..

3-

,41‘r.1 ‘-

1C:-

.Lm,'v4.

 

4‘,
O

:‘x

" YIt";

$9.27»!

{(11.5

It
,4 "
t.9:

..
3'

‘33 i

 

 

1HEEHS
/
(70}!

5981952’7’4

This is to certify that the

thesis entitled

Long distance vocal communication in the spotted
hyena, (Crocuta crocuta)

 

presented by

Keron M. Greene

has been accepted towards fulfillment
of the requirements for

Master's degree in Zoologz

Major profe

Date November 7, 2002

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

' v

vv—‘f

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DaioDuo.p65-p.15

 

 

Long-distanc

 

Long-distance vocal communication in the spotted hyena, Crocuta crocuta

By

Keron M. Greene

A THESIS

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

MASTER OF SCIENCE
Department of Zoology
2002

 

 

 

LONG-l lSll

Spoted
individuals na}
allies. The abii
advertise tI eir
several kilc net
the spotted hye
theacousti pr:

in these ac lust

 

hypotheses 3U

 

ABSTRACT

LONG-DISTANCE VOCAL COMMUNICATION IN THE SPOTTED HYENA,
CROCUTA CROCUTA

By

Keron M. Greene

Spotted hyenas (Crocuta crocuta) live in fission-fusion societies in which '
individuals may frequently exist at variable distances from close kin and other
allies. The ability to produce long-distance vocalizations enables hyenas to
advertise their location and call for assistance from individuals who may be up to
several kilometers away. Here I examined long-distance vocal communication in
the spotted hyena in order to elucidate sources of variation among individuals in
the acoustic properties of the long distance call, and determine whether variation
in these acoustic parameters influences listener response. I tested predictions of
hypotheses suggesting that the use of the whoop vocalization, variation in the
properties of its sound, and the response of conspecifics to this vocalization are
affected by sex, age, and the context in which the call is produced. Hyenas
generally whooped either spontaneously or during social excitement. Adult
females and cubs usually whooped during social excitement, while adult males
most often whooped spontaneously and were less likely to receive a response to
their calls. Most of the observed variation in acoustic parameters was associated
with the context in which the vocalization was emitted, and calls with different

acoustic parameters received differential responses from conspecifics.

 

 

T.

I ti
research.
the Senior

I an
Departmen
was suppor.
Many
Hyena Projei
Stephanie DIt
lsla Graham, '
NUit. Gabe Or
Kim Weibel, er
I aDprec
members. of the
Anne Engh, Ka

Marja MUECke

 

Sofia Wahal. a

Constant Encou

 

ACKNOWLEDGEMENTS

The efforts and assistance of many people made this work possible.

I thank the Office of the President of Kenya for permission to conduct this
research. I also thank the Kenya Wildlife Service, the Narok County Council, and
the Senior Warden of the Masai Mara National Reserve for their cooperation.

I am thankful for funding and logistical support provided by the
Department of Zoology and the Zoology office staff. The Mara Hyena Project
was supported by NSF grant IBN9906445.

Many people have contributed to data collection and support for the Mara
Hyena Project. I would like to thank Nancy Berry, Erin Boydston, Susan Cooper,
Stephanie Dloniak, Martin Durham, Anne Engh, Josh Friedman, Paula Garrett,
lsla Graham, Tyson Harty, Cathy Katona, Toni Lyn Morelli, Keith Nelson, Karen
Nutt, Gabe Ording, Laura Sams, Micaela Sykman, Russ Van Horn, Sofia Wahaj,
Kim Weibel, and Brad White for all their hard work in the field.

I appreciate the help and support provided by the past and present
members of the Holekamp lab. Pat Bills, Erin Boydston, Stephanie Dloniak,
Anne Engh, Karen Hubbard, Joe Kolowski, Sarah Lansing, Terri McElhinny, Eva-
Maria Muecke, Scott Nunes, Micaela Szykman, Jaime Tanner, Russ Van Horn,
Sofia Wahaj, and Heather Watts provided helpful suggestions and discussions,

constant encouragement, and invaluable advice.

iii

 

 

l the
smale for'
extend SDE
suggestion
Holekamp
immeasura
for being rn
graduate w:

Man)
developmen
Schaupp, Cl

and sanity, a
have Constar

and have bee

I thank my guidance committee Tom Getty, Kay Holekamp, and Laura
Smale for their support at all stages and insightful comments on earlier drafts. I
extend special thanks to Laura Smale for her role in data collection, for helpful
suggestions, and encouragement. I am deeply grateful to my advisor Kay
Holekamp for allowing me to work with her and the hyenas. She provided
immeasurable guidance, support, and scientific leadership. I wish to thank her
for being my mentor, and all the encouragement she provided throughout my
graduate work.

Many other people have provided friendship and support throughout the
development of this work. In particular, Steve Danglay, Sarah Lansing, Jen
Schaupp, Chris Varas, and Brad White never failed to help me keep perspective
and sanity, and remind me to laugh. My parents Robert and Pamela Greene
have constantly encouraged me to pursue my interests wherever they take me,

and have been a great source of love and support throughout graduate school.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................ vi

LIST OF FIGURES ...................................................................... vii

CHAPTER 1

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

CHAPTER 2

LONG DISTANCE VOCAL COMMUNICATION

IN THE SPOTTED HYENA .......................................................... 8
Introduction ...................................................................... 8
Methods .......................................................................... 19
Results ........................................................................... 27
Discussion ....................................................................... 48

CHAPTER 3

ESTIMATING TERRITORIAL BOUNDARIES OF SPOTTED HYENA CLANS

USING PLAYBACKS .................................................................. 57
Introduction ..................................................................... 57
Methods ......................................................................... 59
Results ........................................................................... 63
Discussion ...................................................................... 67

LITERATURE CITED ................................................................. 71

 

TABLE 2.1.
physical ChE
After Kruuk i

TABLE 3.1.

coordinates i
broadcast. /
hyenas appn

TABLE 3.2. ;
hyenas. The
map home ra

 

LIST OF TABLES

TABLE 2.1. Descriptions of spotted hyena vocalizations, their associated
physical characteristics, and the contexts in which they are typically observed.
After Kruuk (1972) ........................................................................... 10

TABLE 3.1. Summary of results from call-ins that attracted hyenas. UTM
coordinates refer to the exact geographic location of the vehicle during
broadcast. Arrival direction refers to the general compass direction from which
hyenas approached the vehicle .......................................................... 65

TABLE 3.2. Summary of results from studies using playbacks to attract spotted
hyenas. The present study represents the only attempt to date to use call-ins to
map home ranges; all others used playbacks to census populations .......... 68

 

 

FIGURE 2.‘
and (b) “fas
of (b) is 39.:
MIKE, and I

FIGURE 2.2
whoops. Th
cubs (top), a
were emitter
CUJ; adult n

FIGURE 2.3.
of a whoop.

while standin
(bottom) ......

FIGURE 2.4.
measures obi
intervals were
bout was emit

FIGURE 2.5.

FIGURE 2.5,
gmUp- Adult n
over bars repre

 

LIST OF FIGURES

FIGURE 2.1. Spectrograms illustrating examples of Kruuk’s (1972) (a) “slow”
and (b) “fast” whoop bouts. The whoop rate of (a) is 10.8 whoops/min, while that
of (b) is 39.2 whoops/min. These whoops were produced by (a) adult male
MIKE, and (b) female cub MALI .............................................................. 13

FIGURE 2.2. Spectrograms illustrating Type A (a), Type S (b), and Type T (c)
whoops. Three examples of each whoop type are shown, with samples from
cubs (top), adult males (middle), and adult females (bottom). These whoops
were emitted by nine different individuals. (across from top left: cubs DNA, ALI,
CUJ; adult males RCN, FA, MIKE; adult females JAB, NAV, GER) ................ 15

FIGURE 2.3. Two of the postures assumed by spotted hyenas at the beginning
of a whoop. (top) An adult male whoops while walking. (bottom) A cub whoops
while standing still. Photos courtesy of Anne L. Engh (top) and Kay E. Holekamp
(bottom) ............................................................................................ 18

FIGURE 2.4. Spectrogram of a spotted hyena whoop bout showing four of the
measures obtained. The durations of individual whoops and inter-whoop
intervals were measured and averaged to give mean values for the bout. This
bout was emitted by male cub DYL ......................................................... 26

FIGURE 2.5. Fundamental frequencies of spotted hyena whoops. Most adult
males were excluded since their exact ages were unknown. However, samples
from a few adult natal males who had not yet dispersed were included ........... 28

FIGURE 2.6. Mean (1 SE) fundamental frequencies for hyenas in each age/sex
group. Adult males included both adult natal and immigrant males. Numbers
over bars represent the number of individuals sampled in each age/sex class. 30

FIGURE 2.7. Whoop distributions for each context by sex and age class.
Numbers over bars refer to the number of individuals emitting whoops in the
indicated context ................................................................................. 32

FIGURE 2.8. Frequency distribution for the whoop rates of all sampled bouts
(n=131 ) .............................................................................................. 33

FIGURE 2.9. Relationship between whoop rate and (a) the duration of individual
whoops within the bout, or (b) the duration of inter-whoop intervals. Higher
whoop rates are produced by shortening both whoop durations and inter-whoop
Intervals .................................. , ................ 35

vii

FIGURE 2.10. Relationship between context and whoop rate (mean :t SE) for
adults and cubs. Numbers over bars represent the number of individuals
observed in a given context ................................................................... 37

FIGURE 2.11. Duration (mean. :I: SE) of individual whoops in bouts produced by
adults and cubs in different contexts. Bar labels as in Figure 2.10 ................. 38

FIGURE 2.12. Mean (:l: SE) lengths of inter-whoop intervals in bouts produced
by adults and cubs. Bar labels are as in Figure 2.10 .................................. 40

FIGURE 2.13. Mean (:1: SE) duration of bouts produced by adults and cubs.
Numbers over bars represent the number of individuals emitting whoops in each
context .............................................................................................. 41

FIGURE 2.14. Percent of whoops emitted in each context that received a
response. Numbers over bars refer to the number of individuals receiving a
response after whooping in the indicated context ........................................ 43

FIGURE 2.15. Percent of whoops emitted by each sex/age class that received a
response. Numbers over bars refer to the number of animals receiving a
response ........................................................................................... 45

FIGURE 2.163. Relationship between whoop rate (mean :1: SE) and response.
Numbers over bars represent the number of individuals receiving the indicated
response ........................................................................................... 46

FIGURE 2.16b. Relationship between duration of inter-whoop interval (mean t
SE) and response. Bar labels are as in Figure 2.16a .................................. 47

FIGURE 3.1. Map of the Masai Mara National Reserve showing 62 potential call-
in sites. The shaded area represents the Talek clan home range as defined by

Boydston (2001 ). Dotted lines represent the previously approximated boundaries
of neighboring clans. After Boydston (2001) ............................................. 60

FIGURE 3.2. Speakers were place on platforms mounted in the windows of the
vehicle. Due to the curvature of the speaker horns, they did not need to be
turned to ensure multi-directional sound broadcast. Photo courtesy of Sofia
Wahaj ............................................................................................... 62

FIGURE 3.3. Map of Masai Mara National Reserve showing the locations of call-
ins performed May-August 2000. Sites represented with stars are locations
where territorial behaviors were observed ................................................ 64

Images in this thesis are presented in color.

viii

Chapter 1

GENERAL INTRODUCTION

Communication is generally defined as the provision of information by a
sender that can be utilized by a receiver to make a decision (Bradbury &
Vehrencamp, 1998). A communicatory signal occurs because it benefits the
sender to produce it. The incidental transmission of information resulting from
other activities (i.e. feeding, moving, etc.) is termed a “cue”. Signals may encode .
information about attributes of the sender, such as its motivational state,
behavior, or identity, and about stimuli or events in the environment, such as
location, quality, or quantity of food (Marler, 1967). Recognizing and maintaining
contact with conspecifics is challenging for animals that are often separated in
such a way that visual contact is no longer possible. Although other
communicatory media, such as olfaction, may also be useful for long-distance
communication, in most terrestrial environments sound waves are transmitted
more quickly and with less signal degradation than are odors (Bradbury &
Vehrencamp, 1998). Thus, when the transmission of signals is constrained by

distance, individuals can still communicate efficiently through vocalizations.

Vocal signals can be described in terms of both their acoustic properties
and their behavioral correlates. Acoustic properties can be defined in the
frequency dimension by measuring sound frequency (pitch), amplitude, and
harmonic structure (formants). The temporal structure of a vocalization includes
frequency change over time, repetition of sound components (syllables), and the
duration of the call. Sound can be graphically represented through plots of the
change in amplitude over time (power spectrum), or frequency over time
(Spectrograms). Modern technological advances have made digital
Spectrograms and power spectra efficient methods of sound analysis (Bradbury
& Vehrencamp, 1998). Observations of the behavior of animals as they produce
sound, the stimuli that elicit the vocalization, and the response of receivers to the
sound can offer insight into the functions of calls (Marler, 1967; Hauser, 1996).

