaraRiver1_C08

F igTree3_BO3
SouthemComfort3_A08

MaraRiver2_G01

Total number of

sequences in OTU

2

2

Ix.)

139

Species sharing 97%
sequence identity with OTU

F inegoldia magna

Corynebacterium amycolatum
C. hansenii
C. freneyi

Pantoea agglomerans
P. ananatis

C orynebacterium falsenii
C. auriscanis

C. resistens

C. urealyticum

Cryptosporangium japonicum
C. arvum

C. aurantiacum

C. minutisporangium
Pseudomonas putida

P. rhizosphaerae

P. gingeri

Peptoniphilus olsenii

C andidatus Peptoniphil us
massiliensis

Ignavigranum ruoﬂiae
Se getibacter koreensis

C etobacterium somerae

 

 

 

140 -
y=x

E 120-
I-
d)
>
8 100-
.21
'6
5 80 -
[-1
O
‘3 60 -
I-
3
E 40 .
=
E

20 ~

0 ’ l l L i i l .

0 100 200 300 400 500 600 700

number of sequences sampled

Figure 4.2. Rarefaction curve illustrating the degree of sample coverage in this study for
all sampled hyenas considered collectively. The dotted curves represent the 95%
conﬁdence interval for the plot. To provide perspective when gauging the degree to

which the rarefaction curve is leveling off, the function y = x is also illustrated.

140

220 -l
200 —
180 i
160 -
140 —
120 -
100 l

80-

number of sequences per OTU

60-

40-

201

 

 

 

 

 

0 - l-i-llmnﬂllrrmrrnmnmmnrrmmirrmmnrnrmrrmnnnmrrmmm
1 7 13 19 25 31 37 43 49 55 61 67 73

OTU abundance rank

 

Figure 4.3. Rank abundance curve showing the relative abundance of the 75 OTUS found

in this study.

141

Figure 4.4. Phylogenetic analysis of representatives of the nine OTUS (in bold face) that
each represented more than 1% of all sequences analyzed in the current study. These nine
representatives were incorporated into the phylogenetic tree with their ten nearest
neighbors from the Silva reference library, as determined by SINA. With each
representative sequence, the total number of sequences constituting that OTU is indicated
in brackets. The numbers within branches indicate the number of nearest neighbor
sequences that have been grouped together (i.e. condensed) on that branch. For nearest
neighbor sequences obtained from uncultured bacteria, we have included information

from the Silva database about the environmental source from which they were obtained.

142

 

2.:
23m 5:55 esteem; 3530:: me
$82: _E8V ESSSEV Bea—35. _
E a 5:5 388%? a 8::
$9233 =o_mA_=m_8m sash—2.303 Bahama: 5L
3.6.8 388KB: w J ,

case 88:: Ensues 33:35 h ”1

Ex” 58:; an “.26 88:: 5:203 85.35?

€28 SEE: 83.803 35:55.
3%.?» gouache: 1 _ .

Oman—covestswﬁ c8332.: H will
.2_ ”812:555

 

@820 .3235 83583 wee—8:: a
A883 .3235 Sataugeabeo @2332: 4
Ezemeiﬁue 5353339334 Numb]

553348285 EECBEEQEEK $11.11—

        

EESBEM Eataouﬁgﬁeumbll.

.3. 80.3%?

osmos— 58: :32» e u: o a .5 ................. 11

A ﬁaewouabmw Eatwwowﬁrawu

one 885 5:303 85.85. m

”Ewes SE: E3523 use :25 .
A 5%; BEE 82383 was. as.
SENSE: ESSESEASU

£538? Entauggboo L

.3 So ateEoUEoégm

€25: 288v 8:583 c2338:

7.: 51.8%...

_3._ so are“;

925: 228 a. 82¢ 5:55 5:892 coca—=8:

Amuoa a 888 38551389 3530::

252: 5:55 8:568 BEBE. 1..

930$ 302:9 5:538 ESSSQEBEQ
2:88 5:5 augeﬁgaiemq 1.
18.835.35.83. 1. .
1 man m3. may—Ease .1111 111111 l 11111 1U|1_
—Q— 8< —tOhEOUELO—==Om ._.—@_ gmln==v=tas i l

:u :0 nuances- 1

 

 

 

 

 

 

 

 

 

 

 

 

 

143

30

Mara River ~ I Southern
25 _ LComfort
2o -
15 -

   

  
 
   

o
10 r 00.9%
0009.33/‘V’r0/TT
.. .'.v, I...
0. .9, ““0,”
92". .0‘000

30".

 

   

A
AAAAAAAA

 

 

 

 

25. p 1

number of OTUS discovered

 

 

 

 

 

number of sequences sampled

Figure 4.5. Rarefaction curves illustrating the degree of sample coverage for each hyena
sampled in each clan. The clone libraries have been separated by clan of origin. The

function y = x (dotted lines) appears on each graph for perspective.

144

Table 4.2. The number of sequences analyzed, and the number of OTUS found, in each
clone library. Richness estimates and diversity indices of the bacterial communities
within the anal pouches of the individual hyenas have been included, along with their

95% conﬁdence intervals.

 

 

 

Hyena # seqs OTUS Chaol richness estimate Shannon diversity index
MR1 40 11 18.5 (12.3 - 53.5) 1.85 (1.53 - 2.17)
MR2 44 12 21.5 (13.0 - 65.2) 1.77 (1.46 - 2.08)
MR3 39 10 12.5 (11.2 - 23.5) 2.00 (1.71 - 2.29)
MR4 39 10 19.0 (10.9 - 61.0) 1.44 (1.06 - 1.82)
SCI 41 15 42.5 (21.7 - 128.5) 2.05 (1.67 - 2.42)
SC2 45 12 14.5 (12.4 — 29.0) 1.96 (1.65 - 2.26)
SC3 45 10 15.0 (10.9 - 38.9) 1.40 (1.02 - 1.77)
SC4 42 12 19.0 (13.3 - 48.5) 1.86 (1.52 - 2.20)
EH1 38 6 7.0 (6.1 - 19.7) 1.43 (1.19 - 1.67)
EH2 43 10 17.5 (11.3 - 52.5) 1.43 (1.05 - 1.82)
EH3 39 9 12.0 (9.4 - 32.0) 1.73 (1.43 - 2.03)
EH4 46 6 7.5 (6.2 - 21.1) 0.77 (0.43 - 1.12)
FTl 41 14 21.0 (15.5 - 47.7) 2.08 (1.73 - 2.42)
FT2 39 9 15.0 (10.0 - 46.7) 1.77 (1.49 - 2.05)
FT3 43 17 95.0 (44.2 - 240.5) 2.12 (1.72 - 2.52)
FT4 46 10 17.0 (11.3 - 46.5) 1.06 (0.64 - 1.48)

145

25r

 

 

 

 

 

 

 

 

 

 

 

 

 

16
20-
D
E-t .4-
O
I—
a
15-
8
U
=
0
=
e7
“510»
g 16
'2
= _l_
=
5 l6
-Ll6
4161614
L _1414127633211
0 .....ﬁ mm1.11.11.11.11.11.11.111.

 

 

 

 

 

 

 

 

 

 

12 3 4 5 6 7 8 91011121314151617
OTU abundance rank

Figure 4.6. Rank abundance curve showing the relative abundance of OTUS within the
anal pouch of each sampled hyena. The sample sizes indicate the number of libraries that

contained at least that number of OTUS. Bars indicate standard errors.

146

sequence identity cutoff, our sampling captured the prominent members of these
communities, so evaluating similarities in bacterial community structure among the anal

pouches of individual hyenas was appropriate.

Beta diversity

An effect of clan membership on the structure of bacterial communities in the
anal pouches of hyenas was evident in all three beta diversity analyses (ANOSIM global
R = 0.35, P = 0.001; Figure 4.7; Figure 4.8). Although the samples obtained from Emarti
Hill hyenas did not differ signiﬁcantly in structure from those obtained from Mara River
(R = -0.01, P = 0.568), Southern Comfort (R = 0.40, P = 0.113), or Fig Tree (R = 0.19, P
= 0.087) hyenas, samples from the Mara River, Southern Comfort and Fig Tree hyenas
clearly clustered with other samples from their respective clans (MR vs. SC: R = 0.78, P
= 0.028; MR vs. FT: R = 0.33, P = 0.030; SC vs. FT: R = 0.40, P = 0.028). The high R-
value between the Mara River and Southern Comfort clans indicated that community
structure differed substantially among samples from these two clans. The R-values
between Fig Tree and both Mara River and Southern Comfort clans were more moderate,
indicating greater degrees of overlap in community structure. These patterns were evident
in both visual analyses as well, as the samples from Fig Tree hyenas clustered less tightly

than did those from Mara River and Southern Comfort hyenas.

147

. Mara River 2

 

 

 

O Emarti Hill 3

 

 

 

O Mal-a River 4

 

 

. Mara River 3

 

 

 

0 5mm Hill 1

 

. Mara River 1
I Southern Comfort 3

 

 

 

 

<> Emarti Hill 2

 

I Southern Comfort 1

 

 

 

 

. Southern Comfort 2

 

 

 

I Southern Comfort 4

 

 

 

 

A Fig Tree 3

 

 

 

 

 

A Fig Tree 1

 

 

 

A Fig Tree 2
O Emarti Hill 4

 

 

A Fig Tree 4

 

0.4 0.3 oiz
l - Bray-Curtis

-ll—
'4—

. q j.-

I
0.0

Figure 4.7. Dendrogram illustrating Similarities in the structure of bacterial communities

inhabiting the anal pouches of 16 spotted hyenas representing four distinct clans: Mara

River (circles), Southern Comfort (squares), Emarti Hill (diamonds) and Fig Tree

(triangles).

