Citation
An Analysis of vocalizations of three species of east African cercopithecidae

Material Information

Title:
An Analysis of vocalizations of three species of east African cercopithecidae
Creator:
Walek, Mary Louise, 1946- ( Dissertant )
Maples, William R. ( Thesis advisor )
Fairbanks, Charles H. ( Reviewer )
Markel, Norman N. ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
1972
Language:
English
Physical Description:
xii, 109 leaves. : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Baboons ( jstor )
Chirp ( jstor )
Communications ( jstor )
Forests ( jstor )
Infants ( jstor )
Monkeys ( jstor )
Primates ( jstor )
Signals ( jstor )
Species ( jstor )
Vocalization ( jstor )
Animal communication ( lcsh )
Anthropology thesis Ph. D ( lcsh )
Dissertations, Academic -- Anthropology -- UF ( lcsh )
Sound production by animals ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis -- University of Florida.
Bibliography:
Bibliography: leaves 103-108.
Additional Physical Form:
Also available on World Wide Web
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
022667536 ( AlephBibNum )
13982886 ( OCLC )
ADA5173 ( NOTIS )

Downloads

This item has the following downloads:


Full Text














AN ANALYSIS OF VOCALIZATIONS
OF THREE SPECIES OF EAST AFRICAN CERCOPITHECIDAE







By




iMARY LOUISE WALEK


A DISSERTATION PRESENTED TO THE GRADUATE
COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA
1972



























Copyright by


MEary Louise ",'.alek


--972




























With Love
to
Mom, Dad, and Bruce












AC K? 0', LEDGM ENTS


I would like to thank Dr. William R. Maples, chairman

of my supervisory committee, for his guidance and criticism,

and for his assistance in making this study possible. It was

he who made my trip to East Africa a reality and for this I

will always be grateful. Appreciation must also be extended

to the other members of the committee, Drs. Norman N. Market

and Charles H. Fairbanks, for taking time from their busy

schedules to read and evaluate this study,

I would also like to thank Dr. Arnold Paige of the

Departments of Speech and Electrical Engineering of the Uni-

versity of Florida for giving unselfishly of his time and

for his generosity in making the sonograph and other mater-

ials accessible to me. Without his kind cooperation this

study would not have been possible.

For the use of their laboratory and additional tape

recorders during the editing process, I would like to thank

the Department of Communication Sciences, also of the Univer-

sity of Florida.

To Bruce Walek I extend sincere appreciation for the

photographic reproduction of the sonograms, even though it

meant taking valuable time from his own studies to do so.

Finally, I would like to express my gratitude to the

members of my family for their support and encouragement







which contributed as much to the completion of this sTudy as

the technical assistance received elsewhere.

The study was financed by grants to William R. Maples

from Biomedical Sciences of the University of Florida and

the National Science Foundation (GS 2610).















TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS . . . . . . .

LIST OF TABLES . . . . . . .

LIST OF FIGURES . . . . . .


ABSTRACT

CHAPTER


. * * . . * . 0


. . . .


Jjjj


* . t

* . .


I INTRODUCTION . . . .

The Study of Primate Vocalizations
Objectives of the Present Study
The Study Sites . . .
Methods and Materials of Research
Definition of Terms . . .

II VOCALIZATIONS OF THE SYKES MONKEY

Classification of Calls ...
Discrete and Graded Signals . .
A Structural-Functional Analysis
Taxonomic Implications . . .


III VOCALIZATIONS OF THE COLOBUS MONKEY

Classification of Calls .

IV VOCALIZATIONS OF THE BABOON . .

Vocalizations and Crop-Raiding .
Classification of Calls . . .

V SUMMARY AND CONCLUSIONS . .

The Sykes Study . . . . .
The Colobus Study . . . .
The Baboon Study . . . . .


. . . 1

. . . 2

. . . 6
. . . 7

. . . 7
7
. .


. . . .


* 61

. . . 70

. . . 70
. . . 72


. . 77


PLATES


SYKES MONKEY . . . . . .


e .

e f
t e
* .







TABLE OF CONTENTS (Continued)


Page
PLATES

COLOBUS MONKEY . . . . . . . 95

BABOON . . . . . . . . . . 100

REFERENCES . . . . . . . . . 103

BIOGRAPHICAL SKETCH . . . . . . . . .. 109












































vii














LIST OF TABLES


Table

1

2


Chirp

Multiple


3 Nyah


4 Short N

5 Boom

6 Squeal

7 Shriek

8 Chut

9 Chutter

10 Low, Sh

11 Chutter

12 Growl

13 Err

14 Uh .

15 Snort

16 Snort-C


e . 0 .

e Chirps



yah ..

* * .










ort Chutter

-Squeal .










roak . .


17 Gecker White


Inf


a:


Err White Infant

Gecker Juvenile

Two-Phase Bark .

Bark . . . .

Yak . . . .


viii


* C C
* . C









* C *

* C C





. .C.
* * *
* . .

. . .

. . .















* * *








nt .



* . *



* *

* . .


Page

14

16

18

19

21

22

24

25

27

28

30

32

34,

35

61

64

66

67

68

73

74

76













LIST OF FIGURES


Figure

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Chirp . . .

Chir . . .

Two-Unit Chirp .

Two-Unit Chirp .

Three-Unit Chirp .

Nyah . . .

Short Nyah . ..

Short Nyah . .

Boo . . . .

Scueal . . .

Shriek . . .

Chut . . . .

Chut . . . .

Chutter . . .

Low, Short Chutter

Chutter-Squeal .

Chutter-Squeal (coi

Chutter-Squeal (co

Growl . . .

Growl . . .

Err . . .

Uh . . . .


* *


ntinued)

ntinued)





* .


Page

84

84

85

85

86

86

87

87

88

88

89

89

90

90

91

91

92

92

93

93

94

94






LIST OF FIGURES (Continued)

Figure Page

23 Snort .. .... .. ... 96

2 Snort . . . . . . . . . . 96

25 Snort-Croak . . . . . . 97

26 Snort-Croak . . . .. . . . 97

27 Snort-Croak ... . . . . . . 97

28 Gecker White Infant . . . . . 98

29 Err White Infant . . . . . . 98

30 Gecker Juvenile . . . . . . 99

31 Gecker Juvenile ... . . . . . 99

32 Two-Phase Bark . . . . . . . 101

33 Two-Phase Bark . . . . . . . 101

34 Bark I . . . . . . . . 102

35 Bark I, Yak, Bark II . . . . . . 102






Abstract of Dissertation Presented tc the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



AN ANALYSIS OF VOCALIZATIONS
OF THREE SPECIES OF EAST AFRICAN CERCOPITHECIDAE


By

Mary Louise Walek

August, 1972

Chairman: William R. Maples
Major Department: Anthropology


Vocalizations of three species of Old World monkeys

(Sykes monkey, Cercopithecus mitis; colobus monkey, Colobus

oolvkomos; and baboon, Papio cynocephalus) were collected in

Kenya, East Africa. Sonograms of the vocalizations were made

and analyzed in terms of their physical characteristics (fre-

quency, duration, and tonality). Representative sonograms

are illustrated and tables of all measurements made are pre-

sented. Accompanying behavioral information is also included

whenever available.

Fourteen Sykes vocalizations are catalogued and de-

scribed. An analysis of the calls in terms of discreteness

and gradedness revealed that most of the vocalizations were

of the discrete type.

The Sykes vocalizations were also examined to see if

physical properties, or structure, could be related to func-

tion. It was found that those calls transmitted over dis-

tances which were important for group cohesion and alarm or

xi






threat were structured so as to facilitate locating the caller.

Calls operating at close-range did not have such localizing

properties.

The question of whether Cerconithecus mitis is con-

specific with C. nictitans has been raised in the literature.

This study evaluated the taxonomic status of these two species.

Based on similarities and differences in the vocalizations it

was concluded that separate specific identity should be main-

tained. However, C. mitis was found to be more closely re-

lated to C. nictitans and C. erythrotis-ceDhus than to the

mona superspecies.

Five vocalizations of the colobus monkey are described:

two for infants, one for juveniles, and two for adults. A

nonvocal behavior pattern, the jumping display, was found to

be of communicative significance in alerting a troop to dan-

ger and in indicating the direction of troop movement. It

also functioned as an intimidation display directed at the

observer.

The role which vocalizations play in the crop-raiding

activities of baboons was investigated and the apparent

"decoy" effect of the calls in distracting the attention of

Africans protecting the fields was examined. It was concluded

that the vocalizations did not represent a deliberate and

planned decoy tactic, but instead were a fortuitously diver-

sionary behavior which contributed incidently 7o the success

of the raids. Four types of calls given during the raids

are described.


xii













CHAPTER I


INTRODUCTION



The study of primate communication is assuming an

increasingly important role in the field of primatology. As

more is learned of the behavior of man's closest relatives,

the nonhuman primates, the more apparent it has become that

understanding how behavior patterns are transmitted and in-

terpreted will contribute considerably to understanding the

adaptive mechanisms of social life. Many of the earlier

studies on primate behavior, and even some of the more recent,

either failed to include communication as an aspect of behav-

ior to be investigated or else dealt with it in such a cursory

and unscientific manner as to make it almost worthless for

further investigation and comparison. Part of the problem

is no doubt related to the all-pervasive nature of communica-

tion into all aspects of social life, a fact which is well

illustrated by the following definition: "...social commun-

ication is the process by which the behavior of an individual

affects the behavior of others" (Altmann, 1967:326).

Communication as it has been defined can include many

different forms of behavior occurring in concert and involving

several individuals. It becomes apparent, therefore, that

the study of communication can be a complex undertaking.

1







Furthe- complications become apparent when one considers

that communication can take place via several different chan-

nels. Signals or messages may be olfactory, tactile, visual,

or auditory in nature and for any specific signal transmission

several of these channels of communication may be used simul-

taneously. Each of these channels has been investigated in

studies of communication among the primates (see Andrew,

1963a, 1964; Bolwig, 1964; Hinde and Rowell, 1962; Jolly,

1966; Moynihan, 1967; van Hooff, 1962).



The Study of Primate Vocalizations


The primary concern of the present study, however,

will be with primate vocal communication, the study of which

has taken various forms.

One approach consisted of transcribing the various

calls by using letters of the alphabet in order to give as

close an approximation as possible of the sound of the call

(for an example see Ullrich, 1961). Not only was this sys-

tem largely subjective in nature, but the alphabetical nota-

tion used was generally that of the language of the observer

and hence was not accessible to, or reproducible by, researchers

of all nationalities. In order to overcome some of the sub-

jectivity of such an approach and to insure a certain amount

of uniformity, there was an attempt to use the standard

International Phonetic Alphabet (for an example see Hill and

Booth, 1)57). Howv-er, it soon became apparent that non-

human priates were capable of producing sounds which





3
lacked counterparts in human speech. Ph:retic and alphabet-

ical transcriptions, therefore, could not render an accurate

description of many vocalizations and often gave a false,

human quality to the vocal patterns of primates other than

man.

The most common approach to the study of primate

vocal communication until recently has been the assignment

of a descriptive name to a vocalization, one which depicts

as closely as possible the type of call given. In this way

such terms as 'squeal', 'click', and 'grunt' were used but

were really only of value when as complete an acoustic de-

scription as possible accompanied the assignment of such a

term. Moreover, adjectives such as 'high-pitched', 'shrill',

'harsh', etc., used in such descriptions, were still ambig-

uous and subjective in nature. Hence, comparisons of vocal-

izations reported in different studies remained difficult,

if not impossible, to carry out.

The problem of subjectivity was alleviated to a con-

siderable extent with the development within the last two

decades of sophisticated, sensitive recording devices and

the invention of the sound spectrograph. The spectrogram,

or sonogram, supplies a permanent, visual record of a sound

and serves both as a form of notation and as a method of

precise measurement. Although descriptive terms remain prev-

alent as names or labels for the various calls, they are now

supported by a more concrete, objective, physical description

than was possible before.





4
Receiving its impetus largely fom studies on insects,

anurans, and birds, the use of sonograms in the study of pri-

mate vocalizations has spread rapidly within the last several

years. Most of the work on primate communication has been

with species of Old World monkeys, especially the macaques,

baboons, and vervets. Rowell (1962) used sonograms and de-

scribed nine agonistic noises of the rhesus macaque, relating

each to the various postures and expressions which they ac-

company. Struhsaker (1967) has done an excellent spectro-

graphic analysis of vervet vocalizations combining a detailed

physical description of the calls with behavioral data when-

ever possible. Although numerous studies on baboon communi-

cation have been reported, sonograms for this species are

apparently lacking in the published literature. Communica-

tion studies involving spectrographic analysis have also been

reported on several other species, such as those of Reynolds

(in Marler, 1965) on the chimpanzee, Schaller (1963) on the

gorilla, Moynihan (1964) on the night monkey, Altmann (1967)

on howlers, Ploog (1966) on squirrel monkeys, Petter and

Petter (1966) on the aye-aye, and Petter (1962) on lemurs.

For a more complete listing of studies on primate communica-

tion, spectrographic or otherwise, the reader is referred to

Altmann's (1968) comprehensive review.

Although Old World monkeys have been extensively

studied in terms of behavior and communication, several

genera within the family Cercopithecidae have not been ade-

quately investigated. Among these are the species of the

genera Cercopithecus and Colobus.






Other than Struhsaker's (1967) study on the vervet,

Cercopithecus aethiops, and most recently his investigations

of the vocalizations of several Cercoithecus species (1970),

very little information on vocalizations is available for the

Cercopithecus group. Most of what is known of the auditory

communication of these species has been drawn out of more gen-

eral behavioral studies or from anecdotal accounts. Other

researchers on the vervet include Booth (1962), Gartlan and

Brain (1968), and Hall and Gartlan (1965). None of these

studies presents spectrographic descriptions. The redtail

monkey, C. ascanius, is the only other species investigated

to any extent and the reader is referred to the works of

Buxton (1952) and Haddow (1952). Again, spectrographic data

is not available for this species in these studies.

Very little is kno'n of the behavior of the Colobus
species. While considerable research has been carried out

on the Asiatic counterpart of the Colobinae, the langurs, the

African members of this subfamily have been largely ignored.

Available reports include those of Booth (1957) on the olive

colobus and Hill and Booth (1957) and Ullrich (1961) on the

black and white colobus. However, information as to social

communication is rather sparse and incomplete in these

studies, A more informative sTudy is Marler's (1970) spec-

trographic analysis of the vocalizations of the red colobus.





0
Objectives of the Present Study


In the present study the vocalizations of two species

of these relatively little-studied genera, the Sykes monkey,

Cercooithecus mitis kibonotensis, and the black and white

colobus monkey, Colobus Dolykomos Dalliatus, will be described

and analyzed. The basic aim of this study is to present a

catalogue of the vocal repertoire of these species. Accom-

panying behavioral information will also be presented when-

ever available.

The vocalizations of the baboon, Papio cynocechalus,

will also be described. This investigation formed part of

a larger study being conducted at the time by William R.

Maples, William F. Greenhood, and myself on crop-raiding be-

havior of baboons. The aim of this portion of that study is

to describe those vocalizations given while crop-raiding and

to investigate the role they play in this activity.

All descriptions and classifications of the calls are

based on spectrographic analysis.



The Study Sites


The field observations upon which the following study

is based were conducted in Kenya, East Africa, from September

to December, 1969. The study site for the Sykes and colobus

monkey material was at Diani Beach on the Kenya coast, approx-

imately eighteen miles south of rPiombassa. The site itself

was approximately one-quarter to one-half mile inland. The





7
material on baboon vocalizations was collected at farm sites

primarily at Shimoni, a small fishing village near the

southern tip of the Kenya coast. Some recordings were also

made near Roka along the Mombassa-::Ialindi road, approximately

twenty-four miles south of Malindi.

In general, the climax vegetation community for the

Diani Beach/Shimoni area has been described as lowland dry

forest on coral rag (Combretum schumanii-Cassipourea species)

(Moomaw, 1960). The forests at Diani Beach are broken into

smaller segments of varying size by the interspersal of

lantana thicket and cultivated land. The forest essentially

consists of two layers, an upper canopy composed of tall,

branching trees and a low, dense understory where bushes and

vines predominate. The colobus were observed to favor the

upper canopy but would readily take refuge in the lower layer

when necessary. The Sykes seemed equally at home in both

layers but it was this observer's impression that more time

was spent in the understory. At Roka, the baboons inhabited

a protected forest which has been classified as lowland

woodland (Brachystegia-Afzelia species) by Moomaw (1960).

For a more detailed discussion of the ecology of these areas

and the species of vegetation which predominate, the reader

is referred to Moomaw (1960) and Greenhood (1971).



Methods and Materials of Research


Recordings of the vocalizations ;were made with a

Nagra IV L portable tape recorder at 15 ips with a frequency






response of 30 to 20,000 cps, and a Sennheiser directional

microphone, Miodel MKH 804. Sonograms for analysis were made

on a Kay Electric Sona-Graph, .iIodel 6061-B. A transparent

plastic grid overlay was made showing time on the horizontal

axis and frequency on the vertical axis. The grid vwas placed

over the sonograms and approximate frequency and temporal

measurements were taken.

Field recording techniques at Diani Beach consisted

of approaching as closely and quietly as possible on foot a

Sykes or colobus troop. The thick, tangled vegetation of

this area precluded any possibility of following a troop on

its daily movements; therefore, recordings were made along

roads and footpaths vhich troops regularly frequented and

traversed. At Shimoni, a stationary recording site was set

up within the farm and baboon vocalizations were recorded as

the troops approached and entered the field to raid. The

recording station at the Roka site was within an observation

vehicle which would follow and remain adjacent to the move-

ments of a troop of crop-raiding baboons as they emerged from

the forest and crossed the Mi'alindi-Mombassa road to raid the

farms on the other side. In all cases, accompanying behavioral

observations were recorded in a notebook.

In areas of limited visibility it was the observation

of behavioral context which proved to be the most difficult.

A single individual attempting to operate a tape recorder,

direct a microphone, maintain visual contact with a troop

moving in dense underbrush, and take notes is limited as co






the amount of behavior which can be observed. This linita-

tion was particularly felt in the Diani Beach area. At Shi-

moni and Roka the problem was not as acute as other workers

(notably William R. Maples) were simultaneously observing

crop-raiding behavior and so correlation of the vocalizations

with their observations was possible. It would seem, there-

fore, that perhaps the most fruitful means of studying com-

munication with all its behavioral ramifications within a

somewhat hostile environment would be through a team approach.

Such a system would minimize the physical limitations and

would facilitate the recording of observations. In addition,

it would provide a system of checks and counter-checks on

what was observed. Perhaps in savanna areas, with visibility

relatively unimpaired, these difficulties do not pose as seri-

ous a hinderance as when dealing with forest populations.

In the latter case, the physical limitations imposed by the

very nature of the study itself are only too apparent.



Definition of Terms


So that there may be some form of common reference in

discussions of primate vocalizations, use will be made here

of the terms defined by Thomas T. Struhsaker in his study on

auditory communication of vervet monkeys. These are as

follows:

Unit: The unit is the basic element of a call, and
is represented as a continuous tracing along the
temporal (horizontal) axis of the sonogram.






Phrase: The phrase is a group of units that is
separated from other similar groups by a time in-
terval greater than any time interval separating
the units within a phrase.

Bout: A bout is a grouping of one or more phrases
separated from other similar groupings by a time
interval greater than that separating any of the
phrases within a bout.

Nontonal unit: A nontonal unit is composed of
sound that is more or less continuously developed
over a wide range of frequencies. This has also
been called "noise" by Andrew (1963b) and "harsh
noises" by Rowell and Hinde (1962).

Tonal unit: A tonal unit is composed of sound
characterized by one or more relatively narrow fre-
quency bands and has been referred to as "clear calls"
by Rowell and Hinde (1962) and "sound" by Andrew
(1963b). Units with a harmonic structure are in-
cluded in this category.

Compound unit: A compound unit is composed of both
nontonal and tonal sounds that appear as a continu-
ous tracing on the sonogram.

Mixed unit: Units composed of both tonal and non-
tonal sounds that are rather superimposed on one
another are called mixed units. The tonal and non-
tonal aspects are more or less separated by differ-
ences in frequency.

Distribution of major energy of nontonal sounds:
The distribution of the major energy of a nontonal
sound is represented and thus determined by the
darkest portion of the tracing on the sonogram.
This distribution is generally over a smaller range
than the frequency range of nontonal sounds. The
distribution of major energy of nontonal sounds is
described by an upper frequency (highest pitch of
major energy) and a lower frequency (lowest pitch
of major energy). (Struhsaker, 1967:282-283)

For a brief demonstration of how these terms will be

applied in the present study, the reader is referred to

Figure 26 and Figure 6.

Figure 26 is a spectrographic representation of a

colobus vocalization consisting of three different phrases.






The first is a nontonal, single-unit phrase and is given

only once at the beginning of the call. The second phrase

consists of several tonal units. The third is a nontonal,

single-unit phrase very similar to the first. The second

and third phrases are repeated in alternation. All three

phrases are said to form a bout, part of which is depicted in

the sonogram. The entire bout can be referred to as a "call"

or "vocalization".

Figure 6 represents a call of Cercopithecus mitis

and can be described as a single-unit phrase. It is also a

compound unit since a brief tonal component consisting of

gently arching frequency bands is followed by a noise or

nontonal component with a somewhat irregular appearance on

the soncgram.

For a further demonstration of how these terms are

used, the reader is referred to the descriptions of the vari-

ous calls presented in the text which can be compared with

the appropriate sonogram.












CHAPTER II


VOCALIZATIONS OF THE SYKES MONKEY



The Sykes monkeys (Cercopithecus mitis kibonotensis)

live in social groups of substantial size. Various troop

counts, made while the monkeys passed over an "arboreal

bridge" spanning a roadway, revealed totals of 13, 18, 26,

and 31. However, it never was clear how many troops were in-

volved or if an entire troop had been counted. Sykes are

primarily an arboreal species but, as mentioned previously,

seem to be equally at home in the tall trees as in the lower

brush. They seem not the least bit reluctant to move along

on the ground, especially in those areas where the vines are

quite thick and tangled.

The Sykes would vocalize quite readily. The chirp

was the most frequently heard call and would be elicited by

any sudden movement or noise, or by the approach of humans

or dogs. In the initial stages of the study, the observer

was constantly the object of much threat behavior and chirping.

However, as the Sykes became habituated to the observer's

presence and with the thickening of the foliage due to rain,

the amount of vocalizations given was considerably reduced.







Classification of Calls


Chirp

The chirp (Figs. 1&2 and Table 1) was given by adult

females and immature monkeys and possibly by adult males,

although positive identification of the latter was not possible.

That adult male Sykes might give the call is doubtful in light

of Struhsaker's (1967, 1970) reports for the vervet and other

Cercopithecus species in which he noted that chirps were not

given by adult males.

The chirp was an abrupt, sharp-sounding call of high

frequency and short duration. The frequency range was 0.10 -

16.0+ kHz but the average distribution of the major concentra-

tion of energy was 2.91 6.14 kHz (N=29). Sonograms re-

vealed a single-unit phrase with an abrupt beginning consist-

ing of a narrow, tonal, chevron-shaped component followed by

a noise component of longer duration. The noise component of

approximately half of the chirps was distributed in a two-

band arrangement, while the other half consisted of a single,

broader band (Figs. 1&2). The mean duration of the chirp was

0.18 seconds (N=29).

The chirp apparently functions as a threat-alarm call.

IT is usually given in conjunction with a threat display in

which the monkey either crouches forward with legs stiff and

bobs its head and shoulders up and down, or else lunges for-

wvrd on its forelimbs. In both cases the monkey stares in-

Sently at the intruder. Thic behavior was directed toward

do-s n. lunans but never was observed between one Sykes and





1L
another. A similar observation was made by Haddow (1952) in

his study on the redtail monkey, Cercopithecus ascanius.

TABLE 1

CHIRP


Number (N) Mean (x) Range (R)

Units/Phrase 29/29 1.00 1.00-1.00

Phrase Duration 29 0.18 0.10-0.25
(sees.)

Lower Limit of
Major Energy 29 2.91 1.50-4.00
(kHz)

Upper Limit of
Major Energy 29 6.14 3.00-8.00
(kHz)



Chirps were also given by a feeding animal, the

feeding in some instances being an apparent nervous displace-

ment activity but on other occasions a true feeding activity.

The frequency of repetition of the chirps was vari-

able as was the intensity of the call. Some chirps were

given loudly and in rapid succession; for example in re-

sponse to a loud, sudden noise or to a dog passing nearby.

On other less tense occasions chirps were given infrequently

and softly, almost whisper-like.

Other Sykes responded to the chirping animal by look-

ing in its direction and in the direction of its threat and

by chirping also. Occasionally the other animals would move

deeper into the brush but this was not necessarily a hurried





15
retreat. In some instances the only Sykes remaining visible

was the chirping individual who would soon move off and join

the others.


Multiple Chirps

Included in the Sykes' repertoire in addition to the

single-unit chirps are two- and three-unit chirps (Figs. 3-5

and Table 2). Such multiple chirps were also reported by

Struhsaker (1970) for Cercopithecus cephus, C. erythrotis,

and C. nictitans. As is the case for single chirps, multiple

chirps are apparently restricted to adult females and imma-

ture monkeys.

In the two-unit chirp, the distribution of major

energy in the second unit is over a greater range (2.84 -

5.96 kHz) than in the first (3.80 5.96 kHz). This distinc-

tion was also made by Struhsaker (1970) for the three Cerco-

pithecus species listed above. Struhsaker found that this

shorter frequency range of the first unit was due both to a

higher lowest frequency and a lower upper frequency. However,

in the case of C. mitis only the lowest frequencies were

found to differ considerably. That is, in 88.9 percent of

the double chirps the lowest frequency of Unit I was higher

than the lowest frequency of Unit II, whereas the upper fre-

quencies of the two units were the same in 66.7 percent of

the calls (N=9).

Generalizations concerning energy distribution for

the three units of the triple chirp can not be made since

only one such call was recorded. However, in this one case

energy distribution increased with each subsequent unit.






TABLE 2

MULTIPLE CHIRP


Number (N) Mean (R) Range (R)

Units/Phrase
Double 18/9 2.00 2.00-2.00
Triple 3/1 3.00 3.00-3.00

Unit Duration
secss.)
Double Unit I 9 0.06 0.05-0.08
Unit II 9 0.20 0.12-0.28
Triple Unit I 1 0.05 0.05-0.05
Unit II 1 0.08 0.08-0.08
Unit III 1 0.25 0.25-0.25
Interval Between
Units secss.)
Double 9 0.00 0.00-0.00
Triple 2 0.03 0.05-0.01

Phrase Duration
secss.)
Double 9 0.25 0.18-0.32
Triple 1 0.44 0.44-0.44

Lower Limit of
Major Energy
(kHz)
Double Unit I 9 3.80 3.00-5.00
Unit II 9 2.84 2.00-3.60

Triple Unit I 1 2.00 2.00-2.00
Unit II 1 2.00 2.00-2.00
Unit III 1 2.00 2.00-2.00

Upper Limit of
Major Energy
(kHz)
Double Unit I 9 5.96 5.50-7.00
Unit II 9 5.96 5.00-7.00

Triple Unit I 1 3.00 3.00-3.00
Unit II 1 3.50 3.50-3.50
Unit III 1 5.00 5.00-5.00





17
In addition to frequency differences, the durations

of the two units of the double chirp also differ, with the

first unit being shorter than the second in all cases. The

durations of the units of the three-unit chirp also increase

from first to third, but again any generalization must await

a larger sample.