Considerable attention has been devoted to understanding how natural
selection has shaped the structure and form of animal vocal communication (e.g.
Marten & Marler, 1977; Marten et al., 1977; Wiley & Richards, 1978; Waser &
Brown, 1986), and various theories relating signal structure to the functions of
vocalizations and their evolutionary origins have been proposed (Morton, 1977;
Zahavi, 1979). Signals that contain only information about the sender’s internal
state are considered “motivational” (Marler, 1967). If signals provide receivers
with enough information to determine the context in which the signal is produced,
the signals are regarded as “referential” (Hauser, 1996). Referential signals
permit listeners to predict events in their own environment. Functionally

referential signals have been documented for predator calls (e.g. Seyfarth et al.,

1980a; Macedonia, 1990; Evans et al., 1993) and food calls (e.g. Evans & Evans,
1999; Bugnyar et al., 2001; and review in Hauser, 1996).

In solitary species, long-distance communication might facilitate
successful contacts between conspecifics during the breeding season,
particularly if the information content of the message includes details regarding
the sex and reproductive status of the caller. At relatively close range, brief
encounters might also be facilitated if the message content of the sounds was
sex— or individual- specific (Baker, 1998).

Cohen and Fox (1976) and Schassburger (1993) suggested that
increasing complexity of the vocal repertoire may be indicative of social
complexity, with gregarious animals possessing a larger vocal repertoire than
solitary species. Short distance communication is important in many social
species, and can help facilitate group movements, maintain social bonds, and
provide a mechanism for individual recognition (e.g. Japanese macaques,
Macaca fuscata: Green, 1975; vervet monkeys, Cercopithecus aethiops: Cheney
& Seyfarth, 1982; ring-tailed lemur, Lemur catta: Macedonia, 1986; king
penguins, Aptenodytes patagonicus: Robisson, 1992; gorilla, Gorilla gon'lla:
Seyfarth et al., 1994; Palombit et al., 1999; baboons, Papio cynocephalus:
Rendall et al., 1999; Fischer et al., 2001; European badger, Meles meles: Wong
et al., 1999; white-nosed coati, Nasua nan'ca: Compton et al., 2001 ). Loud, long
distance vocalizations often serve a number of different functions simultaneously.
In many social mammals, these calls are used for territorial defense, mate

attraction, and to maintain contact with widely spaced social companions (e.g.

wolves, Canis lupus: Harrington & Mech, 1979; chimpanzees, Pan troglodytes:
Mitani & Nishida, 1993; lions, Panthera Ieo: McComb et al., 1994; Grinnell &
McComb, 2001).

Here I investigate the functions of, and sources of variation in, the loud
calls of one gregarious carnivore, the spotted hyena (Crocuta crocuta). Spotted
hyenas are ideal for studies of vocal communication because they are highly
vocal animals. They have the largest repertoire of any hyaenid (Kruuk, 1972,
1976; Mills, 1990; Peters & Sliwa, 1997) and a more complex vocal repertoire
than most other social carnivores (i.e. wolves: Harrington & Mech, 1979;
McComb et al., 1993; Schassburger, 1993; lions: Grinnell & McComb, 2001; but
see African wild dogs, Lycaon pictus: Robbins, 2000). Spotted hyenas live in
permanent groups called “clans”, which defend communal territories (Kruuk,
1972; Mills, 1990). A single clan contains multiple adult males and multiple
matrilines of adult females and their offspring, who are raised at a communal den
(Kruuk, 1972; Frank, 1986a; Mills, 1990). Male spotted hyenas disperse from
their natal clans when they are around two years old, while females are generally
philopatric (Mills, 1990; Smale et al., 1997). Each spotted hyena clan is
organized on the basis of a rigid social hierarchy that influences most hyena
social interactions, from feeding and hunting, to greeting ceremonies and sexual
behavior (Frank, 1986b; Holekamp & Smale, 1991; Szykman et al., 2001; Engh
et al., 2002). Hyenas live in fission-fusion societies in which close kin or other
important social allies may not be present in an individual’s immediate vicinity.

The ability to produce long-distance vocalizations enables hyenas to advertise

their location and call for assistance from individuals who may be several
kilometers away. The fluid nature of hyena social groups and the complexity of
the social environment provide ideal conditions for the development of a complex

long-distance communication system.

Chapter overview

Chapter 2 begins with a review of the current literature on spotted hyena
vocalizations, and a description of the vocal repertoire. I then investigated the
sources of variation in spotted hyena whoop vocalizations, and asked whether
the social context in which whoops are emitted affects the structure of the
vocalization produced. I also asked whether whoops with different acoustic
properties elicited different responses from listening conspecifics. This level of
vocal complexity has not been examined in spotted hyenas, and provides
necessary insight into the role of long distance communication in this species.

In Chapter 3, I examined the possibility of exploiting long-distance
vocalizations to determine clan territorial boundaries by using playback
techniques (i.e., call-ins) to attract animals to a particular location and monitor
them for territorial behaviors. Call-ins have been used successfully to determine
the size of hyena populations (Kruuk, 1972; Creel & Creel, 1996; Mills, 1996;
Ogutu & Dublin, 1998; Mills et al., 2001), also might potentially offer an efficient
method of mapping clan divisions within a geographical area.

Data in Chapter 2 were collected from a single spotted hyena clan, the

Talek clan, which defends a territory on the northeastern edge of Kenya’s Masai

Mara National Reserve. The Talek clan has been studied intensively since 1988,
and Dr. Kay Holekamp, Dr. Laura Smale, their field assistants, and graduate
students working on the Mara Hyena Project, collected much of the data
presented here. The audio recordings and field data analyzed here are the result
of years of careful observation by many people. The data presented in Chapter 3
are the result of a series of field experiments I conducted between May and
August 2000, which could not have been completed without the researchers on
the Mara Hyena Project. Erin Boydston provided GIS maps and expertise
essential in documenting hyena responses to the call-ins. Therefore, I use the
term “we” throughout each chapter to indicate that my thesis research was a
collaborative effort.

Each of these chapters examines long-distance communication in the
lives of spotted hyenas. In Chapter 2, I found that much of the variation
observed in the acoustic structure of spotted hyena whoops results from the
influence of the caller’s social environment at the time of the whoop. I also found
a difference in the likelihood of hyenas responding to a whoop based on the
identity of the caller and the context in which the caller vocalized. These data
suggest that spotted hyena whoops are complex signals, which contain extensive
information that can be used by receivers to determine their responses, and
provide a mechanism for maintaining group cohesion over long distances. My
results also suggest a' relationship between whoops and individual vocal

recognition in this species. The results of Chapter 3 indicate that while call-ins

may someday prove to be a useful tool for censusing hyena populations and

determining their space-use patterns, the technique needs refinement.

Chapter 2
LONG DISTANCE VOCAL COMMUNICATION IN THE

SPOTTED HYENA

Introduction

Recent research on animal vocal communication has focused on
determining the functions of calls by elucidating the information they convey to
recipients. Most researchers judge the meaning of calls based on the responses
they elicit from listeners (e.g. Snowdon et al., 1983; Seyfarth et al., 1994; Rendall
et al., 1999). It has been argued that animal signals are both motivational and
referential, containing information about both the internal state of the caller and
the external environment (e.g. Marler et al., 1992). Minor modifications of a
single call type can result in different messages without the need to develop new
call production and reception mechanisms (Cleveland & Snowdon, 1982). Call
subtypes in which different meanings are conveyed by acoustic variation are also
advantageous since they allow recipients to decipher signal meaning without
using visual cues. In many species of nonhuman primates (e.g. Seyfarth et al.,
1980a; Cheney & Seyfarth, 1982; Fischer, 1998), vocalizations have been
described that elicit different responses depending on the call subtype produced,
suggesting that these calls might have meanings associated with their acoustic

structure. In some species, the structure of a particular type of call can be

affected by the circumstances under which it is emitted (Norcross & Newman,
1993; Norcross et al., 1999; Rendall et al., 1999).

While variation in vocalizations and their association with behavioral
contexts has been extensively studied in primates (e.g. Green, 1975; Hauser,
1991; Norcross 8. Newman, 1993; Hammerschmidt & Todt, 1995), relatively little
research of this nature has been done on other animals (Brown & Farabaugh,
1997; Jennings et al., 1997; McCowan & Reiss, 2001; but see lnsley, 1992). The
present study documents variation in acoustic structure and the role of call
context in the long-distance vocalization of one gregarious carnivore, the spotted
hyena (Crocuta crocuta). The spotted hyena is a highly vocal species that
produces a rich repertoire of sounds. These animals live in a fission-fusion
society in which an elaborate vocal communication system appears to allow
group members to maintain social bonds, both within the immediate vicinity and
also with conspecifics out of visual range (Kruuk, 1972). Kruuk (1972) identified
eleven vocalizations in this species, and briefly described the circumstances
under which each call was emitted (Table 2.1).

Spotted hyenas live in social groups (called “clans”) structured by rigid,
linear dominance hierarchies (Kruuk, 1972; Tilson & Hamilton, 1984; Frank,
1986b; Mills, 1990). Although the hierarchical dominance relationships among
Crocuta clan members closely resemble those characteristic of many old-world
primate societies, they are unique among social carnivores (Ewer, 1973;
Holekamp et al., 2000). Kruuk (1972) suggested that most spotted hyena

vocalizations contained either submissive or aggressive messages, and assigned

QC 0.. C x332! E0: BESUDE Emacmac
2.50%»: 03.. >02. 23.5)) C. maxoaioo 3:. 32...... ...u.C:m:€.U.w:...:U _$U.a>£Q U92233I¢3 EDS. .mCC:=N:SUO> 52:): 3.2.9:}... LC th:.:>:t;: h .N D‘QBF

 

 

 

 

c.0000 0.0.0088. 0. 000.
o. 29... 000. .3 .0802, .8. 5.2, 0<

..0.00~..00 0.020: .2. 26.
.050. 0000 .0000 0.09.0 0.00.).

.0000000 .0.0>00
@0000. 000000-26. 0.030: ...00 00.50
00.09 02.0 0..) 00:00 00...-260

26..

 

.0030

005.00 0:03 .0 .000 00. .0 .00 00:0

..00 0. 00.0E0. .6 $0080.00 000005
000:0 000 0.0.00 .00.. 00...Em.

00:00.0
0. 00.00000 0:02, 0000 .0000
0.09.0 .0 0000.0 0.:0E .26. 000...

.000 m. 0. 0: .0. >_.00...E.0.0.
00000. .00000 -EEE- 00000026..

000.0

 

00.005 >0 0026.6. 0000 .0000.
.00600 .0 0000.000 00 00...Em

0.0.000 02000.30 0000.0 0.30.2

.0000000 .0.0>00 05.00.
00:00 00:26.0. 000000-26. 00> .com

.026

 

000.000. 0.0000 0000>0 .0 000.0
0 0.0.00 .00.. 0000 00.0 .006 .0 00..
5.2, =0. .06 0000.02.08 0. 0000
..0000.0 000020 .050 >00E 55>

0000.0 .000 0600
00 0.30.0 60.00:. 0000 000000
0.00 .090 .0 .0.00~..00 =0...

.50 0000.00. .00
000000 0.0.00.0. 000 0:00 .0600 0<

00003 .00“.

 

00:00
.0E0.x0 505.2, 0300002600 00
0. .00000 0000 000003 0.00 0.00.0
0000 102,000.. 00.0w. .>000E00 0.
.0 000.0 .0000 .0000 0.00 >0 00...Em.

0000.0
.000 0_~N:E 5.2, 5009.0 000000
0.:0E 00000 0. :00 0000 .0“.

.00.... .0 65000.0 60.0.02, 00...Em.

2.02.0 Exm 000. 0.90
0.000 00 >0E .000. 00> .000 000000
-26. 00. 000 0. 000300. 0000 00000
.0 :00 .00.. 26.09026. 00 0.0
0:00 0E00 .090 00.000 26. 00.00500
:00 0000 .000. -000. 0:00 .0 0000.0

 

0000>>

 

 

 

02.2.00

 

0.3000

 

00:00 .0 00..0..000n_

 

02.05.0002
.0 0502

 

 

.03: .320. ES. 000.00.). 002020
0.00.0... 0.0 >00. 00.0.5 0. 000.000 00. 000 02.000.00.000 .00.0>00 000.0800 ..00. .000..0~..000> 000>0 000000 .0 0000000000 «.N 0.00.0

10

 

 

.

 

 

 

 

 

02.0.0000 000. .000
.000.>.00. 0260..-..02. 0 00..00E
0002, 00000 00.0.00 .0000 >0

00.0 00 .06 000.0. 0000
00.00 500. 2.09.0 00.... 0000
0000 00.. 0.00 .0000 0.09.0 0.00....

.0000000 .0.0>00 00000.
00000.0 00 .000 .00 000.000 0<

.00000 000

 

0.... 0 E0... 000. .00 0. 0.00.000
0. 000020. 0002, .0 .00....000 0.0.00
0.0.00. 0 00.2.0.6. 0002. 0000 >0

.26. .00.0. 0000 .0260
00.0000 _.0. .0000 0.09.0 0.00.2

0000.0 0000 002,

00.00.E .0. 0000000 >05. .0.00. >.0>
0-00-00.00.00; .00E0_0 90000.0
0 0...; 0000 .000000 -000- .0 0.00000
.00-02.0.0 000. 000000-090 .000..