148

 

 

Figure 4.8. Multidimensional scaling (MDS) plot comparing similarities in the structure
of bacterial communities inhabiting the anal pouches of 16 Spotted hyenas representing
four distinct clans: Mara River (circles), Southern Comfort (squares), Emarti Hill
(diamonds) and Fig Tree (triangles). The plot was based on Bray-Curtis dissimilarity
indices. In MDS plots, sampled communities that are dissimilar are placed far apart
whereas those that are similar are placed close together (Quinn & Keough 2002; Gotelli

& Ellison 2004). The stress ofthe plot is 0.123.

149

DISCUSSION

The microbiome of the spotted hyena anal pouch

In the current study, in which culture-independent microbial sampling techniques
were employed, the vast majority (98%) of analyzed sequences were obtained from
previously uncultured bacterial forms. Therefore, at the species-level, we were unable to
determine the taxonomic identities of most members of the microbiome of the anal pouch
of spotted hyenas. However, at an approximate genus-level, it was clear that the most
prominent members of the anal pouch microbiome were Peptostreptococcus,
Anaerococcus, Propionibacterium, and C orynebacterium. Previous culture—based surveys
of the bacterial contents of the anal scent glands of carnivores yielded the following
genera: Streptococcus, Proteus, Bacillus, Eubacterium, Clostridium, F usobacterium,
Bacteroides, Peptococus, and Peptostreptococcus (Indian mongoose, Herpestes
auropunctatus: Gorman et al. 1974; red fox, Vulpes vulpes: Albone et al. 1978). The only
genus found in both culture-based studies and in the current study was
Peptostreptococcus. This ﬁnding reinforces the idea that culture-independent sampling
techniques are valuable complements to culture-based approaches when investigating the
structure and function of microbial communities (Riesenfeld et al. 2004).

Peptostreptococcus and Anaerococcus (both phylum F irmicutes) are anaerobic
microbes that are frequently associated with the mouth, skin, and gastrointestinal and
urinary tracts of mammals (Murdoch 1998; Ezaki et al. 2006). Importantly,
peptostreptococci have been shown to produce several of the volatile fatty acid
components of mammalian scent marks, including acetic, propionic, and butyric acids

(Gorman et al. 1974). Each of these fatty acids has been identiﬁed in the paste of spotted

150

hyenas (Buglass et al. 1991; Hofer et al. 2001). Propionibacterium (phylum
Actinobacteria) are predominantly anaerobic bacteria that comprise much of the
mammalian skin microbiota, and are also symbionts within mammalian gastrointestinal
and urinary tracts (Stackebrandt et al. 2006). Like peptostreptococci, propionibacteria
have been shown to produce propionic and acetic acids as fermentation by-products
(Stackebrandt et al. 2006; Cogen et al. 2008). Corynebacterium (phylum Actinobacteria)
are aerobic or facultatively anaerobic microbes that are also often found in association
with the skin surfaces and the gastrointestinal and urinary tracts of mammals (Liebl 2006;
von Graevenitz & Bernard 2006). Corynebacteria have been shown to generate short and
medium chain fatty acids when metabolizing large organic molecules associated with
mammalian skin surfaces (James et al. 2004). In addition to the Short chain fatty acids
discussed above, numerous medium chain fatty acids have been identiﬁed in the paste of
spotted hyenas (Buglass et al. 1991; Hofer et al. 2001). The fermentation hypothesis for
the production of social odors among mammals suggests that volatile fatty acids are
critical components of scent marks and that these fatty acids are by-products of bacterial.
fermentation within specialized scent glands (Albone et al. 1974; Gorman et a1. 1974;
Gorman 1976). Here we have provided evidence suggesting that bacteria capable of
producing these fatty acids are present in the anal pouches of hyenas.

In addition to Peptostreptococcus, Anaerococcus, Propionibacterium, and
Corynebacterium, our phylogenetic analyses also revealed the presence of F inegoldia,
Peptoniphilus, C etobacterium, Ignavigranum, Pantoea, Pseudomonas,
Cryptosporangium, and Segetibacter bacteria in the anal pouch microbiome. F inegoldia,

Peptoniphilus, Cetobacterium, and Ignavigranum are obligate or facultative anaerobic

151

bacteria that have been found previously in close association with mammalian bodies
(Collins et al. 1999; Ruoff 2002; F inegold et al. 2003; Ezaki et al. 2006; Schaal et al.
2006). Pantoea and Pseudomonas, although notable plant pathogens (Pujol & Kado
2000; Kersters et al. 2006), have also been identiﬁed as mammalian symbionts (Rolph et
al. 2001; Lim et al. 2006; Kersters et al. 2006). Cryptosporangium and Segetibacter,
however, have not been identiﬁed as mammalian symbionts; rather, both genera are
aerobic bacteria previously isolated from agricultural soils (Tamura et al. 1998; An et al.
2007). Presumably, these bacteria were picked up from the environment when hyenas
pasted on grass stalks or when they lay on the ground. We do not know whether
Cryptosporangium and Segetibacter have a symbiotic relationship with Spotted hyenas or

whether their occurrence in the survey was purely incidental.

Similarities in the structure of bacterial communities within hyena clans

Our analyses demonstrated similarities in the structure of bacterial communities
within the anal pouches of spotted hyenas from the same clan. This was evident in the
non-random clustering of samples from the same clan in both the dendrogram (Figure
4.7) and the MDS plot (Figure 4.8). It was also evident in the ANOSIM, as the null
hypothesis was rejected that Bray-Curtis Similarity indices were no greater within than
between hyena clans. Substantial differences in the structure of bacterial communities
within the anal pouches of Mara River and Southern Comfort hyenas were apparent.
Samples from these two clans clustered tightly with other samples from their respective
clans in both visual analyses, and the ANOSIM R-value was very high (0.78) between the

Mara River and Southern Comfort clans, indicating that the bacterial communities of the

152

two clans were distinct. Differences in the structure of bacterial communities were also
apparent between Fig Tree and Mara River hyenas, as well as between Fig Tree and
Southern Comfort hyenas, but the degrees of difference were more moderate. This was
because less similarity in bacterial community structure existed among members of the
Fig Tree clan than among members of the Mara River or Southern Comfort clans. The
primary reason for there being less similarity in community structure among members of
the Fig Tree clan was that the anal pouch of one female (Fig Tree 4) was dominated by a
single phylotype (35/46 sequences). The anal pouches of other members of the Fig Tree
clan were more diverse, particularly with respect to evenness, as evidenced by the
Shannon diversity indices within this clan.

Although similarities in bacterial community structure were evident within the
Mara River, Southern Comfort and Fig Tree clans, there was little Similarity in
community structure among the anal pouches of Emarti Hill hyenas. Therefore, although
our results Showed that clan membership signiﬁcantly affected the structure of bacterial
communities within the anal pouches of spotted hyenas, clan membership was evidently

not the only factor inﬂuencing community structure.

Conclusions and future directions

It has been postulated that symbiotic bacteria are responsible for the production of
group-speciﬁc odors among social mammals that engage in scent marking (Albone et al.
1978). If symbiotic bacteria are responsible for the production of group-speciﬁc odors,
then the Structure of bacterial communities within the scent glands of social mammals

should be more Similar among group members than among members of different social

153

groups. Our analyses of the microbial contents of the anal pouches of free-living Spotted
hyenas afﬁrm this prediction, providing support for the hypothesis that differences in
bacteria are responsible for the production of group-speciﬁc odors in this species
(Burgener et al. 2008; Theis et al. 2008). However, before accepting this hypothesis,
additional predictions need to be tested. First, the speciﬁc bacterial strains in the anal
pouch microbiome must generate the volatile fatty acids identiﬁed in paste. Although
testing this prediction will be difﬁcult, it can be accomplished using contemporary
ecologically-guided approaches to cultivating previously uncultured anaerobic bacteria
(Stevenson et al. 2004; Eichorst et al. 2007), coupled with gas chromatography / mass
spectrometry analyses. Second, the bacterial proﬁles of paste samples must correlate with
their chemical proﬁles (e. g. Xu et al. 2007). Third, hyenas must demonstrate that relative
abundances of fatty acids alone provide a sufﬁcient basis upon which to discriminate
among samples. This prediction could be systematically tested via bioassays conducted
with captive spotted hyenas housed at the Berkeley Field Station for the Study of
Behavior, Ecology and Reproduction (University of California, Berkeley), and potentially
with free-living spotted hyenas as well (see Chapter Three).

In the current study we broadly controlled for the age, sex and reproductive state
of paste donors. In future studies, we intend to determine whether age, sex and
reproductive state affect the structure of symbiotic bacterial communities in the scent
glands of spotted hyenas, because these factors inﬂuence the structure of microbial
communities associated with other mammalian hosts (age: Palmer et al. 2007; sex: Alexy
et al. 2003; Voigt et al. 2005; reproductive state: Bartlett et al. 1977; but see Eschenbach

et al. 2000; Coolen et al. 2005). Evaluating the effects of sex and reproductive state on

154

the symbiotic bacterial communities in the anal pouches of spotted hyenas would be
particularly interesting given that information about both sex and female reproductive
state are available to conspeciﬁcs in paste secretions (Chapter Three). Lastly, in the
future, we intend to determine whether genotype (e.g. Newton et al. 2001; Lanyon et al.
2007), diet (e. g. Ley et al. 2008), or association patterns among hyenas within clans (e. g.
Palmer et al. 2007), inﬂuence the structure of bacterial communities residing in the anal
pouches of spotted hyenas. Our objectives in doing so are to ascertain the extent of the
role bacteria play in the social lives of hyenas, while bringing together the all too often

disparate ﬁelds of behavioral, microbial and chemical ecology.

155

LITERATURE CITED

Achiraman, S. & Archunan, G. (2006). 1-Iodo-2methylundecane, a putative estrus-
speciﬁc urinary chemo-signal of female mouse (Mus musculus). —
Theriogenology 66, 1913-1920.

Agosta, W. C. (1992). Chemical Communication. — Scientiﬁc American Library, New
York.