It is interesting to note that in both energy dis-

tribution and duration the final units of both the two- and

three-unit chirps fall within the range of values for the

single-unit chirp.


Nyah and Short Nyah

The nyah (Fig. 6 and Table 3) is a loud call consist-

ing of tonal and nontonal components. Although the call

itself consists of a single-unit phrase, it always occurred

in bouts in which it was repeated at intervals of approxi-

mately 0.50 seconds.

On the sonogram the nyah appears as a harmonic series

of bands beginning abruptly and arching into a mixed band of

tonality and noise. In a variant of the nyah, referred to

here as the short nyah (Figs. 7&8 and Table 4), this harmonic

band pattern at the beginning is absent, the call, therefore,

sounding like a sharp "ah". Both the nyah and the short nyah

can be given in a bout, but it seems the latter was usually

given under tense or excitable conditions, such as a pack of

stray dogs passing nearby.

Mean duration for the nyah was 0<31 seconds (N=12)

and 0.19 seconds for the short nyah (N=14). The average






energy distribution for both forms vw.as quite similar, with

0.82 5.46 kHz for the nyah (N=15) and 0.57 5.31 kHz for

the short form (N=l4). Total energy distribution ranged

from 0.10 16.0+ kHz for both calls. Since it was never

known exactly how many Sykes groups were in the area and since

individual groups were not distinguishable, it is possible

that all the nyahs of the sample were made by the same indi-

vidual. This possibility is based on the assumption that there

is but one adult male per group, a common occurrence among

Cercopithecus monkeys according to Struhsaker (1969). The

short nyahs of the sample, however, were given by one indi-

vidual only since they were all recorded on a single occa-

sion with only one troop present.

TABLE 3

NYAH


Number (N) Mean (x) Range (R)

Units/Phrase 15/15 1.00 1.00-1.00

Phrase Duration 12 0.31 0.20-0.38
secss.)

Lower Limit of
Major Energy 15 0.82 0.25-1.80
(kHz)

Upper Limit of
Major Energy 15 5.46 4.00-7.00
(kHz)






TABLE 4

SHORT NYAH


Number (N) Mean (x) Range (R)

Units/Phrase 14/14 1.00 1.00-1.00

Phrase Duration 14 0.19 0.05-0.28
secss.)

Lower Limit of
Major Energy 14 0.57 0.20-2.30
(kHz)

Upper Limit of
Major Energy 14 5.31 4.00-6.50
(kHz)


The animal giving the call was very difficult to
observe; in fact, this was done on only one occasion and quite

by accident. A single Sykes was observed sitting alone in

the topmost branches of a tall tree giving the call at about

18:00. No other Sykes were visible in the immediate area.

The call was made by exhaling forcibly with the mouth open.

This individual was too far away to permit accurate sexual

identification but its large size would indicate it to be

most likely a male. This assumption is further supported by

Haddow's (1952) observation of the redtail monkey, C. ascanius,

in which only the adult male gives a call Haddow described as

"Kyuh". In addition, the nyah and short nyah of the Sykes

are probably equivalent to the "loud calls" reported by

Struhsaker (1970) and Bourliere, Hunkeler, and Bertrand (1970)

for other Cercopithecus species which were also only given by

adult males.






Struhsaker's (1970) "loud calls" are of two types,

the Dow and the hack. Since he does not make explicit the

basis for distinguishing them spectrographically, and since

he does not do so in his table of measurements, it is impos-

sible to compare or correlate one or the other with the nyah

or short nyah.

The nyah call was frequently given prior to group

progression. It may have signaled group withdrawal, for soon

after the call was given no Sykes would be visible in the area,

having moved in the direction from which the call came. In

addition, such calls seem to be evoked by a potentially dis-

ruptive or threatening stimulus as was the case with the pack

of dogs. Bourliere, Hunkeler, and Bertrand report that loud

calls are given by the alpha male in the evening at the

sleeping trees and also "in response to any situation that

was potentially dangerous for the group members, or potenti-

ally disruptive to the cohesion of the group" (1970:313).

Haddow (1952) and Struhsaker (1969) concluded that the loud

calls may function as a central focal point for maintaining

intragroup cohesion on occasions when the group might be dis-

persed and may also function in intergroup spacing.


Boom

On several occasions a low frequency call was heard

which sounded very much like "whoo" but which closely resem-

bles in frequency and duration the booms of the mona super-

species as described by Struhsaker (1970). The boom (Fig. 9

and Table 5) is completely tonal and appears as a very





21
black band situated at the bottom of the sonogram. The call

is given as a single-unit phrase, is seldom repeated, and

appears to be strikingly uniform.

The energy distribution is restricted to such a low

frequency that the lowest distribution of energy on the sono-

gram is difficult to read, being so close to the bottom mar-

gin of the sonogram itself. This lower limit of the energy

range has, therefore, been approximated at 0.01 kHz. The

upper limit of energy distribution is 0.40 kHz (N=2).


TABLE 5

BOOM


Number (N) Mean (x) Range (R)

Units/Phrase 2/2 1.00 1.00-1.00

Phrase Duration 2 0.20 0.20-0.21
secss.)

Lower Limit of
Energy (kHz) 2 0.01 0.01-0.01

Upper Limit of
Energy (kHz) 2 0.40 0.40-0.40




The individual giving the call was never seen, nor

did the animal sound as if it were very close to the observer.

However, among those species (Cercopithecus mona, C. campbelli,

and C. Dogonias) which Struhsaker (1970) studied that had

boom calls, only adult males gave the call.

Apparently, the boom was involved in some way with

coordinating group movement as it was usually given while the






group was moving, or else the group would move soon there-

after. On several occasions the boom w''as heard following

agonistic encounters within the group and so may serve to

facilitate group cohesion by overriding the disruptive and

dispersive effects of antagonism among its members.


Saueal

The squeal (Fig. 10 and Table 6) was a high-pitched,

uniform call. It is entirely tonal and occurs as a single-

unit phrase. On the sonogram the squeal appears as two

slightly undulating bands, one at approximately 3.00 4.00

kHz and the other at 6.00 7.50 kHz. The end of the call

lies on a slightly lower frequency level than the beginning.

TABLE 6

SQUEAL


Number (N) Mean (x) Range (R)

Units/Phrase 7/7 1.00 1.00-1.00

Phrase Duration 7 0.63 0.39-1.20
secss.)

Lower Limit of
Energy (kHz) 7 3.20 2.70-3.50

Upper Limit of
Energy (kHz) 7 7.87 7.00-9.00



The mean duration of the squeal is 0.63 seconds (N=7).

The mean lower frequency is 3.20 ikz and the mean upper fre-

quency is 7.87 kHz (N=7).






Shriek

The shriek (Fig. 11 and Table 7) was a high-pitched,

shrill-sounding call of variable duration. Certain sections

of the call vary in degree of tonality and duration. Since

it appears as a continuous tracing on the sonogram it must be

classed as a compound, single-unit phrase.

The first part of the shriek resembles the squeal and

is completely tonal with a two-band appearance on the sonogram.

However, it is somewhat higher in pitch than is the squeal,

with a mean frequency of 3.46 8.74 kHz (N=5). This first

part continues into, and is superimposed by, a mixed tonal-

nontonal portion which appears as a noisier, broader, double

band with a mean frequency of 3.20 8.64 kHz (N=5). As the

call progresses, this portion may become noisier and its

appearance will become more distorted on the sonogram.

The mean duration of the tonal component is 0.36
seconds (N=5), while that of the mixed portion is 0.65 seconds

(N=5) for a mean total phrase length of 1.01 seconds.

Chut

On one occasion a call was given which can perhaps

best be described as sounding like "chut" (Figs. 12&13 and

Table 8). There were three such chuts or units grouped to-

gether to form a phrase and two phrases were given. There is

some tonal quality to the call but this is overlapped for the

most part by a nontonal component (thus forming a mixed unit)

giving the call a noisy appearance on the sonogram.















TABLE 7

SHRIEK


Number (N) Mean (x) Range (R)

Units/Phrase 5/5 1.00 1.00-1.00

Tonal Portion

Duration (sees.) 5 0.36 0.13-0.55

Lower Limit of
Energy (kHz) 5 3.46 3.10-4.00

Upper Limit of
Energy (kHz) 5 8.47 7.50-9.50

Mixed Portion

Duration secss.) 5 0.65 0.27-0.99

Lower Limit of
Major Energy 5 3.20 3.00-3.50
(kHz)

Upper Limit of
Major Energy 5 8.64 8.00-9.00
(kHi z)





25
The mean unit duration was 0.08 seconds (N=6) and the

average distribution of major energy was 2.95 4.53 hHz (N=6).

However, of the three units the third was the longest (0.12

seconds) and had a greater frequency range. The upper limit

of the total energy range of this third unit was approximately

13.0 kHz. On the sonogram the first unit appears as a single

band while the second and third units consist of two bands.

However, the second of these bands is difficult to discern

and measure since it is considerably obscured by the noise

component. Although there are some spectrographic similari-

ties with the chutter, the calls differ enough acoustically

to warrant their separate classification at this time.

TABLE 8

CHUT


Number (N) M[ean (x) Range (R)

Units/Phrase 6/2 3.00 3.00-3.00

Unit Duration 6 0.08 0.03-0.13
secss.)

Interval Between
Units secss.) 4 0.12 0.05-0.20

Phrase Duration 2 0.50 0.49-0.50
secss.)

Lower Limit of
Major Energy 6 2.95 2.50-3.50
(kHz)

Upper Limit of
Major Energy 6 4.53 3.20-5.50
(kHz)






The chut was given by either a subadult oc small

adult, possibly female. Two such animals were grooming high

up in a tree. Upon seeing the observer, the one being groomed

gave a squeal which was followed by the chut phrases. It then

moved away from the groomer who chirped when it noticed the

observer. Taking into consideration the context in which it

was given and the call with which it was associated, the chut

could be tentatively identified as a fear-alarm call.


Chutter

The chutter (Fig. 14 and Table 9) is a completely

nontonal phrase composed of several units of short duration.

Each unit on the sonogram resembles a column of noise with

some broader than others. There is some variation in chutter

phrase duration, depending upon the number of units included

in the call. One of the three calls sampled contained twelve

units, the others each had four. The significance of phrase

length differences is not yet understood, but perhaps there

is some relationship between the intensity of stimulus

arousal and the number of units per phrase. It is also pos-

sible that the variation reflects individual differences.

Perhaps more than one type of chutter is represented in the

sample. Struhsaker (1967) was able to discern several dif-

ferent kinds of chutters for the vervet monkey. Without a

larger sample size and more information as to the function

cf the cutter calls and the circumstances which evoke them,

any statement as to a possible typology of cutter phrases

woulc be premature.





27
The unit duration and the interval between units was

similar among the three phrases. However, it was noticed

that the units of a phrase tend to increase in duration and

intensity such that those units late in the phrase are the

longest and loudest. The major energy distribution ranged

from 2.00 11.00 kHz with the mean distribution as 2.67 -

7,83 kHz (N=18).

TABLE 9

CUTTER


Number (N) Mean (x) Range (R)

Units/Phrase 20/3 6.33 4.00-12.00

Unit Duration 18 0.06 0.01-0.11
secss.)

Interval Between
Units secss.) 15 0.04 0.01-0.08

Phrase Duration 3 0.58 0.39-0.95
secss.)

Lower Limit of
Major Energy 18 2.67 2.00-4.00
(kHz)

Upper Limit of
Major Energy 18 7.83 4.00-11.00
(kHz)


The individual giving the chutter call was never seen
and the situations in which they were given were seldom ap-

parent. They were sometimes, but not always, heard during

intragroup fights so apparently the cutters play some part

in agonistic encounters. Struhsaker (1969) likewise retorts






having heard cutters during the fight episodes of other

Cerconithecus species.


Low, Short Chutter

On one occasion a low chutter of short duration

(Fig. 15 and Table 10) was given by an adult Sykes. Two

phrases were recorded, each consisting of two units. This

chutter is lower in frequency than the chutter described

above, with a mean major energy distribution of 0.75 2.88

kHz (N=4). However, in terms of total energy distribution,

there is some indication on the sonograms of energy in the

vicinity of 7.50 kHz.

TABLE 10

LOW, SHORT CHUTTER


Number (N) Mean (x) Range (R)


4/2

4



2

2


Units/Phrase

Unit Duration
secss.)

Interval Between
Units secss.)

Phrase Duration
secss.)

Lower Limit of
Major Energy
(kHz)

Upper Limrit of
Major Energy
(kH 2)


2.00

0.04



0.05

0.14



0.75



2.88


2.00-2.00

0.03-0.07



0.05-0.05

0.13-0.15



0.50-1.00



2.75-3.00


------------'-





29
Energy distribution and duration of each of the units

are auite similar, with the units appearing on the sonograms

as thin columns of noise. Mean unit duration is 0.04 seconds

(N=4) and phrase duration averages 0.14 seconds (N=2). This

chutter, like the previous one, is entirely nontonal.


Chutter-Saueal

The chutter-squeal (Figs. 16-18 and Table 11) can

best be described as a multi-phrase bout of long duration.

The following discussion is based on measurements of two

such calls which were both recorded during the same session.

Since the vocalizing individual was not seen, it is possible

that both these calls were given by the same animal.

The first phrase consists of a series of short units

resembling in some ways both the chutter and the chirp.

These units are basically nontonal like the chutter but yet

there is a hint of a chevron-shaped tonal component (as in

the chirp) superimposed over the nontonal element. The

first bout had a phrase consisting of nine chutter units and

the second bout had four chutter units. With a mean unit

duration of 0.07 seconds (N=13), the units are as short as

those of the chutter but do not have as wide a frequency

range.

Shortly following this first phrase are single-unit

phrases which appear on the sonogram as nontonal-tonal undu-

lating bands. There were five of these phrases in the first

bout and two in the second. These phrases are of long dura-

tion; two sampled had a mean duration of 3.93 seconds. The





30
mean of the lower frequency of major energy distribution is

3.75 kHz while the mean upper frequency is 7.00 kHz.

The final phrase of the bout is also a single, mixed,

tonal-nontonal unit but the nontonality or noise is not as

pronounced as in the preceding phrases. The average dura-

tion of this unit is 0.62 seconds (N=2) and the mean fre-

quency range is 3.25 5.65 kHz (N=2).

TABLE 11

CHUTTER-SQUEAL


Number (N) Mean (x) Range (R)


Chutter Portion

Units/Phrase

Unit Duration
secss.)

Interval Between
Units secss.)

Phrase Duration
secss.)

Lower Limit of
Major Energy
(kHz)

Upper Limit of
Major Energy
(kHz)

Mixed Portion I

Interval Between
Chutter & Mixed I
secss.)

Units/Phrase

Phrase Duration
secss.)


13/2

13



11

2



13



13





2


2/2

2


6.50

0.07



o.c4

0.64



3.15


0.02


1.00

3.93


4.oo-9.oo

0.02-0.10



0.02-0.08

0.39-0.88



2.50-4.50



4.00-7.00





0.02-0.03


1.00-1.00

3.08-4.78







TABLE 11 (Continued)


Number (N) Mean (R) Range (R)


Lower Limit of
Major Energy
(kHz)

Upper Limit of
Major Energy
(kHz)

Mixed Portion II

Interval Between
Mixed I & Mixed II
secss.)

Units/Phrase

Phrase Duration
secss.)

Lower Limit of
Major Energy
(kHz)

Upper Limit of
Major Energy
(kHz)

Total Duration


2



2





2


2/2

2



2



2


2


3.75



7.00





0.14


1.00

0.62



3.25



5.65


5.35


3.50-4.oo



6.50-7.50





0.10-0.18


1.00-1.00

0.50-0.73



3.00-3.50



5.50-5.80


4.10-6.60


Neither the final nor preceding phrases show any

great similarity, audibly or spectrographically, to the shriek

or squeal phrases reported earlier.

The sum of the duration of each of the phrases as

well as the intervals between them results in a lengthy call

averaging approximately 5.35 seconds (N=2).






Growl

The growl (Figs. 19&20 and Table 12) is predcminately

a tonal call although there may be some nontonality present,

particularly at the end of the vocalization. The growl con-

sists of a multi-unit phrase which may occur singly or may be

repeated.

TABLE 12

GROWL


Number (N) Mean (x) Range (R)

Units/Phrase 135/17 7.94 3.00-20.00

Unit Duration 135 0.02 0.01-0.03
secss.)

Interval Between
Units secss.) 118 0.02 0.01-0.03

Phrase Duration 17 0.20 0.05-0.59
secss.)

Interval Between
Phrases secss,) 9 0.L9 0.08-0.67

Lower Limit of
Major Energy 17 0.52 0.25-0.75
(kHz)

Upper Limit of
Major Energy 17 1.15 1.00-2.00
(kHz)


The number of units per phrase was variable but aver-

aged 7.94 units per call (N=17). The unit duration and the

interval between units was similar in all the phrases sampled,

so phrase duration was largely dependent on the number of





33
units present. 'Since the number of units varied from three

to twenty, phrase duration varied considerably also. However,

the mean phrase duration was 0.20 seconds (N=17). Despite

their temporal variation, the calls are grouped into a single

category at this time on the basis of their acoustic and

spectrographic similarities.

Energy was distributed on approximately half the sono-

grams in a single band formation (N=8) and in a double (N=5)

or triple (N=4) band arrangement on the remainder. In the

last two cases, however, most of the energy would be concen-

trated in the lowermost band. The average distribution of

the heaviest energy concentration, therefore, was 0.52 1.15

kHz (N=17) but the total energy distribution might extend to

3.50 kHz in some cases.

Identification of the vocalizing individual was dif-

ficult but the call was apparently given by both juveniles

and adults. Examples of instances when the growl was heard

include: during chirping sequences by a Sykes threatening

the observer, as Sykes fed while looking at the observer,

during fight episodes, and while a troop was moving in the

brush.


Err

A call somewhat similar acoustically to the growl

is the "err" (Fig. 21 and Table 13). On the sonogram, how-

ever, each call is broken into four or five short, discrete

units which extend to higher frequency levels than the growl.

The "err" is completely tonal.





34
The mean phrase duration of the "err" is 0.17 seconds

(N=3), with a mean energy range of 0.81 5.25 kHz for the

units (N=14).

TABLE 13

ERR


Number (N) Mean (x) Range (R)

Units/Phrase 14/3 4.67 4.00-5.00

Unit Duration 14 0.02 0.01-0.05
secss.)

Interval Between
Units secss.) 11 0.02 0.01-0.05

Phrase Duration 3 0.17 0.12-0.28
secss.)

Lower Limit of
Energy (kHz) 14 0.81 0.50-1.20

Upper Limit of
Energy (kHz) 14 5.25 4.00-7.50


The individuals vocalizing were not identified and

the function of the "err" call is not understood. However,

the call was heard immediately following a shriek on two of

three occasions.


Uh

A short call best described as sounding like "uh"

(Fig. 22 and Table 14) was given before a shriek three times

out of a sample of five. The call was given in pairs or in

threes with a mean interval between the units of 0.12 seconds






35
(N=7). The mean unit duration was very short at 0.04 seconds

(N=12).


TABLE 14

UH


Number (N) Mean (x)


Units/Phrase

Unit Duration
secss.)

Interval Between
Units secss.)

Phrase Duration
secss.)

Lower Limit of
Major Energy
(kHz)

Upper Limit of
Major Energy
(kHz)


12/5

12



7

5



12



12


2.40

0.04



0.12

0.28



1.78



5.56


Range (R)


2.00-3.00

0.01-0.09



0.04-0.24

0.08-0.40



1.00-3.50



4.10-8.00


The "uh" is lower in frequency than the shriek which

sometimes followed (three out of five times). The mean lower

frequency of major energy is 1.78 kHz and the mean upper fre-

quency is 5.56 kHz (N=12). The call is entirely nontonal and

appears as a narrow column of noise on the sonogram.



Discrete and Graded Signals


In a study such as this in which the vocal reper-

toire of a primate species is being classified and described,

it is customary to examine the nature of the signals as to






whether they are discrete or graded. Discrete signals are

separate and distinguishable whereas graded signals are more

variable and can be arranged to form a continuum with other

such signals.

Marler has pointed out that continuous or graded

sounds are more common in rhesus monkeys, baboons, chimpan-

zees, and gorillas, resulting in sound systems which are

"much more complex in the development of these subtle contin-

uous and multidimensional variations of the sound signals"

(1965:566). The graded nature of the vocalizations of these

terrestrial or semi-terrestrial species which are relatively

isolated from other primate species of similar size and struc-

ture contrasts significantly, says Marler, with the relatively

discrete nature of the sound systems of forest-living primates

in frequent contact with other primate species, such as is

the case with Cercooithecus and Colobus.

In such circumstances the possibility of inter-
specific confusion is greatly increased and, cor-
respondingly, the limited evidence available on
the sound signals of such forms suggests that they
are more highly structured and that this structuring
is used at least in part to maintain species speci-
ficity in a major proportion of the vocal reper-
toire. (Marler, 1965:565)

The vocal repertoire of the Sykes monkeys seems to

comply with Marler's thesis. In the study area there were

three other diurnal species of nonhuman primates observed,

the baboon (Papio cynocephalus), the colobus (Colobus poly-

komos), and the vervet (Cercooithecus aethiops). The Sykes

were at times observed in close proximity to all three

species. Of these species, the vervet is the most closely






related to the Sykes, morphologically and behaviorally.

According to Struhsaker (1967), the majority of the vervet

vocalizations which he studied were of the discrete type.

Similarly, most of the Sykes' calls, w-hile showing some de-

gree of variability within categories, are also of the dis-

crete type. Signals which may prove to be of the continuous

type would include the shriek and the squeal, and the calls

of the chutter group.

For example, the squeal was described as a tonal call

consisting of two parallel bands with a mean energy distribu-

tion of 3.20 7.87 kHz (N=7). The shriek, on the other hand,

was composed of two components, a two-band tonal section fol-

lowed by a mixed portion of tonal and nontonal sound. The

initial tonal component strongly resembles the squeal, but

the mean energy distribution is slightly higher at 3.46 -

8.74 kHz (N=5). However, the ranges for the energy values

for the two calls do overlap (Table 6 and Table 7). It would

seem likely, therefore, that there may be a graded continuum

between squeals and shrieks.

There may also be a continuum within the chutter

group, although the sample is small and hence not extremely

reliable. Perhaps the chut, the chutter, and the short chut-

ter are but points along a graded continuum of such calls.

Also, there may be some functional significance in the number

of units per chutter phrase, creating a gradient in this re-

gard as well.







Future fieldwork and further study may point out

more continuous signals for the Cercoithecus mitis group as

well as modify those tentatively identified here.

This investigation has demonstrated the possible

existence of a graded series of signals or messages. That

this is not sufficient evidence to indicate a graded commun-

ication system has been aptly pointed out by Altmann (1967).

He stresses that signals used in graded or analogical com-

munication must be differentiated from those signals which

may appear graded but merely reflect the normal range of vari-

ation which a sample of calls may be expected to show. Such

variation would be due to the morphological and physiological

variations of the structures involved in sound production.

It is necessary, says Altmann, to demonstrate functional con-

tinuity: "What is needed is (1) a continuous array of mes-

sages, (2) a continuous array of responses to these messages,

and (3) a one-to-one mapping between messages and responses"

(1967:341).

The failure to make the distinction between graded

signals which are associated with graded responses and those

which are not has resulted in some confusion and misunder-

standing concerning the use of the terms 'discrete' and

'graded'. The terms were introduced into the study of pri-

mate communication largely through the work of Hockett (1960)

who proposed thirteen design features, of which discreteness

was one, which he felt were characteristic of human lang-

uages. Apparently in their haste to look for Hockett's de-

sign features among the communication systems of nonhuman






39
primates, many researchers failed to recognize that discrete-

ness and gradedness among the response patterns as well as

discreteness and gradedness among the signals themselves must

be demonstrated in order for a communication system to be

classified as one or the other. Consequently, although these

terms occur frequently in the literature, Altmann (1967)

points out that a graded communication system has never been

demonstrated. Even if such a system were described, Altmann

further emphasizes that discontinuities between the major

categories of communication patterns themselves will make for

a certain amount of discreteness. The classification of the

system as one or the other, therefore, largely becomes a

question of semantics.

It would seem that an understanding of the nature of

communication systems among nonhuman primates lies not in its

comparison, feature for feature, with human language but

rather with the understanding of the adaptive significance of

such features as may be present. That the latter has not yet

been accomplished is due in large part to the fact that pri-

mate communication studies are still in their infancy. Hope-

fully, however, through future field studies and the amassing

of greater amounts of information, certain adaptive trends

and correlations will become apparent which will have signi-

ficance for understanding not only the nature of nonhuman

primate communication but the nature of human language as

well.






A Structural-Functional Analysis


There are many different types of vocalizations among

the primates and these vocalizations serve various functions.

However, the exact nature of the relationship between func-

tion and structure of primate calls has not yet been deter-

mined. Although work along these lines has been carried out

for some insect species (Alexander, 1960, 1968), anuran

species (Bogert, 1960; Littlejohn, 1959), and avian species

(Marler, 1955, 1959; Marler and Hamilton, 1966), most discus-

sions of primate vocalizations in structural-functional terms

have remained generalized.

There are several reasons which may account for these

generalized studies. Firstly, the study of primate communica-

tion is still a relatively new endeavor and the emphasis,

until recently, has been on collection and description rather

than on detailed investigative analysis. Secondly, primate

communication has been shown to be an exceedingly complex pro-

cess involving several channels of communication; namely,

visual, auditory, tactile, and olfactory. The relative con-

tribution which each of these channels makes to a species'

communication system, the degree to which they are interrelated,

and the extent to which they limit and modify communication by

other channels are difficult to assess. Therefore, a discus-

sion of the structure of a particular vocalization must also

account for the possibility of simultaneous transmission of

information by some other means, such as visual signals.

Thirdly, the determination of the function of a call is often





4i
difficult, requiring an evaluation of the behavioral and en-

vironmental contexts in which the call was given as well as

a description of the nature of the responses which it evoked.