00.000

 

000.00 .0 00..
0 .0 0000.000 00. .0 .0 0.... .06
000.. 000 0000>0 0002000 .0 000.0
0002000 00.000000 0. 0.... 0 00
.0 000 00. .0 0002, 0950.0 0002.

000.000

0.00 .090 .0 .0.00N..00 _.0. 5.2,
05000.0 .0 .0. 0.0.000 00.00.. 0.
.0. .0000 0.09.0 .0 0000.0 0.00.2

.0000000 .0.0>00
00000. 00000 0 0. 0.00.0 20000.0
000 000000-26. .0 020000000 0.00m.

0.0E0. E.0.<

 

0.5 0 00.0.0.0. >..00.00 0002,
.00000. .0000 000 0. .0000 0002,
00.00 000.0 000 0000000 0002.

0.0.000 02000.00

.2 c. 2.08 052.0.
0 5.2, 00.00 .0000. -00- 000 -000
- 002, .0000000 .0.0>00 0. 00 05.00.
02.0 26. .00 00.>.0> 5.2, .000-.

.265

 

0.090 5.2, 000. 00.. 0. >000000.
0009.0 000.0 00.00 0.0300
0003 0.0000 0.090 5.2, 0<

0.0.000
02000.00 .0 .0.090 002, 0<

.0.00 0. 0000000. 0. _.0> 000.00
0 0.... .0000000 20.. 0 0. 00 00000.
.02... 00... .2. 02.20., 5.; .23

__0>

 

 

0.0. 0 .06
0000 .000000 .0 000.0000 000.5

00.0000 0000 .0000 0.09.0 0.00.2
0.0.000 00.00.. 0. >020 00.000”.

20.0000. 000...
00E00 .0 000 00.00.E0m .000 m 000.
00000 0.0000 .000-000-000 .0 00000
0.00. 000 000000-090 >.0> 000..

000.0

 

 

00003.0

 

0. 0.000

 

00000 .0 02.0.6000

 

02.05.0002
.0 0.002

 

 

11

possible functions to each call. The loud calls of the spotted hyena (yells,
giggles, whoops) are the most well known and recognizable. Yells and giggles
are usually given after an individual has been attacked and are generally
submissive calls, but the function of whoops and their eliciting stimuli are poorly
understood (Kruuk, 1972; Mills, 1990).

The spotted hyena whoop is a high-amplitude call with fundamental
frequencies ranging between 200 and 375 Hz (East 8. Hofer, 1991a). Multiple
discrete calls are produced in bouts of sound that can travel distances of five
kilometers or more (Kruuk, 1972). Kruuk (1972) identified two types of whoops,
slow and fast, which he distinguished based on the rate at which individual calls
were produced within a bout (Figure 2.1). East and Hofer (1991a) provided the
first systematic technical description of this vocalization, documenting its acoustic
structure, structural variation in whoops among individuals, and ontogenetic
changes in whoop structure within individuals. They identified three distinct types
of whoops, designated “A", “S”, and “T”, which differ from one another in
structure and sound. In high quality recordings, these types are easily
distinguished acoustically and by visual inspection of Spectrograms (Figure 2.2).
Type A (asymmetrical) whoops contain an initial low frequency section followed
by an abrupt rise. Type S (symmetrical) whoops begin with a low frequency
section, rise in pitch, and then return to the initial low frequency. Type T
(terminal) whoops are lowing sounds characterized by little change in frequency,
and they are always found at the end of a bout. Bouts typically contain

' combinations of all three types, however longer bouts tend to contain higher

12

Figure 2.1. Spectrograms illustrating examples of Kruuk’s (1972) (a) “slow” and
(b) “fast” whoop bouts. The whoop rate of (a) is 10.8 whoops/min, while that of
(b) is 39.2 whoops/min. These whoops were produced by (a) adult male MIKE,
and (b) female cub MALI.

13

 

 

0.

a. .

 

0% 6.3%»... r... Waswwwfl .

7;

r...-

 

w.. t.

S.

m. N.

pa... .. .0...

v“... 16 .0

I. or 0

’..¥SVXMu.n ..... ‘0. r.
0....«w‘mflv2 .......u

1...... ..

.. ;
.. 150-... A

  

.qu

. v

m m

.01 .0. 0......
'.W.”.uvru 6...no..._ .7...

u: ...HO\‘7A -...-
a

v
”.I‘ .0) . O
L .a. i . . F ‘. ‘. . p.
.l. 0 .~ . p I

v

    

€9.00

.I .- ”Th‘u \LOM.

 

.0.

14

 

Figure 2.2 Spectrograms illustrating Type A (a), Type S (b), and Type T (c)
whoops. Three examples of each whoop type are shown, with samples from
cubs (top), adult males (middle), and adult females (bottom). These whoops
were emitted by nine different individuals. (across from top left: cubs DNA, ALI,
CUJ; adult males RCN, FA, MIKE; adult females JAB, NAV, GER)

15

(a) (b) (C)

 

 

 

kaz kHz kHz
3.5 - ‘ 3.5 - 3.5 -
3.0- .1 ‘ I 3.0- t 3.0-
2.5 - "f: 2.5 - f ' . 2.5 -

2.0 - f; 2.0 - 2.0 -

1.5- Mfr-I 1.5-
r» '

1.0- W 1‘ 1.0-

0.5- M 0.5-

15. an? N w~~
1.0- M
05.‘;gu~u~,_z;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25 25 25
kHz kHz kHz
35- 35I
30I 10-
(C)
mm 2.5 I 2.5 I
cps 2.0 I 2.0 I
AM‘ 15- 15-
10- LOI.
0-5 " , 0.5 - "
25
kHz kHZ kHZ _
35- 35- 35-
30- 10- 30-
2.5 I 2.5 I 25 I
I

2.0 - 2.0 - ’_" 2.0 -
1.5- ' 1.5- ' 1.5-
1.0- 1.0-

0.5- 0.5-

1.0- .
0.5-

 

 

 

 

 

 

 

 

   

percentages of Type A whoops. Male and female spotted hyenas of all ages
produce all three whoop types. East and Hofer (1991a) suggested that the high
frequency portion found .in whoop Types A and S may allow hyenas to determine
the direction from which the call is coming, while the low frequency portions of
may reduce attenuation and enhance long-distance sound transmission.

Spotted hyenas of both sexes begin to whoop during the first month of life
(East & Hofer, 1991 a), and continue to do so in a variety of circumstances, often
with no observable stimulus prompting them to vocalize. Hyenas can whoop
while standing, walking, loping, or lying down (Kruuk, 1972; Mills, 1989; East &
Hofer, 1991a). Typically a spotted hyena lowers its mouth toward the ground
during the initial low frequency section of a whoop (Figure 2.3), and at most
raises its head to roughly the same plane as the animal’s back as the high
frequency section is produced, regardless of its posture before whooping. The
head is never pointed skyward while the hyena whoops, as occurs during
howling by wolves (Harrington & Mech, 1979).

Sound frequencies change over the course of a single whoop, and there is
a great deal of variation among individuals with respect to temporal patterning.
However, since the basic acoustic structure of the call remains stable within
individuals over time, spotted hyenas may be able to use these variable acoustic
cues to locate and identify specific callers, and to assess the caller’s motivational
state (Waser & Waser, 1977; McGregor & Krebs, 1984; Mills, 1989; East &

Hofer, 1991a, b). Calls with more frequency modulation facilitate better

17

 

0

Figure 2.3 Two of the postures assumed by spotted hyenas at the beginning of
a whoop. (top) An adult male whooping while walking. (bottom) A cub whooping
while standing still. Photos courtesy of Anne L. Engh (top) and Kay E. Holekamp

(bottom).

localization (Bradbury & Vehrencamp, 1998) and the rate at which whoops are
produced, as well as variation in the range of frequencies, may reflect the caller's
motivational state (Kruuk, 1972; Eisenberg, 1976; Morton, 1977; Mills, 1990). If
variation in the acoustic structure of-whoops communicates what is happening to
the caller, then the sound produced ought to vary with the circumstances under
which the vocalization is emitted. Although East and Hofer (1991b) described
the situations in which hyenas whoop, and the average composition of whoop
bouts for each age and sex class, they did not describe how acoustic variation in
the vocalization relates either to the caller's immediate circumstances or to the
responses of conspecific listeners. The present study inquires whether the
context in which a whoop bout is emitted influences the structure of the whoops
within a bout, and whether the responses of conspecifics vary with the acoustic

structure of whoops.

Methods

Study animals

This study was conducted in the Talek area of the Masai Mara National
Reserve, in southwest Kenya. This is an area of open, rolling grasslands grazed
year-round by large concentrations of ungulates. The subject population was one
large Crocuta clan inhabiting a home range of approximately 65 km2 (Frank,
1986b; Boydston, 2001). All members of the Talek clan were identified by their

unique spot patterns and other conspicuous characteristics, such as ear notches,

19

and the sex of each individual was determined from the dimorphic glans
morphology of the erect phallus (Frank et al., 1990). Birth dates were assigned
to cubs by estimating their ages when first observed above ground at natal or
communal dens. Cub ages could be estimated to within seven days based on
their pelage, size, and other aspects of their appearance and behavior. In this
study, hyenas were considered juveniles until they were 24 months of age, and
older animals were classified as adults. Genealogical relationships were as
described by Holekamp et al. (1993), and mother-offspring relationships after
1993 were assigned based on regular nursing associations.

Continuous monitoring of the study clan began in 1988, and the
information used for this study was extracted from a data set of observations
made between 1988 and 1998. Observers were present in the study area at
least 23 days per month, except during April 1991, when observers were present
on only 14 days. The activities of Talek hyenas were monitored from vehicles,
which were used as mobile blinds. An observation session started when one or
more hyenas were found separated from other hyenas by at least 200m, and
ended when these animals moved out of sight (as when entering bushes) or
when observers left the area. Observations were made at natal and communal
dens, at ungulate kills, and away from both dens and kills (e.g., when animals
were traveling or resting). Following Altmann (1974), all occurrences of agonistic
interactions, affiliative behaviors, and arrivals or departures of hyenas were
recorded as critical incidents. Recorded agonistic behavior included both the

aggressive behavior of the initiating individual, and the response of its victim.

20

Contexts in which whocms were emitted

From 1995 to 1998, field researchers studying the Talek clan recorded
whoops during routine observation sessions using a Marantz PMD-22 portable
cassette recorder and a Sennheiser ME66 directional microphone. For each
whoop bout recorded, caller identity, date and time, location, social context (see
below) at the time of the whoop, and maximum sound pressure were noted. The
resulting sound library consisted of recordings of whoops produced by hyenas of
all ages, in various social contexts, and found at all observation locations. For the
majority of recorded vocalizations, detailed information was available describing
events occurring during the time periods immediately before and after each
whoop. We also documented the type and intensity of behavior displayed in
response to specific whoop bouts by all animals present, as well as the identities
of the animals involved. Additional information for whoops that had not been
tape-recorded was collected following the same procedure, using archived field
notes. In all cases, whoop bouts were only examined if there were at least 10
minutes of behavioral observations available both prior to and following the
whoop, and if there were no other vocalizations occurring during the 10 minute
intervals before and after the whoop.

We assigned all whoops emitted by Talek hyenas to one of three general
social contexts: 1) During general excitement. Whoops sometimes occurred
when the caller (and frequently the entire social group) was exhibiting a high
level of excitement. Such situations included instances when the caller behaved

aggressively toward another individual, when the caller had just been the target

21

of aggression by a conspecific, when the caller was a bystander to aggression
directed toward a third hyena, or when the caller’s group was involved in a
confrontation with lions or members of a neighboring clan. Hyenas exhibit
excitement by holding their heads up, their ears forward, and their tails elevated,
erect, and bristled (Kruuk, 1972). 2) Spontaneously. Hyenas were often
observed whooping when there was no apparent stimulus eliciting the call. For
example, hyenas whooped when they were alone or when they were with other
hyenas but no social or interspecific interactions were occurring. Individuals
also frequently whooped when approaching or leaving a group of conspecifics in
which nothing was happening to generate social excitement. Thus these whoops
were also classified as occurring “spontaneously.” 3) Following a greeting.
Spotted hyenas often engage in a greeting ceremony when they encounter
another clan member (Kruuk, 1972; East et al., 1993). Upon meeting, the two
hyenas stand head-to-tail, lift the rear leg closest to the other hyena, and sniff
one another’s genitals. Here we noted whether one member of a greeting pair
whooped immediately following the greeting. Preliminary analyses suggested
that whoops produced after greeting may share similarities to whoops produced
spontaneously and during excitement: with respect to whoop rate, whoop
duration, and inter-whoop interval they were not significantly different than either
spontaneous or excited whoops. Because greeting contexts contained
similarities to contexts involving both spontaneous whoops and those emitted

during social excitement, and because whooping following a greeting was rare

22

(n=13), we excluded whoops following greetings from all analyses involving

context.