Alberts, S. C. (1992). Constraints on the design of chemical communication systems in
terrestrial vertebrates. — Am. Nat. 139, S62-S89.

Albone, E. S. (1984). Mammalian Semiochemistry. — John Wiley, New York.

Albone, E. S., Eglinton, G., Walker, J. M. & Ware, G. C. (1974). The anal sac secretion
of the red fox (Vulpes vulpes): its chemistry and microbiology. A comparison
with the anal sac secretion of the lion (Panthera leo). — Life Sci. 14, 387-400.

Albone, E. S., Gosden, P. E., Ware, G. C., Macdonald, D. W. & Hough, N. G. (1978).
Bacterial action and chemical signalling in the red fox (Vulpes vulpes) and other
mammals. — In: Flavor Chemistry of Animal Foods (Bullard, R. W., ed), p. 78-
91.

Alcock, J. (2005). Animal Behavior: An Evolutionary Approach. — Sinauer Associates,
Inc., Sunderland.

Alexy, K. J ., Gassett, J. W., Osborn, D. A. & Miller, K. V. (2003). Bacterial fauna of the
tarsal tufts of white-tailed deer (Odocoileus virginianus). — Am. Midl. Nat. 149,
237-240.

Altmann, J. (1974). Observational study Of behavior: sampling methods. — Behaviour
49, 227-267.

Altmann, S. A. & Altmann, J. (2003). The transformation of behaviour ﬁeld studies. —
Anim. Behav. 65, 413-423.

An, D., Lee, H., Im, W., Liu, Q. & Lee, S. (2007). Segetibacter koreensis gen. nov., Sp.
nov., a novel member of the phylum Bacteroidetes, isolated from the soil of a
ginseng ﬁeld in South Korea. —— Int. J. Syst. Evol. Microbiol. 57, 1828-1833.

Apollonio, M., Mattioli, L., Scandura, M., Mauri, L., Gazzola, A. & Avanzinelli, E.
(2004). Wolves in the Casentinesi Forests: insights for wolf conservation in Italy

from a protected area with a rich wild prey community. — Biol. Conserv. 120,
249-260.

156

Apps, P. J., Viljoen, H. W., Richardson, P. R. K. & Pretorius, V. (1989). Volatile
components of anal gland secretion of aardwolf (Proteles cristatus). — J. Chem.
Ecol. 15, 1681-1688.

Aubin, T. (2004). Penguins and their noisy world. — Annals of the Brazilian Academy of
Sciences 76, 279-283.

Barbraud, C. & Delord, K. (2006). Population census of blue petrels Halobaena caerulea
at Mayes Islnad, Iles Kerguelen. — Antarct. Sci. 18, 199-204.

Barea-Azcon, J. M., Virgos, E., Ballesteros-Duperon, E., Moleon, M. & Chirosa, M.
(2007). Surveying carnivores at large spatial scales: a comparison of four broad-
applied methods. — Biodivers. Conserv. 16, 1213-1230.

Bartlett, J. G., Onderdonk, A. B., Drude, B, Goldstein, C., Anderka, M., Alpert, S. &
Mccorrnack, W. M. (1977). Quantitative bacteriology of the vaginal ﬂora. — J.
Infect. Dis. 136, 271-277.

Bayly, K. L. & Evans, C. S. (2003). Dynamic changes in alarm call structure: a strategy
for reducing conspicuousness to avian predators? — Behaviour 140, 353-369.

Bearder, S. K. & Randall, R. M. (1978). The use of fecal marking sites by Spotted hyenas
and civets. — Carnivore 1, 32-48.

Beauchamp, G. K., Katahira, K., Yamazaki, K., Mennella, J. A., Bard, J. & Boyse, E. A.
(1995). Evidence suggesting that the odortypes of pregnant women are a

compound of maternal and fetal odortypes. — Proc. Natl. Acad. Sci. U. S. A. 92,
2617-2621.

Beauchamp, G. K. & Yamazaki, K. (2005). Individual differences and the chemical
senses. — Chem. Senses 30, I6-I9.

Beauchamp, G. K., Yamazaki, K., Curran, M., Bard, J. & Boyse, E. A. (1994). Fetal
odour types are evident in the urine of pregnant female mice. —— Immunogenetics
39, 109-113.

Begg, C. M., Begg, K. S., Du Toit, M. & Mills, M. G. L. (2003). Scent-marking
behaviour of the honey badger, Mellivora capensis (Mustelidae), in the southern
Kalahari. — Anim. Behav. 66, 917-929.

Bel, M. C., Coulon, J ., Sreng, L., Allaine, D., Bagneres, A. G. & Clement, J. L. (1999).

Social signals involved in scent-marking behavior by cheek-rubbing in alpine
marmots (Marmota marmota). — J. Chem. Ecol. 25, 2267-2283.

157

 

Blas, J ., Perez-Rodriquez, L., Bortolotti, G. R., Vinuela, J. & Marchant, T. A. (2006).
Testosterone increases bioavailability of carotenoids: Insights into the honesty of
sexual Signaling. — Proc. Natl. Acad. Sci. U. S. A. 103, 18633-18637.

Blaustein, A. R. (1981). Sexual selection and mammalian olfaction. — Am. Nat. 117,
1006-1010.

Bloss, J., Acree, T. E., Bloss, J. M., Hood, W. R. & Kunz, T. H. (2002). Potential use of
chemical cues for colony-mate recognition in the big brown bat, Eptesicusfuscus.
—— J. Chem. Ecol. 28, 819-834.

Blumstein, D. T. & Armitage, K. B. (1997). Does sociality drive the evolution of
communicative complexity? A comparative test with ground-dwelling sciurid
alarm calls. — Am. Nat. 150, 179-200.

Boul, K. E., Funk, W. C., Darst, C. R., Cannatella, D. C. & Ryan, M. J. (2007). Sexual
selection drives speciation in an Amazonian frog. — Proc. R. Soc. Biol. Sci. Ser.
B 274, 399-406.

Boydston, E. E., Kapheim, K. M. & Holekamp, K. E. (2006). Patterns of den occupation
by the Spotted hyaena (Crocuta crocuta). — Afr. J. Ecol. 44, 77-86.

Boydston, E. E., Morelli, T. L. & Holekamp, K. E. (2001). Sex differences in territorial
behavior exhibited by the Spotted hyena (Hyaenidae, Crocuta crocuta). —-
Ethology 107, 369-385.

Bradbury, J. W. & Vehrencamp, S. L. (1998). Principles of Animal Communication. —
Sinauer Associates, Inc., Sunderland, MA.

Bradbury, J. W. & Vehrencamp, S. L. (2000). Economic models of animal
communication. — Anim. Behav. 59, 259-268.

Brennan, P. A. & Kendrick, K. M. (2006). Mammalian social odours: attraction and
individual recognition. — Philos. Trans. R. Soc. Lond. B Biol. Sci. 361, 2061-
2078.

Brown, R. E. (1979). Mammalian social odors: a critical review. — Adv. Study Behav.
10, 104-162.

Brumm, H., Voss, K., Kollmer, I. & Todt, D. (2004). Acoustic communication in noise:
regulation of call characteristics in a New World monkey. -— J. Exp. Biol. 207,
443-448.

Buesching, C. D., Stopka, P. & Macdonald, D. W. (2003). The social function of allo-
marking in the European badger (Meles meles). — Behaviour 140, 965-980.

158

Buesching, C. D., Waterhouse, J. S. & Macdonald, D. W. (2002). Gas-chromatographic
analyses of the subcaudal gland secretion of the European badger (Meles meles).
Part II: time-related variation in the individual-speciﬁc composition. — J. Chem.
Ecol. 28, 57-69.

Buglass, A. J ., Darling, F. M. C. & Waterhouse, J. S. (1991). Analysis of the anal sac
secretion of the Hyaenidae. — In: Chemical Signals in Vertebrates 5 (Macdonald,
D. W., Muller-Schwarze, D. & Natynczuk, S. E., eds). Kluwer Academic /
Plenum Publishers, New York, p. 65-69.

Burgener, N., East, M. L., Hofer, H. & Dehnhard, M. (2006). Do scent marks in spotted
hyenas (Crocuta crocuta) code for clan membership? Poster. Chemical Signals in

Vertebrates 11, University of Chester, England.

Burgener, N., East, M. L., Hofer, H. & Dehnhard, M. (2008). Do spotted hyena scent
marks code for clan membership? — In: Chemical Signals in Vertebrates 1 1
(Hurst, J. L., Beynon, R. J., Roberts, S. C. & Wyatt, T. D., eds). Springer, New
York, p. 169-178.

Burkhardt, R. W. (2005). Patterns of Behavior. — University of Chicago Press, Chicago.

Cavaggioni, A., Mucignat-Caretta, C., Redaelli, M. & Zagotto, G. (2006). The scent of
urine spots of male mice, Mus musculus: changes in chemical composition over
time. — Rapid Commun. Mass Spectrom. 20, 3741-3746.

Cheetham, S. A., Thom, M. D., Beynon, R. J. & Hurst, J. L. (2008). The effect of
familiarity on mate choice. — In: Chemical Signals in Vertebrates 11 (Hurst, J.
L., Beynon, R. J ., Roberts, S. C. & Wyatt, T. D., eds). Springer, New York, p.
271-280.

Clarke, K. R. (1993). Non-parametric multivariate analysis of changes in community
structure. — Aust. J. Ecol. 18, 117-143.

Cogen, A. L., Nizet, V. & Gallo, R. L. (2008). Skin microbiota: a source of disease or
defense? — Br. J. Dermatol. 158, 442-455.

Cole, J. R., Chai, B., Farris, R. J ., Wang, Q., Kulam-Syed-Mohideen, A. S., Mcgarrell, D.
M., Bandela, A. M., Cardenas, E., Garrity, G. M. & Tiedje, J. M. (2007). The
ribosomal database project (RDP-II): introducing myRDP Space and quality
controlled public data. — Nucleic Acids Res. 35, D169-D172.