Consequently, among those primate species which have been

studied, there remain numerous calls whose functions are un-

known. Lastly, it has not yet been determined which aspects

of a primate vocalization, if any, are most significant for

the transmission of information. That some components of a

call, such as frequency, intensity, duration, and timing,

are differentially informative has been shown to be the case

among arthropods (Alexander, 1968), anurans (Bogert, 1960),

and birds (Falls, 1963; Marler, 1955, 1959). Experimental

designs, such as play-back techniques both in the laboratory

and in the field, will probably be of considerable impor-

tance in evaluating the communicative significance of the

various components.

The present investigation will attempt to apply a

structural-functional approach to the study of Sykes vocaliza-

tions by focusing on several principles of sound localization

as outlined by Marler (1955, 1959) in his studies on bird

vocalizations. Structural comparisons will be made between

those Sykes calls used to communicate over long distances

and those which operate at close range. The discussion will

be limited to those calls whose functions are known.

As Marler (1955, 1959) points out, sound localization

in vertebrates relies on binaural comparisons of differences

in phase, intensity, and time of arrival. Intensity differ-

ences become apparent when the head acts as an obstruction,






thus hindering the sound from reaching one ear. The direc-

tion of the sound can be inferred due to the resulting dif-

ference in intensity at the two ears. Since this "shadowing

effect" is significant only when the obstruction is of the

same order of size as the sound wavelength or larger, locali-

zation by intensity differences is most useful with high fre-

quency sounds (Marler, 1959).

On the other hand, localization by phase differences

is confined to low-pitched sounds (Marler, 1959). Phase dif-

ferences are the result of pressure changes occurring at

different moments in the two ears. In high frequency sounds

where the wavelength is less than the distance between the

ears several wave oscillations may occur, resulting in ambig-

uous phase comparisons. In addition, the nerve leading from

the ear is unresponsive for about a thousandth of a second

after each oscillation, so a sound with more than a thousand

vibrations per second would not be completely represented by

nerve impulses (Marler, 1959).

Accurate comparisons of differences in time of arri-

val of sound at the two ears is dependent on the way a sound

begins and ends rather than on frequency ranges. It has been

found that sharp discontinuities are most easily perceived,

especially if frequently repeated (PMarler, 1959).

Marler concludes that the ideal sound permitting

source localization "will include a high pitch for location

by intensity difference, a low pitch for location by phase

difference, and it will be sharply broken and repetitive for

location by time difference" (1959:175). In man it has been





43
found that pure tones in the region of 2.0 5.0 kHz are dif-

ficult to locate by comparisons of phase and intensity differ-

ences, being too high for the former and too low for the

latter to be effective (Marler, 1955). For this frequency

range only time differences can be used to successfully locate

sounds. The lower limit of this frequency range is probably

similar in the Sykes monkey due to the refractory period of

the auditory nerve. However, the upper limit is probably

higher (exactly how much so is yet to be determined) since

the diameter of the Sykes' head is smaller than man's and

the "shadowing effect" will thus not take place until a smaller

wavelength or higher frequency is reached.

Five Sykes calls, about whose functions something is

known, were chosen for investigation. These vocalizations

will be examined in terms of whether they can or can not be

easily localized by phase, intensity, or time differences.

The five calls are the nyah, boom, chirp, squeal, and shriek.

It has been mentioned elsewhere that the nyah call

given by adult males functions to promote group cohesion and

coordinate group movements. Struhsaker (1970) emphasizes

that by doing so, such calls may also serve to maintain re-

productive isolation in areas of mixed species associations.

When the call is given the group members, who have been dis-

persed at the time, begin to move in the direction from which

the call came. The call is loud and carries over considerable

distances. Visual contact does not seem to be of importance

here so one would expect the nyah, in keeping with its func-

tion, to be a call which is easily localized. Analysis of its






structure shows 7his to be the case. The average range of

major energy distribution for the nyah was shown to be between

0.82 and 5.46 kHz (Table 3) which would apparently include a

low enough pitch for localization by phase differences. The

upper limit of this range would appear at first glance to be

too low to facilitate localization by intensity differences

since it most likely falls within the "hard-to-locate" range

discussed earlier. However, since in total energy distribu-

tion the upper frequencies of the nyah may extend to 16.0 kHz

or more, localization by means of intensity differences is

indeed possible. The call is of a short duration (0.31 sec-

onds), has an abrupt beginning, and is almost always repeated

thus enabling localization by temporal differences. In short,

the nyah call fulfills all three of Marler's requirements for

an ideal sound permitting source localization. Indeed, this

is what one would expect of a call which apparently plays

such a vital role in group maintenance and cohesion.

Another call whose function is apparently concerned

with group cohesion without requiring visual contact with the

caller is the boom vocalization. This call, however, differs

from the nyah in several respects. The boom is a very low-

pitched call with an average energy distribution of 0.01 -

0.40 kHz (Table 5). This low frequency would permit locali-

zation by phase differences but not by intensity differences.

Although the call is of a short duration (0.20 seconds), it

is not repeated, nor does it begin or end abruptly, thus

making localization by temporal differences difficult. There

are several possibilities which may account for the structure






of this call. Although the boom would not appear as ideal

for localizing a sound source as the nyah, perhaps it is given

under circumstances where an extreme degree of localization

would be disadvantageous to the caller or the group, such as

in reaction to the observer or a potential predator. Another

possible explanation is that perhaps phase difference alone

is an efficient means of sound localization as long as the

call is restricted to very low frequencies. In addition, a

low-pitched group cohesion call might be used because low

frequency sounds carry over greater distances than sounds of

higher frequencies (Moynihan, 1967). Also, there is less

chance that a low-pitched call will be refracted by trees or

other objects as is the case with high-pitched sounds (rMarler,

1955), which is obviously an important consideration for an

arboreal species. The key to explaining the nature of the

boom clearly lies with a better understanding of its function

as well as with the formulation of an experimental approach

which would test the localization properties of the call.

The chirp acts as a threat/alarm call and its phys-

ical properties would seem to facilitate localization. Al-

though the average major energy distribution of 2.91 6.14

kHz (Table 1) falls within the "hard-to-locate" zone, the

total energy range of 0.10 16.0+ kHz most likely permits

localization by means of phase and intensity differences. The

call has an abrupt beginning, is of short duration (0.18 sec-

onds), and is usually repeated. With increasing intensity or

levels of excitement, the call is repeated more frequently

and may have an even more abrupt beginning. Localization by






comparison of temporal differences is clearly possible and

may be of considerable importance. One might expect an alarm

call to be given in such a way as to prohibit localization of

its source for safety's sake, as is the case for some avian

calls (izarler, 1955). That this is not the case for the chirp

is supported by the observation that the other group members

look toward the caller upon hearing the chirp and may then

chirp also. In addition, the caller is frequently exposed

in the open when giving the call and may not move back into

the brush for some time; in fact, other Sykes may actually

move up to be near the caller as they join in chirping.

Since the chirp is easily located it aids in indicating the

source of danger to other group members and in enlisting their

support in threat.

The last two calls to be considered are the squeal

and shriek. Both seem to be associated with agonistic encoun-

ters although their exact function is undetermined. Their

energy distributions of 3.30 7.87 kHz for the squeal (Table 6)

and 3.20 8.74 kHz for the shriek (Table 7) are apparently

within the "hard-to-locate" range. The calls are of a longer

duration than the others which have been discussed, 0.63

seconds (squeal) and 1.01 seconds (shriek), and both start

and end gradually. Precise localization of the sound source

would appear to be difficult. However, the most important

difference between the shriek and saueal and the other calls

which have been described is that the functions of the laCzer

require transmission over distances, whereas the shriek and

squeal apparently function in close-range a3onistic encounters





47
Consequently, other channels of communication, such as facial

expressions or touch, may contribute to the message content

and indicate the location of the vocalizer.

It is difficult to explain the physical properties

of the remaining calls in the Sykes' repertoire until more

is known of their function. Since most of these calls seem

to function on a close-range basis, one would expect other

channels of communication to be of some importance in signal

transmission.

It should be emphasized that a discussion of the

structural and functional components of a call in terms of

its localization properties is an oversimplification of the

nature of primate vocalizations. Other factors, such as re-

quirements for species specificity and individual identifi-

cation or the nature of the physical and social environments,

may affect the physical structure of a primate call. For

example, Marler (1965) points out that in those calls used

over distances to maintain spacing between groups, the re-

quirements of species specificity necessitates use of a

larger range of sound properties than in close-range commun-

ication. These spacing calls, says Marler, are therefore

more highly structured and purer in tone. It is possible,

therefore, that requirements for species specificity might

be partly responsible for the highly structured nyah call of

the Sykes monkey.

Oversimplified generalizations about primate vocali-

zations can be dangerous and misleading. A statement that

arboreal species living under conditions of low visibility






have developed elaborate high-pitched calls (Cartlan and

Brain, 1968) gives the misleading impression that a high

pitch alone makes such calls functionally significant and

fails to recognize and account for the existence of low-

pitched calls among such species. What is needed instead is

a structural-functional analysis to furnish clues as to why

a particular structure or form may predominate. It is hoped

that the analysis presented in this section has demonstrated

the feasibility of such an approach.



Taxonomic Imolications


The genus CercoDithecus is said to have more species

and forms than any other primate genus in Africa (Struhsaker,

1970). Its species have usually been distinguished taxonomi-

cally on the basis of pelage coloration. Unfortunately, wide

variability in pelage coloration within the genus and within

its species has resulted in some disagreement over the inter-

pretation of phylogenetic affinities. In cases where morpho-

logical characters are inadequate or equivocal, the taxono-

mist has been urged to make use of data from studies in be-

havior, ecology, physiology, and biochemistry (Lanyon, 1969).

Through the use of improved recording equipment and

the development of the sound spectrograph, one aspect of be-

havior, i.e., vocal communication, has been shown to be of

some significance in taxonomic studies of orthopteran insects

(Alexander, 1962), anuran amphibians (Bogert, 1960; Blair,

1962, 1964), and birds (Marler, 1957; Lohrl, 1963). Studies





49
such as that of Struhsaker (1970) have indicated that vocal-

izations, when used with the same precautions and care appli-

cable to any potential taxonomic character, can be of consid-

erable importance in primate taxonomy as well, Struhsaker

examined the taxonomic position of several Cercooithecus

species, basing his decisions largely on similarities or dif-

ferences in certain vocalizations.

One of the species studied by Struhsaker (1970) was

Cercopithecus nictitans. Several authors have pointed out

the strong physical resemblance between C. nictitans and

C. mitis (Booth, 1956; Haddow, 1952; Kingdon, 1971; and Tap-

pen, 1960). Hill (1966) placed C. nictitans in the mitis

superspecies and the 1969 Wenner Gren Conference raised the

question that perhaps C. nictitans and C. mitis might be best

considered as conspecific (Thorington and Groves, 1970).

A comparison of the vocalizations of these two species might

shed some light on their phylogenetic relationship.

Of the two species, C. nictitans is the more restricted

in distribution, being largely confined to the West African

area. Two subspecies are generally recognized. C. n. martini

is the most westerly located (Ghana, Nigeria, Cameroun).

C. n. nictitans is also on the western coast (Gabon, Cameroun)

but extends eastwards into the Central African Republic area.

C. mitis, characterized by numerous subspecies, is much more

extensive in its range, being found in central, east, and

south Africa. C. mitis is said to be the only forest guenon

having such a wide distribution in southern and eastern

Africa (Kingdon, 1971). It apparently inhabited these areas






before climatic changes isolated many of the forests, thus

closing them to those Cerconithecus species which have devel-

oped more recently (Kingdon, 1971). In fact, C. mitis is

believed to be near the stock from which the main guenon ra-

diation has derived (Kingdon, 1971). For a more detailed

discussion of the distribution of the various C. mitis sub-

species, the reader is referred to Rahm (1970).

The subspecies which will serve as the basis for the

present discussion of vocalizations are C. n. martini and

C. n. nictitans studied by Struhsaker (1970) in Cameroun and

C. m. kibonotensis from the southeast Kenya coast whose vo-

calizations have been described in this paper.

The reader should note that the C. mitis and C. nic-

titans populations here being considered are allopatric, a

situation which can make for considerable difficulty in as-

sessing phylogenetic relationships (Mayr, 1969). In this

regard, it is the examination of sympatric species which is

most conclusive, since any mechanisms assuring species iden-

tity through reproductive isolation can be more reliably

recognized in the sympatric state. However, Struhsaker

(1970) has shown that interpopulation variability in the

basic structure of vocalizations within a species appears to

be minimal. The species he considered included C. aethiops

(populations from East and West Africa), C. mona, C. nicti-

tans (populations from both subspecies), and C. pogonias

(populations from the subspecies DoEonias and Eravi). Within

each of these species and across subspecies lines Struhsaker

fo:ond that the vocalizations snowed a considerable amount of





51
stability and uniformity. Therefore, it will be assumed that

the vocalizations of C. mitis are relatively uniform through-

out its range also. In this way, it will be assumed that the

vocalizations of C. mitis kibonotensis of East Africa would

be basically similar to those C. m. stuhlmanni and C. m.

maesi, two western subspecies of C. mitis which are reported

to be sympatric with C. n. nictitans at the eastern limits of

its range (Rahm, 1970; Kingdon, 1971). Ideally, a study to

determine whether C. nictitans and C. mitis are consDecific

should be carried out in this area of sympathy. Until such

studies are available, however, and until studies indicative

of the extent of variation of the vocalizations within C.

mitis are carried out, the investigation presented here will

proceed on the assumption which has been made based on Struh-

saker's analysis (1970). No attempt is being made, therefore,

to present the findings of this study as a final taxonomic

statement.

Hill (1966) apparently feels there is enough dis-

tinction between the western and eastern subspecies of the

C. mitis group to warrant their separation into two different

species, C. mitis and C. albosularis respectively. However,

other workers (Rahm, 1970; Kuhn, 1967; Kingdon, 1971) have

classified these as a single polytypic species. That hybrids

have been reported between C. m. stuhlmanni and C. m. albo-

gularis (Booth, 1968) not only tends to support the latter

approach but also adds more weight to the assumption that

isolating mechanisms, including any vocalizations which may

function as isolating mechanisms, are probably uniform





52
throughout the range of C. mitis. It should be pointed out

that since the eastern subspecies of C. mitis are relatively

isolated from other forest cuenons while those in the west

may be sympatric with several, such as, C. mona, C. ascanius,

C. l'hoesti, and C. neglectus, there may be selective pressures

operating in the west which may not be operating in the east.

These differences may be reflected in the vocalizations of

these sDecies.

Struhsaker (1970) believes that the loud calls given

by adult male Cercopithecus monkeys function to maintain re-

productive isolation between species by enhancing group

coherency, especially in response to alarming stimuli. In

the sympatric state, therefore, the loud calls of two species

would-be different; Struhsaker, working with several species,

found this to be the case.

A comparison of the loud calls which Struhsaker re-

ports for C. nictitans (1970:407, 427) with those reported

here for C. m. kibonotensis (Figs. 6-8 and Tables 3&4) re-

veals some differences. The nyahs of C. mitis cover a higher

frequency range than do the pows of C. nictitans and have a

more elaborate development of harmonics. The tonality of the

C. mitis calls also has an arching configuration, rising and

then falling in frequency, as opposed to the descending

quality of the tonality in C. nictitans. The short nyahs

have some spectrographic resemblance to the hacks of C. nic-

titans, but again a higher frequency is reached in C. mitis.

No calls equivalent to the pow-hacks which Struhsaker found

to be intermediate between pcws and hacks were found for





53
C. mitis. Struhsaker could not discern any significant dif-

ference in the loud calls of the two subspecies of C. nicti-

tans, nor could he distinguish the hacks of C. eryvhrotis

and C. cephus (1970:408, 429) which he subsequently placed

in a single species. The loud calls of C. mitis and C. nic-

titans are apparently not as similar as those of these other

conspecific populations.

In addition, booms were recorded for C. mitis but not

for C. nictitans and in this regard C. mitis resembles the

mona superspecies. The C. mitis booms (Fig. 9 and Table 5)

resemble the C. mona calls (Struhsaker, 1970:409, 431) very

closely. However, since this type of call is restricted to

such low frequencies, variability may be expected to be

limited so that similarities may not be significant. Booms

are typically given before a series of hacks in C. mona.

Booms were also frequently heard in conjunction with nyah

calls in C. mitis. However, these similarities in boom calls

should not be taken to indicate an especially close relation-

ship between C. mona and C. mitis. The hacks of C. mona

(1970:410, 423) do not appear to be any more similar to those

of C. mitis than the latter was to C. nictitans. In addition,

only hack-type units were recorded for C. mona whereas C. mitis

definitely shows two types, i.e., nyahs and short nyahs. No

calls similar to the "ooo" contact calls reported by Struh-

saker for the mona supersDecies were found in C. mitis.

There are also several differences between other calls of

C. mona and C. mitis and these will be discussed below.






The chirp (alarm) calls of C. nitis (Figs. 1-5 and

Tables I&2) bear some resemblance to the chirps of C. nicti-

tans and C. erythrotis-ceohus (Struhsaker, 1970:411, 44,.

435, 436). In terms of the spectrographic form of the chirps

the three species are very similar, but those of C. mitis

are of somewhat higher frequencies as well as of longer dura-

tion. Multi-unit chirps were present in all three species.

Struhsaker (1970) found the chirps of C. erythrotis and C.

cephus to be very similar. Some of these chirps were iden-

tical to those given by C. nictitans. Similarity in alarm

calls among sympatric species faced with common predators

may be explained as a case of parallelism. Struhsaker (1970)

points out, however, that C. mona and C. pogonias also asso-

ciate freely with C. nictitans and C. erythrotis-ceohus but

maintain very distinct alarm calls. Therefore, Struhsaker

concludes that C. nictitans and C. erythrotis-cephus may be

more closely related to each other than to C. mona.

Struhsaker (1970) reported grunts and chutters for

C. nictitans and C. erythrotis-cephus but found that no such

calls were given by members of the mona superspecies. The

grunts are thought to function in maintaining group cohesion

while the chutters were given during agonistic encounters.

C. mitis has vocalizations similar to these grunts and chut-

ters, although the samples for C. mitis are too small to be

conclusive. A low, short chutter given by C. mitis (Fig. 15

and 'able 10) resembles very closely the C. nictitans chut-

ter2 Jscri:'bed by Struhsaker ( 970:419, 1-41) both in energy

disT ribu tlon and duration. However, the relationship of this





55
cutter to others of higher frequencies which have been re-

ported for C. mitis in this paper is not clear. Struhsaker

does not mention any other chutter types for C. nictitans

and C. erythrotis-cephus but his 1970 study was apparently

not intended to be descriptive of the complete sound reper-

toires. Therefore, it remains to be seen whether the exis-

tence of these other chutter-types in C. mitis is indicative

of a significant species difference. The "err" reported here

for C. mitis (Fig. 21 and Table 13) may be equivalent to the

grunts; if so, some differences are apparent. While all the

C. nictitans grunts (Struhsaker, 1970:418, 441) were single-

unit phrases and the C. erythrotis grunts (Struhsaker, 1970:

418, 441) were either single- or triple-unit phrases, the

"err" of C. mitis was made up of four or five units. Although

unit and phrase duration for all three species are basically

similar, the C. mitis call extends to a much higher frequency.

Again, Struhsaker found that the grunts and chutters of C.

erythrotis and C. cephus were very similar to each other and

also to those of C. nictitans; the grunts, in fact, vere indis-

tinguishable.

Of the vocalizations discussed above, the one which

plays the most important role in helping to maintain species

identity through reproductive isolation is the loud call.

Struhsaker points out that vocalizations having different

functions evolve at different rates: "Calls serving to alarm

or coordinate groups may evolve more slowly than those enhan-

cing reproductive isolation through the maintenance of conspe-

cific intragroup cohesion around one male" (1970:370). If





56
this assumption is correct, then one would expect the chirps,

grunts, and chutters to not be under the same pressures for

species specificity that influence the loud calls. Further-

more, species showing similarities in these calls may be more

closely related to each other than to species which do not

have such similar calls (Struhsaker, 1970).

It has been shown that the loud calls of C. mitis are

readily distinguishable spectrographically from those of C.

nictitans. Since these differences are greater than those

found among conspecifics, a separate specific status for C.

mitis and C. nictitans seems warranted at this time. This

conclusion is further supported by the observation that the

chirps, grunts, and chutters of C. mitis are also not as

similar to those of C. nictitans as are these calls within

the C. nictitans subspecies and within the C. erythrotis-

cephus species. According to Struhsaker (1970), the fact

that grunts, chutters, and chirps are absent in the mona su-

perspecies would seem to indicate that these calls are not

to be interpreted as conservative, widespread vocalizations

but rather may be indicative of close phylogenetic relation-

ships. Following Struhsaker's reasoning, one can conclude

that the similarities of the C. mitis calls to those of C.

nictitans and C. erythrotis-cephus, while not indicative of

conspecific status, would indicate that C. mitis, C. nicti-

tans, and C. erythrotis-ceohus are more closely related to

each other than to the mona superspecies.

It should be emphasized again that future studies

may reveal that the vocalize ions of the more westerly





57
distributed C. mitis populations may be considerably differ-

ent from those reported here for C. m. kibonotensis. Infor-

mation as to the amount of variation in vocalizations through-

out C. mitis' range is needed to either support or refute the

assumption made here concerning their supposed stability and

uniformity. Information is also needed as to the extent to

which C. mitis and C. nictitans may associate in their zones

of sympatry. If the two species should be found to hybridize

freely in these zones, then serious consideration of a con-

specific status would be called for.

Although the samples collected for C. mitis were often

too small to permit rigorous statistical comparison and test-

ing, it is recognized here that a true assessment of the de-

gree of relationship between two species must be based on

careful statistical measures before any great reliability in

the data may be assumed. It is also recognized that taxonomic

conclusions should not be based on one or a few characters

alone, but should depend instead on the total assessment of

morphology, physiology, behavior, karyotypes, etc. However,

if Struhsaker's observation is true that "of all the single

classes of characters so far used, vocalizations seem the most

stable and thus the single most reliable indicator of phylo-

genetic affinities within this genus" (1970:402), then fu-

ture studies of vocalizations of the Cerconithecus species

may very well prove to be of considerable importance in the

assessment of phylogenetic relationships within this highly

variable genus. The analysis of vocalizations, therefore,

should merit considerable attention by primate taxonomists.












CHAPTER III


VOCALIZATIONS OF THE COLOBUS MONKEY



In sharp contrast to the relatively vocal Sykes mon-

keys are the taciturn colobus monkeys (Colobus polykomos cal-

liatus). These beautiful, black and white primates sit for

hours at a time without making a sound.

Colobus live in small groups; those in the study area

ranged in size from three to eight animals per group. They

maintain a relatively small territory in comparison to the

Sykes. Each troop was observed to occupy the same general

area for the duration of the study and, as a result, they

could be located quite readily.

The taxonomic designation, C. Dolykomos, being used

here follows that of Schouteden (1947), Tappen (1960), Napier

and Napier (1967), and Kuhn (1967). However, others, i.e.,

Schwarz (1929), Dandelot (1968), and Rahm (1970), differen-

tiate C. angolensis from C. polykomos and under this system

the colobus of this study would fall within the former desig-

nation. It is generally agreed that before the taxonomic

Problems can be resolved more information is needed about the

adaptations of these monkeys and areas of distribution over-

lap need to be investigated (Tappen, 1960).






The colobus have onr habit which, while not vocal,

definitely serves as a means of communication. The colobus

take fantastic leaps, land with a resounding thump against

tree limbs and branches, and continue in this way jumping

from tree to tree. The other colobus are alerted (probably

by both the sight and the sound) and look in its direction,

watching the leaping colobus closely. These displays are

quite impressive and seemed to be given in response to the

observer's presence, probably functioning in part as an

attempt to intimidate the intruder.

This ability to take leaps of considerable distance

also serves to indicate danger and the direction of troop

movement. If a colobus is startled, it will vault itself

seemingly into midair only to come crashing down into the

thicker brush below. The other colobus are immediately alerted

and also leap dowm into the lower level. A similar observa-

tion has been made by Schenkel and Schenkel-Hulliger (1967).

In a short while, when the danger has apparently passed, the

colobus will gradually move back up into the higher branches.

Marler (1968) also emphasizes the reliance of the colobus on

visual cues for maintaining intragroup spacing during move-

ment, pointing out that the conspicuous color of the colobus

and their habit of keeping high in the trees facilitate vi-

sual coordination of their activities.

Colobus tend to remain silent unless sufficiently
disturbed. Such disturbing circumstances included the close

proximity of the observer, the milling around of a group of

four or five dogs in the woods directly below the colobus





60
troop, and the circling overhead of a pair of Bateleur Eagles

(Terathooius ecaudatus).

On the latter occasion, a group of eight colobus were

sitting relaxed in the tall trees. In this troop was a female

with a small white infant. The female had been sitting unper-

turbed with the infant squirming about in her lap but as a pair

of Bateleur Eagles slowly circled closer she became quite tense

and attentive, craning her neck to watch the birds. After a

few moments she clutched the infant to her and moved over to

a more thickly leaved tree. As she did so, a male colobus

gave a croaking, snorting sound and bounded over to the fe-

male's side, also watching the eagles intently. Moments later,

the male was observed holding the infant so he must have ta-

ken it from the female when he moved near her. However, inci-

dents such as this were infrequent and other than the vocal-

izations of infants or juveniles, the troop was generally

silent.

The white infant was a focal point of interest for

the entire group. In fact, all the adult females were so

interested in the infant that at times it was very difficult

to determine who the actual mother was. This was further

complicated by the frequent practice of passing the infant

from one adult to another when it would begin to squirm and

scueal.

The colobus were frequently found in the same immed-

iate vicinity with the Sykes monkeys. The two species could

of-cen be seen feeding in the same trees or even on the same

branch just a few feet apart.






Classification of Calls


Snort

A loud, snort-like sound (Figs. 23&2L and Table 15)
was given by adult colobus on several occasions. The call is

most likely an alarm and/or threat call as it was evoked by

the sudden appearance of the observer, by a group of dogs,

and by the Bateleur Eagles. A similar call was reported for

C. polykomos by Hill and Booth who described it as an "explo-

sive snort, uttered partly through the nose" which was given

by adults of both sexes (1957:311).

TABLE 15

SNORT


Number (N) Mean (x) Range (R)

Units/Phrase 4/3 1.33 1.00-2.00

Unit Duration 4 0.13 0.06-0.15
(secs.)

Lower Limit of
Major Energy 4 0.25 0.25-0.25
(kHz)

Upper Limit of
Major Energy 4 10.45 3.30-13.00
(kHz)



The snort begins abruptly and appears primarily as a
nontonal vocalization on the sonogram. The call averages

approximately 0.13 seconds in length (N=4). The major energy

distribution is about 0.25 13.00 kHz with the greatest con-

centration of noise showing on the sonogram in the vicinity of





62
4.0 8.0 kHz. The snort usually consisted of a single-unit

phrase, although on one occasion a two-unit snort phrase was

given, the two units being separated by an interval of 0.L4

seconds. The second snort was approximately only half as long

as the first and its energy distribution was confined to a

lower frequency range (0.25 3.30 kHz).