Responses to whoops

All hyenas within a 5 km radius from the caller were considered to be
within range of the sound (Kruuk, 1972). The events we observed following a ‘
bout were assigned to one of two general categories, those in which’there was
an unambiguous reaction exhibited by one or more listeners, and those in which
there was no significant response from any observable conspecifics within
acoustic range. Behavior of conspecifics that heard whoops was variable. An
approach to within 1m of the animal that just whooped, social interaction with the
caller, or the arrival on the scene of hyenas not originally present was considered
a reaction to the vocalization. However a mere change in body or head
orientation toward the animal that just whooped was categorized together with
cases in which all listeners appeared to ignore the caller completely.

In order to determine whether the arrival of new hyenas to the scene,
and/or the approach of a hyena to the caller, could indeed be considered
responses to a whoop, we compared the rates of arrivals and approaches
observed before and after each whoop using the Wilcoxon Signed-Ranks test.
We examined archived field notes containing observations made between
January 1995 and December 2001 for all instances in which an identified hyena
emitted a whoop. We examined behaviors occurring during ten-minute intervals

before and after the beginning of the whoop bout. The identity of each

23

responding hyena was noted, as well as the nature of any interaction it had with
the caller, if any occurred. We rejected any instance in which there were less
than ten continuous minutes of observations before or after the vocalization,
when another whoop occurred during the ten minutes both before and after, and
if the “pre-whoop” period overlapped with the “post-whoop” period of a previously
occurring vocalization. We also eliminated any instance in which the caller
whooped as it was arriving to or departing from a scene. While hyenas
frequently whooped in these contexts, these instances were disregarded for the
purposes of this analysis in order to minimize any potential bias resulting from
the absence of the caller from the scene.

In total 85 instances were examined from archived field notes. We
recorded the caller's id, the context of the vocalization (excitement, greeting, or
spontaneous as previously described), the time relative to the whoop of an arrival
or approach, and the identity of the arriving or approaching animal. Approaches
included conspecifics moving to within 1 meter of either the caller or the
individuals who had been interacting with the caller (as in maternal interventions
or assistance against an aggressive individual).

Previous studies have shown that vocal recognition of maternal kin occurs
among hyenas with r-values as small as 0.125 (Holekamp et al., 1999).
Therefore, for analyses of the relationship of each responding hyena to the caller,
we defined kin as maternal relatives having an r-value of 0.125 or greater. Since
paternity information was not available for all individuals sampled here, we did

not consider paternal kin in this analysis.

24

Digital sound analysis

Recorded whoops were digitized at a sampling frequency of 12 kHz using
a 16-bit mono audio format. Spectrograms were generated using the Avisoft-
SASLab Pro software package (Specht, 2000) with a Hamming window and an
FFT length of 512. We measured six acoustic features: bout duration, number of
whoops per bout, the duration of each whoop, the duration of each inter-whoop
interval, and fundamental frequency (Figure 2.4). Bout duration was defined as
the number of seconds between the beginning of the first whoop in the series
and the end of the last whoop in the series. The number of whoops per bout was
determined by counting the discrete calls within each bout. Whoop duration (in
seconds) was measured for each call within the bout, and the inter-whoop
interval was determined by measuring the number-of seconds elapsed between
the end of one call in a bout and the beginning of the next call. The acoustic
frequency varies from the beginning to the end of a whoop, and each whoop type
(A, S, T) has an inherently different pattern of frequency change. Therefore, the
fundamental frequency was considered to be the lowest frequency at the
beginning of a whoop. We measured this frequency using digital spectrograms
(Figure 2.4). Measurements were not taken from partial recordings or those that
produced indistinct spectrograms. We calculated the whoop rate (number of
whoops per minute) within a bout, the mean duration of whoops within a single

bout, and the mean duration of inter-whoop intervals within each bout.

25

 

.._>n_ nae m_mE >9 35.5 mm;
Son £5 .53 05 .8 $29 cmoE can 3 vow—29m new 3539: 9oz, $9.95 30:3sz can 302.; .3229:

.o €03.93 9:. .uoEmEo mosmmoE 9: Co 52 9.562.». Son noon; new? .33on a .6 Emaobooaw v.“ 2:2“.

m m_. 3 2 N_. E. or m w n m m v m N _.

. L” . l '0‘.‘ . 9...... .... .b .— .4. 4”...

32.25 t 5553:...“— .

I .7 . f .. . s u .
. . c .. .. . , .
D ’0 a. J. t .l. . . r. F
. . ... . .... .....
.. ... ,,.4 .... 4. 4:4,. J ... .. v . . . u n
. ...... .. u 4 . .. h... .
... .... .. . .. .i. . . . .4 .
.. l. .. . A .. oz . .4 . .
”In . . . v .. . ... v . .
. . . ., . . . r .. ,e.
. . .. .. .. . . . 4.. ....
r V . r I
. . . .. . . . . . 4 . . . . .
. . 4 .....,.... .., .. ...
.. .p . ~ .. . s .v. . n. .. p
., .. . i.
. . . . a .
. . \ a .
.
~
. .0 . n
.
A
r
. w
I
.. n i
. . .
‘ . 4
r
_ .
. ‘

    

._.....4..... pf... .. .. 44.4.3.
.40.: .5... 4+: 49...... a:

. no
4 3
;\ 3
ed

  
 

~.~.-r«'--V".§

 

3225 Qoo:§...25 noes: .. .. . . ,. od

Son , an
NT...

26

Data analysis

A context, as well as whether or not any response occurred was recorded
for each of 112 digitized whoop bouts, as were measurements of their acoustic
parameters. Behavioral observations of 85 whoops drawn from additional
sessions in archived field notes but not associated with a sound recording were
used in analyses of context and response. Data were analyzed using Chi-
square, Mann-Whitney U, or Wilcoxon Signed-Ranks tests. Mean values were
presented :l: standard errors. Statistical significances for multiple comparison
tests were calculated using the sequential Bonferroni method (Rice, 1989), and
the resulting adjusted critical values presented with the test statistic. All other
differences between groups were considered significant when PS 0.05. All
analyses were performed using SYSTAT version 8.0 software package (SPSS,

Inc.).

Results

Variation in fundamental freguency

The fundamental frequencies of whoops from Talek hyenas ranged from
597 Hz in the youngest cubs to 128 Hz in adult females (Figure 2.5). As cubs
mature, their voices deepen steadily until at least 24 months of age (R2=0.678,
F1,71=149.226, p<0.0001). As females age, their voices deepen, reaching adult

pitch around 50 months of age. Male hyenas disperse into a new clan sometime

27

600 t 0 Female

 

 

N C;
.I, 500 1.2:
>. :1.
o
C m
s
o- 400 *0-
e a;
U- a:
a s;
'E 300 ’30 C:
Q) . .7
E . w:»_ «A, e . .
.8 (J..C2’
c 200 — ',
3 ee
LL e e e .
100 l l .1 . l I l l
O 25 5O 75 100 125 150 175
Age (mo)

Figure 2.5 Fundamental frequencies of spotted hyena whoops. Most adult
males were excluded since their exact ages were unknown. However, samples
from a few adult natal males who had not yet dispersed were included.

28

after reaching reproductive maturity at 2 years of age (Smale et al., 1997), and
therefore few of our recorded whoops were emitted by adult males of known age
(n=5). However, we were able to compare the fundamental frequencies of adult
males and females using recordings of adult immigrant males (Figure 2.6).
Among both adults and cubs, whoops produced by females had significantly
lower fundamental frequencies than whoops produced by male peers (adults:
Mann-Whitney U=142, p<0.0001; cubs: U=378, p=0.016; Figure 2.6). There was
no effect of context on fundamental frequency for any sex/age group (adult
female: U=22.5, p=0.145; adult males: U=48, p=0.769; female cubs: U=13.5,

p=0.202; male cubs: U=97, p=0.824).

Contexts in which whoops Occurred

Behavioral data and clear digital spectrograms were available for 131
whoop bouts representing samples from 62 individual hyenas (mean = 2.113 t
0.25 bouts per individual). The context in which the whoop bout was emitted
could not be determined with certainty in 19 cases. Data from these 19
recordings were used only in analyses that did not require contextual information.
For the remaining 112 recordings, average values for all 5 acoustic parameters
(number of whoops, bout duration, whoop duration, mean inter-whoop interval,
and whoop rate) were calculated when there were multiple whoop bouts emitted
in the same context by any individual hyena. This resulted in 77 cases with
known context and a sound recording. For analyses of whoop context that

required behavioral information but not spectrographic data, archived

29

500 — I Female

E p=0.016 [3 Male
,>,~ 400 —-
C
(D
D
g p<0.0001
p 300 —
To
“.5,
E 200 -
(U
'0
C
.2
C 100 _'
(U
G)
2
0 l

 

 

 

 

 

 

 

 

Adult Cub

Figure 2.6 Mean (:1: SE) fundamental frequencies for hyenas in each age/sex
group. Adult males included both adult natal and immigrant males. Numbers
over bars represent the number of individuals sampled in each age/sex class.

30

observations (n=85 individuals) were examined to determine the context of
additional bouts in order to supplement the information gained from the audio
recordings.

The probability of whooping in a given context varied among sex/age
groups (n=145, X23=32.211, p<0.0001, P5001; Figure 2.7). Male and female
cubs were equally likely to whoop in both contexts (n=77, X21=0.978, p=0.323,
P50.025), and were grouped together for subsequent analyses of whoop
contexts. Adult male hyenas were more likely to whoop spontaneously, and less
likely to whoop during social excitement, than either adult females (n=68,
X21=22.838, p<0.0001, P50.017) or cubs (n=116, X21=27.905, p<0.0001,
P50.013). There was no significant difference between adult females and cubs
in the likelihood of whooping spontaneously or during excitement (n=106,

x=.=o.04o p=0.842, P5005).

V_ari_a_ti_gn in whoopfltg

The frequency distribution (Figure 2.8) of the whoop rates of all bouts
sampled (n=131) demonstrates a continuous range of whoop rates, rather than
the bimodal “fast” and “slow" whoop types suggested by Kruuk (1972). The rate
at which whoops are produced in a whooping bout (whoop rate = number of
whoops/min) might be affected by changes in either the duration of the whoops

themselves, the duration of the interludes between the whoops (inter-whoop

31

100 r- I Excitement
90 _ r- : Spontaneous

7O
60
50

% whoops emitted
.5
O

 

 

 

 

 

 

 

 

 

Figure 2.7 Whoop distributions for each context by sex and age class. Numbers
over bars refer to the number of individuals emitting whoops in the indicated
context.

32

Number of bouts

 

l l

0 10 20 30 40 50 60 70
Whoops/min

 

Figure 2.8. Frequency distribution for the whoop rates of all sampled bouts
(n=131)

33

intervals), or both parameters. In fact, higher whoop rates were associated both
with shorter whoop durations and with shorter inter-whoop intervals
(F2,124=260.665, p<0.0001, whoop duration: rp= -0.718; inter-whoop interval:

rp= -0.730; Figure 2.9).

Effects of sexI ageI and context on bout parameters

There were significant effects of age and context on whoop rate (age:
Mann-Whitney U=246, p<0.0001; context: U=878, p<0.0001; Figure 2.10). There
was no effect of the sex of the caller on whoop rate (U=629, p=0.232). Cubs
whooped faster than adults in spontaneous contexts (U=30, p=0.001), but there
was no significant difference based on age during social excitement (U=96,
p=0.072). Whoop rates were faster during social excitement than in spontaneous
contexts for both adults (U=245, p=0.002) and cubs (U=169, p=0.015).

The mean duration of whoops within a bout was affected by both age and
context (age: U=811, p=0.001; context: U=276, p=0.001 Figure 2.11), but not by
sex (U=547, p=0.892). Adults produced longer whoops than cubs when
whooping spontaneously (U=169, 'p=0.015), but there was no significant
difference between adults and cubs for whoops produced during social
excitement (U=200, p=0.096). The duration of adult whoops were shorter when
whooping during social excitement than spontaneously (U=87, p=0.036), but
there was no significant difference in the length of whoops produced by cubs in

the two contexts (U=65, p=0.063).

34

Figure 2.9. Relationship between whoop rate and (a) the duration of individual
whoops within the bout, or (b) the duration of inter-whoop intervals. Higher
whoop rates are produced by shortening both whoop durations and inter-whoop

intervals.

35

 

_
0.
QC

(a)

_ _ _ _ _
5. 0. 5. 0. 5.
2 2 1 1 O
863 cosmSn goon; cows.

10 20 30 4O 50 60 7O

 

0.0
O

Whoops/min

V
6

(b)

  

. at
. s snip cw

r _ _ _ _
5 4 3 2 4|

Aoomv _m>._oE_ 30:33.95 cams.

 

O

30 40 50 60 70
Whoops/min

10 20

0

36

p=0.015

 

 

 

 

 

 

35 7"1"”?- Adult
20 [j C b

30 _ u
25 — 15 p=0.001

.E

E

a 20 —

Q.