Collins, M. D., Lawson, P. A., Monasterio, R., Falsen, E., Sjoden, B. & Facklam, R. R.

(1999). Ignavigranum ruoﬂiae Sp. nov., isolated from human clinical specimens.
—— International Journal of Systematic Bacteriology 49, 97-101.

159

Colwell, R. K. (2006). EstimateS: Statistical estimation of species richness and shared
species from samples. Version 8. — Persistent URL <purl.oclc.org/estimates>

Coolen, M. J ., Post, E., Davis, C. C. & Forney, L. J. (2005). Characterization of microbial
communities found in the human viagina by analysis restriction fragment length
polymorphisms of 16S rRNA genes. — Appl. Environ. Microbiol. 71, 8729-8737.

Cooper, S. M., Holekamp, K. E. & Smale, L. (1999). A seasonal feast: long-term analysis
of feeding behaviour in the Spotted hyaena (Crocuta crocuta). — Afr. J. Ecol. 37,
149-160.

Darling, F. F. (1960). An ecological reconnaissance of the Mara plains in Kenya colony.
—— The Wildlife Society, Washington, DC.

Darwin, C. (1872). The Expression of the Emotions in Man and Animals. — John
Murray, London.

Dethlefsen, L., Eckburg, P. B., Bik, E. M. & Relman, D. A. (2006). Assembly of the
human intestinal microbiota. —— Trends Ecol. Evol. 21, 517-523.

Dethlefsen, L., Mcfall-Ngai, M. J. & Relman, D. A. (2007). An ecological and
evolutionary perspective on human-microbe mutualism and disease. —— Nature
449, 811-818.

Doty, R. L. (1995). Introduction and historical perspective. — In: Handbook of Olfaction
and Gustation (Doty, R. L., ed). Marcel Dekker, Inc., New York, p. 1-32.

Drea, C. M. & Scordato, E. S. (2008). Olfactory communication in the ringtailed lemur
(Lemur catta): form and function of multimodal signals. —— In: Chemical Signals
in Vertebrates 11 (Hurst, J. L., Beynon, R. J ., Roberts, S. C. & Wyatt, T. D., eds).
Springer, New York, p. 91-102.

Drea, C. M., Vignieri, S. N., Kim, H. S., Weldele, M. L. & Glickman, S. E. (2002).
Responses to olfactory stimuli in spotted hyenas (Crocuta crocuta): II.
Discrimination of conspeciﬁc scent. — J. Comp. Psychol. 116, 342-349.

Drickamer, L. C. & Brown, P. L. (1998). Age-related changes in odor preferences by
house mice living in seminatural enclosures. — J. Chem. Ecol. 24, 1745-1756.

Dugatkin, L. A. (2004). Principles of Animal Behavior. — W.W. Norton & Company,
New York.

Dusenbery, D. B. (1992). Sensory Ecology. — W.H. Freeman and Company, New York.

East, M. L., Burke, T., Wilhelm, K., Greig, C. & Hofer, H. (2003). Sexual conﬂicts in
spotted hyenas: male and female mating tactics and their reproductive outcome

160

with respect to age, social status and tenure. — Proc. R. Soc. Biol. Sci. Ser. B
270, 1247-1254.

East, M. L. & Hofer, H. (1991a). Loud calling in a female-dominated mammalian society
.1. Structure and composition of whooping bouts of spotted hyaenas, Crocuta
crocuta. — Anim. Behav. 42, 637-649.

East, M. L. & Hofer, H. (1991b). Loud calling in a female-dominated mammalian society
.11. Behavioral contexts and functions of whooping of spotted hyaenas, Crocuta
crocuta. — Anim. Behav. 42, 651-669.

East, M. L. & Hofer, H. (1995). Serengeti nights. — Wildlife Conservation July/August, 5
38-45.

East, M. L. & Hofer, H. (2001). Male spotted hyenas (Crocuta crocuta) queue for status i-
in social groups dominated by females. — Behav. Ecol. 12, 558-568.

 

East, M. L., Hofer, H. & Wickler, W. (1993). The erect penis is a ﬂag of submission in a
female-dominated society - Greetings in Serengeti Spotted hyenas. — Behav. F'
Ecol. Sociobiol. 33, 355-370.

Eichorst, S. A., Breznak, J. A. & Schmidt, T. M. (2007). Isolation and characterization of
soil bacteria that deﬁne T erriglobus gen. nov., in the phylum Acidobacteria. ———-
Appl. Environ. Microbiol. 73, 2708-2717.

Eisenberg, J. F. & Kleiman, D. G. (1972). Olfactory communication in mammals. —
Annu. Rev. Ecol. Syst. 13, 1-32.

Elkinton, J. S. & Carde, R. T. (1984). Odor dispersion. — In: Chemical Ecology of
Insects (Bell, W. J. & Carde, R. T., eds). Chapman and Hall, Ltd., p. 73-91.

Elwood, R. W., Pothanikat, R. M. E. & Briffa, M. (2006). Honest and dishonest displays,
motivational state and subsequent decisions in hermit crab Shell ﬁghts. — Anim.
Behav. 72.

Engh, A. L., Esch, K., Smale, L. & Holekamp, K. E. (2000). Mechanisms of maternal
rank 'inheritance' in the spotted hyaena, Crocuta crocuta. — Anim. Behav. 60,
323-332.

Engh, A. L., Funk, S. M., Van Horn, R. C., Scribner, K. T., Bruford, M. W., Libants, S.,

Szykman, M., Smale, L. & Holekamp, K. E. (2002). Reproductive skew among
males in a female-dominated mammalian society. — Behav. Ecol. 13, 193-200.

Eschenbach, D. A., Thwin, S. S., Patton, D. L., Hooton, T. M., Stapleton, A. E., Agnew,
K., Winter, C., Meier, A. & Starnm, W. E. (2000). Inﬂuence of the normal

161

menstrual cycle on vaginal tissue, discharge, and microﬂora. —— Clin. Infect. Dis.
30, 901-907.

Estes, R. D. (1992). The Behavior Guide to African Mammals. — University of
California Press, Berkeley.

Ezaki, T., Li, N. & Kawamura, Y. (2006). The anaerobic gram-positive cocci. —
Prokaryotes 4, 795-808.

Fell, R. J ., Buesching, C. D. & Macdonald, D. W. (2006). The social integration of
European badger (Meles meles) cubs into their natal group. — Behaviour 143,
683-700.

F erkin, M. H., Lee, D. N. & Leonard, S. T. (2004). The reproductive state of female voles
affects their scent marking behavior and the responses of male conspeciﬁcs to
such marks. — Ethology 110, 257-272.

F erkin, M. H. & Pierce, A. A. (2007). Perspectives on over-marking: is it good to be on
top? — J. Ethol. 25, 107-116.

Ferkin, M. H., Pierce, A. A., Sealand, R. O. & Delbarco-Trillo, J. (2005). Meadow voles,
Microtus pennsylvanicus, can distinguish more over-marks from fewer over-
marks. — Anim. Cogn. 8, 182-189.

F erkin, M. H., Sorokin, E. S., Johnston, R. E. & Lee, C. J. (1997). Attractiveness of
scents varies with protein content of the diet in meadow voles. — Anim. Behav.
53, 133-141.

Finegold, S. M., Vaisanen, M.-L., Molitoris, D. R., Tomzynski, T. J ., Song, Y., Liu, C.,
Collins, M. D. & Lawson, P. A. (2003). Cetobacterium somerae Sp. nov. from
Human Feces and Emended Description of the Genus Cetobacterium. — Syst.

Appl. Microbiol. 26, 177-181.

Fisher, H. S., Swaisgood, R. R. & F itch-Snyder, H. (2003). Odor familiarity and female
preferences for males in a threatened primate, the pygmy loris Nycticebus

pygmaeus: applications for genetic management of small populations. —
Naturwissenschaften 90, 509-512.

Fisher, J. (1954). Evolution and bird sociality. — In: Evolution as a Process (Huxley, J .,
Hardy, A. C. & Ford, E. B., eds). Allen and Unwin, London, p. 71-83.

F orsman, A. & Hagman, M. (2006). Calling is an honest indicator of paternal genetic
quality in poison frogs. — Evolution 60, 2148-2157.

Frank, L. G. (1986a). Social organization of the spotted hyaena (Crocuta crocuta). I.
Demography. — Anim. Behav. 34, 1500-1509.

162

 

Frank, L. G. (1986b). Social organization of the spotted hyaena Crocuta crocuta .Il.
Dominance and reproduction. — Anim. Behav. 34, 1510-1527.

Frank, L. G., Glickman, S. E. & Powch, I. (1990). Sexual Dimorphism in the Spotted
Hyena (Crocuta crocuta). — J. 2001. 221, 308-313.

Frank, L. G., Holekamp, K. E. & Smale, L. (1995). Dominance, demography, and
reproductive success of female Spotted hyenas. — In: Serengeti II: Dynamics,
management, and conservation of an ecosystem (Sinclair, A. R. E. & Arcese, P.,
eds). University of Chicago Press, Chicago, p. 364-3 84.

Freeberg, T. M. (2006). Social complexity can drive vocal complexity. — Psychol. Sci.
17, 557-561.

French, J. A. & Cleveland, J. (1984). Scent-marking in the tamarin, Saguinus oedipus:
sex differences and ontogeny. — Anim. Behav. 32, 615-623.

Geissmann, T. & Nijman, V. (2006). Calling in wild silvery gibbons (Hylobates moloch)

in Java (Indonesia): Behavior, phylogeny, and conservation. — Am. J. Primatol.
68, 1-19.

Gorman, M. & Mills, M. G. L. (1984). Scent marking strategies in hyaenas (Mammalia).
— J. Zool. 202, 535-547.

Gorman, M. L. (1976). A mechanism for individual recognition by odour in Herpestes
auropunctatus (Camivora: Viverridae). — Anim. Behav. 24, 141-145.