Snort-Croak

The snort may stand by itself as a call or it may pre-

cede another call identified here as the snort-croak (Figs.

25-27 and Table 16). Marler (1970) mentions that C. gueraza

has a distinctive roaring call which sometimes may be preceded

by a snort. This may be equivalent to the snort-croak de-

scribed here, but this writer hesitates to use the term 'roar'

because such a term would seem to imply a single, long, loud

call which definitely wvas not the case in the snort-croak.

The snort-croak is probably equivalent to the "full call" re-

ported by Hill and Booth (1957) as given by adult male C.

polykomos. These authors also noted that the snort-like alarm

call may precede the full call.

A jumping-roaring display said to be highly ritualized

and which may function as a mechanism of intergroup spacing

has been described by Ullrich (1961) and Schenkel and Schenkel-

Hulliger (1967). The roaring is reported to spread from group

to grouD so that a chorus develops. Although the snort-croak

was never heard under these circumstances while this observer

was in the field, it is possible that it is equivalent to this

roaring display. Likewise, whether the "full call" described

oy Hill and Booth (1957) is the same as the roaring display







can not be determined from their report. Based on the ver-

bal descriptions of .he calls, they all seem to be similar;

confirmation of this must await spectrographic comparison.

The snort-croak is actually a long bout consisting

of three different phrase-types. Besides the initial single-

unit snort phrase, the snort-croak includes a multi-unit

phrase of a croaking, gutteral sound given in alternation

with a faint, short, inhalation, single-unit phrase.

The initial snort of the snort-croak is very similar

acoustically and spectrographically to the snort phrase which

has already been described. The phrase duration of the snorts

of the snort-croak averages 0.11 seconds and the major energy

is distributed over a range of 0.25 12.50 kHz (N=4).

Each croak phrase is composed of several short tonal

units. Those measured had anywhere from eight to sixteen

units (mean of 10.21 units; N=24). Each unit is of a very

short duration (0.01 or 0.02 seconds) but together as a phrase

the mean duration is 0.24 seconds (N=24). The croak is low

in frequency and has a deep, gutteral sound. The average dis-

tribution of the heaviest concentration of energy is from

0.26 0.97 kHz. The number of croak phrases per bout is

variable, averaging 5.38 phrases per bout (N=13) but ranging

from two to seventeen. It should be pointed out that the

seventeen croak phrases represent an extreme; the next highest

number of phrases is nine. The bout with seventeen phrases

was given during the encounter with the eagles and was given

more rapidly than the others. In addition, the croak phrases

were shorter, averaging 0.16 seconds as compared to an average






of 0.34 seconds (N=11) for the other croak phrases. A cor-

tion of this seventeen-phrase bout is pictured in Figure 27.

In between the croak phrases is a faint sound made as

the colobus inhales. This unit is nontonal and averages 0.05

seconds in duration (N=22). The average major energy distri-

bution of this single-unit phrase is 0.27 1.25 kHz (N=22).


TABLE 16

SNORT-CROAK


Number (N) Mean (x) Range (R)


Snort Phrase

Units/Phrase

Phrase Duration
secss.)

Lower Limit of
Major Energy
(kHz)

Upper Limit of
Major Energy
(kHz)

Croak Phrase

Units/Phrase

Unit Duration
(secs.)

Interval Between
Units secss.)

Phrase Duration
secss.)

Lower Limit of
Energy (kHz)

Uncer Limit of
Energy (kHz)


4/4

4



4



1




2L5,/24

245



221

24


1.00

0.11



0.25



12.50




10.21

0.01



0.01

0.24



0.26


1.00-1.00

0.10-0.12



0.25-0.25



12.50-12.50




8.00-16.00

0,01-0.02



0.01-0.02

0.15-0.43



0.25-0.40


0.97 0.75-1.00







TABLE 16 (Continued)



Number (N) Mean (x) Range (R)

Inhalation Phrase

Units/Phrase 22/22 1.00 1.00-1.00

Phrase Duration 22 0.05 0.04-0.07
secss.)

Lower Limit of
Major Energy 22 0.27 0.25-0.40
(kHz)

Upper Limit of
Major Energy 22 1.25 0.70-2.00
(kHz)



Gecker White Infant

The gecker (Fig. 28 and Table 17) is a high-pitched,

squeaky sound and was given quite frequently by a white in-

fant colobus. Usually the infant would be squirming and

wriggling around while giving the call but it was impossible

to tell if it was trying to nurse. The adult holding the

infant would either shift it in its arms or else let another

colobus take it.

The frequent, at times almost constant, geckering by

the infant was in sharp contrast to the usual nonvocal atti-

tude of the rest of the troop. As a result, the somewhat

noisy infant seemed to be able to solicit attention and be an

important focal point for the rest of the group.

Energy distribution of the gecker ranged from 0.25

to 13.00 kHz, but on the sonogram the darkest portion averaged






3.64 6.69 kHz (N=11). The mean phrase duration was 0.32

seconds (N=11) and several such single-unit phrases were com-

bined to give bouts of variable length. It is difficult to

assess the duration of a bout or the number of phrases per

bout since the call varies throughout in loudness and gives

an impression of randomness and irregularity rather than of

a fixed pattern of phrase sequences.

TABLE 17

GECKER WHITE INFANT


Number (N) Mean (x) Range (R)


Units/Phrase 11/11 1.00 1.00-1.00

Phrase Duration 11 0.32 0.27-0.40
secss.)

Interval Between
Phrases secss.) 0 0.13 0.05-0.28

Lower Limit of
Major Energy 11 3.64 2.00-4.50
(kHz)

Upper Limit of
Major Energy 11 6.69 6.10-7.70
(kHz)


The gecker is both tonal and nontonal, with the tonal

section first forming an arch-like pattern on the sonogram

which then levels out into the noise component. As mentioned

above, one of these "arches" is considerably darker than the

rest; the others occur above and below it in a harmonic

pattern.






Err White Infant

On one occasion a phrase consisting of three "err"
sounds was heard at the end of a geckering sequence (Fig. 29

and Table 18). The exact relationship of this call to the

gecker is not understood; it may be a part of the gecker or

it may be an individual vocalization. However, on the sono-

gram the call is different in appearance from the gecker.

TABLE 18

ERR WHITE INFANT


Number (N) Mean (x) Range (R)

Units/Phrase 3/1 3.00 3.00-3.00

Unit Duration 3 0.11 0.03-0.15
secss.)

Interval Between
Units secss.) 2 0.05 0.05-0.05

Phrase Duration 1 0.35 0.35-0.35
secss.)

Lower Limit of
Energy (kHz) 3 0.63 0.50-0.90

Upper Limit of
Energy (kHz) 3 2.13 1.80-2.50


The mean duration of the three tonal units was short
at 0.11 seconds (N=3). The "err" was lower in pitch than the

gecker, with a mean energy range from 0.63 to 2.13 kHz (N=3).

The duration of the phrase as a whole was 0.35 seconds.






Gecker Juvenile

A call similar acoustically and spectrographically to

the gecker of the white infant was given by a juvenile (Fig.

30 and Table 19). This individual had already attained its

black and white coloration and, except for its smaller size,

it was indistinguishable from the adult colobus. The juven-

ile no longer seemed to be very dependent on its mother al-

though it was carried occasionally, apparently when frightened

by the observer.

The juvenile gecker is also composed of tonal arch-

like bands passing into nontonal components. Harmonics are

present but the bands are closer together and more numerous

than in the infant gecker.

TABLE 19

GECKER JUVENILE


Number (N) Mean (x) Range (R)

Units/Phrase 13/13 1.00 1.00-1.00

Phrase Duration 13 0.11 0.04-0.14
secss.)

Interval Between
Phrases secss.) 11 0.02 0.01-0.05

Lower Limit of
Major Energy 13 2.60 1.90-3.80
(kH z)

Upper Limit of
Major Energy 13 8.05 6.00-10.10
(kHz)






The individual single-unit phrases were of shorter

duration than those given by the white infant, with an aver-

age length of 0.11 seconds (N=13). The average energy dis-

tribution ranged from 2.60 to 8.05 kHz (N=13), but again one

band was somewhat darker than the rest at approximately 4.0 -

5.0 kHz. As in the infant gecker, it is very difficult to

judge the duration of a bout or to determine the number of

phrases per bout.

Occurring in the geckering bout on one occasion was

a phrase which appeared on the sonogram as a single, wavy,

linear unit, 0.54 seconds in duration (Fig. 31). The unit

is basically tonal and occupies a frequency range of approx-

imately 2.90 6.00 kHz. The darkest portion was again in the

vicinity of 4.0 5.0 kHz so perhaps this phrase is but a

series of geckering phrases given close together and lacking

harmonic development.

The gecker was given by the juvenile as it was startled

by the observer and ran to an adult female. Upon being em-

braced by the female, the juvenile stopped vocalizing.













CHAPTER IV


VOCALIZATIONS OF THE BABOON



Vocalizations and Crop-Raiding


The information collected on the baboon vocalizations

formed part of a larger study on baboon crop-raiding behavior.

Unfortunately, conditions were somewhat less than ideal at the

Shimoni site during the time this observer was in the field.

This site promised to be the best suited for making recordings

and observing the baboons' behavior. However, due to seasonal

conditions and a long drought in the area, no significant

crops were being grown other than a few remnants gone to seed

from the last growing season. In addition, the Africans who

usually sit up in observation platforms in order to spot ba-

boons and chase them from the field were not present.

When in the area a few years earlier, Maples (per-

sonal communication) noted what, superficially at least,

appeared to be diversionary behavior. It seemed that one

group of baboons would vocalize and thus attract and keep

the attention of the shamba, or farm, guards while the rest

of the troop would silently move into the farm a short dis-

tance away and raid the crops growing there. If this behav-

ior were indeed coordinated and intentional, it would shed

an interesting and new light on the nature of communication

70





71
systems among nonhuman primates. It wa's the purpose of this

part of the study to investigate and record those vocaliza-

tions given in conjunction with this behavior.

Although the normal raiding conditions were absent

during the course of this observer's investigations, the

baboons did come to the shamba practically every day to for-

age for whatever remains could be found in the field. As a

result, enough information was collected to allow at least a

tentative explanation as to the nature of this seemingly de-

liberate "decoy" behavior.

Although shamba guards were lacking and consequently

the raiding situation not as tense as it might have been under

normal circumstances, it was noted that the baboons, in reac-

tion to an observer moving about rapidly in the field, would

move into the forest edge and gradually begin to move silently

around the perimeter of the farm. The baboons at the rear of

this troop movement would be up in the trees giving the

threat/alarm bark. It is to these vocalizing baboons that

the observer's attention is drawn. Consequently, he remains

nearer to this group of baboons and, presumably, the shamba

guards would be inclined to do the same. The vocalizing

animals tend to minimize their own exposure to danger by

sounding the alarm when the human intruder is moving away or

looking in another direction, a practical behavioral adap-

tation since shamba guards may be armed with rocks, sticks,

or bows and arrows. With the humans' attention thus drawn

to this vocalizing group, the rest of the troop is relatively

free of human intervention and has the opportunity to enter

the shamba again to carry out additional raids.





72
It would seem, therefore. that even though the vocal-

izations given during the raids are diversionary in nature,

they are probably not deliberately intended to be so. Instead,

the calls given in reaction to the intruding human help to set

up a fortuitous chain of events which allows part of the troop

to continue raiding.

At Roka, similar observations were made although

the conditions were somewhat different. Instead of a farm

surrounded on three sides by forest, as was the case at Shi-

moni, the farms at Roka were separated from the forest by a

fairly well-traveled road. When raiding, the baboons would

have to come out of the forest and cross the road to enter

the farms. Unlike Shimoni, crops were being grown in most

of the fields and some farms were protected by shamba guards,

usually women or children. In general, these baboons were not

very vocal at all but would carry out their raids in silence.

But again, it was observed that vocalizations of any kind or

any movement in the trees would tend to attract the guards

to that area while further down the road other baboons would

be free to enter the farms undisturbed. Although not inten-

tional, such behavior functions very well as a decoy mechanism.

The calls described in the following sections were

recorded during encounters such as those described above.



Classification of Calls


Two-Pnase Bark

e two-ohase bark (PFis. 32&33 and Table 20) was re-

corded on one occasion at Roka. This was just after a shamba





73
raid had been completed and all the baboons were on the for-

est side of the road. The observation vehicle was driven

adjacent to the point of entry into the forest. It was then

that a large, fully adult male conspicuously sitting in a tall

tree at the forest edge emitted two two-phase barks. The two-

phase bark has been reported by Hall and DeVore (1965) as an

attack/threat vocalization given in reaction to humans and

large predators.

TABLE 20

TWO-PHASE BARK


Number (N) Mean (x) Range (R)

Units/Phrase 2/2 1.00 1.00-1.00

Phrase Duration 2 0.62 0.51-0.73
secss.)

Lower Limit of
Energy (kHz) 2 0.25 0.25-0.25

Upper Limit of
Energy (kHz) 2 4.10 3.20-5.00




The two-phase bark is a loud, tonal, single-unit

phrase of low frequency and long duration. The energy dis-

tribution for the first two-phase bark is from 0.25 to 5.00

kHz and the distribution for the second is 0.25 3.20 kHz.

However, the heaviest concentration of energy for both calls

is in the 0.50 1.50 kHz range. The duration of the calls

is 0.73 seconds for the first and 0.51 seconds for the second.




74
On the sonogram the call appears as a very dark band

in the lower frequencies with some harmonic development.


Bark

Humans moving about in the farm at Shimoni would usu-

ally provoke bark-like vocalizations (Figs. 34&35 and Table

21) from nearby baboons who would then withdraw into the

forest.

There seemed to be two types of barks distinguishable

both acoustically and spectrographically; however, this may

be due to sex and/or age differences. Basically, the differ-

ence seems to be one of pitch, with one bark type noticeably

lower and less intense. Both may be described as single-unit

phrases.


TABLE 21

BARK


Number (N) Mean (x) Range (R)


Bark I

Units/Phrase

Phrase Duration
secss.)

Lower Limit of
Energy (kHz')

Upper Limit of
Energy (kHz)

Bark II

Units/Phrase

Phrase Duration
secss.)


12/12

12



12


7/7

6


1.00

0.32



0.27


1.00

0.32


1.00-1.00

0.20-0.40



0.25-0.50


3.70-5.00



1.00-1.00

0.25-0.45






TABLE 21 (Continued)


Number (N) Mean (x) Range (R)

Lower Limit of
Energy (kHz) 7 0.48 0.40-0.70

Upper Limit of
Energy (kHz) 7 2.88 2.50-3.00


The average energy distribution for Bark I is 0.27 -
4.13 kHz (N=12). The call is tonal and harmonics are well

developed. The mean duration is 0.32 seconds (N=12).

Bark II is lower in pitch with an average energy dis-
tribution of 0.48 2.88 kHz (N=7). The harmonic pattern

does not seem to be as well developed here as in Bark I and

the upper frequencies of the call are somewhat weaker. The

call averages 0.32 seconds in length (N=6).


Yak

The yak (Fig. 35 and Table 22) was believed to have

been given by a juvenile. It was given under the same condi-

tions as the barks and was given in conjunction with them.

Structurally the yak is very similar to the barks
(especially Bark I) on the sonogram. It too consists of a

single-unit phrase. The main difference is that the yak is

higher in pitch and has a very well-developed harmonic pat-

tern. The yak is also acoustically distinguishable from the

barks.






TA3LM 22

YAK


Number (N) Mean (x) Range (R)

Units/Phrase 3/3 1.00 1.00-1.00

Phrase Duration 2 0.30 0.30-0.31
secss.)

Lower Limit of
Energy (kHz) 3 0.50 0.50-0.50

Upper Limit of
Energy (kHz) 3 6.40 5.00-8.0+



The average energy distribution of the yak is approx-

imately 0.50 6.40 kHz (N=3); however, the harmonic pattern

on one occasion extended beyond 8.0 kHz. The average dura-

tion of the call is 0.30 seconds (N=2).

On the occasion when the calls were recorded the yak

occurred in a repeated sequence of Bark I, Yak, Bark II

given while an observer moved about in the shamba at Shimoni.

The baboons giving the calls were at the forest edge. While

they were vocalizing, the rest of the troop was quietly cir-

cling the farm a short distance within the forest.













CHAPTER V


SUMMARY AND CONCLUSIONS



The study of primate behavior developed rapidly in

the years following World War II to become a popular disci-

pline within the fields of zoology and physical anthropology.

However, one aspect of its study, primate communication, has

only recently begun to receive the attention which it rightly

deserves. That as yet no single study has been able to deal

with social communication in all its aspects is testimony to

the complexity of the task. Instead, individual researchers

have chosen to study some specific aspect or charnel of com-

munication, usually of a single species, in the hope that

with enough information collected in this manner a larger,

more comprehensive overview of the communication process as

a whole may eventually be possible. It is with this intent

that the present study has been conducted.

By means of sonograms the vocalizations of three

species of Old World monkeys (Sykes monkey, colobus monkey,

and baboon) were illustrated, described, and analyzed. The

conclusions and findings are summarized in the following

sections.






The Sykes Study


I) Fourteen Sykes vocalizations were catalogued and

described in terms of their physical characteristics (fre-

quency, duration, tonality, etc.). The possible functional

significance of the calls and any accompanying behavior was

also presented.

2) The vocalizations were examined in terms of dis-

creteness vs. gradedness. Most of the calls were found to be

of the discrete type but calls forming a possible gradient

were also discussed. It was emphasized that demonstrating

the existence of graded signals does not imply the existence

of a graded communication system.

3) The Sykes vocalizations show considerable variety

in pitch or frequency and timing. An attempt was made to

correlate these differences in structure with differences in

functional requirements, particularly the need for localiza-

tion. It was found that the properties of calls concerned

with group cohesion and alarm or threat which were transmitted

over distances without necessarily the aid of visual contact

were such as to facilitate localization of the vocalizer.

On the other hand, the structure of calls which were trans-

mitted at close-range where visual or other cues may be oper-

able did not show those features which would facilitate local-

ization.

4) The taxonomic position of the Sykes monkey, Cer-

cocithecus mitis, in relation to C. nictitans has been a

matter of some speculation. A preliminary analysis and





79
comparison of the vocalizations of the two species indicated

that separate specific status should be maintained at this

time. However, analysis revealed that C. mitis was more closely

related to C. nictitans and C. erythrotis-ceohus than to the

mona superspecies.



The Colobus Study


1) Two calls were described for adult colobus as well

as one for juveniles and two for infants. When possible,

their probable functions were discussed and any accompanying

behavior was described.

2) Although vocalizations were the primary concern

in this study, a nonvocal communication pattern, the jumping

display, seemed to play an important role in the life of the

relatively nonvocal colobus. This behavior was described

and its possible functional significance was discussed.



The Baboon Study


1) The physical characteristics of four types of

calls given during crop-raiding were described.

2) The apparent "decoy" effect of the calls in

attracting the attention of farm guards protecting the crops

while other baboons silently entered the fields to raid was

described and analyzed. It was concluded that the calls

were fortuitously diversionary in nature rather than repre-

senting a deliberate and intentional plan of behavior.






There had been a tendency in studies of primate be-

havior and communication to accept as generalizations for all

species that which had been described for but a few well-

studied populations. That the behavior of all species can

not be seen in a baboon-rhesus context or that the behavior

of all baboons is not typified by the Nairobi Park population,

for example, has become increasingly apparent as more and more

primate species and populations in various ecological settings

are being studied.

Before the formation of encompassing generalizations

then, if indeed such should prove to be possible, the primary

need in communication studies is one of collection of data

on as many species as possible. Ideally, information on com-

munication systems should be accompanied by information on

behavioral context and functional significance. Through ob-

servation of the behavioral context accompanying a communi-

catory act, something can be learned regarding the functional

significance of a particular vocalization, facial expression,

etc. Analysis of the relationship between structure and

function may indicate why certain structural traits have

proved adaptive and have been maintained through the process

of natural selection. Comparison of such findings from sev-

eral studies on separate species or on different populations

of the same species may reveal ways in which the ecological

setting can influence which adaptive mechanisms will predom-

inate. In addition, an understanding of the degree of vari-

ation to be found in communication patterns within species

will aid in taxonomic interpretations and in the assessment





81
of vocalizations and other channels of communication as tax-

onomic characters. In this way, one can begin to understand

something of the relationship between communication, organ-

ized social life, and ecology: of man's closest relatives,

the nonhuman primates.

Finally, it should be emphasized that primate commun-

ication should not be studied solely in terms of how many fea-

tures of human language may or may not be present, but also

as an entity worthy of study in its own right. However, an

understanding of the nature of communication systems among

nonhuman primates and how these systems relate to social or-

ganization and environment may eventually shed light on the

circumstances accompanying the origin of language in man and

on the nature of this early system.

That primate communication is a behavior which has

proved adaptive and evolutionarily successful can not be dis-

puted, but the mechanisms behind that adaptation are as yet

not fully understood. It is hoped that through further study

and fieldwork more and more information may be collected and

correlated to reveal something of the adaptive significance

of communication systems in general and of how these systems

are integrated into the larger whole of primate social life.

It is this intricate weaving and integration of behavioral

systems, of which communication is but one, which makes the

coordination of individuals into a society possible, and

which in turn insures the survival of the species.






























PLATES






























SYKES MONKEY









16


12


8 iZ'j.


4
0i i~~


,9 LYI1
r


i I i





"' i. L 'I


4 jt -
'illef I ll, 1151( Ift s Ieritt 1 i

0.20


ft 1
S.4sra II

0.40


Time secss.)


Figure 1.


J I l .

1 i- pi'
,r~i^1
'I 'i l '


ll^^''j^


p*


0.20


.4
0.40


Time secss.)


Figure 2. Chirp


Chirp


I.
Si


O ,l


*


tl; -:


fiahtlliM
~a me
EMMAA









16


12 R
K~t


0 0.20 0.40


Figure 3. Two-Unit Chirp


Time secss.)


i" I ;

i ** 5" ,'V. , ; '" d
^^^&:i^{^i^M


0 0.20 0.40


Figure 4. Two-Unit Chirp


Time secss.)


16 .

12 S -












12
3MLilal


0.20 0.40


Figure 5.


Time secss.)


Three-Unit Chirp


i .1





0 0.20 0.40


Time secss.)


Figure 6. Nyah


0








16

hKa

12 -L:4MR
EAsiM
8^M


tt!


F i I


Ia
1..rrtc~l


0.20


0.40


Time


(sees.)


Figure 7. Short Nyah

(short nyah followed by chirp of another Sykes)


16 ,. .

12 ;n A:
' -8..



0 .I'r
IaX h


II

'il :.




S1 0-.


0.20


0.40


0.60


Time secss.)


Figure 8. Short Nyah


(short nyahs followed by chirp of another Sykes)


-' y
*I 'I


0


i













jam
Vii,





.. I i I i H-



-^W~tMKHM^^^^^^


0.20


0.40


Time (sees.)


Figure 9. Boom

(ignore background noise at 3 kHz and 8-10 kHz)


I



ai 2 7 -- _-....- ..... ... _-.Qr,. ,..,..


0.60 Time


secss.)


Figure 10. Scueal


- I

16 rm

12 B










o


Fuir


0.20


0.4C


t,


I


.s~rdc,:




Full Text

PAGE 1

AN ANALYSIS OF VOCALIZATIONS OF THREE SPECIES OF EAST AFRICAN CERCOPITHECIDAE By MARY LOUISE WALEK A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1972

PAGE 2

Copyright by Mary Louise V/alek 972

PAGE 3

With Love to lorn, Dad, and Bruce

PAGE 4

ACKN OWLEDGMENTS I would like to thank Dr. William R. Maples, chairman of my supervisory committee, for his guidance and criticism, and for his assistance in making this study possible. It was he who made my trip to East Africa a reality and for this I will always be grateful. Appreciation must also be extended to the other members of the committee, Drs . Norman N. Markel and Charles H. Fairbanks, for taking time from their busy schedules to read and evaluate this study, I would also like to thank Dr. Arnold Faige of the Departments of Speech and Electrical Engineering of the University of Florida for giving unselfishly of his time and for his generosity in making the sonogra ph and other materials accessible to me. Without his kind cooperation this study would not have been possible. For the use of their laboratory and additional tape recorders during the editing process, I would like to thank the Department of Communication Sciences, also of the University of Florida. To Bruce Walek I extend sincere appreciation for the photographic reproduction of the sonograms, even though it meant taking valuable time from his own studies to do so. Finally, I would like to express my gratitude to the members of my family for their support and encouragement iv

PAGE 5

which contributed as much to the completion of this study as the technical assistance received elsewhere. The study was financed by grants to William R. Maple; from Biomedical Sciences of the University of Florida and the National Science Foundation (G3 2610).