8

c 15 _

3

.—l
O
l

01
l

 

 

 

 

 

 

Figure 2.10 Relationship between context and whoop rate (mean :l: SE) for
adults and cubs. Numbers over bars represent the number of individuals
observed in a given context.

37

”0°36 , p=0.015

 

N
o
l

I Adult
Ci Cub

_x
01
l

Mean whoop duration (sec)
0 .3
'01 o
l l

 

 

 

,0 .
O

 

 

 

Figure 2.11 Duration (mean :1: SE) of individual whoops in bouts produced by
adults and cubs in different contexts. Bar labels as in Figure 2.10.

38

There were also significant effects of both age and context on the duration
of inter-whoop intervals within a bout (age: U=745, p=0.009, context: U=225,
p<0.0001; Figure 2.12). Although there was an apparent effect of sex on this
bout parameter (U=337, p=0.039), there was no effect of sex when controlled for
age and context (adults: excitement U=15, p=0.734, spontaneous U=22,
p=0.099; cubs: excitement U=36, p=0.452, spontaneous U=3, p=0.066). There
was no difference between inter-whoop interval length in adults and cubs during
social excitement (U=175, p=0.405), however adults produced significantly
longer inter-whoop intervals than did cubs when whooping spontaneously
(U=168, p=0.017). While adults produced shorter inter-whoop intervals during
social excitement than when whooping spontaneously (U=49, p=0.001), there
was no difference between contexts for cubs (U=66, p=0.069).

There was a significant effect of age (U=794, p=0.001) and context
(U=368, p=0.025), but no effect of sex (U=535, p=0.985), on the average
duration of the entire whoop bout. Bouts produced by adults were significantly
longer than bouts produced by cubs both during social excitement (U=208,

=0.053) and when whooping spontaneously (U=158, p=0.048; Figure 2.13).
There was no difference in bout duration between contexts for either age group
(adults: U=106, p=0.142; cubs: U=86, p=0.322). There were no effects of age,
context, or sex on the number of whoops per bout (age: U=572.5, p=0.698;

context: U=623, p=0.298; sex: U=635.5, p=0.198).

39

p=0.001

A
1

 

_ I Adult
{193" a Cub

00
I

 

 

 

_x
I

Mean inter-whoop interval (sec)
N
l

 

 

 

 

 

 

O

 

Figure 2.12 Mean (1 SE) lengths of inter-whoop intervals in bouts produced by
adults and cubs. Bar labels are as in Figure 2.10.

40

p=0.048

(JO
0
l

I Adult
Cl Cub

p=0.053

N
01
l

 

N
O
l

.3
O
l

Bout duration (sec)
3
l

 

 

 

 

 

 

Figure 2.13 Mean (:1: SE) duration of bouts produced by adults and cubs.
Numbers over bars represent the number of individuals emitting whoops in each
context.

41

Responses to whoops

In total, 85 instances from archived data were examined in which we could
determine whether or not listeners responded to whoops. We found no
difference between the number of arrivals before and after a whoop (Wilcoxon
Signed Ranks Z=0.330, p=0.741), however significantly more approaches
occurred after the whoop than before (Z=2.953, p=0.003). Therefore, although
the arrival of an individual to the scene apparently does not necessarily represent
a response to a whoop, the approach of an individual to the caller and/or
interaction with the caller was considered here to be a response to the
vocalization. For all subsequent analyses we considered a whoop to have
received a response only if hyenas approached the caller or the individuals who
had been interacting with the caller at the time the whoop occurred.

The reaction to whoops by listening hyenas could be identified for 105
audio recordings. Average values for all 5 acoustic parameters (number of
whoops, bout duration, mean inter-whoop interval, mean whoop duration, whoop
rate) were calculated when there were multiple whoop bouts emitted by an
individual hyena that received the same response. This resulted in 67 cases in
which a sound recording was available and the response could be identified. For
the following analyses, data concerning context and caller identity from these
recorded bouts were combined with data from the 85 archived observations of
whooping which were examined previously. Whoops produced spontaneously
were less likely to receive a response than were whoops produced during social

excitement (n=127, X21=4.547, p=0.033; Figure 2.14). Male and female cubs

42

A
O
I

% whoops emitted
N 0)
O O
l l

_\
O
l

 

 

 

Figure 2.14 Percent of whoops emitted in each context that received a
response. Numbers over bars refer to the number of animals receiving a
response after whooping in the indicated context.

43

were equally likely to receive a response (n=88, X21=0.063, p=0.802; Figure
2.15), so these groups were combined in subsequent analyses. Of those
instances in which cubs received a response, 37.5% of the responses were by
kin (15 responses by kin in 40 cases in which there was a response). Adult
females and cubs were equally likely to receive a response (n=115, X21=0.017,
p=0.896), although calls by adult females were somewhat more likely to receive a
response than were calls by adult males, this difference was not statistically
significant (n=64, X21=3.114, p=0.078). However, adult males received

responses less often than did cubs (n=125, X21=7.471, p=0.006; Figure 2.15).

Bout parameters and response

For analyses of bout parameters, only calls with both behavioral
observations and acoustic recordings were examined. Average values were
calculated for any individual with multiple whoop bouts in a particular response
category, resulting in 27 cases in which one or more hyenas responded to the
caller within 10 minutes of whoop onset, and 40 cases in which no change was
observed in the social environment. There was no difference in whoop duration,
bout duration, or number of whoops, between calls that received responses and
those that did not (whoop length: Mann-Whitney U=492.5, p=0.111; bout
duration: U=576, p=0.51 1; number of whoops: U=751, p=0.188). However, calls
that received a response were characterized by higher whoop rates (U=883,
p=0.005; Figure 2.16a), and shorter mean inter-whoop intervals (U=311,

p<0.0001; Figure 2.16b), than were calls to which no response was observed.

44

 

% whoops emitted
_x N 00 h 01 O) \l (I)
O O O O O O O O
l l

 

O

 

Figure 2.15 Percent of whoops emitted by each sex/age class that received a
response. Numbers over bars refer to the number of animals receiving a
response.

45

CT”

 

3O '7 p=0.016

 

 

Whoops/min
—¥ N N
(11 O 01
l l l

_\
O
l

 

 

 

No Change Response

Figure 2.16a Relationship between whoop rate (mean 1: SE) and response.
Numbers over bars represent the number of individuals receiving the indicated
response.

46

p=0.001

 

4O

9’
01
j

s»
‘?

 

.N
‘1‘

.N
S”

27

A—A
C: a”

.o
3"

Mean inter-whoop interval (sec)

 

 

 

.0
o

No Change Response

Figure 2.16b Relationship between duration of inter-whoop interval (mean 1
SEM) and response. Bar labels as in 2.16a.

47

 

Effects of whooptym

All recorded bouts contained either entirely Type A whoops or some
combination of Type A and Type S. Type T whoops occurred at the end of 42 of
131 (32.1%) bouts. There was no difference between sex/age groups in the
likelihood that a Type T whoop would occur at the end of a bout (n=131, X23:
0.284, p=0.963). There was no effect of context on the likelihood of a bout
including a Type T whoop (n=99, X22: 0.005, p= 0.943), nor were bouts
containing a Type T whoop more or less likely to receive a response than bouts

not containing a Type T whoop (n=103, X21= 0.016, p= 0.901 ).

Discussion

Previous work by other researchers has shown that acoustic features of
spotted hyena whoops vary with age, sex, and individual identity (East & Hofer,
1991 a, b; Holekamp et al., 1999). Holekamp et al. (1999) used playback
experiments to determine whether spotted hyenas can identify individual
conspecifics by their whoops. Mothers were more likely to respond to the calls of
their own offspring than to those of unrelated juveniles, and they also gave
stronger responses to the calls of younger cubs than to those of older offspring,
indicating that whoops contain information about a caller’s age in addition to

communicating individual identity (see also East & Hofer, 1991a).

48

East and Hofer (1991a) found that calls produced by cubs typically
contained fewer harmonics, wider spacing between harmonics, a higher
fundamental frequency, a shorter duration of the low frequency section, and
fewer individual whoops per bout than did calls produced by adults. Additionally,
these earlier works found that calls given by adult females had lower
fundamental frequencies than those given by adult males. Here we replicated
their earlier findings that cubs produced whoops with higher fundamental
frequencies than adults, and males produced higher fundamental frequencies
than did their female peers. We also found that as cubs mature, the fundamental
frequencies of their whoops decrease. In most mammals, the fundamental
frequency of the call is primarily dependent on morphological factors such as
glottal width, vocal cord length, and length of the resonating tube (Michelson,
1983). Ontogenetic changes in these parameters are the most likely sources of
frequency variation among calls produced by spotted hyenas, since we found no
relation between social context and the fundamental frequency of the whoops.
Increased lung capacity and chest girth may also explain the differences we
observed in whoop length between cubs and adults. Although East and Hofer
(1991a) found Type T whoops more likely to be found in male bouts than female
bouts, as well as more likely in adult bouts than in cub bouts, we found no
difference between sexes or age classes in the use of this whoop type. Since
the composition of bouts does not appear to vary with sex, age, or context, or
affect the probability of receiving a response, the functional significance of Type

A, S, and T whoops remains unknown.

49

Here we have demonstrated that sound features vary with the
circumstances under which each whoop bout is emitted. Hyenas in the present
study whooped in socially interactive contexts, but also whooped for no apparent
reason. Both adults and cubs whooped faster when they were stimulated by
social excitement than when they whooped spontaneously. Whoops given in
these two types of situations most likely have different functions, as indicated by
the greater likelihood of receiving a response when whooping during excitement.
Hyenas occasionally whooped following greetings, which might simply offer an
alternative source of social excitement. The whoop rate may function to alert
conspecifics to the caller’s internal state, while other acoustic features signal the
caller's identity. Although whoop rates did not differ between age groups in
contexts involving social excitement, cubs whooped faster than adults when
there was no apparent stimulus for their vocalizations. Our results thus indicate
that spotted hyenas vary the rate at which they emit whoops according to the
circumstances prompting them to vocalize. and they do so by changing both the
duration of the individual whoops and the duration of the intervals between them.

As occurs in hyenas, subtle acoustic variations specific to the social
context and function of the call are produced by Japanese macaques (Macaca
fuscata, Green, 1975), cotton top tamarins (Sanguinus oedipus, Snowdon et al.,
1983), and vervet monkeys (Cercopithecus aethiops, Cheney & Seyfarth, 1982;
Seyfarth & Cheney, 1984). Common marrnoset (Callithrixjacchus) ‘phee’ calls
can function to reunite a group, locate mates, or assess the reproductive

condition of a prospective mate. Call subtypes specific to these different

50

 

 

functions consistently differ in both acoustic frequency and temporal parameters
(Norcross & Newman, 1993; Norcross et al., 1999). Some non-primate
mammals change features of their alarm calls based on response urgency, as
seen in ground squirrels (Spennophilus beecheyi, Owings & Virginia, 1978) and
marmots (Marmota flaviventn's, Blumstein 8. Annitage, 1997). Brant’s whistling
rats (Parotomys brantsir) modify the duration of their alarm calls (Le Roux et al.,
2001) and Richardson’s ground squirrels (Spennophilus n'chardsonir) vary the
rate of alarm calling (Warkentin et al., 2001) based on the degree of perceived
threat. Whistling rats produce shorter calls in high-risk than in low-risk situations,
while ground squirrels produce a faster call when the predator is closer. Other
species vary the acoustic properties of their alarm calls based on the type of
predator threat. Aerial and terrestrial predators elicit acoustically different alarm
calls, and receive different responses from listeners in chickens (Gallus
domesticus) (Gyger et al., 1987), vervet monkeys (Seyfarth et al., 1980b), and
ring-tailed lemurs (Lemur catta) (Macedonia, 1990). Suricates (Suricata
suricatta) integrate both information about response urgency and predator type
into their alarm calls (Manser, 2001). In contrast to the whoop vocalizations of
spotted hyenas, alarm calls are typically only given in the presence of a predator,
and are usually simple in structure (Bradbury 8. Vehrencamp, 1998).

Spider monkey (Ateles geoffroyl) ‘whinny’ calls are loud calls with
subtypes varying in the frequency dimension (T eixidor & Byrne, 1999).
Playbacks of calls recorded in different contexts elicit consistently different

responses, but determination of call context by recipients first requires

51

 

identification of the caller (Teixidor & Byrne, 1999). In this case, calls are
context-specific, but individual variation apparently requires listeners to be
familiar with the individual vocal signatures of all group members to discern call
meaning. It is not yet known what role, if any, familiarity with individual vocal
patterns plays in spotted hyena communication.

The contextdependent vocalizations described above are examples of
long-distance calls. In short-distance calls emitted by some mammals, acoustic
features do not appear to vary with context. Mountain gorilla (Gorilla gorilla)
double-grunts consist of at least two subtypes differing only in the frequency of
the second formant (Seyfarth et al., 1994). For this short-distance call, only the
acoustic structure of the double-grunt influences listener response, and the
behavioral context in which the call is emitted does not appear to play a role in
determining the parameters of the sound (Seyfarth et al., 1994). Acoustically
different subtypes also occur in baboon (Papio cynocephalus) grunts, but
listener's responses vary with context and the caller’s identity (Rendall et al.,
1999). It may be that the effectiveness of calls functioning to convey information
to animals in the caller’s immediate vicinity varies with the receivers’ perception
of the circumstances, whereas calls designed to communicate with distant
conspecifics must transmit contextual information to individuals who have no
knowledge of the caller’s current condition.