Gorman, M. L., Nedwell, D. B. & Smith, R. M. (1974). An analysis of the contents of the
anal scent pockets of Herpestes auropunctatus (Camivora: Viverridae). — J.
Zool. 172, 389-399.

Gorman, M. L. & Trowbridge, B. J. (1989). The role of odor in the social lives of
carnivores. — In: Carnivore Behavior, Ecology and Evolution (Gittleman, J. L.,
ed). Cornell University Press, Ithaca, p. 57-88.

Gosling, L. M. (1982). A reassessment of the function of scent marking in territories. —
Z. Tierpsychol. 60, 89-1 18.

Gosling, L. M. & Roberts, S. C. (2001). Scent-marking by male mammals: cheat-proof
signals to competitors and mates. — Adv. Study Behav. 30, 169-217.

Gosling, L. M., Roberts, S. C., Thornton, E. A. & Andrew, M. J. (2000). Life history

costs of olfactory status signalling in mice. — Behav. Ecol. Sociobiol. 48, 328-
332.

163

Gotelli, N. J. & Ellison, A. M. (2004). A Primer of Ecological Statistics. — Sinauer
Associates, Inc., Sunderland, MA.

Gould, L. & Overdorff, D. J. (2002). Adult male scent-marking in Lemur catta and
Eulemurfulvus rufus. — Int. J. Primatol. 23, 575-586.

Gould, S. J. & Lewontin, R. C. (1979). The spandrels of San Marco and the Panglossian
paradigm: a critique of the adaptationist programme. — Proc. R. Soc. Biol. Sci.
Ser. B 205, 581-598.

Grafe, T. U. (2005). Anuran choruses as communication networks. —— In: Animal
Communication Networks (Mcgregor, P. K., ed). Cambridge University Press,
Cambridge, p. 277-299.

Haine, E. R. (2008). Symbiont-mediated protection. — Proc. R. Soc. Biol. Sci. Ser. B
275, 353-361.

Hammer, 0., Harper, D. A. T. & Ryan, P. D. (2001). PAST: paleontological statistics
software package for education and data analysis. —- Palaeontologia Electronica
4, 1-9.

Hauser, B., Deschner, T. & Boesch, C. (2008). DeveIOpment of a liquid-chromatography-
tandem mass spectrometry method for the determination of 23 endogenous
steroids in small quantities of primate urine. — J. Chromatogr. B 862, 100—112.

Hauser, M. D. (1996). The Evolution of Communication. — The MIT Press, Cambridge,
MA.

Hayes, R. A., Richardson, B. J. & Wyllie, S. G. (2001). Increased social dominance in
male rabbits, Oryctolagus cuniculus, is associated with increased secretion of 2-
phenoxy ethanol from the chin gland. —— In: Chemical Signals in Vertebrates 9
(Marchlewska-Koj, A., Lepri, J. L. & Muller-Schwarze, D., eds). Kluwer
Academic / Plenum Publishers, New York, p. 335-341.

Henschel, J. R. & Skinner, J. D. (1991). Territorial behaviour by a clan of spotted
hyaenas, Crocuta crocuta. — Ethology 88, 223-235.

Heymann, E. W. (2000). Spatial patterns of scent marking in wild moustached tamarins,
Saguinus mystax: no evidence for a territorial function. — Anim. Behav. 60, 723-
730.

Hofer, H. & East, M. L. (1993a). The commuting system of Serengeti spotted hyaenas:

how a predator copes with migratory prey. 1. Social organization. —— Anim.
Behav., 547-557.

164

Hofer, H. & East, M. L. (1993b). The commuting system of Serengeti spotted hyaenas:
how a predator copes with migratory prey. II. Intrusion pressure and commuters'
space use. — Anim. Behav. 46, 559-574.

Hofer, H. & East, M. L. (1993c). The commuting system of Serengeti spotted hyaenas:
how a predator copes with migratory prey. III. Attendance and maternal care. —
Anim. Behav. 46, 575-589.

Hofer, H. & East, M. L. (2003). Behavioral processes and costs of co-existence in female
spotted hyenas: a life history perspective. — Evol. Ecol. 17, 315-331.

Hofer, H., East, M. L., Sammang, I. & Dehnhard, M. (2001). Analysis of volatile
compounds in scent-marks of spotted hyenas (Crocuta crocuta) and their possible
function in olfactory communication. — In: Chemical Signals in Vertebrates 9
(Marchlewska-Koj, A., Lepri, J. L. & Muller-Schwarze, D., eds). Kluwer
Academic / Plenum Publishers, New York, p. 141-148.

Holekamp, K. E., Boydston, E. E., Szykman, M., Graham, 1., Nutt, K. J ., Birch, S.,
Piskiel, A. & Singh, M. (1999a). Vocal recognition in the spotted hyaena and its
possible implications regarding the evolution of intelligence. — Anim. Behav. 58,
383-395.

Holekamp, K. E., Cooper, S. M., Katona, C. 1., Berry, N. A., Frank, L. G. & Smale, L.
(1997a). Patterns of association among female spotted hyenas (Crocuta crocuta).
— J. Mammal. 78, 55-64.

Holekamp, K. E., Ogutu, J. 0., Dublin, H. T., Frank, L. G. & Smale, L. (1993). Fission of
a spotted hyena clan: consequences of prolonged female absenteeism and causes
of female emigration. — Ethology 93, 285-299.

Holekamp, K. E., Sakai, S. T. & Lundrigan, B. L. (2007). Social intelligence in the
spotted hyena (Crocuta crocuta). — Phil. Trans. R. Soc. B 362, 523-538.

Holekamp, K. E. & Smale, L. (1991). Dominance acquisition during mammalian social
development: the 'inheritance' of maternal rank. — Am. Zool. 31, 306-317.

Holekamp, K. E. & Smale, L. (1993). Ontogeny of dominance in free-living spotted
hyaenas: juvenile rank relations with other immature individuals. — Anim.
Behav. 46, 451-466.

Holekamp, K. E. & Smale, L. (1998). Behavioral development in the spotted hyena. —
Bioscience 48, 997-1005.

Holekamp, K. E. & Smale, L. (2000). Feisty females and meek males: reproductive

strategies in the spotted hyena. — In: Reproduction in Context (Wallen, K. &
Schneider, J. E., eds). MIT Press, Cambridge, MA, p. 257-285.

165

Holekamp, K. E., Smale, L., Berg, R. & Cooper, S. M. (1997b). Hunting rates and
hunting success in the spotted hyena (Crocuta crocuta). — J. Zool. 242, 1-15.

Holekamp, K. E., Smale, L. & Szykman, M. (1996). Rank and reproduction in the female
spotted hyaena. — J. Reprod. F ertil. 108, 229-237.

Holekamp, K. E., Szykman, M., Boydston, E. E. & Smale, L. (1999b). Association of
seasonal reproductive patterns with changing food availability in an equatorial
carnivore, the spotted hyaena (Crocuta crocuta). — J. Reprod. Fertil. 116, 87-93.

Honer, O. P., Wachter, B., East, M. L., Streich, W. J ., Wilhelm, K., Burke, T. & Hofer,
H. (2007). Female mate-choice drives the evolution of male-biased dispersal in a
social mammal. — Nature 448, 798-802.

Hurst, J. L. (2005). Scent marking and social communication. — In: Animal
Communication Networks (Mcgregor, P. K., ed). Cambridge University Press,
Cambridge, p. 219-243.

Hurst, J. L. & Beynon, R. J. (2004). Scent wars: the chemobiology of competitive
signalling in mice. — BioEssayS 26, 1288-1298.

Hurst, J. L., Fang, J. & Barnard, C. J. (1993). The role of substrate odours in maintaining
social tolerance between male house mice (Mus musculus domesticus):
relatedness, incidental kinship effects and the establishment of social status. —
Anim. Behav. 48, 157-167.

Ibeas, F., Gallego, D., Diez, J. J. & Pajares, J. A. (2007). An operative kairomonal lure
for managing pine sawyer beetle Monochamus galloprovincialis (Coleoptera:
Cerymbycidae). — J. Appl. Entomol. 131, 13-20.

James, A., Casey, J ., Hyliands, D. & Mycock, G. (2004). Fatty acid metabolism by
cutaneous bacteria and its role in axillary malodour. —— World J. Microbiol.
Biotechnol. 20, 787-793.

Jeans-Williams, N. L. & Borden, J. H. (2006). Evaluation of two pheromone baits for
containment and concentration of attack by the western balsam bark beetle,
Dryocoetes confusus (Coleoptera: Curculionidae). — West. J. Appl. For. 21, 27 -
32.

Jog, M. & Watve, M. (2005). Role of parasites and commensals in shaping host
behaviour. — Current Science 89, 1184-1191.

J ohansson, B. G. & Jones, T. M. (2007). The role of chemical communication in mate
choice. — Biological Reviews 82, 265-289.

Johnson, R. P. (1973). Scent marking in mammals. — Anim. Behav. 21 , 521-535.

166

Johnston, R. E. (2003). Chemical communication in rodents: from pheromones to
individual recognition. — J. Mammal. 84, 1141-1162.

Johnston, R. E., Chiang, G. & Tung, C. (1994). The information in scent over-marks of
golden hamsters. — Anim. Behav. 48, 323—330.

Johnston, R. E., Sorokin, E. S. & Ferkin, M. H. (1997). Female voles discriminate males'
over-marks and prefer top-scent males. — Anim. Behav. 54, 679-690.

J ohnstone, R. A. (1997). The evolution of animal signals. — In: Behavioral Ecology: An
Evolutionary Approach (Krebs, J. R. & Davies, N. B., eds). Blackwell Science,
Malden, MA, p. 155-178.

Jordan, N. R., Cherry, M. I. & Manser, M. B. (2007). Latrine distribution and patterns of
use by wild meerkats: implications for territory and mate defence. — Anim.
Behav. 73, 613-622.