PAGE 6

TABLE OF CONTENTS LIST OF TABLES Page ACKNOWLEDGMENTS iv V ill LIST OF FIGURES , j_ x ABSTRACT xi CHAPTER I INTRODUCTION 1 The Study of Primate Vocalizations ...... 2 Objectives of the Present Study 6 The Study Sites " 6 Methods and Materials of Research 7 Definition of Terms 9 II VOCALIZATIONS OF THE SYKE3 MONKEY 12 Classification of Calls 13 Discrete and Graded Signals ... 35 A Structural-Functional Analysis ....... i^o Taxonomic Implications ............ i|8 III VOCALIZATIONS OF THE C0L03US MONKEY 58 Classification of Calls . 61 IV VOCALIZATIONS OF THE 3ABC0N 70 Vocalizations and Crop-Raiding 70 Classification of Calls . . , , , 72 V SUMMARY AND CONCLUSIONS 77 The Sykes Study . . . * 78 The Colobus Study , 79 The Baooon Study ............... 79 PLATES SYKES MONKEY 63 vi

PAGE 7

TABLE 0? CONTENTS (Continue J age PLATES COLOBUS MONKEY 95 BABOON 100 REFERENCES 103 BIOGRAPHICAL SKETCH 109 vi 1

PAGE 8

Table

PAGE 9

LIST OF FIGURES Fi ^ re Page 1 Chirp Qk 2 Chirp 8 ^ 3 Two-Unit Chirp q? 4 Two-Unit Chirp or 5 Three-Unit Chirp 35 6 Nyah 86 7 Short Nyah go 8 Short Nyah go 9 Boom 00 10 Squeal 88 11 Shriek gg 12 Chut 89 13 Chut 90 14 Chutter aQ 15 Low, Short Chutter Q1 16 Chutter-Squeal Q1 17 Chutter-Squeal (continued) Q2 18 Chutter-Squeal (continued) ... 92 19 Growl 93 20 Growl g ~ » 21 Srr 94 22 Uh « q4 IX

PAGE 10

LIST OF FIGURE (Continued) Figure Page 23 Snort 96 24 Snort 96 25 Snort-Croak ........ ..... 97 26 Snort-Croak 97 27 Snort-Croak 97 28 Gecker White Infant 98 29 Err White Infant 98 30 Gecker Juvenile 99 31 Gecker Juvenile 99 32 TwoPhase Bark 101 33 Two-Phase Bark 101 3^ Bark I , 10 > 35 Bark I, Yak, Bark II 102

PAGE 11

5 If C T t ?* Dlss ertation Presented +c the Graduate Council of ,ne University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AN ANALYSIS OF VOCALIZATIONS OF THREE SPECIES OF EAST AFRICAN CERCOPITHECIDAE By Mary Louise Walek August, 1972 Chairman: William R. Maples Major Department: Anthropology Vocalizations of three species of Old World monkeys (Sykes monkey, Cercopithecus mitis; colobus monkey, Colobus JPQ lykor nos; and baboon, Papio otocephalus ) were collected in Kenya, East Africa. Sonograms of the vocalizations were made and analyzed in terms of their physical characteristics (frequency, duration, and tonality). Representative sonograms are illustrated and tables of all measurements made are presented, Accompanying behavioral information is also included whenever available. Fourteen Sykes vocalizations are catalogued and described. An analysis of the calls in terms of discreteness and gradedness revealed that most of the vocalizations were of the discrete type. The Sykes vocalizations were also examined to see if physical properties, cr structure, could be related to function. It was found that those calls transmitted over distances which were important for group cohesion and alarm or xi

PAGE 12

threat were structured so as to facilitate locating the caller. Calls operating at close-range did not have such localizing prooerties . The question of whether Cercooithecus mitis is conspecific with C. nictitans has been raised in the literature. This study evaluated the taxonomic status of these two species. Based on similarities and differences in the vocalisations it was concluded that separate specific identity should be maintained. However, C. mitis was found to be more closely related to C. nictitans and C. erythrotis ceohus than to the mona superspecies . Five vocalizations of the colobus monkey are described: two for infants, one for juveniles, and two for adults. A nonvocal behavior pattern, the jumping display, was found to be of communicative significance in alerting a troop to danger and in indicating the direction of troop movement. It also functioned as an intimidation display directed at the observer. The role which vocalizations play in the cropraiding activities of baboons was investigated and the apparent "decoy" effect of the calls in distracting the attention of Africans protecting the fields was examined. It was concluded that the vocalizations did not represent a deliberate and planned decoy tactic, but instead were a fortuitously diversionary behavior which contributed incidently to the success of the raids. Four types of calls giver, during the raids are described. XII

PAGE 13

CHAPTER I INTRODUCTION The study of primate communication is assuming an increasingly important role in the field of primatology. As more is learned of the behavior of man's closest relatives, the nonhuman primates, the more apparent it has become that understanding how behavior patterns are transmitted and interpreted will contribute considerably to understanding the adaptive mechanisms of social life. Many of the earlier studies on primate behavior, and even some of the mere recent, either failed to include communication as an aspect of behavior to be investigated or else dealt with it in such a cursory and unscientific manner as to make it almost worthless for further investigation and comparison. Fart of the problem is no doubt related to the allpervas ive nature of communication into all aspects of social life, a fact which is well illustrated by the following definition: "...social communication is the process by which the behavior of an individual affects the behavior of others" (Altmann, 1967:326). Communication as it has been defined can include many different forms of behavior occur ing in concert and involving several individuals. It becomes apparent, therefore, that the study of communication can be a complex undertaking. 1

PAGE 14

2 Further complications become apparent when one considers that, communication can take place via several different channels. Signals or messages may be olfactory, tactile, visual, or auditory in nature and for any specific signal transmission several of these channels of communication may be used simultaneously. Each of these channels has been investigated in studies of communication among the primates (see Andrew, 1963a, 1964; Bolwig, 196^; Hinde and Rowell, 1962; Jolly, 1966; Moynihan, 1967; van Hooff, 1962). The Study of Primate Vocalizations The primary concern of the present study, however, will be with primate vocal communication, the study of which has taken various forms. One approach consisted of transcribing the various calls fay using letters of the alphabet in order to give as close an approximation as possible of the sound of the call (for an example see Ullrich, 1961). Not only was this system largely subjective in nature, but the alphabetical notation used was generally that of the language of the observer and hence was not accessible to, or reproducible by, researchers of all nationalities. In order to overcome some of the subjectivity of such an approach and to insure a certain amount of uniformity, there was an attempt to use the standard International Phonetic Alphabet (for an example see Hill and Booth, 1957)« However, it soon became" apparent that nonhuman primates were capable of producing sounds which

PAGE 15

lacked counterparts in human speech. Phonetic and alphabetical transcriptions, therefore, could not render an accurate description of many vocalizations and often gave a false, human quality to the vocal patterns of primates other than man The most common approach to the study of primate vocal communication until recently has been the assignment of a descriptive name to a vocalization, one which depicts as closely as possible the type of call given. In this way such terms as 'squeal', 'click', and 'grunt' were used but were really only of value when as complete an acoustic description as possible accompanied the assignment of such a term. Moreover, adjectives such as 'high-pitched', 'shrill', •harsh', etc., used in such descriptions, were still ambiguous and subjective in nature. Hence, comparisons of vocalizations reported in different studies remained difficult, if not impossible, to carry out. The problem of subjectivity was alleviated to a considerable extent with the development within the last two decades of sophisticated, sensitive recording devices and the invention of the sound spectrograph. The spectrogram, or sonogram, supplies a permanent, visual record of a sound and serves both as a form of notation and as a method of precise measurement. Although descriptive terms remain prevalent as names or labels for the various calls, they are now supported by a more concrete, objective, physical description than was possible before.

PAGE 16

4 Receiving its impetus largely fom studies on insects, anurans, and birds, the use of sonograms in the study of primate vocalisations has spread rapidly within the last several years. Most of the work on primate communication has been with species of Old World monkeys, especially the macaques, baboons, and vervets. Rowell (1962) used sonograms and described nine agonistic noises of the rhesus macaque, relating each to the various postures and expressions which they accompany. Struhsaker (1967) has done an excellent spectrograph^ analysis of vervet vocalizations combining a detailed physical description of the calls with behavioral lata whenever possible. Although numerous studies on baboon communication have been reported, sonograms for this species are apparently lacking in the published literature. Communication studies involving spectrograph 1c analysis have also been reported on several other species, such as those of Reynolds (in Marler, 1965) on the chimpanzee, Schaller (1963) on the gorilla, Moynihan (1964) on the night monkey, Altmann (1967) on howlers, Ploog (1966) on squirrel monkeys, Fetter and Petter (1966) on the ayeaye, and Fetter (1962) on lemurs. For a more complete listing of studies on primate communication, spectrographic or otherwise, the reader is referred to Altmann's (1968) comprehensive review. Although Old World monkeys have been extensively studied in terms of behavior and communication, several genera within the family Cercopithecidae have not been adequately investigated. Among these are the species of the genera Cercopithecus and Colobus.

PAGE 17

Other than Struhsaker's (1967) study on the vervet, Gercooithecus aetniops, ana most recently his investigations of the vocalizations of several Gercooithecu s species (1970), very little information on vocalizations is available for the Cercopithecus group. Most of what is known of the auditory communication of these species has been drawn out of more gen eral behavioral studies or from anecdotal accounts. other researchers on the vervet include Booth (l 9 62), Gartlan and Brain (1968), and Hall and Gartlan (1965). None of these studies presents spectrograph^ descriptions. The redtail monkey, C. as_canius, is the only other species investigated to any extent and the reader is referred to the works of Buxton (1 952 ) and Haddow (1952)> rtgain> spectrograph . c data is not available for this species in these studies. Very little is known of the behavior of the Colobus species. While considerable research has been carried out on the Asiatic counterpart of the Colobinae, the langurs, the African members of this subfamily have been largely ignored. Available reports include those of Booth (1 957 ) on the olive colobus and Hill and Booth (1957) and Ullrich (l 9 6l) on the black and white colobus. However, information as to social communication is rather sparse and incomplete in these studies. A more informative sxudy is Marler-s (1970) spectrograph^ analysis of the vocalizations of the red colobus.

PAGE 18

Objectives of the present Study In the present study the vocalizations of two species of these relatively little-studied genera, the Sykes monkey, Cer co oith ecus mitis kibonotensis . and the black and white colobus monkey, Colobus polykomos palliatus . will be described and analyzed. The basic aim of this study is to present a catalogue of the vocal repertoire of these species. Accompanying behavioral information will also be presented whenever available. The vocalizations of the baboon, Papio cynocechalus . will also be described. This investigation formed part of a larger study being conducted at the time by William R. Maples, William F. Greenhood, and myself on crop-raiding behavior of baboons. The aim of this portion of that study is to describe those vocalizations given while crop-raiding and to investigate the role they play in this activity. All descriptions and classifications of the calls are based on spectrographs analysis. The Study Sites The field observations upon which the following study is based were conducted in Kenya, East Africa, from September to December, 1969. The study site for the Sykes and colobus monkey material was at Diani Beach on the Kenya coast, approximately eighteen miles south of Mombassa. The site itself was approximately one-quarter to one-half mile inland. The

PAGE 19

? material on baboon vocalizations was collected at farm sites primarily at Shimoni, a small fishing village near the southern tip of the Kenya coast. Some recordings were also made near Roka along the Mombassa-Malindi road, approximately twentyfour miles south of Malindi. In general, the climax vegetation community for the Diani Beach/Shimoni area has been described as lowland dry forest on coral rag ( Combretum schumanii Cassipourea species) (Moomaw, i960). The forests at Diani Beach are broken into smaller segments of varying size by the interspersal of lantana thicket and cultivated land. The forest essentially consists of two layers, an upper canopy composed of tall, branching trees and a low, dense understory where bushes and vines predominate. The colobus were observed to favor the upper canopy but would readily take refuge in the lower layer when necessary. The Sykes seemed equally at home in both layers but it was this observer's impression that more time was spent in the understory. At Roka, the baboons inhabited a protected forest which has been classified as lov/land woodland (Bxachysj2e^i^-Aj^eiia species) by Moomaw (i960). For a more detailed discussion of the ecology of these areas and the species of vegetation which predominate, the reader is referred to Moomaw (i960) and Greenhood (1971). Methods ,ari d_g;:aterials of Research Recordings of the vocalisations were made with a Nagra IV L portable tape recorder at 15 i p3 with a frequency

PAGE 20

8 response of 30 to 20,000 cps, and a Sennheiser directional microphone, Model MKH 80^. Sonograms for analysis were made on a Kay Electric Sona-Graph, IVIodel 6o6l-B. A transparent plastic grid overlay was made showing time on the horizontal axis and frequency on the vertical axis. The grid was placed over the sonograms and approximate frequency and temporal measurements were taken. Field recording techniques at Diani Beach consisted of approaching as closely and quietly as possible on foot a Sykes or colobus troop. The thick, tangled vegetation of this area precluded any possibility of following a troop on its daily movements; therefore, recordings were made along roads and footpaths which troops regularly frequented and traversed. At Shimoni, a stationary recording site was set up within the farm and baboon vocalizations were recorded as the troops approached and entered the field to raid. The recording station at the Roka site was within an observation vehicle which would follow and remain adjacent to the movements of a troop of crop-raiding baboons as they emerged from the forest and crossed the iMalindi-Mombassa road to raid the farms on the other side. In all cases, accompanying behavioral observations were recorded in a notebook. In areas of limited visibility it was the observation of behavioral context which proved to be the most difficult. A single individual attempting to operate a tape recorder, direct a microphone, maintain visual contact with a troop moving in dense underbrush, and take notes is limited as -co

PAGE 21

9 the amount of behavior which can he observed. This limitation was particularly felt in the Diani Beach area. At Shimoni and Roka the problem was not as acute as other workers (notably William R. Maples) were simultaneously observing crop-raiding behavior and so correlation of the vocalizations with their observations was possible. It would seem, therefore, that perhaps the most fruitful means of studying communication with all its behavioral ramifications within a somewhat hostile environment would be through a team approach. Such a system would minimize the physical limitations and would facilitate the recording of observations. In addition, it would provide a system of checks and counter-checks on what was observed. Perhaps in savanna areas, with visibility relatively unimpaired, these difficulties do not pose as serious a hinderance as when dealing with forest populations. In the latter case, the physical limitations imposed by the very nature of the study itself are only too apparent. Definition of Terms So that there may be some form of common reference in discussions of primate vocalizations, use will be made here of the terms defined by Thomas T. Struhsaker in his study on auditory communication of vervet monkeys. These are as follows : Mit: The unit is the basic element of a call, and is represented as a continuous tracing along the Temporal (horizontal) axis of the sonogram.

PAGE 22

10 Phrase: The phrase is a grouo of units that is separated from other similar groups by a time~irberval greater than any time interval separating the units within a phrase. Bout: A bout is a grouping of one or mor^ phrases separated from other similar grouoings bv a time interval greater than that separating any of the phrases within a bout. Nontonal unit: A nontonal unit is comoosed of sound that is more or less continuously developed over a wide range of frequencies. This has also oeen called "noise" by Andrew (1963b) and "harsh noises" by Rowell and Hinde (1962). Tonal unit: A tonal unit is composed of sound characterized by one or more relatively narrow frequency bands and has been referred to as "clear calls" by Rowell and Hinde (1962) and "sound" by Andrew (1963b). Units with a harmonic structure are included in this category. Compound unit: A compound unit is composed of both nontonal ana tonal sounds that appear as a continuous tracing on the sonogram. Mixed unit: Units comoosed of both tonal and nontonal sounds that are rather superimposed on one another are called mixed units. The tonal and nontonal aspects are more or less separated by differences in frequency. Distribut ion of major energy of nontonal sounds : The distribution of the major energy of a nontonal sound is represented and thus determined by the darkest portion of the tracing on the sonogram. This distribution is generally over a smaller range than the frequency range of nontonal sounds. The distribution of major energy of nontonal sounds is described oy an upper frequency (highest pitch of major energy) and a lower freauency (lowest pitch of major energy) . (Struhsaker, 1967:282-263) For a brief demonstration of how these terms will be applied in the present study, the reader is referred to Figure 26 and Figure 6. Figure 26 is a spectrographs representation of a colobus vocalization consisting of three different phrases.

PAGE 23

11 The first is a nontonal, single-unit phrase and is given only once at the beginning of the call. The second phrase consists of several tonal units. The third is a nontonal, single-unit phrase very similar to the first. The second and third phrases are repeated in alternation. All three Phrases are said to form a bout, part of which is depicted in the sonogram. The entire bout can be referred to as a -callor "vocalization". Figure 6 represents a call of Cexcejdthecus jutis and can be described as a single-unit phrase. It is alsc a compound unit since a brief tonal component consisting of gently arching frequency bands is followed by a noise or nontonal component with a somewhat irregular appearance on the sonogram. For a further demonstration of how these terms are used, the reader is referred to the descriptions of the various calls presented in the text which can be compared with the appropriate sonogram.

PAGE 24

CHAPTER II VOCALIZATIONS OF THE SYKES MONKEY The Sykes monkeys ( Cercopithecus mitis kibonotensis ) live in social groups of substantial size. Various troop counts, made while the monkeys passed over an "arboreal bridge" spanning a roadway, revealed totals of 13, 18, 26, and 31. However, it never was clear how many troops were involved or if an entire troop had been counted. Sykes are primarily an arboreal species but, as mentioned previously, seem to be equally at home in the tall trees as in the lower brush. They seem not the least bit reluctant to move along on the ground, especially in those areas where the vines are quite thick and tangled. The Sykes would vocalize quite readily. The chirp was the most frequently heard call and would be elicited by any sudden movement or noise, or by the approach of humans or dogs. In the initial stages of the study, the observer was constantly the object of much threat behavior and chirping. However, as the Sykes became habituated to the observer's presence and with the thickening of the foliage due to rain, the amount of vocalizations given was considerably reduced. 12

PAGE 25

13 Classification of Calls C hirp The chirp (Figs. 1&2 and Table 1) was given by adult females and immature monkeys and possibly by adult males, although positive identification of the latter was not possible, That adult male Sykes might give the call is doubtful in light of Struhsaker's (1967, 1970) reports for the vervet and other Cercopithecus species in which he noted that chirps were not given by adult males. The chirp was an abrupt, sharp-sounding call of high frequency and short duration. The frequency range was 0.10 16.0+ kHz but the average distribution of the major concentration of energy was 2.91 6.14 kHz (N=2 9 ) . Sonograms revealed a single-unit phrase with an abrupt beginning consisting of a narrow, tonal, chevron-shaped component followed by a noise component of longer duration. The noise component of approximately half of the chirps was distributed in a twoband arrangement, while the other half consisted of a single, broader band (Figs. 1&2). The mean duration of the chirp was 0.1 q seconds (N=29) . The chirp apparently functions as a threat-alarm call. It is usually given in conjunction with a threat display in which the monkey either crouches forward with legs stiff and bobs its head and shoulders up and down, or else lunges forward on its foreli-abs. in both cases the monkey stares intentl2 at the intruder. This behavior was directed toward ""''' : and humans but never ^ observed between one Sykes ana

PAGE 26

14 another, A similar observation was made by Haddow (1952) in his study on the redtail monkey, Cercooithecus ascanius. TABLE 1 CHIRP Number (N) Mean (x) Range (R) Units/Phrase 29/29 1.00 1.00-1.00 Phrase Duration 29 0.18 0.10-0.2^ (sees.) Lower Limit of MaJ °('kHSr gy ^ 2 ' 91 1-50-4.00 Upper Limit of MaJC 7v?? rgy 29 6 lk 3.00-8.00 Chirps were also given by a feeding animal, the feeding in some instances being an apparent nervous displacement activity but on other occasions a true feeding activity. The frequency of repetition of the chirps was variable as was the intensity of the call. Some chirps were given loudly and in rapid succession; for example in response to a loud, sudden noise or to a dog passing nearby. On other less tense occasions chirps were given infrequently and softly, almost whisper-like. Other Sykes responded to the chirping animal hy looking in its direction and in the direction of its threat and by chirping also. Occasionally the other animals would move deeper into the brush but this was not necessarily a hurried

PAGE 27

15 retreat. In some instances the only Sykes remaining visible was xhe chirping individual who would soon move off and join the others. Multiple Chi res Included in the Sykes* repertoire in addition to the single-unit chirps are twoand three-unit chirps (Figs. 3-5 and Table 2). Such multiple chirps were also reported by Struhsaker (1970) for Cercooithecus ceohus . C. erythrotis . and C. nictitans . As is the case for single chirps, multiple chirps are apparently restricted to adult females and immature monkeys . In the two-unit chirp, the distribution of major energy in the second unit is over a greater range (2.64 5.96 kHz) than in the first (3. 80 5 -96 kHz). This distinction was also made by Struhsaker (1970) for the three Cerco£ithecus species listed above. Struhsaker found that this shorter frequency range of the first unit was due both to a higher lowest frequency and a lower upper frequency. However, in the case of C. mitis only the lowest frequencies were found to differ considerably. That is, in 88.9 percent of the double chirps the lowest frequency of Unit I was higher than the lowest frequency of Unit II, whereas the upper frequencies of the two units were the same in 66.7 percent of the calls (N=9) . Generalizations concerning energy distribution for the three units of the triple chirp can not be made since only one such call was recorded. However, in this one case energy distribution increased with each subsequent unit.

PAGE 28

16 TA3LE 2 MULTIPLE CHIRP

PAGE 29

17 In addition to frequency differences, the durations of the two units of the double chirp also differ, with the first unit being shorter than the second in all cases. The durations of the units of the three-unit chirp also increase from first to third, but again any generalization must await a larger sample. It is interesting to note that in both energy distribution and duration the final units of both the twoand three-unit chirps fall within the range of values for the single-unit chirp. Nyah and Short Nyah The nyah (Fig. 6 and Table 3) is a loud call consisting of tonal and nontonal components. Although the call itself consists of a single-unit phrase, it always occurred in bouts in which it was repeated at intervals of approximately 0.50 seconds. On the sonogram the nyah appears as a harmonic series of bands beginning abruptly and arching into a mixed band of tonality and noise. In a variant of the nyah, referred to here as the short nyah (Figs. 7 &8 and Table k) , this harmonic band pattern at the beginning is absent, the call, therefore, sounding like a sharp "ah". Both the nyah and the short nyah can be given in a bout, but it seems the latter was usually given under tense or excitable conditions, such as a pack of stray dogs passing nearby. Mean duration for the nyah was 0,31 seconds (N=l2) and 0.19 seconds for the short nyah (N=l4). The average

PAGE 30

18 energy distribution for both forms was quite similar, with 0.82 $.l].6 kHz for the nyah (N=15) and O.57 5.31 kHz for the short form (N=l4) . Total energy distribution ranged from 0.10 16.0+ kHz for both calls. Since it was never known exactly how many Sykes groups were in the area and since individual groups were not distinguishable, it is possible that all the nyahs of the sample were made by the same individual. This possibility is based on the assumption that there is but one adult male per group, a common occurrence among Cercopithecus monkeys according to Struhsaker (1969). The short nyahs of the sample, however, were given by one individual only since they were all recorded on a single occasion with only one troop present. TABLE 3 NYAH Number (N) Mean (x) Range (R) Units/Phrase 15 / 15 lmQQ 1.00-I.OO Phrase Duration l2 . 3 1 0.20-0. 38 Lower Limit of Major Energy 15 0>82 (kHz) 0.25-1.80 Upper Limit of Major Energy 15 5 . 46 ^ 00 -?.0< \ j£n Z )

PAGE 31

19 SHORT NYAH fumber (N) Mean (x) Range (R) Units/Phrase 1/1/1/1 1 nn 1V1^ 1.00 1.00-1.00 Phrase Duration 14 n 10 (sees.) 14 0<1 9 0.05-0.28 Lower Limit of Major Energy iu _ _, (kHz) ^ °'57 0.20-2.30 Upper Limit of Major Energy 1it _, (kHz) ^ 5.31 ^.00-6.50 The animal giving the call was very difficult to observe; in fact, this was done on only one occasion and q uite by accident. A single Sykes was observed sitting alone in the topmost branches of a tall tree giving the call at about 18:00. No other Sykes were visible in the immediate area, The call .as m ade by exhaling forcibly with the mouth open. This individual was too far away to permit accurate sexual identification but its large size would indicate it to be most likely a male. This assumption is further supported by Haddow's (1952) observation of the redtail monkey, C. ascanius, in -hich only the adult male gives a call Haddow described as "Kyuh". m addition, the nyah and short nyah of the Sykes are probably equivalent to the "loud calls" reported by Struhsaker (l 9? o) and Bourliere, Hunkeler, and Bertrand (l 97 o) for other C^o^ithecus species which were also only give n by adult males.

PAGE 32

20 Struhsaker »s (1970) "loud calls" are of two types, the pow and the hack. Since he does not make explicit the basis for distinguishing them spectrographically, and sinehe does not do so in his table of measurements, it is impossible to compare or correlate one or the other with the nyah or short nyah. The nyah call was frequently given prior to group progression. It may have signaled group withdrawal, for soon after the call was given no Sykes would be visible in the area, having moved in the direction from which the call came. In addition, such calls seem to be evoked by a potentially disruptive or threatening stimulus as was the case with the pack of dogs. Bourliere, Hunkeler, and Bertrand report that loud calls are given by the alpha male in the evening at the sleeping trees and also "in response to any situation that was potentially dangerous for the group members, or potentially disruptive to the cohesion of the group" (1970:313). Haddow (1952) and Struhsaker (1969) concluded that the loud calls may function as a central focal point for maintaining intragroup cohesion on occasions when the group might be dispersed and may also function in intergroup spacing. Boom On several occasions a low frequency call was heard which sounded very much like "whoo" but which closely resembles in frequency and duration the booms of the mona superspecies as described by Struhsaker (1970). The boom (Fig. 9 and Table 5) is completely tonal and appears as a very

PAGE 33

21 black band situated at tha bottom of the sonogram. The call is given as a single-unit phrase, is seldom repeated, and appears to be strikingly uniform. The energy distribution is restricted to such a low frequency that the lowest distribution of energy on the sonogram is difficult to read, being so close to the bottom margin of the sonogram itself. This lower limit of the energy range has, therefore, been approximated at 0.01 kHz. The upper limit of energy distribution is 0.^0 kHz (N=2) . TABLE 5 BOOM Number (N) Mean (x) Range (R) Units/Phrase Phrase Duration (sees . ) Lower Limit of Energy (kHz) Upper Limit of Energy (kHz) 2/2 2 1.00 0.20 0.01 0.40 1.00-1.00 0.20-0.21 0.01-0.01 0.^0-0.^0 The individual giving the call was never seen, nor did the animal sound as if it were very close to the observer. However, among those species (Cercopithecus mpna, c. campbelli, and c. ^ogpnias) which Struhsaker (l 9 ?0) studied that had boom calls, only adult males gave the call. Apparently, the boom was involved in some way with coordinating group movement as it was usually given while the

PAGE 34

22 group was moving, or else the group would move soon thereafter. On several occasions the boom was heard following agonistic encounters within the group and so may serve to facilitate group cohesion by overriding the disruptive and dispersive effects of antagonism among its members. Squeal The squeal (Fig. 10 and Table 6) was a high-pitched, uniform call. It is entirely tonal and occurs as a singleunit phrase. On the sonogram the squeal appears as two slightly undulating bands, one at approximately 3. 00 4.00 kHz and the other at 6.00 7-50 kHz. The end of the call lies on a slightly lower frequency level than the beginning. TABLE 6 SQUEAL Number (N) Mean (x) Range (R) Units/Phrase 7/7 1#00 1.00-I.OO Phrase Duration 7 . 63 0.39-1.20 (sees.) Lower Limit of Energy (kHz) 7 3 . 2o 2.70-3.50 Upper Limit of Energy (kHz) 7 ? . 8? 7 .00-9.00 The mean duration of the squeal is O.63 seconds (N=7). The mean lower frequency is 3.20 kHz and the mean upper frequency is 7.87 kHz (N=7).

PAGE 35

23 Sn riek The shriek (Fig. n and Table 7) was a high-pitched, shrill-sounding call of variable duration. Certain sections of the call vary in degree of tonality and duration. Since it appears as a continuous tracing on the sonogram it must be classed as a compound, single-unit phrase. The first part of the shriek resembles the squeal and is completely tonal with a two-band appearance on the sonogram, However, it is somewhat higher in pitch than is the squeal, with a mean frequency of 3.^6 8.7^ kHz (N= 5 ) . This first part continues into, and is superimposed by, a mixed tonalnontonal portion which appears as a noisier, broader, double band with a mean frequency of 3. 20 8.64 kHz ( N =5). As the call progresses, this portion may become noisier and its appearance will become more distorted on the sonogram. The mean duration of the tonal component is 0.36 seconds (N=5), whilp +ha+ ^-f +u~ d>, wniie that 01 the mixed portion is 0.65 seconds (N=5) for a mean total phrase length of 1.01 seconds. Chut On one occasion a call was given which can perhaps besx be described as sounding like "chut" (Pigs. 12 ^ 3 an a Table 8). There were three such chuts or units grouped together to form a phrase and two phrases were given. There is some tonal quality to the call but this is overlapped for the most cart by a nontonal component (thus forming a mixed unit) giving the call a noisy appearance on the sonogram.