East and Hofer (1991b) inquired what prompted hyenas of each age/sex
class to whoop while at the communal den, and suggested possible functions for

these calls. They observed that most cub whoops occurred either spontaneously

52

(with no apparent preceding stimulus), or when the individual had been attacked
or was othenivise agitated. These authors suggested that spontaneous cub
whoops function as self-advertisement, but that agitated whoops function to elicit
support. They observed adults of both sexes whooping spontaneously more
frequently than in any other context. Adults also whooped after being attacked,
or after witnessing an attack on another clan member, perhaps to request
support or to remind listeners of the whooping animal’s identity. Mothers often
whooped to call cubs, and adults of both sexes whooped in reply to distant calls,
apparently indicating their current location. Adults whooped to recruit conspecific
aid during interactions with lions or rival clans, and males sometimes directed
whoops at other clan members or whooped while scent-marking, possibly as an
inter- or intra- sexual display (East 8 Hofer, 1991b). East and Hofer (1991b)
found that adult males called more frequently than adult females, and that high-
ranking adults vocalized at higher rates than did same-sex subordinates.

In the present study, cubs and adult females did not whoop more often in
any one context, but adult males most often whooped spontaneously. Since
whoops produced spontaneously typically have slower whoop rates, indicating
low levels of excitement, whoops produced for no apparent reason probably
function to advertise identity and location. Adult males are almost always
immigrants (Smale et al., 1993) and therefore might be expected to advertise
their identity and presence within the home range more frequently than natal
animals. Doing so may signal to females that a male is in the home range at that

moment, possibly decreasing the rates of aggression he experiences, or

53

increasing his chances of mating, by familiarizing females with his presence in
the home range. Additionally, whooping may announce to prospective
immigrants that there is at least one adult male already present in the clan (East
& Hofer, 1991b), and in fact may inform listeners about male queue length in the
clan (East & Hofer, 2001; Engh et al., 2002).

Both the length of individual whoops and duration of inter-whoop intervals
affect the rate at which whoops are produced within a bout. The data presented
here did not show dichotomous “slow” and “fast” whoop subtypes as Kruuk
(1972) suggested, but rather a range of whoop rates with no distinct group
demarcation. Modulation of whoop rate may provide listeners with functionally
different whoop subtypes that convey different information. Calls here that
received responses generally had higher whoop rates than those that did not.
Hyena audiences may be better able to discern the current general state of the
caller as indicated by the acoustic properties of the sound, than the specific
circumstances under which the call is emitted. Since variations in whoop rate
and inter-whoop interval are associated with variations in context, it is not
surprising that the response of listeners varies with these cues. Interestingly,
cubs were not more likely than adult females to receive responses to their calls.
The data presented here also indicate that male and female cubs are equally
likely to receive responses to their calls, suggesting that listeners do not
discriminate based on the sex of the calling cub.

Variation in conspecific response to acoustic signals has been studied in a

variety of social mammals. Chacma baboons produce different bark calls in

54

different contexts, listeners respond differently to bark variations given in high-
risk and low-risk contexts, and these calls exist in graded variants (Fischer et al.,
2001). Scream subtypes in rhesus monkeys (Macaca mulatta) and pigtail
macaques (Macaca nemestrina) differ with respect to several measures
(including bandwidth, number of pulses, and frequency modulation), and
conspecifics respond differentially to different scream subtypes (Gouzoules et al.,
1984; Gouzoules & Gouzoules, 1989). Fischer (1998) correlated the acoustic
structure of Barbary macaque (Macaca sylvanus) ‘shrill barks’ with their eliciting
stimuli, and a series of playback experiments determined that receivers
perceived two distinct categories of calls. Within each call category, recipient
responses were consistent regardless of whether inter-individual differences
between calls were small or large. Thus, individual variation within a subtype
may not necessarily affect the meaning of the call.

Acoustic features of spotted hyena calls vary with age, sex, and context.
Furthermore, particular acoustic features of hyena calls appear to elicit particular
types of responses from listeners. Additional research is needed to document
the relationship between call structure and different intensities of response. In
this study, sample sizes were too small to examine the relationship between call
structure and response intensity. Due to the high potential for vocal recognition
in this species, field experiments could be used to determine the extent to which
recognition occurs. While it is clear that individual recognition occurs in spotted
hyenas (Kruuk, 1972; East & Hofer, 1991a, b; Holekamp et al., 1999), the

specific acoustic features and mechanisms (genetic relatedness, familiarity, etc.)

55

mediating individual recognition are still unclear. Similarities between the calls of
mother and cub, siblings, or within a matriline, might offer a mechanism for
identifying possible allies from a distance. Since spotted hyenas are highly
territorial, the existence of acoustic features specific to particular clans could
potentially facilitate the identification of alien intruders without the necessity of
visual sightings. Hofer et al. (2001) demonstrated that hyenas could identify clan
members by scent. Vocal cues might offer yet another method for hyenas to

determine the clan affiliation of unseen conspecifics.

56

Chapter 3
ESTIMATING TERRITORIAL BOUNDARIES OF SPOTTED HYENA CLANS
USING PLAYBACKS

Introduction

The use of pre-recorded sounds as a method of attracting animals (“call-
ins”) has been used to estimate the density of spotted hyena populations (Kruuk,
1972; Creel 8: Creel, 1996; Mills, 1996; Ogutu & Dublin, 1998; Mills et al., 2001),
but has never been applied to determining defended territory boundaries.
Traditionally, territorial boundaries have been defined through connecting the
outermost points at which individually distinguished hyenas are resighted (i.e.
Whateley & Brooks, 1978; Frank, 1986a; Boydston et al., 2001). Conventional
techniques rely on long-term, intensive monitoring of multiple individuals per clan,
which may be impractical if researchers seek to chart the boundaries of more
than one territory, or rapidly determine the number of clans inhabiting a
geographical area. This study sought to determine whether established call-in
procedures could be used to delineate the boundaries of spotted hyena clans by
inciting territorial behaviors, and then connecting the points representing
locations where such behaviors have been observed.

Spotted hyenas are gregarious carnivores that defend group territories

(Kruuk, 1972; Frank, 1986a). Boundaries are maintained through border patrols

57

in which groups of individuals engage in scent-marking via the deposition of anal
gland secretions (‘pasting’ see Kruuk, 1972; Boydston et al., 2001), as well as
establishment of communal “latrines” at which multiple hyenas defecate (Kruuk,
1972). In instances where large groups of hyenas from two neighboring clans
are present in territorial boundary areas, clashes, or clan “wars”, may occur
during which both sides participate in coordinated rushes and attacks and give
frequent long-distance vocalizations (whoops) to recruit allies from afar (Kruuk,
1972; East & Hofer, 1991b; Boydston et al., 2001).

Spotted hyenas generally forage within their own territories, hunting
medium and large-bodied ungulates. and territorial defense is often associated
with competition for food resources (Kruuk, 1972; Henschel & Skinner, 1991).
Even in areas of the Serengeti where aliens in transit are tolerated within a
territory, commuters who feed on prey outside of their home territories are
routinely attacked by resident hyenas (Hofer & East, 1993). Here I hoped to
exploit the well-documented aggression directed by hyenas towards intruders to
determine where clan territorial boundaries occurred throughout the Masai Mara
National Reserve in Kenya. Specifically, I hoped to stimulate such agonistic
interactions by attracting hyenas from multiple clans to selected sites using

recorded vocalizations.

58

 

E" _'...¢'_.

Methods

Study area and call-in site selection

This study was conducted between May and August 2000 in the Masai
Mara National Reserve (MMNR) in southwest Kenya. The MMNR covers
approximately 1, 530 km2 of rolling plains consisting of short and tall grasslands
with scattered Acacia woodlands, and thickets of Croton sp. and Euclea sp.
Dense forests border the rivers, and multiple seasonal streams crisscross the
reserve. Spotted hyenas in MMNR are organized into clans that defend discrete

territories (Frank, 1986a; Boydston et al., 2001).

Previous studies have suggested that hyenas are attracted to playbacks
from a maximum of 2.5-3.5 km (Mills, 1996; Ogutu & Dublin, 1998; Mills et al.,
2001). Using a map of the Reserve, we identified 62 potential call-in sites that
were approximately 5km apart. In contrast to Ogutu and Dublin (1998), who
used random call-in locations to minimize the effects of territoriality on the
numbers of hyenas responding, we sought to exploit the tendency of hyenas to
respond in groups (i.e. if one hyena responds, all nearby hyenas will respond).
Some locations were chosen to coincide with peripheral areas of hyena clans
whose borders had been approximately identified by Boydston et al. (2001), and
others were randomly chosen in areas where hyenas have been observed but

clan distinctions were not known (Figure 3.1 ).

59

 

 

.388 coumgom 55. .250 9:09:90:
3 35953 acumgxoaaa 230326 05 “commie. was. 3:00 éoouv coumgom 3 3.5% mm coca. chc 5.0 5.2.2.
05 8:32am. moi pounce on... 623 54.8 .3233 mm 9.36...» 9:83. .95sz Bus. Enos. 9: “.o no.2 é.» 2:2“.

 

 

60

Call-in stimulus and presentation

We prepared four call-in tapes containing sounds known to attract hyenas:
vocalizations of hyenas feeding at a kill, calls from hyenas mobbing lions
(Panthera Ieo), an inter-clan altercation, and the sounds of a dying ungulate.
Each stimulus tape had been recorded in a different location, and only one
contained the sounds of MMNR hyenas. This tape was recorded in the Talek
clan home range, whose boundaries were defined by Boydston et al. (2001) (see
Figure 3.1). No call-ins in the current study were performed inside the Talek clan
borders.

Call-in tapes were broadcast for 6 minutes at full volume using a Marantz
PMD-22 portable cassette recorder connected to two 20-watt amplified speakers
(Optimus model 32-1240, Tandy Corporation) powered by our vehicle’s 12-volt
battery. The speakers were mounted in the front windows of the vehicle, facing
away from one another (Figure 3.2). Previous studies have suggested that any
responding hyenas should arrive within 30 minutes of sound onset, andthat they
should typically respond in groups (Ogutu & Dublin, 1998; Mills et al., 2001). The
area around the call-in station was continuously scanned in all directions by
multiple observers for 30 minutes after the beginning of the broadcast. We
considered any hyena approaching to within 300m of the vehicle to be
responding to the call-in. The information recorded at each call-in included: (i)
date, time, and location of the call-in; (ii) number of hyenas arriving and the

direction of from which each hyena arrived; (iii) latency of hyena arrival relative to

61

 

 

Figure 3.2 Speakers were placed on platforms mounted in the windows of the
vehicle. Due to the curvature of the speaker horns, they did not need to be
turned to ensure multi—directional sound broadcast . Photo courtesy of Sofia
Wahaj.

62

 

sound onset; (iv) behaviors of arriving hyenas (i.e. territorial marking, clan war,
etc.); (v) maximum visibility in each compass direction.

All call-ins were broadcast between 0600h and 1000b, and only in the
absence of rain and high winds to maximize sound transmission (Larom et al.,
1997; Bradbury & Vehrencamp, 1998). In order to avoid habituation, call-ins at
adjacent locations were performed at least 3 days apart, and with different
recordings. All statistical analyses were performed using SYSTAT software
(version 8.0, SPSS, Inc.). All means are reported with standard errors. Maps

were generated using ArcView GIS software.

Results

In total, we performed 38 call-ins during the study period (Figure 3.3).
When played at high volume, some of our tapes produced distorted sounds.
Therefore, the two tapes with the best sound quality were used for the majority of
call-ins (n=19 and n=14). Hyenas were attracted to 25 of 38 (65.8%) of call-ins
(Table 3.1). Of those 25, the mean number of hyenas seen at a site was 7.480
11.224, with only 9 (23.7%) attracting 10 or more hyenas, and a maximum of 25.
The mean latency to arrival of the first responding hyena was 5.840 1:1 .360
minutes from the onset of sound and 85% of all respondents arrived within 20

minutes. Although lions are known to respond to playbacks of hyena

63

 

69830 203 82223 .mtanc. 26:3 25:30. 2m 23» 55, 353652
8.5 doom fiam:<.>m.2 upstate... 21.8 S 25:80. 65 @5265 9.031 .96sz «is. =4me 3 ans. .nd 2:2“.

 

 

Table 3.1 Summary of results from call-ins that attracted hyenas. UTM
coordinates refer to the exact geographic location of the vehicle during
broadcast. Arrival direction refers to the general compass direction from which
we observed hyenas approaching the vehicle.