Kappeler, P. M. (1990). Social status and scent-marking behaviour in Lemur catta. —
Anim. Behav. 40, 774-776.

Kersters, K., De Vos, R, Gillis, M., Swings, J ., Vandamme, P. & Stackebrandt, E.
(2006). Introductionto the Proteobacteria. — Prokaryotes 5, 3-37.

Kingston, T. & Rossiter, S. J. (2004). Harrnonic-hopping in Wallacea's bats. — Nature
429, 654-657.

Kolowski, J. M., Katan, D., Theis, K. R. & Holekamp, K. E. (2007). Daily patterns of
activity in the spotted hyena. — J. Mammal. 88, 1017-1028.

Krebs, J. R. & Dawkins, R. (1984). Animal signals: mind-reading and manipulation. —
In: Behavioural Ecology: An Evolutionary approach (Krebs, J. R. & Davies, N.
B., eds). Sinauer Associates, Inc., Sunderland, MA, p. 380-402.

Kruuk, H. (1972). The Spotted Hyena: A Study of Predation and Social Behavior. —
University of Chicago Press, Chicago, IL.

Kruuk, H. (2003). Niko's Nature. — Oxford University Press, Oxford.

Kruuk, H., Gorman, M. & Leitch, A. (1984). Scent-marking with the subcaudal gland by
the European badger, Meles meles L. — Anim. Behav. 32, 899-907.

Lam, K., Babor, D., Duthie, B., Babor, E., Moore, M. & Gries, G. (2007). Proliferating

bacterial symbionts on house ﬂy eggs affect oviposition behaviour of adult ﬂies.
— Anim. Behav. 74, 81-92.

167

Lanyon, C. V., Rushton, S. P., O'donnell, A. G., Goodfellow, M., Ward, A. C., Petrie, M.,
Jensen, S. P., Gosling, L. M. & Penn, D. (2007). Murine scent mark microbial
communities are genetically determined. — FEMS Microbiol. Ecol. 59, 576-5 83.

Lazaro-Perea, C., Snowdon, C. T. & De Fatima Arruda, M. (1999). Scent-marking
behavior in wild groups of common marmosets (Callithrixjacchus). — Behav.
Ecol. Sociobiol. 46, 313-324.

Leger, D. W. (1993). Contextual sources Of information and responses to animal
communication signals. — Psychol. Bull. 113, 295-304.

Lehmann, J ., Korstjens, A. H. & Dunbar, R. I. M. (2007). Group size, grooming and
social cohesion in primates. — Anim. Behav. 74, 1617-1629.

Lewis, R. J. (2006). Scent marking in sifaka: no one function explains it all. — Am. J.
Primatol. 68, 622-636.

Ley, R. E., Hamady, M., Lozupone, C., Turnbaugh, P. J ., Ramey, R. R., Bircher, J. S.,
Schlegel, M. L., Tucker, T. A., Schrenzel, M. D., Knight, R. & Gordon, J. I.
(2008). Evolution of mammals and their gut microbes. — Science 320, 1647-
1651.

Leyden, J. L., Mcginley, K. J., Holzle, E., Labows, J. N. & Kligman, A. M. (1981). The
microbiology of the human axilla and its relationship to axillary odor. — The J.
Invest. Dermatol. 77, 413-416.

Liebl, W. (2006). Corynebacteriurn - Nonmedical. —- Prokaryotes 3, 796-818.

Lim, P., Chen, S., Tsai, C. & Pai, M. (2006). Pantoea peritonitis in a patient receiving
chronic ambulatory peritoneal dialysis. — Nephrology 11, 97-99.

Lombardo, M. P. (2008). Access to mutualistic endosymbiotic microbes: an
underappreciated beneﬁt of group living. — Behav. Ecol. Sociobiol. 62, 479-497.

Lopez, P., Amo, L. & Martin, J. (2006). Reliable signaling by chemical cues of male
traits and health state in male lizards, Lacerta monticola. — J. Chem. Ecol. 32,
473-488.

Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar, Buchner, A.,
Lai, T., Steppi, S., Jobb, G., Forster, W., Brettske, I., Gerber, S., Ginhart, A. W.,
Gross, O., Grumann, S., Hermann, S., Jost, R., Konig, A., Liss, T., Lussmann, R.,
May, M., Nonhoff, B., Reichel, B., Strehlow, R., Stamatakis, A., Stuckmann, N.,
Vilbig, A., Lenke, M., Ludwig, T., Bode, A. & Schleifer, K. H. (2004). ARB: a
software environment for sequence data. — Nucleic Acids Res. 32, 1363-1371.

168

Macdonald, D. W. (1985). The carnivores: order Camivora. -— In: Social Odours in
Mammals (Brown, R. E. & Macdonald, D. W., eds). Clarendon Press, Oxford, p.
619-722.

Magurran, A. E. (2004). Measuring Biological Diversity. — Blackwell Publishing,
Malden, MA.

Marler, P. (1967). Animal communication Signals. — Science 157, 769-774.

Mateo, J. M. (2006). The nature and representation of individual recognition odours in
Belding's ground squirrels. — Anim. Behav. 71, 141-154.

Matthews, L. H. (1939). Reproduction in the spotted hyaena, Crocuta crocuta. — Phil.
Trans, B 230, 1-78.

Maynard Smith, J. & Harper, D. (2003). Animal Signals. — Oxford University Press,
New York.

Mccomb, K. & Reby, D. (2005). Vocal communication networks in large terrestrial
mammals. — In: Animal Communication Networks (Mcgregor, P. K., ed).
Cambridge University Press, Cambridge, p. 372-3 89.

Mccomack, R. L. (1965). Extended tables of the Wilcoxon matched pair signed rank
statistic. — Journal of the American Statistical Association 60, 864-871.

Mccowan, B., Anderson, K., Heagarty, A. & Cameron, A. (2008). Utility of social
network analysis for primate behavioral management and well-being. — Appl.
Anim. Behav. Sci. 109, 396-405.

Mcfall-Ngai, M. J. & Ruby, E. G. (1991). Symbiont recognition and subsequent
morphogenesis as early events in an animal-bacterial mutualism. — Science 254,
1491-1494.

Mcgregor, P. K. (1993). Signalling in territorial systems: a context for individual
identiﬁcation, ranging and eavesdropping. — Phil. Trans, B 340, 237-244.

Mcgregor, P. K., ed. (2005). Animal Communication Networks. — Cambrdige
University Press, Cambridge.

Mech, S. G., Dunlap, A. S. & Wolff, J. O. (2003). Female prairie voles do not choose
males based on their frequency of scent marking. — Behav. Process. 61, 101-108.

Meyer, J. M., Goodwin, T. E. & Schulte, B. A. (2008). Intrasexual chemical

communication and social responses of captive female African elephants,
Loxodonta africana. — Anim. Behav. 76, 163-174.

169

Michael, R. P. (1975). Hormonal steroids and sexual communication in primates. — J.
Steroid Biochemist. 6, 161-170.

Miller, K. E., Laszlo, K. & Dietz, J. M. (2003). The role of scent marking in the social
communication of wild golden lion tamarins, Leontopithecus rosalia. — Anim.
Behav. 65, 795-803.

Mills, M. G. L. (1990). Kalahari Hyenas: Comparative Behavioral Ecology of Two
Species. — Chapman & Hall, New York, NY.

Mills, M. G. L. & Gorman, M. L. (1987). The scent-marking behaviour of the spotted
hyaena Crocuta crocuta in the Southern Kalahari. — J. 2001. 212, 483-497.

Moran, G. & Sorensen, L. (1986). Scent marking behavior in a captive group of meerkats
(Suricata suricatta). — J. Mammal. 67, 120-132.

Muller, C. A. & Manser, M. B. (2007). 'Nasty neighbours' rather than 'dear enemies' in a
social carnivore. — Proc. R. Soc. Biol. Sci. Ser. B 274, 959-965.

Muller, C. A. & Manser, M. B. (2008). Scent-marking and intrasexual competition in a
cooperative carnivore with low reproductive skew. — Ethology 114, 174-185.

Muller-Schwarze, D. (2001). From individuals to populations: Field studies as proving
grounds for the role of chemical signals. —- In: Chemical Signals in Vertebrates 9
(Marchlewska-Koj, A., Lepri, J. L. & Muller-Schwarze, D., eds). Kluwer
Academic / Plenum Publishers, New York, p. 1-10.

Muller-Schwarze, D. (2006). Chemical Ecology of Vertebrates. —— Cambridge University
Press, Cambridge.

Mundry, R. & Fischer, J. (1998). Use of statistical programs for nonparametric tests of
small samples often leads to incorrect P values: examples from Animal
Behaviour. — Anim. Behav. 56, 256-259.

Murdoch, D. A. (1998). Gram-positive anaerobic cocci. —-— Clin. Microbiol. Rev. 11, 81-
120.

Newton, E. R., Piper, J. M., Shain, R. N., Perdue, S. T. & Peairs, W. (2001). Predictors of
the vaginal microﬂora. — Am. J. Obstet. Gynecol. 184, 845-855.

Olsson, I. A. S., Nevison, C. M., Patterson-Kane, E. G., Sherwin, C. M., Van De Weerd,
H. A. & Wurbel, H. (2003). Understanding behaviour: the relevance of

ethological approaches in laboratory animal science. — Appl. Anim. Behav. Sci.
81, 245-264.

170

Ord, T. J ., Peters, R. A., Clucas, B. & Stamps, J. A. (2007). Lizards speed up visual
displays in noisy motion habitats. — Proc. R. Soc. Biol. Sci. Ser. B 274, 1057-
1062.

O'riain, M. J. & Jarvis, J. U. M. (1997). Colony member recognition and xenophobia in
the naked mole-rat. -— Anim. Behav. 53, 487-498.

Palagi, E., Gregorace, A. & Borgognini Tarli, S. M. (2002). Development of olfactory
behavior in captive ring-tailed lemurs (Lemur catta). —— Int. J. Primatol. 23, 587-
599.