PAGE 36

Units/Phrase Tonal Portion Duration (sees.) Lov/er Limit of Energy (kHz) Upper Limit of Energy (kHz) Mixed Portion Duration (sees.) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) TABLE 7 SHRIEK Number (N) Mean (x) Range (R) 5/5 5 5 5 5 5 1.00 1.00-1.00 0.36 0.13-0.55 3.46 3.10-4.00 8.47 7.50-9.50 0.65 0.27-0.99 3.20 3.00-3.50 8.64 8.00-9.00

PAGE 37

25 The mean unit duration was 0.06 seconds (N=6) and the average distribution of major energy was 2.95 ^.53 mz ( N=6 ) However, of the three units the third was the longest (0.12 seconds) and had a greater frequency range. The upper limit of the total energy range of this third unit was approximately 13.0 kHz. On the sonogram the first unit appears as a single band while the second and third units consist of two bands. However, the second of these bands is difficult to discern and measure since it is considerably obscured by the noise component. Although there are some spectrograph^ similarities with the chutter, the calls differ enough acoustically to warrant their separate classification at this time.

PAGE 38

26 The chut was given by either a subadult or snail adult, possibly female. Two such animals were grooming high up in a tree. Upon seeing the observer, the one being groomed gave a squeal which was followed by the chut phrases. It then moved away from the groomer who chirped when it noticed the observer. Taking into consideration the context in which it was given and the call with which it was associated, the chut could be tentatively identified as a fear-alarm call. Chutter The chutter (Fig. 14 and Table 9) is a completely nontonal phrase composed of several units of short duration. Each unit on the sonogram resembles a column of noise with some broader than others. There is some variation in chutter phrase duration, depending upon the number of units included in the call. One of the three calls sampled contained twelve units, the others each had four. The significance of phrase length differences is not yet understood, but perhaps there • is some relationship between the intensity of stimulus arousal and the number of units per phrase. It is also possible that the variation reflects individual differences. Perhaps more than one type of chutter is represented in the sample. Struhsaker (1967) was able to discern several different kinds of chutters for the vervet monkey. Without a larger sample size and more information as to the function of the chutter calls and the circumstances which evoke them, any statement as tc a possible typology of chutter phrases wou Id b e pr ema tur e .

PAGE 39

2? The unit duration and the interval between units was similar among the three phrases. However, it was noticed that the units of a phrase tend to increase in duration and intensity such that those units late in the phrase are the longest and loudest. The major energy distribution ranged from 2.00 11.00 kHz with the mean distribution as 2.67 7,83 kHz (N=18). TABLE 9 CHUTTER Number (N) Mean (5c) Range (R) 4.00-12.00 Units/Phrase 20/3 6.33 ^(secs^? 10 " 18 °' 06 0.01-0.11 Interval Between UnltS (secS ) 15 0.0^ 0.01-0.03 Phr (s% e o30 ati0n 3 °58 0-39-0.95 Lower Limit of Hajorjnergy 18 2 . 6? 2>00 . 4 . 00 Upper Limit of Major Energy iq n Qo ,, „ (kHz) ?e83 4.00-11,00 The individual giving the chutter call was never seen and the situations in which they were given were seldom apparent. They were sometimes, but not always, heard during intragroup fights so apparently the chutxers play some parx in agonistic encounters. Struhsaker (1969) likewise reports

PAGE 40

28 having heard chutters during the fight episodes of other Cercooithecus species,, Low, Short Ghutter On one occasion a lov/ chutter of short duration (Fig. 15 and Table 10) was given by an adult Sykes. Two phrases were recorded, each consisting of two units. This chutter is lower in frequency than the chutter described above, with a mean major energy distribution of O.75 2.88 kHz (N=*0 . However, in terms of total energy distribution, there is some indication on the sonograms of energy in the vicinity of 7-50 kHz. TABLE 10 LOW, SHORT CHUTTER Units/Phrase Unit Duration (sees . ) Interval Between Units (sees.) Phrase Duration (sees. ) Lower Limit of Major Energy (kHz) Upper Limit of 3 or Energy (v< ; j) Number (N) Mean (x) Range (R) k/2 k 2 2 k 0.75 2.88 2.00 2.00-2.00 o.o4 0.03-0.07 0.05 0.05-0.05 o.i4 0.13-0.15 0.50-1.00 2.75-3.00

PAGE 41

29 Energy distribution and duration of each of the units are quite similar, with the units appearing on the sonograms as thin columns of noise. Mean unit duration is 0.0^+ seconds (N=*0 and phrase duration averages 0.14 seconds (N=2) . This chutter, like the previous one, is entirely non-tonal. Chutter-Squeal The chutter-squeal (Figs. 16-18 and Table 11) can best be described as a multi-phrase bout of long duration. The following discussion is based on measurements of two such calls which were both recorded during the same session. Since the vocalizing individual v/as not seen, it is possible that both these calls were given by the same animal. The first phrase consists of a series of short units resembling in some ways both the chutter and the chirp. These units are basically nontonal like the chutter but yet there is a hint of a chevron-shaped tonal component (as in the chirp) superimposed over the nontonal element. The first bout had a phrase consisting of nine chutter units and the second bout had four chutter units. With a mean unit duration of 0.0? seconds (N=l 3 ), the units are as short as those of the chutter but do not have as wide a frequency range. Shorxly following this first phrase are single-unit phrases which appear on the sonogram as nontonaltonal undulating bands. There were five of these phrases in the first bout and two in the second. These phrases are of long duration; two sampled had a mean duration of 3.93 seconds. The

PAGE 42

30 mean of the lower frequency of major energy distribution is 3.75 kHz while the mean upper frequency is 7. 00 kHz. The final phrase of the bout is also a single, mixed, tonal-nontonal unit but the nontonality or noise is not as pronounced as in the preceding phrases. The average duration of this unit is 0.62 seconds (N=2) and the mean frequency range is 3.25 5.65 kHz (N=2). TABLE 11 CHUTTERSQUEAL Number (N) Mean (x) Range (R) Chutter Portion Units/Phrase Unit Duration ( sees . ) Interval Between Units (sees.) Phrase Duration (sees . ) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) Mixed Portion I Interval Between Chutter & Mixed I (sees . ) Units/Phrase Phrase Duration (sees . ) 13/2 13 11 2 13 13 6.50 4.00-9.00 0.07 0.02-0.10 0.C4 0.02-0.03 0.64 O.39-O.88 3.15 5.94 2.50-4.50 4.00-7.00 0.02 0.02-0.03 2/2 2 1.00 1.00-1.00 3-93 308-4. 78

PAGE 43

TABLE 11 (Continued) 31 Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) Mixed Portion II Interval Between Mixed I & Mixed II (sees . ) Units/Phrase Phrase Duration (sees . ) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) Total Duration Number (N; Mean ( x ) Range (R) 2/2 2 3-75 7.00 0.14 3.25 5.65 5.35 3.50-4.00 6.50-7.50 0.10-0.18 1.00 1.00-1.00 0.62 O.5O-O.73 3.00-3.50 5.50-5.80 4.10-6.60 Neither the final nor preceding phrases show any great similarity, audibly or spectrographically, to the shriek or squeal phrases reported earlier. The sum of the duration of each of the phrases as well as the intervals between them results in a lengthy call averaging approximately 5-35 seconds (N=2) .

PAGE 44

32 Grov;l The growl (Figs. 1 9 &20 and Table 12) is predominately a tonal call although there may be some nontonality present, particularly at the end of the vocalization. The growl consists of a multi-unit phrase which may occur singly or may be repeated. Units/Phrase Unit Duration (sees . ) Interval Between Units (sees.) Phrase Duration (sees. ) Interval Between Phrases (sees.) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) TABLE 12 GROWL Number (N) Mean (x) Range (R) 135/17 135 118 17 17 17 7-9^ 3.00-20.00 0.02 C. 01-0. 03 0.02 0.01-0. 03 0.20 0.05-0.59 0.^9 0.08-0. 67 0.52 0.25-0.75 1.15 1.00-2.00 The number of units per phrase was variable but averaged 7.9^ units per call (N=17) . The unit duration and the interval between units was similar in all the phrases sampled, so phrase duration was largely dependent on the number of

PAGE 45

33 units present. 'Since the number of units varied from three to twenty, phrase duration varied considerably also. However, the mean phrase duration was 0.20 seconds (N=17) . Despite their temporal variation, the calls are grouped into a single category at this time on the basis of their acoustic and spectrographic similarities. Energy was distributed on approximately half the sonograms in a single band formation (N=8) and in a double (N=5) or triple (N=4) band arrangement on the remainder. In the last two cases, however, most of the energy would be concentrated in the lowermost band. The average distribution of the heaviest energy concentration, therefore, was O.52 1.15 kHz (N=l?) but the total energy distribution might extend to 3>50 kHz in some cases. Identification of the vocalizing individual was difficult but the call was apparently given by both juveniles and adults. Examples of instances when the growl was heard include: during chirping sequences by a Sykes threatening the observer, as Sykes fed while looking at the observer, during fight episodes, and while a troop was moving in the brush . Err A call somewhat similar acoustically to the growl is the "err" (Fig. 21 and Table 13). On the sonogram, however, each call is broken into four cr five short, discrete units which extend to higher frequency levels than the growl. The "err" is completely tonal.

PAGE 46

3^ The mean phrase duration of the "err" is 0.17 seconds (N=3), with a mean energy range of 0.81 5.25 kHz for the units (N=l4). TABLE 13 ERR Number (N) Mean (x) Range (R) 4.00-5.00 Units/Phrase 14/3 ^,67 "^(secs?) 10 " lk °' 02 0.01-0.05 Interval Betv/een Units (secs.) n . 02 0.OI-O.05 Phrase Duration 3 . 1? 0.12-0.28 ^ S GCS • j Lower Limit of Energy (kHz) lk 0#8l 0. 50-1.20 Ucmer Limit of Energy (kHz) lk 5 . 25 L. o7 . 5 o The individuals vocalising were not identified and the function of the "err" call is not understood. However, the call was heard immediately following a shriek on two of three occasions. Uh A short call best described as sounding like "uh" (Fig. 22 and Table !**•) was given before a shriek three times out of a sample of five. The call was given in pairs or in threes with a mean interval between the units of 0.12 seconds

PAGE 47

35 (N=7). The mean unit duration was very short at 0.04 second (N=12). s TABLE Ik UH Number (N) Mean (x) Range (R) Units/Phrase 12/5 2>if0 2 . 00 _ 3 . 00 Unit Duration 1? n nh (sees.) 12 °' Qli 0.01-0.09 Interval Between Un " S (3eCS -» 7 0.12 0.04-0.21, ^Tle'csT' 10 " 5 °28 0.08-0.40 Lower Limit of "-Jorawrgy 12 1>78 ^^.^ Upper Limit of Major Energy o (kHz) 5-56 ^.10-8.00 The »uh" is lower in frequency than the shriek which sometimes followed (three out of five times). The mean lower frequency of major energy is 1.78 kHz and the mean upper frequency is 5.56 kHz (N=12). The call is entirely nontonal and appears as a narrow column of noise on the sonogram. Discrete and Graded Signals In a study such as this in which the vocal repertoire of a primate species is being classified and described, it is customary to examine the nature of the signals as to

PAGE 48

36 whether they are discrexe or graded. Discrete signals are separate and distinguishable whereas graded signals are more variable and can be arranged to form a continuum with other such signals. Marler has pointed out that continuous or graded sounds are more common in rhesus monkeys, baboons, chimpanzees, and gorillas, resulting in sound systems which are "much more complex in the development of these subtle continuous ana multidimensional variations of the sound signals" (1965:566). The graded nature of the vocalizations of these terrestrial or semi-terrestrial species which are relatively isolated from other primate species of similar size and structure contrasts significantly, says Marler, with the relatively discrete nature of the sound systems of forest-living primates in frequent contact with other primate species, such as is the case with Cercordthecus and Colobus. In such circumstances the possibility of interspecific confusion is. greatly increased and, correspondingly, the limited evidence available on the sound signals of such forms suggests that thev are more mghly structured and that this structuring is used_ax least m part to maintain species specificity in a major proportion of the vocal repertoire. (Marler, 1965:565) ~ P The vocal repertoire of the Sykes monkeys seems to comply with Marler «s thesis. In the study area there were three other diurnal species of nonhuman primates observed, xhe baboon (Papio pyocephalus ) . the colobus ( Colobus colykomos), and the vervet ( Cercocithecus aethiops ). The Sykes were at times observed in close proximity to all three species. Of these species, the vervet is the most closely

PAGE 49

37 relaxed to the Sykes, morphologically and behaviorally. According to Struhsaker (1967). the majority of the vervet vocalisations which he studied were of the discrete type. Similarly, most of the Sykes' calls, while showing some degree of variability within categories, are also of the discrete type. Signals which may prove to be of the continuous type would include the shriek and the squeal, and the calls of the chutter group. For example, the squeal was described as a tonal call consisting of two parallel bands with a mean energy distribution of 3.20 7.87 kHz (N= 7 ). The shriek, on the other hand, was composed of two components, a two-band tonal section followed by a mixed portion of tonal and nontonal sound. The initial tonal component strongly resembles the squeal, but the mean energy distribution is slightly higher at 3.46 8.74 kHz (N=5). However, the ranges for the energy values for the two calls do overlap (Table 6 and Table 7). It would seem likely, therefore, that there may be a graded continuum between squeals and shrieks. There may also be a continuum within the chutter group, although the sample is small and hence not extremely reliable. Perhaps the chut, the chutter, and the short chutter are but points along a graded continuum of such calls. Also, there may be some functional significance in the number of units per chutter phrase, creating a gradient in this regard as well.

PAGE 50

33 Future fieldwork and further study may point out more continuous signals for the Cerco oith ecus mitis group as well as modify those tentatively identified here. This investigation has demonstrated the possible existence of a graded series of signals or messages. That this is not sufficient evidence to indicate a graded communication system has been aptly pointed out by Altmann (1967). He stresses that signals used in graded or analogical communication must be differentiated from those signals which may appear graded but merely reflect the normal range of variation which a sample of calls may be expected to show. Such variation would be due to the morphological and physiological variations of the structures involved in sound production. It is necessary, says Altmann, to demonstrate functional con tinuity : "What is needed is (1) a continuous array of messages, (2) a continuous array of responses to these messages, and (3) a one-to-one mapping between messages and responses" (1967:3^1). The failure to make the distinction between graded signals which are associated with graded responses and those which are not has resulted in some confusion and misunderstanding concerning the use of the terms 'discrete' and 'graded'. The terms were introduced into the study of primate communication largely through the work of Hockett (i960) who proposed thirteen design features, of which discreteness was one, which he felt were characteristic of human languages. Apparently in their haste to look for Hockett's design features among the communication systems of nonhuman

PAGE 51

39 primates, many researchers failed to recognize that discreteness and gradedness among the response patterns as well as discreteness and gradedness among the signals themselves must be demonstrated in order for a communication system to be classified as one or the other. Consequently, although these terms occur frequently in the literature, Altmann (1967) points out that a graded communication system has never been demonstrated. Even if such a system were described, Altmann further emphasizes that discontinuities between the major categories of communication patterns themselves will make for a certain amount of discreteness. The classification of the system as one or the other, therefore, largely becomes a question of semantics. It would seem that an understanding of the nature of communication systems among nonhuman primates lies not in its comparison, feature for feature, with human language but rather with the understanding of the adaptive significance of such features as may be present. That the latter has not yet been accomplished is due in large part to the fact that primate communication studies are still in their infancy. Hopefully, however, through future field studies and the amassing of greater amounts of informaxion, certain adaptive trends and correlations will become apparent which will have significance for understanding not only the nature of nonhuman primate communication but the nature of human language as well .

PAGE 52

A Structural-Functional Analysis There are many different types of vocalizations among the primates and these vocalizations serve various functions. However, the exact nature of the relationship between function and structure of primate calls has not yet been determined. Although work along these lines has been carried out for some insect species (Alexander, i960, 1968), anuran species (Bogert, i960; Little John, 1959), and avian species (Marler, 1955, 1959; Marler and Hamilton, 1966), most discussions of primate vocalizations in structural-functional terms have remained generalized. There are several reasons which may account for these generalized studies. Firstly, the study of primate communication is still a relatively new endeavor and the emphasis, until recently, has been on collection and description rather than on detailed investigative analysis. Secondly, primate communication has been shown to be an exceedingly complex process involving several channels of communication; namely, visual, auditory, tactile, and olfactory. The relative contribution which each of these channels makes to a species' communication system, the degree to which they are interrelated, and the extent to which they limit and modify communication by other channels are difficult to assess. Therefore, a discussion of the structure of a particular vocalization must also account, for the possibility of simultaneous transmission of information by some other means, such as visual signals. Thirdly, the determination of the function of a call is often

PAGE 53

to difficult, requiring an evaluation of the behavioral and environmental contexts in which the call was given as well as a description of the nature of the responses which it evoked. Consequently, among those primate species which have been studied, there remain numerous calls whose functions are unknown. Lastly, it has not yet been determined which aspects of a primate vocalization, if any, are most significant for the transmission of information. That some components of a call, such as frequency, intensity, duration, and timing, are differentially informative has been shown to be the case among arthropods (Alexander, 1968), anurans (Bogerx, i960), and birds (Falls, 1963; Marler, 1955, 1959). Experimental designs, such as play-back techniques both in the laboratory and in the field, will probably be of considerable importance in evaluating the communicative significance of the various components. The present investigation will attempt to apply a structural-functional approach to the study of Sykes vocalizations by focusing on several principles of sound localization as outlined by Marler (1955, 1959) in his studies on bird vocalizations. Structural comparisons will be made between those Sykes calls used to communicate over long distances and those which operate at close range. The discussion will be limited to those calls whose functions are known. As Marler (1955, 1959) points out, sound localization in vertebrates relies on binaural comparisons of differences in phase, intensity, and time of arrival. Intensity differences become apparent when the head acts as an obstruction,

PAGE 54

42 thus hindering the sound iron reaching one ear. The direction of the sound can be inferred due to the resulting difference in intensity at the two ears. Since this "shadowing effect" is significant only when the obstruction is of the same order of size as the sound wavelength or larger, localization by intensity differences is most useful with high frequency sounds (Marler, 1959). On the other hand, localization by phase differences is confined to low-pitched sounds (Marler, 1959) . Phase differences are the result of pressure changes occurring at different moments in the two ears. In high frequency sounds where the wavelength is less than the distance between the ears several wave oscillations may occur, resulting in ambiguous phase comparisons. In addition, the nerve leading from the ear is unresponsive for about a thousandth of a second after each oscillation, so a sound with more than a thousand vibrations per second would not be completely represented by nerve impulses (Marler, 1959). Accurate comparisons of differences in time of arrival of sound at the two ears is dependent on the way a sound begins and ends rather than on frequency ranges. It has been found that sharp discontinuities are most easily perceived, especially if frequently repeated (Marler, 1959) . Marler concludes that the ideal sound permitting source localization "will include a high pitch for location by intensity difference, a low pitch for location by phase difference, and it will be sharply broken and repetitive for location by time difference" (1959:175). In man it has oeen

PAGE 55

^3 found that pure tones in the region of 2.0 5.0 kHz are difficult to locate by comparisons of phase and intensity differences, being too high for the former and too low for the latter to be effective (Marler, 1955). For this frequency range only time differences can be used to successfully locate sounds. The lower limit of this frequency range is probably similar in the Sykes monkey due to the refractory period of the auditory nerve. However, the upper limit is probably higher (exactly how much so is yet to be determined) since the diameter of the Sykeshead is smaller than man's and the "shadowing effect" will thus not take place until a smaller wavelength or higher frequency is reached. Five Sykes calls, about whose functions something is known, were chosen for investigation. These vocalisations will be examined in terms of whether they can or can not be easily localized by phase, intensity, or time differences. The five calls are the nyah, boom, chirp, squeal, and shriek. It has been mentioned elsewhere that the nyah call given by adult males functions to promote group cohesion and coordinate group movements. Struhsaker (1970) emphasizes that by doing so, such calls may also serve to maintain reproductive isolation in areas of mixed species associations. When the call is given the group members, who have been dispersed at the time, begin to move in the direction from which the call came. The call is loud and carries ever considerable distances. Visual contact does not seem to be of importance here so one would expect the nyah, in keeping with its function, to be a call which is easily localized. Analysis of its

PAGE 56

44 structure shov;s xhis to be the case. The average range of major energy distribution for the nyah was shown to be between 0,82 and 5.^6 kHz (Table 3) which would apparently include a low enough pitch for localization by phase differences. The upper limit of this range would appear at first glance to be too low to facilitate localization by intensity differences since it most likely falls within the "hard-to-locate" range discussed earlier. However, since in total energy distribution the upper frequencies of the nyah nay extend to lo.O kHz or more, localization by means of intensity differences is indeed possible. The call is of a short duration (O.31 seconds), has an abrupt beginning, and is almost always repeated thus enabling localization by temporal differences. In short, the nyah call fulfills all three of Marler's requirements for an ideal sound permitting source localization. Indeed, this is what one would expect of a call which apparently plays such a vital role in group maintenance and cohesion. Another call whose function is apparently concerned with group cohesion without requiring visual contact with the caller is the boom vocalization. This call, however, differs from the nyah in several respects. The boon is a very lowpitched call with an average energy distribution of 0.C1 0.^-0 kHz (Table 5). This low frequency would permit localization by phase differences but not by intensity differences. Although the call is of a short duration (0.20 seconds), it is not reoeated, nor does it begin or end abruptly, thus making localization by temporal differences difficult. There are several possibilities which nay account for the structure

PAGE 57

m-5 of this call. Although the boom would not appear as ideal for localizing a sound source as the nyah, perhaps it is given under circumstances where an extreme degree of localization would be disadvantageous to the caller or the group, such as in reaction to the observer or a potential predator. Another possible explanation is that perhaps phase difference alone is an efficient means of sound localization as long as the call is restricted to very low frequencies. In addition, a low-pitched group cohesion call might be used because low frequency sounds carry over greater distances than sounds of higher frequencies (Moynihan, 1967) . Also, there is less chance that a low-pitched call win be refracted by trees or other objects as is the case with high-pitched sounds (Marler, 1955), which is obviously an important consideration for an arboreal species. The key to explaining the nature of the boom clearly lies with a better understanding of its function as well as with the formulation of an experimental approach which would test the localization properties of the call. The chirp acts as a threat/alarm call and its physical properties would seem to facilitate localization. Although the average major energy distribution of 2.91 _ 6tlk kHz (Table 1) falls within the "hard-to-locate" zone, the total energy range of 0.10 16. 0+ kHz most lively permits localization by means of phase and intensity differences. Th call has an abrupt beginning, is of short duration (0.18 seconds'), and is usually repeated. With increasing intensity or levels of excitement, the call is repeated more frequently and may have an even more abrupt beginning. Localization by :ne

PAGE 58

comparison of temporal differences is clearly possible and may be of considerable importance. One might expect an alarm call to be given in such a way as to prohibit localisation of its source for safety's sake, as is the case for some avian calls (Marler, 1955). That this is not the case for the chirp is supported by the observation that the other group members look toward the caller upon hearing the chirp and may then chirp also. In addition, the caller is frequently exposed in the open when giving the call and may not move back into the brush for some time; in fact, other Sykes may actually move up to be near the caller as they join in chirping. Since the chirp is easily located it aids in indicating the source of danger to other group members and in enlisting their support in threat. The last two calls to be considered are the squeal and shriek. Both seem to be associated with agonistic encounters although their exact function is undetermined. Their energy distributions of 3. 30 7.87 kHz for the squeal (Table 6) and 3.20 8.74 kHz for the shriek (Table 7) are apparently within the "hard-to-locate" range. The calls are of a longer duration than the others which have been discussed, O.63 seconds (squeal) and 1.01 seconds (shriek), and both start and end gradually. Precise localization of the sound source would appear to be difficult. However, the most important difference between the shriek and squeal and the other calls which have been described is that the functions of the latter require transmission over distances, v/hez 3 the shriek and squeal apparently function in close-range agonistic encounters.