 

Latency to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Call-in UTM N um E Date 12:33:": minimal 7:15;: 6111:2163“)
1 9834669 743415 5/19/2000 2 3 6.5 N, Nw
2 9829411 746566 5/22/2000 3 4.5 SE, Nw
3 9839935 735633 5/23/2000 11 0 9.9 N, Nw
4 9833456 728859 5/24/2000 2 2 4.5 NW, w
5 9820972 748643 5/26/2000 27 27 E
6 9848699 735723 5/27/2000 25 3 7.6 s, w, sw
9 9830913 753072 5/31/2000 14 4 11.2 w, sw
10 9839291 730266 6/2/2000 2 6.9 NW, NE
11 9844259 737647 6/5/2000 11 16.5 sw, N
13 9828861 740978 6/9/2000 7 16.8 sw
14 9842251 733236 6/10/2000 13 1 8.3 E, sw, Nw
16 9852273 731704 6/12/2000 3 2 12.3 s, Nw
17 9821207 763157 6/13/2000 8 7 18.9 sw, s
18 9846081 726311 6/14/2000 3 3 w
19 9840113 719362 6/16/2000 11 1 6 s,sw
20 9837368 730146 6/18/2000 3 3 4.3 s, SE
21 9827300 752593 6/19/2000 14 2 9.9 E, sw
25 9846240 744009 6/24/2000 10 3 16.9 sw, NE
26 9824475 751894 6/26/2000 1 24 24 N
29 9829839 747173 6/29/2000 12 2 10.6 NW
30 9831697 728207 6/30/2000 2 13 13 Nw
33 9847670 748136 7/2/2000 ' 10 15 s, SE
35 9838495 734683 7/4/2000 8 8 sw
37 9846853 731927 7/5/2000 7 2 16.1 N, NW
38 9846708 731546 7/18/2000 17 3 4 N,W

 

65

 

L .g

vocalizations, lions were attracted to our sites only once; two Iionesses walked
past our vehicle, but did not stop. During one call-in, we attracted a jackal
(Cam's mesomelas.) in addition to several hyenas. Although no previous study
has reported jackals responding to hyena call-ins, Mills et al. (2001) suggested it

might be possible to adapt the call-in method for use with this species.

Description of behaviors

Most hyenas arrived at the call-in site running or Ioping, those who walked
were typically following other individuals. Hyenas would often approach to within
50-100m of the vehicle, and tended to search the area surrounding the vehicle,
particularly if the sound was still being broadcast.

On two occasions (Call-ins #3 and #20, Figure 3.3), hyenas exhibited
possible territorial behaviors upon arrival. During Call-in #3, seven individuals
exhibited excitement on arrival (e.g., tails bristled, ears erect and forward) and
began pasting as a group. The group then moved off to the south, still bristle-
tailed, in what appeared to be a border patrol. Behaviors exhibited during Call-in
#20 were less clear: one adult arrived within two minutes of sound onset, and
two minutes later additional adults (n=2) ran in from the opposite direction. The
second group was bristle-tailed and chased the first hyena for a short distance,
then Ioped away in the direction from which they arrived. Since the number of
individuals involved was so small, it is difficult to determine whether the first

hyena was an intruder, or simply subordinate to the others.

66

On all other occasions, responding hyenas exhibited non-territorial
behaviors once they arrived at the site. That is, they engaged in greetings,
aggressive, and affiliative behaviors associated with intraclan interactions, or
simply rested. Hyenas who arrived as a group typically also left the site as a
group, and often hyenas departed within 10 minutes of arrival. Here playbacks

never elicited a clan war.

Discussion

Two of the recordings used in this study appear to be as effective at
attracting hyenas as those used by other studies (Table 3.2), but although Ogutu
and Dublin (1998) reported that their call-ins incited clan wars in MMNR, we were
unable to stimulate inter-clan interactions. Similarly, Whateley and Brooks
(1978) were unable to observe clan wars at any of their call-in stations in
Hluhluwe Game Reserve in South Africa, even those on known clan boundaries.
While it may be possible to do so, current call-in procedures do not appear to be
an effective method of reliably eliciting clan wars.

The combination of sounds used here for call-ins is thought to be effective
because the sounds of a kill attract hyenas who are hungry and the sounds of
encroaching alien hyenas attract residents even when sated, thus potentially

attracting all hyenas within range of the sound. Lions are able to

67

Table 3.2 Summary of results from studies using playbacks to attract spotted
hyenas. The present study represents the only attempt to date to use call-ins to
map home ranges; all others used playbacks to census populations.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Mean
Location Number of Cari-unearth Hyenas per Clan Wars
Call-ins Resy onse Successful Observed
p Call-in
Masai Mara National 0
Present Study Reserve, Kenya 38 65.8 /o 7.5 NO
. Kruger National Park, 0
Mrlls et al. (2001) South Africa 346 65.6 /o 4.1 n/a
Ogutu and Dublin Masai Mara National ,,
(1998) Reserve, Kenya 192 n/a 10'7 YES
Whately and Hluhluwe Game Reserve, 0
Brooks (1978) South Africa 105 83 /° 3'4 ”0

 

** Mean approximated from total hyenas observed and the number of call-ins performed. The
number of successful call-ins versus those to which there was no response was not reported.
After Ogutu and Dublin (1998).

68

 

 

 

assess group size based on acoustic signals (McComb et al., 1994), but it is
unknown whether hyenas can make this distinction. Spotted hyenas may not
respond to a call-in if they judge their own social group to be smaller than the
perceived Size of the intruders’ group. To date, no study has examined the
ability of hyenas to discriminate between the sounds of clan members and alien
hyenas, although this has been demonstrated in other species (lions, McComb et
al., 1993; Mexican jays, Aphelocoma ultraman'na, Hopp et al., 2001). If vocal
recognition at this level does not occur, hyenas may be responding only to the
sounds of group activity. Spotted hyena societies are characterized by rigid
linear dominance hierarchies (Frank, 1986b; Smale et al., 1993; Holekamp 8
Smale, 1998) in which high-ranking animals typically have more maternal
relatives, and hence more potential allies than do low-ranking individuals (Frank
et al., 1995; Holekamp et al., 1996). Boydston (2001) determined that low-
ranking females spent much of their time near the edges of the territory, while
high-ranking females were more often located in the central core area.
Therefore, call-ins near a territorial border may be more likely to reach low-
ranklng individuals who might be most reluctant to approach the sounds of a
large group. -

Social structure and social behaviors must be considered when examining
the results of call-in studies. While call-ins may be an effective, low-cost method
of estimating the size of spotted hyena populations, more research is needed
before these procedures can be used to map territory boundaries. Rather than

attempting to incite clan wars or other territorial behaviors, call-ins could be used

69

in conjunction with mark-resighting or radio-tracking to increase the probability of
locating known individuals. Traditional methods of determining territorial borders

could then be applied to the resulting set of locations.

70

 

Literature Cited

Altmann, J. 1974. Observational study of behavior: sampling methods.
Behaviour, 49, 227-262.

Baker, C. M. 1998. Communication in marsh mongooses (Atilax paludinosus):
anal gland secretion and scat discrimination in adults, and individual
variation in vocalizations of juveniles. South African Journal of Zoology,
33, 49-51.

Blumstein, D. T. 8 Annitage, K. B. 1997. Alarm calling in yellow-bellied marmots:
l. The meaning of Situationally variable alarm calls. Animal Behaviour, 53,
143-171 .

Boydston, E. E. 2001. Socio-Ecological Determinants of Space Utilization in the
Spotted Hyena, Crocuta crocuta. PhD dissertation. Michigan State
University, East Lansing.

Boydston, E. E., Morelli, T. L. 8 Holekamp, K. E. 2001. Sex differences in
territorial behavior exhibited by the spotted hyena (Hyaenidae, Crocuta
crocuta). Ethology, 107, 369-385.

Bradbury, J. W. 8 Vehrencamp, S. L. 1998. Principles of Animal Communication.
Sunderland, Massachusetts: Sinauer Associates, Inc.

Brown, E. D. 8 Farabaugh, S. M. 1997. What birds with complex social
relationships can tell us about vocal learning: vocal sharing in avian
groups. In: Social Influences on Vocal Development (Ed. by Snowdon, C.
T. 8 Hausberger, M.), pp. 98-127. Cambridge: Cambridge University
Press.

Bugnyar, T., Kijne, M. 8 Kotrschal, K. 2001. Food calling in raven: are yells
referential signals? Animal Behaviour, 61.

Cheney, D. 8 Seyfarth, R. 1982. How vervet monkeys perceive their grunts: field
playback experiments. Animal Behaviour, 30, 739-751.

Cleveland, J. 8 Snowdon, C. T. 1982. The complex vocal repertoire of the adult

cotton-top tamarin. Zeitschn'ft Fur Tie/psychologie-Joumal of Comparative
Ethology. 58, 231 -270.

71

 

 

 

Cohen, J. A. 8 Fox, M. W. 1976. Vocalizations of wild canids and possible effects
of domestication. Behavioural Processes, 1, 77-92.

Compton, L. A., Clarke, J. A., Seidensticker, J. 8 Ingrisano, D. R. 2001. Acoustic
characteristics of white-nosed coati vocalizations: A test of motivation-
structural rules. Journal of Mammalogy, 82, 1 054-1 058.

Creel, S. 8 Creel, N. M. 1996. Limitation of African wild dogs by competition with
larger carnivores. Conservation Biology, 10, 526-538.

East, M. L. 8 Hofer, H. 1991a. Loud calling in a female-dominated mammalian
society I: Structure and composition of whooping bouts of spotted hyaenas
(Crocuta crocuta). Animal Behaviour, 42, 637-649.

East, M. L. 8 Hofer, H. 1991b. Loud calling in a female-dominated mammalian
society II: Behavioral contexts and functions of whooping of spotted
hyaenas (Crocuta crocuta). Animal Behaviour, 42, 651-669.

East, M. L. 8 Hofer, H. 2001. Male spotted hyenas (Crocuta crocuta) queue for
status in social groups dominated by females. Behavioral Ecology, 12,
558-568.

Eisenberg, J. F. 1976. Communication mechanisms and social integration in the
black spider monkey, Ate/es fuscipes robustus, and related species.
Smiths. Contr. Zool., 213, 1-108.

Engh, A. L., Funk, S. M., Van Horn, R. C., Scribner, K. T., Bruford, M. W.,
Libants, S., Szykman, M., Smale, L. 8 Holekamp, K. E. 2002.
Reproductive skew among males in a female-dominated mammalian
society. Behavioral Ecology, 13, 193-200.

Evans, C. S. 8 Evans, L. 1999. Chicken food calls are functionally referential.
Animal Behaviour, 58, 307-319.

Evans, C. S., Evans, L. 8 Marler, P. 1993. On the Meaning of Alarm Calls -
Functional Reference in an Avian Vocal System. Animal Behaviour, 46,
23-38.

Ewer, R. F. 1973. The Camivores. Ithaca: Cornell University Press.

Fischer, J. 1998. Barbary macaques categorize shrill barks into two call types.
Animal Behaviour, 55, 799-807.

Fischer, J., Metz, M., Cheney, D. L. 8 Seyfarth, R. M. 2001. Baboon responses
to graded bark variants. Animal Behaviour, 61, 925-931.

72

 

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

Frank, L. G. 1986b. Social organization of the spotted hyaena (Crocuta crocuta)
II. Dominance and reproduction. Animal Behaviour, 34, 1510-1527.

Frank, L. G., Glickman, S. E. 8 al., e. 1990. Sexual dimorphism in the spotted
hyaena (Crocuta crocuta). Journal of Zoology, 221, 308-313.

Frank, L. G., Holekamp, K. E. 8 Smale, L. 1995. Dominance, demography, and
reproductive success of female spotted hyenas. In: Serengeti II: dynamics,
management, and conservation of an ecosystem (Ed. by Sinclair, A. R. E.
8 Arcese, P.), pp. 364-384. Chicago: University of Chicago Press.

Gouzoules, H. 8 Gouzoules, S. 1989. Design features and developmental
modification of pigtail macaque, Macaca nemestrina agonistic screams.
Animal Behaviour, 37, 383-401.

Gouzoules, S., Gouzoules, H. 8 Marler, P. 1984. Rhesus monkey (Macaca
mulatta) screams: Representational signalling in the recruitment of
agonistic aid. Animal Behaviour, 32, 1 82-193.

Green, S. 1975. Communication by a graded vocal system in Japanese
monkeys. In: Primate Behavior (Ed. by Rosenblum, L. A.), pp. 1-102. New
York: Academic Press.

Grinnell, J. 8 McComb, K. 2001. Roaring and social communication in African
lions: the limitations imposed by listeners. Animal Behaviour, 62, 93-98.

Gyger, M., Marler, P. 8 Pickert, R. 1987. Semantics of an avian alarm call
system: the male domestic fowl, G. domesticus. Behaviour, 102, 15-40.

Hammerschmidt, K. 8 Todt, D. 1995. Individual differences in in vocalizations of
young Barbary macaques (Macaca sylvanus): a multi-parametric analysis
to identify critical cues in acoustic signalling. Behaviour, 132, 381-399.

Harrington, F. H. 8 Mech, L. D. 1979. Wolf howling and its role in territory
maintenance. Behaviour, 68, 207-249.

Hauser, M. 1991. Sources of acoustic variation in Rhesus macaque (Macaca
mulatta) vocalizations. Ethology. 89, 29-46.