Palagi, E., Telara, S. & Borgognini Tarli, S. M. (2004). Reproductive strategies in Lemur
catta: Balance among sending, receiving, and countermarking scent signals. —
Int. J. Primatol. 25, 1019-1031.

Palmer, C., Bik, E. M., Digiulio, D. B., Relman, D. A. & Brown, P. O. (2007).
Development of the human infant intestinal microbiota. — PLOS Biol. 5, 1556-
1573.

Palphramand, K. L. & White, P. C. L. (2007). Badgers, Meles meles, discriminate
between neighbour, alien and self scent. — Anim. Behav. 74, 429-436.

Pardo, I. & Armitage, K. B. (1997). Species assemblages as descriptors of mesohabitats.
— Hydrobiologia 344, 111-128.

Parekh, J. C. (2002). Axillary odor; its physiology, microbiology and chemistry. —
Cosmetics and Toiletries 117, 53-62.

Parker, M. (2006). Making scents: Learning the "words" in the wild dog's chemical
language. — In: Wildlife Conservation, p. 28-31.

Penn, D. & Potts, W. K. (1998). Chemical signals and parasite-mediated sexual selection.
— Trends Ecol. Evol. 13, 391-396.

Petrulis, A. & Johnston, R. E. (1997). Causes of scent marking in female golden hamsters
(Mesocricetus auratus): speciﬁc signals or classes of information? — J. Comp.
Psychol. 111, 25-36.

Poesel, A., Dabelsteen, T. & Pedersen, S. B. (2007). Implications of conspeciﬁc
background noise for features of blue tit, C yanistes caeruleus, communication
networks at dawn. — J. Omithol. 148, 123-128.

Pruesse, E., Quast, C., Knittel, K., Fuchs, B. M., Ludwig, W., Peplies, J. & Glockner, F.
O. (2007). SILVA: a comprehensive online resource for quality checked and

aligned ribosomal RNA sequence data compatible with ARB. — Nucleic Acids
Res. 35, 7188-7196.

171

Pujol, C. J. & Kado, C. I. (2000). Genetic and biochemical characterization of the
pathway in Pantoea citrea leading to pink disease of pineapple. — J. Bacteriol.
182, 2230-2237.

Quinn, G. P. & Keough, M. J. (2002). Experimental Design and Data Analysis for
Biologists. — Cambridge University Press, Cambridge.

Rajanarayanan, S. & Archunan, G. (2004). Occurrence of ﬂehmen in male buffaloes
(Bubalus bubalis) with special reference to estrus. — Theriogenology 61, 861-
866.

Ralls, K. (1971). Mammalian scent marking. — Science 171 , 443-449.

Rasa, O. A. E. (1973). Marking behaviour and its social signiﬁcance in the African dwarf
mongoose, Helogale undulata rufula. — Z. Tierpsychol. 32, 293-318.

Rasmussen, L. E. L., Krishnamurthy, V. & Sukumar, R. (2005). Behavioural and
chemical conﬁrmation of the preovulatory pheromone, (Z)-7-dodecenyl acetate,
in wild Asian elephants: its relationship to musth. — Behaviour 142, 351-396.

Rawls, J. F., Samuel, B. S. & Gordon, J. I. (2004). Gnotobiotic zebraﬁsh reveal
evolutionary conserved responses to the gut microbiota. — Proc. Natl. Acad. Sci.
USA. 101, 4596-4601.

Reiger, I. (1979). Scent rubbing in carnivores. — Carnivores 2, 17-25.

Remage-Healey, L., Nowacek, D. P. & Bass, A. H. (2006). Dolphin foraging sounds
suppress calling and elevate stress hormone levels in a prey species, the Gulf
toadﬁsh. — J. Exp. Biol. 209, 4444-4451.

Rennie, P. J ., Gower, D. B., Holland, K. T., Mallet, A. I. & Watkins, W. J. (1990). The
skin microﬂora and the formation of human axillary odour. — Int. J. Cosmet. Sci.
12, 197-207.

Rich, T. J. & Hurst, J. L. (1998). Scent marks as reliable signals of the competitive ability
of mates. — Anim. Behav. 56, 727-735.

Rich, T. J. & Hurst, J. L. (1999). The competing countermarks hypothesis: reliable
assessment of competitive ability by potential mates. — Anim. Behav. 58, 1027-
1037.

Riesenfeld, C. S., Schloss, P. D. & Handelsman, J. (2004). Metagenomics: genomic
analysis of microbial communities. — Annu. Rev. Genet. 38, 525-552.

172

Roberts, S. C. & Gosling, L. M. (2004). Manipulation of olfactory signaling and mate
choice for conservation breeding: a case study of harvest mice. — Conserv. Biol.
18, 548-556.

Rolph, H. J ., Lennon, A., Riggio, M. P., Saunders, W. P., Mackenzie, D., Coldero, L. &
Bagg, J. (2001). Molecular identiﬁcation of microorganisms from endodontic
infections. — J. Clin. Microbiol. 39, 3282-3289.

Rosell, F. & Bjorkoyli, T. (2002). A test of the clear enemy phenomenon in the Eurasian
beaver. —— Anim. Behav. 63, 1073-1078.

Rostain, R. R., Ben-David, M., Groves, P. & Randall, J. A. (2004). Why do river otters
scent-mark? An experimental test of several hypotheses. —— Anim. Behav. 68,
703-711.

Ruoff, K. L. (2002). Miscellaneous catalase-negative, gram-positive cocci: emerging
opportunists. — J. Clin. Microbiol. 40, 1129-1133.

Ruther, J., Sieben, S. & Schricker, B. (2002). Nestmate recognition in social wasps:
manipulation of hydrocarbon proﬁles induces aggression in the European hornet.
— Naturwissenschaften 89, 111-114.

Ryan, M. J ., Tuttle, M. D. & Rand, A. S. (1982). Bat predation and sexual advertisement
in a neotropical anuran. — Am. Nat. 119, 136-139.

Ryan, P. G., Dorse, C. & Hilton, G. M. (2006). The conservation status of the spectacled
petrel Procellaria conspicillata. — Biol. Conserv. 131, 575-583.

Saﬁ, K. & Kerth, G. (2003). Secretions of the interaural gland contain information about
individuality and colony membership in the Bechstein's bat. — Anim. Behav. 65,
363-369.

Savage, D. C. (1977). Microbial ecology of the gastrointestinal tract. — Annu. Rev.
Microbiol. 31, 107-133.

Schaal, K. P., Yassin, A. F. & Stackebrandt, E. (2006). The Family Actinomycetaceae:
The genera Actinomyces, Actinobaculum, Arcanobacterium, Varibaculum,
Mobiluncus. — Prokaryotes 3, 430-537.

Schaller, G. B. (1993). The Last Panda. — University of Chicago Press, Chicago, IL.

Schloss, P. D. & Handelsman, J. (2004). Status of the microbial census. — Microbiol.
Mol. Biol. Rev. 68, 686-691.

173

Schloss, P. D. & Handelsman, J. (2005). Introducing DOTUR, a computer program for
deﬁning operational taxonomic units and estimating species richness. —— Appl.
Environ. Microbiol. 71, 1501-1506.

Schulte, B. A., Freeman, E. W., Goodwin, T. E., Hollister-Smith, J. & Rasmussen, L. E.
L. (2007). Honest signalling through chemicals by elephants with applications for
care and conservation. — Appl. Anim. Behav. Sci. 102, 344-363.

Scordato, E. S. & Drea, C. M. (2007). Scents and sensibility: information content of
olfactory signals in the ringtailed lemur, Lemur catta. —— Anim. Behav. 73, 301-
314.

Setchell, J. M., Charpentier, M. J. E., Bedjabaga, I. 8., Reed, P., Wickings, E. J. &
Knapp, L. A. (2006). Secondary sexual characters and female quality in primates.
— Behav. Ecol. Sociobiol. 61, 305-315.

Shaffer, J. P. (1995). Multiple hypothesis testing. — Annu. Rev. Psychol. 46, 561-584.

Sharpe, L. L. (2005). Play does not enhance social cohesion in a cooperative mammal. —
Anim. Behav. 70, 551-558.

Slade, B. E., Schulte, B. A. & Rasmussen, L. E. L. (2003). Oestrous state dynamics in
chemical communication by captive female Asian elephants. — Anim. Behav. 65,
813-819.

Sliwa, A. (1996). A functional analysis of scent marking and mating behaviour in the
aardwolf, Proteles cristatus. — Doctoral Dissertation, University of Pretoria.

Sliwa, A. & Richardson, P. R. K. (1998). Responses of aardwolves, Proteles cristatus,
Sparrman 1783, to translocated scent marks. — Anim. Behav. 56, 137-146.

Smale, L., Frank, L. G. & Holekamp, K. E. (1993). Ontogeny of dominance in free-living
spotted hyaenas: juvenile rank relations with adult females and immigrant males.
— Anim. Behav. 46, 467-477.

Smale, L., Nunes, S. & Holekamp, K. E. (1997). Sexually dimorphic dispersal in
mammals: patterns, causes and consequences. -— Adv. Study Behav. 26, 181-250.

Smith, J. E., Kolowski, J. M., Graham, K. E., Dawes, S. E. & Holekamp, K. E. (2008).
Social and ecological determinants of ﬁssion-fusion dynamics in the spotted
hyaena. — Anim. Behav. 76, 619-636.

Smith, J. E., Memenis, S. K. & Holekamp, K. E. (2007). Rank-related partner choice in

the ﬁssion-fusion society of the spotted hyena. — Behav. Ecol. Sociobiol. 61,
753-765.

174

Smith, T. E., Abbott, D. H., Tomlinsom, A. J. & Mlotkiewicz, J. A. (1997). Differential
display of investigative behavior permits discrimination of scent signatures from
familiar and unfamiliar socially dominant female marmoset monkeys (Callithrix

jacchus). — J. Chem. Ecol. 23, 2523-2546.