PAGE 59

^7 Consequently, other channels of communication, such as facial expressions or touch, nay contribute to the message content and indicate the location of the vocalizer. It is difficult to explain the physical properties of the remaining calls in the Sykes • repertoire until more is know of their function. Since most of these calls seem to function on a close-range basis, one would expect other channels of communication to be of some importance in signal transmission. It should be emphasized that a discussion of the structural and functional components of a call in terms of its localization properties is an oversimplification of the nature of primate vocalizations. Other factors, such as requirements for species specificity and individual identification or the nature of the physical and social environments, may affect the physical structure of a primate call. For example, Marler (1965) points out that in those calls used over distances to maintain spacing between groups, the requirements of species specificity necessitates use of a larger range of sound properties than in close-range communication. These spacing calls, says Marler, are therefore more highly structured and purer in tone. It is possible, therefore, that requirements for species specificity might be partly responsible for the highly structured nyah call of the Sykes monkey. Oversimplified generalizations about primate vocalizations can be dangerous and misleading. A statement that arboreal species living under conditions of low visibility

PAGE 60

48 have developed elaborate high-pitched calls (Gartlan and Brain, 1968) gives the misleading impression that a high pitch alone makes such calls functionally significant and fails to recognize and account for the existence of lowpitched calls among such species. What is needed instead is a structural-functional analysis to furnish clues as to whv. a particular structure or form may predominate. It is hoped that the analysis presented in this section has demonstrated the feasibility of such an approach. Taxonomic Implications The genus Cercopithecus is said to have more species and forms than any other primate genus in Africa (Struhsaker, 1970). Its species have usually been distinguished taxonomically on the basis of pelage coloration. Unfortunately, wide variability in pelage coloration within the genus and within its species has resulted in some disagreement over xhe interpretation of phylogenetic affinities. In cases where morphological characters are inadequate or equivocal, the taxonomist has been urged to make use of aata from studies in behavior, ecology, physiology, and biochemistry (Lanyon, 1969) . Through the use of improved recording equipment and the development of the sound spectrograph, one aspect of behavior, i.e., vocal communication, has been shown to be of some significance in taxonomic studies of orthopteran insects (Alexander, 1962) , anuran amphibians (Bogert, 19oC; Blair, 1962, 1964), and birds (Marler, 195?; Lohrl, 1963). Studies

PAGE 61

4 9 such as that of Struhsaker (1970) have indicated that vocalizations, when used with the same precautions and care applicable to any potential taxonoraie character, can be of considerable importance in primate taxonomy as well, Struhsaker examined the taxonomic position of several Cercopixhecus species, basing his decisions largely on similarities or differences in certain vocalizations. One of the species studied by Struhsaker (1970) was Cercopithecus nictitans. Several authors have pointed out the strong physical resemblance between C. nictitans and C. mitis (Booth, 1956; Haddow, 1952; Kingdon, 1971; and Tappen, I960). Hill (1966) placed C. nictitans in the mitis superspecies and the 1969 Wenner Gren Conference raised the question that perhaps C. nictitans and C. mitis might be best considered as conspecific (Thorington and Groves, 1970) . A comparison of the vocalizations of these two species might shed some light on their phylogenetic relationship. Of the two species, C. nictitans is the more restricted in distribution, being largely confined to the West African area. Two subspecies are generally recognized. C. n. martini is the most westerly located (Ghana, Nigeria, Cameroun) . C. n. nictitans is also on the western coast (Gabon, Cameroun) hut extends eastwards into the Central African Republic area. C. mitis, characterized by numerous subspecies, is much more extensive in its range, being found in central, east, ana south Africa. C. mitis is said to be the only forest guenon having such a wide distribution in southern, and eastern Africa (Kingdon, 1971). it apparently inhabited these areas

PAGE 62

50 before climatic changes isolated many of the forests, thus closing them to those Cerco pith ecus species which have developed more recently (Kingdon, 1971). In fact, C. mitis is believed to be near the stock from which the main guenon radiation has derived (Kingdon, 1971). For a more detailed discussion of the distribution of the various C. mitis subspecies, the reader is referred to Rahm (1970). The subspecies which will serve as the basis for the present discussion of vocalizations are C. n. martini and C. n. nictitans studied by Struhsaker (1970) in Caraeroun and £• E« kibonotensis from the southeast Kenya coast whose vocalizations have been described in this paper. The reader should note that the C. mitis and C. nictitans populations here being considered are allopatric, a situation which can make for considerable difficulty in assessing phylogenetic relationships (Mayr, 1969). In this regard, it is the examination of sympatric species which is most conclusive, since any mechanisms assuring species identity through reproductive isolation can be more reliably recognized in the sympatric state. However, Struhsaker (1970) has shown that interpopulation variability in the basic structure of vocalizations within a species appears to be minimal. The species he considered included C. aethiops (populations from East and West Africa), G. mona, C. nicti tans (populations from both subspecies), and C. pogonias (populations from the subspecies pogonias and gray! ) . Within each of these species and across subspecies lines Struhsaker found that the vocalizations snowed a considerable amount of

PAGE 63

51 stability and uniformity. Therefore, it will be assumed that the vocalizations of C. mitis are relatively uniform throughout its range also. In this way, it will be assumed that the vocalisations of G. mitis kibonotensis of East Africa would be basically similar to those C. m. s tuhlmanni and C. m. maesi, two western subspecies of C. mitis which are reported to be sympatric with C. n. nic titans at the eastern limits of its range (Rahm, 1970; Kingdon, 1971). Ideally, a study to determine whether C. nictitan s and C. mitis are conspecific should be carried out in this area of sympatry. Until such studies are available, however, and until studies indicative of the extent of variation of the vocalizations within C. mitis are carried out, the investigation presented here will proceed on the assumption which has been made based on Struhsaker's analysis (1970). No attempt is being made, therefore, to present the findings of this study as a final taxonomic statement. Hill (1966) apparently feels there is enough distinction between the western and eastern subspecies of the C. mitis group to warrant their separation into two different species, C. mitis and C. albogularis respectively. However, other workers (Rahm, 1970; Kuhn, 1967; Kingdon, 1971) have classified these as a single polytypic species. That hybrids have been reported between c. m. stuhlmanni and C. m. albo^aaris . (Booth, 1968) not only tends to support the latter approach but also adds more weight to the assumption that isolating mechanisms, including any vocalisations which may function as isolating mechanisms, are probably uniform

PAGE 64

s9 throughout the range of C. mitis . It should be pointed out that since the eastern subspecies of C. mitis are relatively isolated from other forest guenons while those in the west may be sympatric with several, such as, C. mona , C. ascanius , C. 1'hoesti , and C. neglectus . there may be slective pressures operating in the west which may not be operating in the east. These differences may be reflected in the vocalizations of these species. Struhsaker (1970) believes that the loud calls given by adult male Cercopithecus monkeys function to maintain reproductive isolation between species by enhancing group coherency, especially in response to alarming stimuli. In the sympatric state, therefore, the loud calls of two species would be different; Struhsaker, working with several species, found this to be the case. A comparison of the loud calls which Struhsaker reports for C. nictitans (1970:40?, 427) with those reported here for C. m. kibonotensis (Figs. 6-8 and Tables 3&4) reveals some differences. The nyahs of C. mitis cover a higher frequency range than do the pows of C. nictitans and have a more elaborate development of harmonics. The tonality of the G. mitis calls also has an arching configuration, rising and then falling in frequency, as opposed, to the descending quality of the tonality in C. nictitans . The short nyahs have some spectrographs resemblance to the hacks of C. nictitans, but again a higher frequency is reached in C, mitis. No calls equivalent to the pow-hacks which Struhsaker found to be intsrmediate between pews and hacks were found for

PAGE 65

53 £• yitis. Struhsaker could not discern any significant difference in the loud calls of the two subspecies of C. nictitans, nor could he distinguish the hacks of C. er ythrotis and C. cephus (1970:408, 429) which he subsequently placed in a single species. The loud calls of C. mitis and 0. nictitans are apparently not as similar as those of these other conspecific populations. In addition, booms were recorded for C. mitis but not for C. nictitans and in this regard C. mitis resembles the mona superspecies . The C. mitis booms (Fig. 9 and Table 5) resemble the C. mona calls (Struhsaker, 1970:409, 431) ver y closely. However, since this type of call is restricted to such low frequencies, variability may be expected to be limited so that similarities may not be significant. Booms are typically given before a series of hacks in C. mona . Booms were also frequently heard in conjunction with nyah calls in C. mitis. However, these similarities in boom calls should not be taken to indicate an especially close relationship between C. mona and C. mitis . The hacks of C. mona (1970:410, 423) do not appear to be any more similar to those of C. mitis than the latter was to C. nictitans . In addition, only hack-type units were recorded for G. mona whereas C. mitis definitely shows two types, i.e., nyahs and short nyahs. No calls similar to the "000" contact calls reported by Struhsaker for the mona superspecies were found in C. mitis. There are also several differences between other calls of G. mona and C. mitis and these will be discussed below.

PAGE 66

5^ The chirp (alarm) calls of C, nitjs (Figs, 1-5 and Tables i&2) bear some resemblance to the chirps of C. nictitans and C. erythrotis ceohus (Struhsaker, 1970:411, klk, ^35» *+3o) . In terms of the spectrographic form of the chirps the three species are very similar, but those of C. mitis are of somewhat higher frequencies as well as of longer duration. Multi-unit chirps were present in all three species. Struhsaker (1970) found the chirps of C. erythrotis and C. cephus to be very similar. Some of these chirps were identical to those given by C. nictitans. Similarity in alarm calls among sympatric species faced with common predators may be explained as a case of parallelism. Struhsaker (1970) points out, however, that C. mona and C. pogonias also associate freely v/ith C. nictitans and C. erythrotis ceohus but maintain very distinct alarm calls. Therefore, Struhsaker concludes that C. nictitans and C. erythrotis cephus may be more closely related to each other than to C. mona. Struhsaker (1970) reported grunts and chutters for C. n ictitans and C. erythrotis cephus but found that no such calls were given by members of the mona superspecies . The grunts are thought to function in maintaining group cohesion while the chutters were given during agonistic encounters. C. mitis has vocalizations similar to these grunts and chutters. although the samples for C. mitis are too small to be conclusive. A low, short chutter given by C. mitis (Fig. 15 and Table 1.0) resembles very closely .the C. nictita ns chutter' id by Struhsaker , kkl) both in energy Lstri Ion and duration. However, the relationship of this

PAGE 67

55 chutter to others of higher frequencies which have been reported for C. mitis in this paper is not clear. Struhsaker does not mention any other chutter types for C. nic titans and C. erythrotis-c_ephus but his 1970 study was apparently not intended to be descriptive of the complete sound repertoires. Therefore, it remains to be seen whether the existence of these other chuttertypes in C. mitis is indicative of a significant species difference. The "err" reported here for C. mitis (Fig. 21 and Table 13) may be equivalent to the grunts; if so, some differences are apparent. While all the C. nictitans grunts (Struhsaker, 1970:418, kkl) were singleunit phrases and the C. erythrotis grunts (Struhsaker, 197O: ^18, kkl) were either singleor triple-unit phrases, the "err" of G. mitis was made up of four or five units. Although unit and phrase duration for all three species are basically similar, the C. mitis call extends to a much higher frequency, Again, Struhsaker found that the grunts and chutters of C. er ythrotis and C. cephus were very similar to each other and also to those of C. nictitans; the grunts, in fact, were indistinguishable. Of the vocalizations discussed above, the one which plays the most important role in helping to maintain species identity through reproductive isolation is the loud call. Struhsaker points out that vocalizations having different functions evolve at different rates: "Calls serving to alarm or coordinate groups may evolve more slowly than those enhancing reproductive isolation through the maintenance of cons pecific intragroup cohesion around one male" (1970:370), If

PAGE 68

56 this assumption is correct, then one would expect the chirps, grunts, and emitters to not be under the same pressures for species specificity that influence the loud calls. Furthermore, species showing similarities in these calls may be more closely related to each other than to species which do not have such similar calls (Struhsaker, 1970). It has been shown that the loud calls of G. mitis are readily distinguishable spectrographically from those of G. nictitans . Since these differences are greater than those found among conspecifics, a separate specific status for C. mills a nd C. nictitans seems warranted at this time. This conclusion is further supported by the observation that the chirps, grunts, and chutters of C. mitis are also not as similar to those of C. nictitans as are these calls within the C. nictitans subspecies and within the C. ervthrotiscephus species. According to Struhsaker (1970), the fact that grunts, chutters, and chirps are absent in the mona superspecies would seem to indicate that these calls are not to be interpreted as conservative, widespread vocalizations but rather may be indicative of close phylogenetic relationships. Following Struhsaker's reasoning, one can conclude that the similarities of the C. mitis calls to those of C. nictitans and C. eryth r o t i s c e ohu s , while not indicative of conspecific status, would indicate that C. mitis , C. nicti tans , and C. erythrotis ceohus are more closely relaxed to each other than to the mona superspecies . It should be emphasized again that future studies may reveal that the vocalizations of the more westerly

PAGE 69

57 distributed C. mitis populations may be considerably different from those reported here for C. m. kibonotensis , Information as to the amount of variation in vocalizations throughout C. mitis' range is needed to either support or refute the assumption made here concerning their supposed stability and uniformity. Information is also needed as to the extent to which C. mitis and C. nictitans may associate in their zones of sympatry. If the two species should be found to hybridize freely in these zones, then serious consideration of a conspecific status would be called for. Although the samples collected for C. mitis were often too small to permit rigorous statistical comparison and testing, it is recognized here that a true assessment of the degree of relationship between two species must be based on careful statistical measures before any great reliability i the data may be assumed. It is also recognized that taxonomi conclusions should not be based on one or a few characters alone, but should depend instead on the total assessment of morphology, physiology, behavior, karyotypes, etc. However, if Struhsaker's observation is true that "of all the single classes of characters so far used, vocalizations seem the most stable and thus the single most reliable indicator of phylogenetic affinities within this genus" (1970:402), then future studies of vocalizations of the ^exc^.cithecus species may very well prove to be of considerable importance in the assessment of phylogenetic relationships within this highly variable genus. The analysis of vocalizations, therefore, should merit considerable attention by primate taxonomists. n ' c

PAGE 70

CHAPTER III VOCALIZATIONS OF THE COLOBUS MONKEY In sharp contrast to the relatively vocal Sykes monkeys are the taciturn colobus monkeys ( Colobus polykomos palliatus). These beautiful, black and white primates sit for hours at a time without making a sound. Colobus live in small groups; those in the study area ranged in size from three to eight animals per group. They maintain a relatively small territory in comparison to the Sykes. Each troop was observed to occupy the same general area for the duration of the study and, as a result, they could be located quite readily. The taxonomic designation, C. polykomos , being used here follows that of Schouteden (19^-7), Tappen (i960), Napier and Napier (1967), and Kuhn (1967). However, others, i.e., Schwarz (1929), Dandelot (1968), and Rahm (1970), differentiate C. angolensis from C. polykomos and under this system the colobus of this study would fall within the former designation. It is generally agreed that before the taxonomic problems can be resolved more information is needed about the adaptations of these monkeys and areas of distribution overlap need to be investigated (Tappen, i960). 58

PAGE 71

59 The colobus have one habit which, while not vooal, definitely serves as a means of communication. The colobus take fantastic leaps, land with a resounding thump against tree limbs and branches, and continue in this way jumping from tree to tree. The other colobus are alerted (probably by both the sight and the sound) and look in its direction, watching the leaping colobus closely. These displays are ' quite impressive and seemed tc be given in response to the observer's presence, probably functioning in part as an attempt to intimidate the intruder. This ability to take leaps of considerable distance also serves to inoioate danger and the direction of troop movement. If a colobus is startled, it win vault itself seemingly into midair only to come crashing down into the thicker brush below. The other colobus are immediatelv ale^d anc also leap do™ into the lower level. A similar observation has been made by Schenkel and Schenkel-Hulliger (1967). In a short while, when the danger has apparently passed, the oolobus will gradually move back up intQ ^ n . g . ^ ^^ Marler ( 19 68) also emphasizes the reliance of the colobus on vasual cues for maintaining intragroup spacing during movement, pointing out that the conspicuous color of the colobus and their habit of keeoine hi?h in +h, + * . seeping nign m the trees facilitate visual coordination of their activities. Colobus tend to remain silent unless sufficiently disturbed. Such disturbing circumstances included the cloSe proximity of the observer, the milling around of a group of four or five dogs in the woods directly below the colobus

PAGE 72

60 troop, and the circling overhead of a pair of Eateleur Eagles (T erathopius ecaudatus ) . On the latter occasion, a group of eight cblobus were sitting relaxed in the tall trees, in this troop was a female with a small white infant. The female had been sitting unperturbed with the infant squirming about in her lap but as a pair of Bateleur Eagles slowly circled closer she became quite tense and attentive, craning her neck to watch the birds. After a few moments she clutched the infant to her and moved over to a more thickly leaved tree. As she did so, a male colobus gave a croaking, snorting sound and bounded over to the female's side, also watching the eagles intently. Moments later, the male was observed holding the infant so he must have taken it from the female when he moved near her. However, incidents such as this were infrequent and other than the vocalizations of infants or juveniles, the troop was generally silent. The v/hite infant was a focal point of interest for the entire group. In fact, all the adult females were so interested in the infant that at times it was very difficult to determine who the actual mother was. This was further complicated by the frequent practice of passing the infant from one adult to another when it would begin to squirm and squeal. The colobus were frequently found in the same immediate vicinity with the Sykes monkeys. The two species could often be seen feeding in the same trees or even on the same branch just a few feet apart.

PAGE 73

6l Classification of Calls Snort A loud, snort-like sound (Figs, ly^u an d Table 15) was given by adult colobus on several occasions. The call is most likely an alarm and/or threat call as it was evoked by the sudden appearance of the observer, by a group of dogs, and by the Bateleur Eagles. A similar call was reported for C. polykomos by Hill and Booth who described it as an "explosive snort, uttered partly through the nose" which was given by adults of both sexes ( 1957:311).

PAGE 74

62 4.0 8.0 kHz, The snort usually consisted of a single-unit phrase, although on one occasion a two-unit snort phrase was given, the two units being separated by an interval of O.^lj. seconds. The second snort was approximately only half as long as the first and its energy distribution was confined to a lower frequency range (0.25 3-30 kHz). Snort-Croak The snort may stand by itself as a call or it may precede another call identified here as the snort-croak (Figs. 25-27 and Table 16) . Marler (1970) mentions that C. gueraza has a distinctive roaring call which sometimes may be preceded by a snort. This may be equivalent to the snort-croak described here, but this writer hesitates to use the term 'roar' because such a term would seem to imply a single, long, loud call which definitely was not the case in the snort-croak. The snort-croak is probably equivalent to the "full call" reported by Hill and Booth (195?) as given by adult male C. p olykomos . These authors also noted that the snort-like alarm call may precede the full call. A jumpingroaring display said to be highly ritualized and which may function as a mechanism of intergroup spacing has been described by Ullrich (1961) and Schenkel and SchenkelHulliger (1967). The roaring is reported to spread from group to group so that a chorus develops. Although the snort-croak was never heard under these circumstances while this observer was in the field, it is possible that it is equivalent to this roaring display. Likewise, whether the "full call" described Hill and Booth (195?) is the same as the roaring display

PAGE 75

63 can not be determined from their report. Based on the verbal descriptions of the calls, they all seem to be similar; confirmation of this must await spectrograph^ comparison. The snortcroak is actually a long bout consisting of three different phrasetypes . Besides the initial singleunit snort phrase, the snort-croak includes a multi-unit phrase of a croaking, gutteral sound given in alternation with a faint, short, inhalation, single-unit phrase. The initial snort of the snort-croak is very similar acoustically and spectrographically to the snort phrase which has already been described. The phrase duration of the snorts cf the snortcroak averages 0.11 seconds and the major energy is distributed over a range of O.25 12.50 kHz (N=4) . Each croak phrase is composed of several short tonal units. Those measured had anywhere from eignt to sixteen units (mean of 10.21 units; N=24) . Each unit is of a very short duration (0.01 or 0.02 seconds) but together as a phrase the mean duration is 0.24 seconds (N=24). The croak is low in frequency and has a deep, gutteral sound. The average distribution of the heaviest concentration of energy is from 0.26 0.97 kHz. The number of croak phrases per bout is variable, averaging 5. 38 phrases per bout (N=13) but ranging from two to seventeen. It should be pointed out that the seventeen croak phrases represent an extreme; the next highest number of phrases is nine. The bout with seventeen phrases was given during the encounter with the eagles and was given more rapidly than the others. In addition, the croak phrases were shorter, averaging 0.16 seconds as compared to an average

PAGE 76

of 0.34 seconds (N=ll) for the other croak phrases. A portion of this seventeenphrase bout is pictured in Figure 27. In between the croak phrases is a faint sound made as the colobus inhales. This unit is nontonal and averages 0.05 seconds in duration (N=22). The average major energy distribution of this single-unit phrase is O.27 1.25 kHz (N=22) . TABLE 16 SNORT-CROAK Snort Phrase Units/Phrase Phrase Duration (sees . ) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy ' kHz) Croak Phrase Units/Phrase Unit Duration (sees . ) Interval Between Units (sees. ) Phrase Duration (sees . ) Lower Limit of Energy (kHz) Upper Limit of Energy (kHz) Number (N) Mean (x) Range (R) 4 1.00 1.00-1.00 0.11 0.10-0.12 0.25 0.25-0.25 1 12.50 12.50-12.50 245/24 10.21 8.00-16.00 2 ^5 0.01 0,01-0.02 221 24 24 0.01 0.24 0.26 0.97 0.01-0.02 0.15-0.43 0.25-0.40 0.75-1.00

PAGE 77

65 TABLE 16 (Continued) Inhalation Phrase Units/Phrase Phrase Duration (sees. ) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) Number (N) Mean (5c) Range (R) 22 22 0.27 1.25 22/22 1.00 1.00-1.00 22 o.OS 0.04-0.0? 0.25-0. 40 0.70-2.00 C-ecker White Infant The gecker (Fig. 28 and Table 1?) is a high-pitched, squeaky sound and was given quite frequently by a white infant colobus. usually the infant would be squirming and wriggling around while giving the call but it was impossible to tell if it was trying to nurse. The adult holding the infant would either shift it in its arms or else let another colobus take it. The frequent, at times almost constant, geckering by the infant was in sharp contrast to the usual nonvocal attitude of the rest of the troop. As a result, the somewhat noisy infant seemed to be able to solicit attention and be an important focal point for the rest of the group. Energy distribution of the gecker ranged from O.25 to 13.00 kHz, but on the sonogram the darkest portion averaged

PAGE 78

^6 o 3.6^ 6.69 kHz (N=il). The mean phrase duration was Q.32 seconds (N=ll) ana several such single-unit phrases were combined to give bouts of variable length. It is difficult to assess the duration of a bout or the number of phrases per bout since the call varies throughout in loudness and gives an impression of randomness and irregularity rather than of a fixed pattern of phrase sequences. TABLE I? GECKER WHITE INFANT Number (N) Mean (>;) Range (R) Units/Phrase ll/ll 1.00 1.00-1.00 Phrase Duration 11 0.32 0.27-0.^0 (sees . ) Interval Between Phrases (sees.) 9 0.13 0.05-0.23 Lower Limit of Major Energy 11 3.6^ 2.00-^.50 (kHz) Upper Limit of Major Energy 11 6.69 6.10-7.70 (kHz) The gecker is both tonal and nontonal, with the tonal section first forming an arch-like pattern on the sonogram which then levels out into the noise component. As mentioned above, one of these "arches" is considerably darker than the rest; the others occur above and below it in a harmonic patx err. .

PAGE 79

6? Err White Infant On one occasion a phrase consisting of three "err" sounds was heard at the end of a geckering sequence (Fig. 29 and Table 13). The exact relationship of this call to the gecker is not understood; it nay be a part of the gecker or it may be an individual vocalization. However, on the sonogram the call is different in appearance from the gecker. TABLE 18 ERR WHITE INFANT Number (N) Mean (x) Range (R) Units/Phrase 3/1 3.00 3.00-3.00 Unit Duration (sees. ) Interval Between Units (sees.) Phrase Duration (sees . ) Lower Limit of Energy (kHz) 3 ^^ Upper Limit of Energy (kHz) 3 2jl3 3 0.11 0.03-0.15 2

PAGE 80

65 Gecker Juvenile A call similar acoustically and spectrograph! cally to the gecker of the white infant was given by a juvenile (Fig. 30 and Table 19) . This individual had already attained its black and white coloration and, except for its smaller size, it was indistinguishable from the adult colobus. The juvenile no longer seemed to be very dependent on its mother although it was carried occasionally, apparently when frightened by the observer. The juvenile gecker is also composed of tonal archlike bands passing into nontonal components. Harmonics are present but the bands are closer together and more numerous than in the infant gecker. TABLE 19 GECKER JUVENILE Number (N) Mean (5c) Range (R) Units/Phrase Phrase Duration (sees . ) Interval Between Phrases (sees.) Lower Limit of Major Energy (kHz) Upper Limit of Major Energy (kHz) 13/13 13 11 13 13 1.00 1.00-1.00 0.11 0.04-0.14 0.02 0.01-0.05 2.60 1.90-3.80 8.05 6.00-10.10

PAGE 81

6 9 The individual single-unit phrases were of shorter duration than those given by the white infant, with an average length of 0.11 seconds (N=13). The average energy distribution ranged from 2.60 to 8.05 kHz (N=13), but again one band was somewhat darker than the rest at approximately k.O 5.0 kHz. As in the infant gecker, it is very difficult to judge the duration of a bout or to determine the number of phrases per bout. Occurring in the geckering bout on one occasion was a phrase which appeared on the sonogram as a single, wavy, linear unit, 0.5^ seconds in duration (Fig. 31). The unit is basically tonal and occupies a frequency range of approximately 2.90 6.00 kHz. The darkest portion was again in the vicinity of fc.o 5-0 kHz so perhaps this phrase is but a series of geckering phrases given close together and lacking harmonic development. The gecker was given by the juvenile as it was startled by the observer and ran to an adult female. Upon being braced by the female, the juvenile stopped vocalizing. em-

PAGE 82

CHAPTER IV VOCALIZATIONS OF THE BABOON Vocalizations and Crop-Raiding; The information collected on the baboon vocalizations formed part of a larger study on baboon cropraiding behavior. Unfortunately, conditions were somewhat less than ideal at the Shimoni site during the time this observer was in the field. This site promised to be the best suited for making recordings and observing the baboons' behavior. However, due to seasonal conditions and a long drought in the area, no significant crops were being grown other than a few remnants gone to seed from the last growing season. In addition, the Africans who usually sit up in observation platforms in order to spot baboons and chase them from the field were not present. When in the area a few years earlier, Maples (personal communication) noted what, superficially at least, appeared to be diversionary behavior. It seemed that one group of baboons would vocalize and thus attract and keep the attention of the shamba , or farm, guards while the rest of the troop would silently move into the farm a short distance away and raid the crops growing there. If this behavior were indeed coordinated and intentional, it would shed an interesting and new light on the nature of communication 70

PAGE 83

71 systems among nonhuman primates. It was the purpose of this part of the study to investigate and record those vocalizations given in conjunction with this behavior. Although the normal raiding conditions were absent during the course of this observer's investigations, the baboons did come to the shamba practically every day to forage for whatever remains could be found in the field. As a result, enough information was collected to allow at least a tentative explanation as to the nature of this seemingly deliberate "decoy" behavior. Although shamba guards were lacking and consequently the raiding situation not as tense as it might have been under normal circumstances, it was noted that the baboons, in reaction to an observer moving about rapidly in the field, would move into the forest edge and gradually begin to move silently around the perimeter of the farm. The baboons at the rear of this troop movement would be up in the trees giving the threat/alarm bark. It is to these vocalizing baboons thax the observer's attention is drawn. Consequently, he remains nearer to this group of baboons and, presumably, the shamba guards would be inclined to do the same. The vocalizing animals tend to minimize their own exposure to danger bv sounding the alarm when the human intruder is moving away or looking in another direction, a practical behavioral adaptation since shamba guards may be armed with rocks, sticks, or bows and arrows. With the humans' attention thus drawn to this vocalizing group, the rest of the troop is relatively free of human intervention and has the opportunity to enter the shamba again to carry out additional raids.