Hauser, M. 1996. The Evolution of Communication. Cambridge, Massachusetts:
MIT Press.

73

 

Henschel, J. R. 8 Skinner, J. D. 1991. Territorial behavior by a clan of spotted
hyaenas Crocuta crocuta. Ethology, 88, 223-235.

Hofer, H. 8 East, M. L. 1993. The Commuting System of Serengeti Spotted
Hyaenas - How a Predator Copes with Migratory Prey .2. Intrusion
Pressure and Commuters Space Use. Animal Behaviour, 46, 559-574.

Hofer, H., East, M. L., Sammang, l. 8 Denhard, 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 (Ed. by Marchlewska-Koj, A., Lepri, J. J. 8 Muller-Schwarze,
D.), pp. 141-148. New York: Kluwer Academic/Plenum Publishers.

Holekamp, K. E., Boydston, E. E. 8 Smale, L. 2000. Group travel in social
carnivores. In: On the Move: How and why animals travel in groups (Ed.
by Boinsky, S. 8 Garber, P. A.), pp. 587-627. Chicago: University of
Chicago Press.

Holekamp, K. E., Boydston, E. E., Szykman, M., Graham, l., Nutt, K. J., Birch, S.,
Piskiel, A. 8 Singh, M. 1999. Vocal recognition in the spotted hyaena and
its possible implications regarding the evolution of intelligence. Animal
Behaviour, 58, 383-395.

Holekamp, K. E. 8 Smale, L. 1991. Dominance Acquisition During Mammalian
Social-Development - the Inheritance of Maternal Rank. American
Zoologist, 31, 306-317.

Holekamp, K. E. 8 Smale, L. 1993. Ontogeny of Dominance in Free-Living
Spotted Hyaenas - Juvenile Rank Relations with Other Immature
Individuals. Animal Behaviour, 46, 451-466.

Holekamp, K. E. 8 Smale, L. 1998. Behavioral development in the spotted
hyena. Bioscience, 48, 997-1005.

Holekamp, K. E., Smale, L. 8 Szykman, M. 1996. Rank and reproduction in the
female spotted hyaena. Journal of Reproduction and Fertility, 108, 229-
237.

Hopp, S. L., Jablonski, P. 8 Brown, J. L. 2001. Recognition of group membership
by voice in Mexican jays, Aphelocoma ultramarina. Animal Behaviour, 62,
297-303.

lnsley, S. J. 1992. Mother-Offspring Separation and Acoustic Stereotypy - a

Comparison of Call Morphology in 2 Species of Pinnipeds. Behaviour,
120, 103-122.

74

 

 

Jennings, D. J., Hayden, T. J. 8 Kent, J. P. 1997. Changes in the structure of the
peep vocalization of female domestic chicks (Gallus gallus domesticus)
with age. lntemational Journal of Comparative Psychology, 10, 57-73.

Kruuk, H. 1972. The Spotted Hyena: a Study of Predation and Social Behavior.
Chicago: Chicago University Press.

Kruuk, H. 1976. Feeding and social behaviour of the striped hyaena (Hyaena
vulgaris Desmarest). East African Wildlife Joumal, 14, 91-111.

Larom, D., Garstang, M., Payne, K., Raspet, R. 8 Lindeque, M. 1997. The
influence of surface atmospheric conditions on the range and area
reached by animal vocalizations. Joumal of Experimental Biology, 200,
421-431.

Le Roux, A., Jackson, T. P. 8 Cherry, M. I. 2001. Does Brants' whistling rat
(Parotomys brantsii) use an urgency- based alarm system in reaction to
aerial and terrestrial predators? Behaviour, 138, 757-773.

Macedonia, J. M. 1986. Individuality in a Contact Call of the Ringtailed Lemur
(Lemur- Catta). American Journal of Primatology, 11, 163-179.

Macedonia, J. M. 1990. What is communicated in the anti-predator calls of
lemurs: evidence from antipredator call playbacks to ringtailed and ruffed
lemurs. Ethology, 86, 177-190.

Manser, M. B. 2001. The acoustic structure of suricates' alarm calls varies with
predator type and the level of response urgency. Proceedings of the Royal
Society of London Series B-Biological Sciences, 268, 2315-2324.

Marler, P. 1967. Animal communication Signals. Science, 157, 769-774.

Marler, P., Evans, C. 8 Hauser, M. D. 1992. Animal vocal signals: reference,
motivation or both? In: Nonverbal Vocal Communication (Ed. by
Papousek, H., Jiirgens, U. 8 Papousek, M.), pp. 65-86. Cambridge:
Cambridge University Press.

Marten, K. 8 Marler, P. 1977. Sound transmission and its significance for animal
vocalizations: temperate habitats. Behavioral Ecology and Sociobiology, 3,
21 7-290.

Marten, K., Quine, D. 8 Marler, P. 1977. Sound transmission and its significance

for animal vocalizations: tropical forest habitats. Behavioral Ecology and
Sociobiology. 3, 291-302.

75

 

 

McComb, K., Packer, C. 8 Pusey, A. 1994. Roaring and numerical assessment
in contests between groups of female lions, Panthe/a [80. Animal
Behaviour, 47, 379-387. .

McComb, K., Pusey, A., Packer, C. 8 Grinnell, J. 1993. Female lions can identify
potentially infanticidal males from their roars. Proceedings of the Royal
Society of London Series B-Biological Sciences, 252, 59-64.

McCowan, B. 8 Reiss, D. 2001. The fallacy of "signature whistles" in bottlenose
dolphins: a comparative perspective of "signature information" in animal
vocalizations. Animal Behaviour, 62, 1 1 51 -1 162.

McGregor, P. K. 8 Krebs, J. R. 1984. Song degradation as a distance cue in
great tit Pan/s major song. Behavioral Biology and Sociobiology, 16, 49-
56.

Michelson, A. 1983. Biophysical basis of sound communication. In: Bioacoustics:
A Comparative Approach (Ed. by Lewis, B.), pp. 3-38. London: Academic
Press.

Mills, M. G. L. 1989. The comparative behavioral ecology of hyenas: the
importance of diet and food dispersion. In: Carnivore Behavior, Ecology,
and Evolution (Ed. by Gittleman, J. L.), pp. 125-142. Ithaca, New York:
Cornell University Press.

Mills, M. G. L. 1990. Kalahari Hyenas : The Comparative Behavioral Ecology of
Two Species. New York: Chapman 8 Hall.

Mills, M. G. L. 1996. Methodological Advances in Capture, Census, and Food-
Habits Studies of Large African Camivores. In: Carnivore behavior,
ecology, and evolution (Ed. by Gittleman, J. L.), pp. 223-242. Ithaca, New
York: Cornell University Press.

Mills, M. G. L., Juritz, J. M. 8 Zucchini, W. 2001. Estimating the Size of spotted
hyaena (Crocuta 'crocuta) populations through playback recordings
allowing for non-response. Animal Conservation, 4, 335-343.

Mitani, JO 8 Nishida, T. 1993. Contexts and social correlates of long-distance
calling by male chimpanzees. Animal Behaviour, 45, 735-746.

Morton, E. S. 1977. On the occurrence and significance of motivational-structural

rules in some bird and mammal sounds. American Naturalist, 111, 855-
869.

76

 

 

Norcross, J. L. 8 Newman, J. D. 1993. Context and Gender-Specific Differences
in the Acoustic Structure of Common Marrnoset (Callithrix-Jacchus) Phee
Calls. American Joumal of Primatology. 30, 37-54.

Norcross, J. L., Newman, J. D. 8 Cofrancesco, L. M. 1999. Context and sex
differences exist in the acoustic structure of phee calls by newly-paired
common marmosets (Callithrix jacchus). American Journal of Primatology,
49, 165-181.

Ogutu, J. O. 8 Dublin, H. T. 1998. The response of lions and spotted hyaenas to
sound playbacks as a technique for estimating population size. African
Journal of Ecology. 36, 83-95.

Owings, D. H. 8 Virginia, R. A. 1978. Alarm calls of California ground squirrels
(Spennophilus beecheyr). Zeitschrift Fur Tie/psychologie-Joumal of
Comparative Ethology. 46, 58-70.

Palombit, R. A., Cheney, D. 8 Seyfarth, R. 1999. Male gmnts as mediators of
social interaction with females in wild chacma baboons (Papio
cynocephalus ursinus). Beha viour, 136, 221 -242.

Peters, G. 8 Sliwa, A. 1997. Acoustic communication in the aardwolf, Proteles
cristatus (Camivora: Hyaenidae). Zeitschrift Fur Saugetierkunde-
Intemational Journal of Mammalian Biology, 62, 219-238.

Rendall, D., Seyfarth, R. M., Cheney, D. L. 8 Owren, M. J. 1999. The meaning
and function of grunt variants in baboons. Animal Behaviour, 57, 583-592.

Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution, 43, 223-225.

Robbins, R. L. 2000. Vocal communication in free-ranging African wild dogs
(Lycaon pictus). Behaviour, 137, 1271-1298.

Robisson, P. 1992. Roles of pitch and duration in the discrimination of the mate's
call in the king penguin Aptenodytes patagonicus. Bioacoustics, 4, 25—36.

Schassburger, R. M. 1993. Vocal communication in the timber wolf, Canis lupus,
Linnaeus. Berlin, Germany: Paul Parey Scientific Publishers.

Seyfarth, R. M. 8 Cheney, D. L. 1984. The Acoustic Features of Vervet Monkey
Grunts. Journal of the Acoustical Society of America, 75, 1 623-1 628.

Seyfarth, R. M., Cheney, D. L., Harcourt, A. H. 8 Stewart, K. J. 1994. The

acoustic features of gorilla double grunts and their relation to behavior.
American Journal of Primatology, 33, 31 -50.

77

 

Seyfarth, R. M., Cheney, D. L. 8 Marler, P. 1980a. Monkey responses to three
different alarm calls: evidence for predator classification and semantic
communication. Science, 210, 801-803.

Seyfarth, R. M., Cheney, D. L. 8 Marler, P. 1980b. Vervet monkey alarm calls:
semantic communication in a free-ranging primate. Animal Behaviour, 28,
1070-1094.

Smale, L., Frank, L. G. 8 Holekamp, K. E. 1993. Ontogeny of dominance in free-
living spotted hyaenas: Juvenile rank relations with adult females and
immigrant males. Animal Behaviour, 46, 467-477.

Smale, L., Nunes, S. 8 Holekamp, K. E. 1997. Sexually dimorphic dispersal in
mammals: Patterns, proximal causes, and consequences. In: Advances in
the Study of Behavior (Ed. by Slater, P. J. B., Rosenblatt, J. S., Snowdon,
C. T. 8 Milinski, M.). San Diego, California: HBJ Publishers, Inc.

Snowdon, C. T., Cleveland, J. 8 French, J. A. 1983. Responses to context- and
individual- specific cues .in cotton-top tamarin long calls. Animal
Behaviour, 31, 92-101 .

Specht, R. 2000. Avisoft-SASLab Pro. Berlin, Germany: Avisoft.

Szykman, M., Engh, A. L., Van Horn, R. 0., Funk, S. M., Scribner, K. T. 8
Holekamp, K. E. 2001. Association patterns among male and female
spotted hyenas (Crocuta crocuta) reflect male mate choice. Behavioral
Ecology and Sociobiology, 50, 231-238.

Teixidor, P. 8 Byrne, R. W. 1999. The "whinny" of spider monkeys: individual
recognition before Situational meaning. Behaviour, 136, 279-308.

Tilson, R. T. 8 Hamilton, W. J. I. 1984. Social dominance and feeding patterns of
spotted hyaenas. Animal Behaviour, 32, 715-724.

Warkentin, K. J., Keeley, A. T. H. 8 Hare, J. F. 2001. Repetitive calls of juvenile
Richardson's ground squirrels (Spen'nophilus richardsonii) communicate
response urgency. Canadian Journal of Zoology-Revue Canadienne De
Zoologie, 79, 569-573.

Waser, P. M. 8 Brown, C. H. 1986. Habitat acoustics and primate
communication. American Joumal of Primatology, 10, 135-154.

Waser, P. M. 8 Waser, M. S. 1977. Experimental studies of primate

vocalizations: specializations for long-distance propagation. Zeitschrift Fur
Tierpsychologie-Joumal of Comparative Ethology. 43, 239-263.

78

 

Whateley, A. 8 Brooks, P. M. 1978. Numbers and movements of spotted
hyaenas in Hluhluwe Game Reserve. The Lammergeyer, 26, 44-52.

Wiley, R. H. 8 Richards, D. B. 1978. Physical constraints on acoustic
communication in the atmosphere: implications for the evolution of animal
vocalizations. Behavioral Ecology and Sociobiology, 3, 69-94.

Wong, J., Stewart, P. D. 8 Macdonald, D. W. 1999. Vocal repertoire in the
European badger (Meles meles): Structure, context, and function. Journal
of Mammalogy, 80, 570-588.

Zahavi, A. 1979. Ritualization and the evolution of movement signals. Behaviour,
72, 77-81.

79

   

Ilililljllllllljllljljlllllsllllljll