Sorenson, M. D., Sefc, K. M. & Payne, R. B. (2003). Speciation by host switch in brood
parasitic indigobirds. — Nature 424, 928-931.

Stackebrandt, E., Cummins, C. S. & Johnson, J. L. (2006). Family Propionibacteriaceae:
The Genus Propionibacterium. — Prokaryotes 3, 400-418.

Stackebrandt, E. & Goebel, B. M. (1994). Taxonomic note: a place for DNA-DNA
reassociation and 16S rRNA sequence analysis in the present Species deﬁnition in
bacteriology. —— Int. J. Syst. Bacteriol. 44, 846-849.

Stevenson, B. S., Eichorst, S. A., Wertz, J. T., Schmidt, T. M. & Breznak, J. A. (2004).
New strategies for cultivation and detection of previously uncultured microbes. -—
Appl. Environ. Microbiol. 70, 4748-4755.

Subchev, M., Toshova, T., Toth, M., Voigt, E., Mikulas, J. & Francke, W. (2004).
Catches of vine bud moth T heresimima ampellophaga (Lep., Zygaenidae:
Procridinae) males in pheromone traps: effect of the purity and age of baits,

design, colour and height of the traps, and daily sexual activity of males. —— J.
Appl. Entomol. 128, 44-50.

Sussman, R. W. (1992). Male life-history and intergroup mobility among ringtailed
lemurs (Lemur catta). —— Int. J. Primatol. 13, 395-413.

Swaisgood, R. R., Lindburg, D., White, A. M., Zhang, H. & Zhou, X. (2004). Chemical
communication in giant pandas. —— In: Giant Pandas: Biology and Conservation

(Lindburg, D. & Baragona, K., eds). University of California Press, Berkeley, p.
106-120.

Swaisgood, R. R., Lindburg, D. G. & Zhang, H. (2002). Discrimination of oestrous status

in giant pandas (A iluropoda melanoleuca) via chemical cues in urine. —— J. Zool.
257, 381-386.

Szykman, M., Engh, A. L., Van Horn, R. C., Boydston, E. E., Scribner, K. T. &
Holekamp, K. E. (2003). Rare male aggression directed toward females in a

female-dominated society: baiting behavior in the spotted hyena. — Aggress.
Behav. 29, 457-474.

Szykman, M., Engh, A. L., Van Horn, R. C., Funk, S. M., Scribner, K. T. & Holekamp,
K. E. (2001). Association patterns among male and female spotted hyenas

(Crocuta crocuta) reﬂect male mate choice. — Behav. Ecol. Sociobiol. 50, 231-
238.

175

Tamura, K., Dudley, J ., Nei, M. & Kumar, S. (2007). MEGA4: Molecular evolutionary
genetics analysis (MEGA) software version 4.0. — Mol. Biol. Evol. 24, 1596-
1599.

Tamura, T., Hayakawa, M. & Hatano, K. (1998). A new genus of the order
Actinomycetales, Cryptosporangium gen. nov., with descriptions of
Cryptosporangium arvum sp. nov. and Cryptosporangium japonicum sp. nov. ——

Int. J. Syst. Bacteriol. 48, 995-1005.

Temeles, E. J. (1994). The role of neighbors in territorial systems - when are they dear
enemies? — Anim. Behav. 47, 339-350.

Theis, K. R., Heckla, A. L., Verge, J. R. & Holekamp, K. E. (2008). The ontogeny of
pasting behavior in free-living spotted hyenas, Crocuta crocuta. — In: Chemical
Signals in Vertebrates 11 (Hurst, J. L., Beynon, R. J ., Roberts, S. C. & Wyatt, T.
D., eds). Springer, New York, p. 179-188.

Theis, K. R., Heckla, A. L., Verge, J. R., Smashey, J. D. & Holekamp, K. E. (submitted).
Intragroup functions of scent-marking by adult spotted hyenas. — Behaviour.

Thiessen, D. & Rice, M. (1976). Mammalian scent gland marking and social behavior. —
Psychol. Bull. 83, 505-539.

Thom, M. D., Stockley, P., Beynon, R. J. & Hurst, J. L. (2008). Scent, mate choice and
genetic heterozygocity. ——- In: Chemical Signals in Vertebrates 11 (Hurst, J. L.,
Beynon, R. J ., Roberts, S. C. & Wyatt, T. D., eds). Springer, New York, p. 291-
301.

Thomas, S. A. & Kaczmarek, B. K. (2002). Scent-marking behaviour by male prairie
voles, Microtus ochrogaster, in response to the scent of opposite- and same-sex
conspeciﬁcs. — Behav. Process. 60, 27-33.

Trinkel, M., Fleischmann, P. H. & Kastberger, G. (2007). Comparison of land-use
strategies of spotted hyenas (Crocuta crocuta, Erxleben) in different ecosystems.
— Afr. J. Ecol. 44, 537-539.

Van Horn, R. C., Engh, A. L., Scribner, K. T., Funk, S. M. & Holekamp, K. E. (2004).
Behavioural structuring of relatedness in the Spotted hyena (Crocuta crocuta)
suggests direct ﬁtness beneﬁts of clan-level cooperation. — Mol. Ecol. 13, 449-
458.

Van Horn, R. C., Mcelhinny, T. L. & Holekamp, K. E. (2003). Age estimation and
dispersal in the spotted hyena (Crocuta crocuta). — J. Mammal. 84, 1019-1030.

Van Horn, R. C., Watts, H. E. & Holekamp, K. E. (2008). Do female hyaenas choose
mates based on tenure? — Nature 454, E1.

176

Vander Meer, R. K., Saliwanchik, D. & Lavine, B. (1989). Temporal changes in colony
cuticular hydrocarbon patterns of Solenopsis invicta. — J. Chem. Ecol. 15, 2115-
2125.

Vanpe, C., Gaillard, J. M., Kjellander, P., Mysterud, A., Magnien, P., Delorrne, D., Van
Laere, G., Klein, F., Liberg, O. & Hewison, A. J. M. (2007). Antler size provides
an honest signal of male phenotypic quality in roe deer. — Am. Nat. 169, 481-
493.

Velando, A., Beamonte-Barrientos, R. & Torres, R. (2006). Pigment-based Skin colour in
the blue-footed booby: an honest signal of current condition used by females to
adjust reproductive investment. —- Oecologia 149, 535-542.

Villasenor, E. & Drummond, H. (2007). Honest begging in the blue-footed booby:
signaling food deprivation and body condition. —— Behav. Ecol. Sociobiol. 61,
1133-1142.

Voigt, C. C., Caspers, B. & Speck, S. (2005). Bats, bacteria, and bat smell: sex-speciﬁc
diversity of microbes in a sexually selected scent organ. — J. Mammal. 86, 745-
749.

Von Graevenitz, A. & Bernard, K. (2006). The Genus Corynebacterium - Medical. ——
Prokaryotes 3, 819-842.

White, A. M., Swaisgood, R. R. & Zhang, H. (2003). Chemical communication in the
giant panda (Ailuropoda melanoleuca): The role of age in the signaller and
assessor. — J. Zool. 259, 171-178.

White, A. M., Swaisgood, R. R. & Zhang, H. (2004). Urinary chemosignals in giant
pandas (A iluropoda melanoleuca): seasonal and developmental effects on signal
discrimination. — J. 2001. 264, 231-238.

White, P. A. (2007). Costs and strategies of communal den use vary by rank for spotted
hyaenas, Crocuta crocuta. —— Anim. Behav. 73, 149-156.

Willis, C. K. R. & Brigham, R. M. (2004). Roost switching, roost sharing and social
cohesion: forest-dwelling big brown bats, Eptesicusfuscus, comforrn to the
ﬁssion-fusion model. — Anim. Behav. 68, 495-505.

Wilson, E. O. & Bossert, W. H. (1963). Chemical communication among animals. ——
Recent Prog. Horm. Res. 19, 673-716.

Wittemyer, G. & Getz, W. M. (2007). Hierarchical dominance structure and social

organization in African elephants, Loxodonta africana. — Anim. Behav. 73, 671-
681.

177

Wolff, J. O., Mech, S. G. & Thomas, S. A. (2002). Scent marking in female prairie voles:
a test of alternative hypotheses. — Ethology 108, 483-494.

Woodmansee, K. B., Zabel, C. J., Glickman, S. E., Frank, L. G. & Keppel, G. (1991).
Scent marking (pasting) in a colony of immature spotted hyenas (Crocuta
crocuta): a developmental study. — J. Comp. Psychol. 105, 10-14.

Wyatt, T. D. (2003). Pheromones and Animal Behaviour — Cambridge University Press,
Cambridge.

Wynne-Edwards, V. C. (1962). Animal Dispersion in Relation to Social Behaviour. —
Oliver and Boyd, Edinburgh.

Xu, Y., Dixon, S. J ., Brereton, R. G., Soini, H. A., Novotny, M. V., Trebesius, K.,
Bergmaier, I., Oberzaucher, E., Grammer, K. & Penn, D. J. (2007). Comparison
of human axillary odour proﬁles obtained by gas chromatography/mass
spectrometry and Skin microbial proﬁles obtained by denaturing gradient gel
electrophoresis using multivariate pattern recognition. — Metabolomics 3, 427-
437.

Yahr, P. (1983). Hormonal inﬂuences on territorial marking behavior. — In: Hormones
and Aggressive Behavior (Svare, B. 8., ed). Springer, New York, p. 145-175.

Zala, S. M., Potts, W. K. & Penn, D. J. (2004). Scent-marking displays provide honest
signals of health and infection. — Behav. Ecol. 15, 338-344.

Zhang, G., Swaisgood, R. R. & Zhang, H. (2004). Evaluation of behavioral factors

inﬂuencing reproductive success and failure in captive giant pandas. — Zoo Biol.
23, 15-31.

178