PAGE 84

72 It would seem, therefore, that even though the vocalizations given during the raids are diversionary in nature, they are probably not deliberately intended to be so. Instead; the calls given in reaction to the intruding human help to set up a fortuitous chain of events which allows part of the troop to continue raiding. At Roka, similar observations were made although the conditions were somewhat different. Instead of a farm surrounded on three sides by forest, as was the case at Shimon! , the farms at Roka were separated from the forest by a fairly well-traveled road. When raiding, the baboons would have to come out of the forest and cross the road to enter the farms. Unlike Shimoni, crops were being grown in most of the fields and some farms were protected by shamba guards, usually women or children. In general, these baboons were not very vocal at all but would carry out their raids in silence. But again, it was observed that vocalizations of any kind or any movement in the trees would tend to attract the guards to that area while further down the road other baboons would be free to enter the farms undisturbed. Although not intentional, such behavior functions very well as a decoy mechanism. The calls described in the following sections were recorded during encounters such as those described above. Classification of Calls TwoPha se 3ark The two-phase bark (Pigs. 32&33 and Table 20) was recorded on one occasion at Roka. This was just after a shamba

PAGE 85

73 raid had been completed and all the baboons were on the forest side of the road. The observation vehicle was driven adjacent to the point of entry into the forest.' It was then thax a large, fully adult male conspicuously sitting in a tall tree at the forest edge emitted two two-phase barks. The twophase bark has been reported by Hall and DeVore (1965) as an attack/threat vocalization given in reaction to humans and large predators. Units/Phrase Phrase Duration (sees . ) Lower Limit of Energy (kHz) Upper Limit of Energy (kHz) TABLE 20 TWOPHASE BARK Number (N) Mean (x) Range (R) 2/2 2 1.00 0.62 0.25 1.00-1.00 0.51-0.73 0.25-0.25 3.20-5.00 The two-phase bark is a loud, tonal, single-unit phrase of low frequency and long duration. The energy distribution for the first two-phase bark is from 0.25 to 5. 00 kHz and the distribution for the second is 0.25 3. 20 kHz. However, the heaviest concentration of energy for both calls is in the 0.50 I.50 kHz range. The duration of the calls is 0.73 seconds for the first and 0.51 seconds for the second.

PAGE 86

Oh. On the sonogram the call appears as a very dark band in the lower frequencies with some harmonic development. Bark Humans moving about in the farm ax Shimoni would usually provoke bark-like vocalizations (Figs. 34*35 and Table 21) from nearby baboons who would then withdraw into the forest. There seemed to be two types of barks distinguishable both acoustically and spectrographically; however, this may be due to sex and/or age differences. Basically, the difference seems to be one of pitch, with one bark type noticeably lower and less intense. Both may be described as single-unit phrases . TABLE 21 BARK Number (N) Mean (x) Range (R) Bark I Units/Phrase 12/12 1.00 1.00-1.00 Phrase Duration 12 0.32 0.20-0.40 (.sees . ) Lower Limit of Energy (kHz) 12 Upper Limit of Energy (kHz) 12 i+ wl r, O.25-O.50 3.70-5.00 j -\ II Units/Phrase Phrase Duration (sees . ) 7/7

PAGE 87

75 TABLE 21 (Continued) Number (N) Mean (x) Range (R) Lower Limit of Energy (kHz) ? Q>48 Upper Limit of Energy (kHz) ? 2
PAGE 88

76 TABLE 22 YAK Number (N) Mean (x) Range (R) Units/Phrase 3/3 i.qq 1 .00-1.00 Phrase Duration 2 0.30 0.30-0.31 (sees.) Lower Limit of Energy (kHz) 3 0.50 O.5O-O.50 Upper Limit of Energy (kHz) 3 6 . 4o 5.0O-8.O+ The average energy distribution of the yak is approximately 0.50 6. .'40 kHz (N=3); however, the harmonic pattern on one occasion extended beyond 8.0 kHz. The average duration of the call is 0.30 seconds (N=2). On the occasion when the calls were recorded the yak occurred in a repeated sequence of Bark I, Yak, Bark II given while an observer moved about in the shamba at Shimoni. The baboons giving the calls were at the forest edge. While they were vocalizing, the rest of the troop was quietly circling the farm a short distance within the forest.

PAGE 89

CHAPTER V SUMMARY AND CONCLUSIONS The study of primate behavior developed rapidly in the years following World War II to become a popular discipline within the fields of zoology and physical anthropology. However, one aspect of its study, primate communication, has only recently begun to receive the attention which it rightly deserves. That as yet no single study has been able to deal with social communication in all its aspects is testimony to the complexity of the task. Instead, individual researchers have chosen to study some specific aspect or channel of communication, usually of a single species, in the hope that with enough information collected in this manner a larger, more comprehensive overview of the communication process as a whole may eventually be possible. It is with this intent that the present study has been conducted. By means of sonograms the vocalizations of three species of Old World monkeys (Sykes monkey, colobus monkey, and baboon) were illustrated, described, and analyzed. The conclusions and findings are summarized in the following sections . 77

PAGE 90

?8 The Svkes Studv ---*— — , i ,j^_ 1 ^ Fourteen Sykes vocalisations were catalogued and described in terms of their physical characteristics (frequency, duration, tonality, etc.). The possible functional significance of the calls and any accompanying behavior was also presented. 2) The vocalizations were examined in terms of discreteness vs. gradedness. Most of the calls were found to be of the discrete type but calls forming a possible gradient were also discussed. It was emphasized that demonstrating the existence of graded signals does not imply the existence of a graded communication system. 3) The Sykes vocalisations show considerable variety in pitch or frequency and timing. An attempt was made to correlate these differences in structure with differences in functional requirements, particularly the need for localization. It was found that the properties of calls concerned with group cohesion and alarm or threat which were transmitted over distances without necessarily the aid of visual contact were such as to facilitate localization of the vocalizer. On the other hand, the structure of calls which were transmitted at close-range where visual or other cues may be operable did not show those features which would facilitate localization. The taxonomic position of the Sykes monkey, Cercopithecus mitis, in relation to C. nic titans has been a matter of some speculation. A preliminary analysis and

PAGE 91

79 comparison of the vocalisaTions of the two species indicated that separate specific status should be maintained at this time. However, analysis revealed that C. ir.it is was more closely related to C. nictitans and C, e rythrotis -ceohus than to the mona superspecies . The Colob us Studv 1) Two calls were described for adult colobus as well as one for juveniles and two for infants. When possible, their probable functions were discussed and any accompanying behavior was described. 2) Although vocalizations were the primary concern in this study, a nonvocal communication pattern, the jumping display, seemed to play an important role in the life of the relatively nonvocal colobus-. This behavior was described and its possible functional significance was discussed. The Baboon Study 1) The physical characteristics of four types of calls given during cropraiding were described. 2) The apparent "decoy" effect of the calls in attracting the attention of farm guards protecting the crops while other baboons silently entered the fields to raid was described and analyzed. It was concluded that the calls were fortuitously diversionary in nature rather than reprssenting a deliberate and intentional plan of behavior.

PAGE 92

80 There had oeen a tendency in studies of primate behavior and communication to accept as generalizations for all species that which had been described for but a few wellstudied populations. That the behavior of all species can not be seen in a baboon-rhesus context or that the behavior of all baboons is not typified by the Nairobi Park population, for example, has become increasingly apparent as more and more primate species and populations in various ecological settings are being studied. Before the formation of encompassing generalizations then, if indeed such should prove to be possible, the primary need in communication studies is one of collection of data on as many species as possible. Ideally, information on communication systems should be accompanied by information on behavioral context and functional significance. Through observation of the behavioral context accompanying a communicatory act, something can be learned regarding the functional significance of a particular vocalization, facial expression, etc. Analysis of the relationship between structure and function may indicate why certain structural traits have proved adaptive and have been maintained through the process of natural selection. Comparison of such findings from several studies on separate species or on different populations of the same species may reveal ways in which the ecological setting can influence which adaptive mechanisms will predominate. In addition, an understanding of the degree of variation to be found In communication patterns within species will aid in taxonomic interpretations and in the assessment

PAGE 93

Bl of vocalisations and other channels of communication as taxonomic characters. In this way, one can begin to understand something of the relationship between communication, organized social life, and ecology of man's closest relatives, the nonhuman primates. Finally, it should be emphasized that primate communication should not be studied solely in terms of how many features of human language may or may not be present, but also as an entity worthy of study in its own right. However, an understanding of the nature of communication systems among nonhuman primates and how these systems relate to social organization and environment may eventually shed light on the circumstances accompanying the origin of language in man and on the nature of this early system. That primate communication is a behavior which has proved adaptive and evolutionarily successful can not be disputed, but the mechanisms behind that adaptation are as yet not fully understood. It is hoped thax through further study and fieldwork more and more information may be collected and correlated to reveal something of the adaptive significance of communication systems in general and of how these systems are integrated into the larger whole of primate social life. It is this intricate weaving and integration of behavioral systems, of which communication is but one, which makes the coordination of individuals into a society possible, and which in turn insures the survival of the species.

PAGE 94

PLATES

PAGE 95

SYKE3 MONKEY

PAGE 96

m E<1 X a< QJ u it • El ' u # i f I I ! ! ! 1 I 0.20 O.iJ-0 Time (sees.) Figure 1 . Chirp Cs! x o fit US am* * mmm%mmta&mm*ki 0.20 0.^0 Tine (sees.) Figure 2. Chirp

PAGE 97

85 16 i 12 « x x CD &4 8 4 Ji^-^^^. i . I . .• ... 1 0.20 0.40 Time (sees.) Figure 3. Two-Unit Chiri a 1 cd . . i? \n 1 F 7 0.^0 Time (sees.) Figure 4. Two-Unit Chiro

PAGE 98

86 16 12 C3 3 CD 8 k 0.20 0.^0 Time (sees.) Figure 5. Three-Unit Chirp m^= I I i 0.20 0.40 Time (sees.) Figure 6. Nyah

PAGE 99

16 1 2 S3 CD k :i MWiwWri ;'•? i J 0.20 i I Time (sees.) 6? 0.40 Figure 7. Short Nyah (short nyah followed by chirp of another Sykes) T 12 ?at ^ rat I|,-: ; 'iJli^ . < *' » » a «>» L*. 0.20 0.^0 Time (sees.) Figure 8. Short Nyah (short nyahs followed by chirp of another Sykes)

PAGE 100

88 i^^i^ .Ti^fi^v^M^ i*,ini * -.mm. fpM#ttm** a»«-ti ttitttt&HMt 0.20 0.40 Time (sees.) Figure 9« Boom (ignore background noise at 3 kHz and 3-10 kHz,) , 16 ~ 12 » •i0.20 0.^0 0.60 Time (sees.) Figure 10. Squeal

PAGE 101

16 N 12 x a — 8 ft ! » i * r.vVvAW' Mitt' ,•'„ $!SM li i 1 1 li v 111 ikkikmiM in « ,i tttMrw^ii B 69 0.20 0.40 0.60 0.80 Time (sees.) Figure 11. Shriek (ignore background noise at 2-3 kHz) .. u laiS i , J HI Iv 73 **«*'• 0.20 0.40 0.60 Time (sees.) Figure 12. Chut

PAGE 102

90 16 12 sa ki^Vivy •&. •?•'. k<\\ ' 'rff, • tilt, i ( i\' ife V . .',;" * < i*' |v , ( ! L! WVH 0.20 O.iJ-0 Time (sees.) Figure 13. Chut (ignore background noise at 7-8 kHz) 16 £12 3 8 j fj'i^|j»»«./ i •!« i ii u •• ii >— M a 0.20 0.^-0 0.60 0.80 Time (sees.) Figure 14. Chutter

PAGE 103

91 rf CO u 0.80 Time (sees.) Figure 15 . Low, Short Chutter (ignore background noise at 7-8 kHz and at 0.80 sees.) 0.20 0.^0 Time (sees.) Figure lb. Chutter-Squeal (chirp of another Sykes superimposed at 0.60 sees.)

PAGE 104

92 * fHtMi^jm^ -. iv ' t. I 0.20 0.40 0.60 0.80 Time (sees.) Figure 17. Chutter-Squeal (continued) (chirp of Fig. 16 superimposed at 0.10 sees.) '««» |W '„ " , * < ** 0.20 l«»M*8*! .**«#* >«w»w»**f*i 0.40 o.6c Time (sees.) Figure 18. Chutter-Squeal (continued)

PAGE 105

93 16 12 i tsi 8 £»#. l»*M * Lifctoi, 0.20 . i iff*!* 1 S I t| 0.40 0.60 Time (sees.) Figure 19 • Growl (ignore background noise at 7-8 kHz) tsl a 4 cu ' Uf iilii I I 0.40 ir? ! 0.80 Time (sees.) Figure 20. Growl (ignore background noise at 7-8 kHz)

PAGE 106

9^ 0.20 i! Time (sees.) Figure 21. Err o I ft sstsaaaa 0.20 0.^0 o.6o Time (sees.) Figure 22. Uh (ignore background noise at 0-0.07 sees, and 0. 50-0. 65 sees.)

PAGE 107

C0L0BU3 MONKEY

PAGE 108

96 i " If .•: i mlteiw >**&&.**, 0.20 i I I ; 0.40 Time (sees.) Figure 23. Snort 1 J'lfcJlMMKttoil*t nit 5t?c c t !ij httti 0.20 0.40 Time (sees.) Figure 24. Snort

PAGE 109

97 Figure 25. Snort-Croak II ' I I * ft I «•• t-v o U 2 S iH tri 1 liri Mllllff .;
PAGE 110

!! I * ''Sty Vi' p^33™» ^''' "'" ' ••wjp^lW^'fc *'' *-1 < 95 0.20 0.^0 0.60 Time (sees.) Figure 28. Gecker-White Infant (ignore background noise at 10 kHz) C3 L mm* ,, Vi '!; . . .. ' , -\ if ' ^ 1 ,. k IH i .. • 1 ; !''« i f-'l-i *?< -'i in ' ; i i r cr o U fe 2 i ' 1 'W'Wta r -iUlg Figure 29. Err-White Infant (ignore background noise at 3-4 kHz)

PAGE 111

99 1 t < -^ 1 ° i-i S 1 '. i«M^W"^ir -Sp , fi4 "^ ,sr <,V "'«> !-...*> r 0.20 0.40 0.60 0.80 Time (sees.) Figure 30. GeckerJuvenile , |jg?*fc*.:*«*»iCvtr''«* ' ,f BMiUMKuiMiUMuaiuaaiAMii, 0.20 0.40 MtMMUfMf^ 0.60 0,80 Time (sees.) Figure 31. GeckerJuvenile

PAGE 112

BABOON

PAGE 113

101 •n in I , I r 0.^0 0.80 Time (sees.) Figure 32. Two-Phase Bark 8 N a 1 U u m -**? Ill ! I .40 0.80 Time (sees.) Figure 33. TwoPhase Bark

PAGE 114

io; o I urn i .( . m p i ' ; 1" fit 0.40 0.80 Time (sees.) Figure 3^. Bark I (ignore background noise at 0.Q0 sees.) E j f i t;l if II I * ill: 0.40 0.80 Tine (sees.) Figure 35Bark I, Yak, Bark II

PAGE 115

REFERENCES Alexander, R. D., i960. Sound communication in O^thopte^? 3 and Cicadidae. In: Animal Sounds and Communication. V. E. Lanyon and w . N. Tavolga, eds . American Institute of Biological Sciences, Publ. no. 7, Washington, D. C, pp. 38-92. , 1962. The role of behavioral study in cricket classification. Sys. Zool., 11: 53-72. » 1968. Arthropods. In: Animal Communication: Techniques of Study and Results of Research. T. A Sebeok, ed. Indiana University Press, Bloomington, pp. 167-216. Altmann, S. A., 1967. The structure of orimate social communication. In: Social Communication Among Primates. S. A. Altmann, ed . The University of Chicago Press Chicago, pp. 325-362. , 1968. Primates. In: Animal Communication: Techniques of Study and Results of Research. T. A. Sebeok ed. Indiana University Press, Bloomington, pp. L166-' Andrew, R. j., 1963a. Evolution of facial expression. Science, 142: 1034-1041. ' 1963b. The origin and evolution of the calls and facial expressions of the primates. Behavior 201-109. , 1964. The displays of primates. In: Evolutionary ana Genetic Biology of Primates, Vol. II. J. BuettnerJanus ch, ed . Academic Press, flew York, pp. 227309 • Blair, W. F., 1962. Non-morphological data in anu^a^ classification. Sys. Zool., 11: 72-84. » 1964. Isolating mechanisms and interspecies interactions in anuran amphibians. Quart. Rpv. Biol. '^Q33i|,_3ij4, ->?' .03

PAGE 116

104 Bogert, C. M., i960. The influence of sound en the behavior of amphibians and reptiles, in: Animal Sounds and Communication. W. E. Lanyon and V,'. N. Tavolga, eds. American Institute of Biological Sciences ', Publ. no! 7, Washington, D. C, pp. 1 37-320. Bolwig, N., 1964. Facial expression in primates with remarks on a parallel development in certain carnivores. Eehavior, 22: 167-I93. Booth, A. H., 1956. The Gercopithecidae of the Gold and Ivory Coasts: geographic and systematic observations. Annals and Magazine of Natural History, Ser. 12*, o : 476480 • > 1957Observations on the natural history of the olive colobus monkey, Frocolobus verus (van Beneapn) . Proc. Zool. Soc. London, 129: 421-430. Booth, C. P., 1962. Some observations on behaviour of Cercopithecus monkeys. Ann. N. Y. Acad. Sci., 102: 477V5T. ' ' 1968. Taxonomic studies of Cercopithecus mitis Wolf (East Africa). Nat. Geogr. Soc. Research Rooorts (1963 Projects), pp. 37-51. Bourliere, F., C. Hunkeler, and M. 3ertrand, 1970. Ecology and behavior of Lowers guenon ( Cercopithecus campbelli lowei ) in the Ivory Coast. In: Old World Monkeys: Evolution, Systematics, and Behavior. J. R. Napier and P. H. Napier, eds. Academic Fress, New York, pp. 297-350. Buxton, A. P., 1952. Observations on the diurnal behaviour of the redtail monkey ( Cercopithecus ascanius schmidtl Matschie) in a small forest in Uganda. J. Anim. Ecol., 21: 25-58. Dandelot, P., 1968. Preliminary Identification Manual for African Mammals, Primates: Anthropoidea. Smithsonian Institution Fress, Washington, D. C. Falls, J. B., 1963Properties of bird song eliciting responses from territorial males. Proc. 13th Int. Orn. Congr., l: 259-271. Gartlan, J. S., and C. K. Brain, 1968. Ecology and social variability in Cercopithecus aethiops and" C. mitis . In: Primates: Studies in Adaptation and Variability. P. C. Jay, ed . Holt, Rinehart and Winston, New York, po. 253-292.

PAGE 117

105 Gr "-'• F«» 1971. Diets and ecology of cropraid in.? baboons. Thesis, University of Florida, Gainesville. Haddow, A, J., 1952. Field and laboratory studies on an African monkey, Cercopithecus a scanius schmidti Matchie. Proc. Zool, Soc. London, 122: 297-394"; Hail, K. R. L. and I. DeVore, I965. Baboon social behavior. In: Primate Behavior: Field Studies of Monkeys and Apes. I. DeVore, ed . Holt, Rinehart and Winston. New York, pp. 53-110. Hall, K._R. L. and J. S. Gartlan, 1965. Ecology and behaviour _ of the vervet monkey, Cercopithecus aetniops, ^?i U1 * S i and ' Lake Victoria. Proc. Zool. Soc. London, i 4 5: 37-56. Hill, W. c. 0., 1966. Primates: Comparative Anatomy and Taxonomy, Vol. VI. Cercopithecoidea. University Press Edinburgh. * and A. H. Booth, 1957. Voice and larynx in African ana Asiatic Colobidae. J. Bombay Nat. Hist. Soc. 5^: 309-321. Hinde ; R. A. and T. E. Rowell, 1962. Communication by postures and facial expressions in the rhesus monkey (ijacaca mulatta ). Proc. Zool. Soc. London, 136:1-21. Hockett, 0. P., i960. Logical considerations in the study of animal communication. In: Animal Sounds and Communication. W. E. Lanyon and W. N. Tavolga, eds, American^ Institute of Biological Sciences, Publ . no. 7. Wasmngton, D. c, pp. 392-^30. Jolly, A., i960. Lemur Behavior: A Madagascar Field Study. The University of Chicago Press, Chicago. Kingdon, J., 1971. East African Mammals: An Atlas of Evolution in Africa. Academic Press, New York. Kuhn, H. J., 1967. Zur Systematik der Cercopithecidae. InNeue Ergenbnisse der Primatologie (Progress in Prima oology). D. Starck, R. Schneider, and H. J. Kuhn, eas. Gustav Fischer, Stuttgart, pp. 25-^6. Lanyon, W. E., 1969. Vocal characters and avian systematics. • in: bird /ocalizations: Their Relations to Curren+ Prooiems m Biology and Psychology. R. A. Hind^, ed. Cambridge University Press, Cambridge, p V , 291-310. Ltt3 ^ oh -l' M J "1 -959Call differentiation in a complex of seven species of Crinia (Anura, Leotodactyi idae) . Evolution, 13: ^-52-^6cT "

PAGE 118

106 Lbhrl, ri., 1963* The use of bird calls to clarify taxonomic relationships. Proc. 13th Int. Orn. Congr., 1:^4552. Marler, P., 1.955. The characteristics of some animal calls. Nature, 176: 6-8. i 1957* Specific distinctiveness in the communication signals of birds. Behavior, 11: 13-39. > 1959* Developments in the study of animal communication. In: Darwin's Biological Work. P. R. Bell, ed. John Wiley and Sons, New York, pp. 150-206. » 1965« Communication in monkeys and apes. In: Primate Behavior: Field Studies of Monkeys and Apes. I. DeVore, ed. Holt, Rinehart and Winston, New Yo^k. Pp. 5^4-584. » 1968. Aggregation and dispersal: two functions in primate communication. In: Primates: Studies in Adaptation and Variability. P. Jay, ed . Holt, Rinehart and Winston, New York, pp. 420-438. > 1970. Vocalizations of East African monkeys: I. Red colobus. Folia Frimatologica, 13: 81-91. and W. J. Hamilton III, 1966. Mechanisms of Animal Behavior. John Wiley and Sons, Mew York. Mayr, 5., 1969. Principles of Systematic Zoology. McGrawHill Book Company, New York. Moomaw, J. C, i960. ' A Study of the Plant Ecology of the Coast Region of Kenya Colony, British East Africa. The Government Printer, Nairobi. Moynihan, M. f 1964. Some behavior patterns of platyrrhine monkeys. I. The night monkey ( Aotus trivirgatus ) . Smiths. Misc. Publ . , 146: 1-84. > 1967 « Comparative aspects of communication in New World primates. In: Primate Ethology. D. Morris, ed. Aldine Publishing Co., Chicago, pp. 236-266. Napier, J. R. and P. H. Napier, 1967. A Handbook of Living Primates. Academic rress, New York. Patter, J. J., 1962. Ecological and behavioral studies of Madagascar lemurs in the field. Ann. Ni 1. Acad. Sci., 102: 267-281. 195-205.

PAGE 119

10? Ploog, D. W., 1966. The behavior of squirrel monkeys (Saimiri sciureus) as revealed by sociometry, bioacoustics, " and brain stimulation. In: Social Communication Among Primates. 5. A. Altmann, ed . The University of Chicago Press, Chicago, pp. 1^9-184. Rahm, U. H., 1970. Ecology, zoogeography, and systematic of some African forest monkeys, in: Old World Monkevs: Evolution, Systematics, and Behavior. J. R. Napier and P. H. Napier, eds. Academic Press, New York pp. 365-W. * Rowell, T. E., 1962. Agonistic noises of the rhesus monkey ( Macaca mulatta ) . Symp. Zool. Soc. London, 8: 91-96. and R. A. Hinde, 1962. Vocal communication by the rhesus monkey ( Macaca mulatta ) . Proc. Zool. Soc." London, 138: 279-294. Schaller, G. B., 1963. The Mountain Gorilla: Ecology and Behavior. The University of Chicago Press.,' Chicago. Schenkel, R. and L. Schenkel-Hulliger, 1967. On the sociolo^ of free-ranging Colobus . In: Neue Ergenbnisse der Primatologie (Progress in Primatology) . D. Starck, R. Schneider, and H. J. Kuhn, eds. ^Gustav Fischer Stuttgart, pp. 185-194. Schouteden, H., 1947. De Zoogdieren van Belgisch-Congo en van Ruanda-Urundi. Annales Muse'e du Con^o 3el°-e~2, Mo. 3» Schwarz, E., 1929. On the local races and distribution of the black and white colobus monkeys. Pr 0c . Zool. Soc. London, 99: 585-598. Struhsaker, T. T., 1967. Auditory communication anonc vervet monkeys ( Cercooithecus aethiops ) . In: Social Communication Among Primates. S. A. Altmann, ed". The University of Chicago Press, Chicago, pp. 28132^1 • , 1969. Correlates of ecology and social organization among African cercopithecines . Folia Primatologica, 1 1 : 80118. » 1 970. Phylogenetic implications of some vocalizations of Cercooithecus monkeys. In: Old World Monkeys: Evolution, Systematics, and 3ehavior. J. R. Napier and P. H. Napier, eds. Academic Press, New York, pp. 365-^^4. Tappen, N. C, i960. Problems of distribution and adaptation of the African monkeys. Current Anth^opolo^." 1 • 01120. &J '

PAGE 120

108 Thorington, R. W. Jr. and C. P. Groves., 1970. An annotated classification of the Gercopithecoidea. In: Old World monkeys: Evolution, Systenatics, and Behavior* J. R. Napier and P. H. Napier, eds . Academic Press, New York, pp. 62964-7. Ullrich, V,'., 1961. Zur Biologie und Soziologie der Colobusaffen ( Colobus guereza caudatus Thomas 1885). Der Zool. Garten, 25 (6): 305-368. van Hooff, J. A. R. A. M . , 1962. Facial expressions in higher primates. Symp. Zool. Soc. London, 8: 97-125 .

PAGE 121

BIOGRAPHICAL SKETCH Mary Louise Walek was bom in Miami, Florida, on May 12, 19^6. There she attended private and public schools and graduated from Miami Senior High School in 1964. She entered the University of Florida in September, 1964, and received the degree of Bachelor of Arts with Honors in June, 1968, with a major in anthropology. She entered the Graduate School of the University of Florida in September, 1968, where she has pursued studies in physical anthropology until the present. She has been the recipient of a NDSA Title IV fellowship (1968-1971) and a Graduate School fellowship from the University of Florida. 109

PAGE 123

I certify that I have read this study and that in mv opinion it conforms to acceptable standards of scholarly Vr-p sentation and is fully adequate, in scooe and Quality, as a dissertation for the degree of Doctor of Philosophy William R . Maples, Chairman Associate Professor of Anthropology I certify that I have read this studv and that in mv opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scooe and Quality, as a dissertation for the degree of Doctor of Philosophy/ Charles H. Professor of Anthropology . I certify that I have read this study and that in mv opinion ix conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and Quality, as a dissertation for the degree of Doctor of Philosophy. Norman N. n . r iarV:el Associate Professor of Speech, Psychology, and Anthropology This dissertation was submitted to the Department of Anthropology m the College of Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1972 Dean, Graauate Scnool


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EVVUFQ9AK_UA5J2X INGEST_TIME 2017-07-14T23:03:54Z PACKAGE UF00097647_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES