I Association patterns in spider monkeys

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted.

Also, if unauthorized

copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy.

Higher quality 6” x 9” black and white

photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

Bell & Howell Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.


P A T T E R N S OF A S S O C I A T I O N ,

FEEDING COMPETITION

AND V O C A L C O M M U N I C A T I O N IN S P I D E R MONKEYS, A T E L E S G E O F FROYI

Ga b r i e l R a m o s - F e r n a n d e z

A DISSERTATION in Biology P r e s e n t e d to the F a c u l t i e s of the U n i v e r s i t y of P e n n s y l v a n i a in Partial F u l f i l l m e n t of the Requirements

for the D e g r e e of D o c t o r of Philosophy

2001 ,J o

Q

Supervisor of Dissertation

%

Graduate Group Chairperson


UMI Number: 3003685

__

®

UMI

UMI Microform 3003685 Copyright 2001 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346


A mis papas, Andrea Fernandez y Eugenio Ramos, en reconocimiento a la formacion que me dieron.

A mi abuelito Ramon Fernandez y Fernandez, que siempre quiso que el hermoso mancebo fuera a Chapingo.

To the monkeys in Punta Laguna, who have withstood much worse disturbances than my constant presence.

11


ACKNOWLEDGMENTS I thank my advisors, Dorothy Cheney and Robert Seyfarth, for their constant support and critical stance during my studies. This includes those times when, not knowing my whereabouts in the field, their best guess was that I had joined an obscure religious sect. Peter Petraitis contributed many ideas to the analyses presented here. All of them visited Punta Laguna and shared some of the experiences that I cannot transmit in this dissertation. John Smith provided office space and an inspiration in his unabated passion for his study subjects and ideas. Mark Liberman demonstrated how much fun it is to create bridges between disciplines. Dan Janzen always brought up the practical issues and convinced me that only someone as energetic as him can have one foot in academia and another in conservation.

Joann Andrews kindly shared her beautiful house in Merida and sent me out to Punta Laguna knowing much more than she told me. Her dedicated love for conservation and for the Maya people will always be an inspiration. My field assistants and friends, Eulogio and Macedonio Canul, taught me as much as I taught them, and it was always great fun to iii


work with them. The rest of the people from Punta Laguna, who perhaps will never understand what on earth were we really doing with all that equipment, enriched my life in the forest with their crazy wisdom.

Laura Vick and David Taub, who began the spider monkey project in Punta Laguna, were very helpful in keeping the study going in times of hardship. The staff at Pronatura Peninsula de Yucatan, especially Sonja Macys, was proof that Joann's inspiration did not only reach me.

My fellow graduate students, Katherine Hardy, Klaus Zuberbuhler, Jon Sullivan, Christine Hawkes, Julie GrosLouis, Christel Lutz, Jesse Snedeker and Jane Kauer turned my visits to Philadelphia into a pleasure. I especially thank Dr. Albert E. Kim, in whose kitchen I cooked the final drafts of this dissertation. Friederike Range did the acoustic analyses presented in chapter 3.

Other members of the animal behavior group at Penn, Robert Harding, Martha Manser, Julia Fischer, Drew Rendall, Michael Owren and John Crawford provided useful comments during various stages of my research. In Mexico, Hugh iv


Drummond, Constantino Macias, Jose Luis Osorno and Magda Giordano did the same. Hal Whitehead advised me on the analyses presented in chapter 1 and kindly shared his interesting software.

Barbara Ayala, Danielle Oliva, Martha Bonilla, Monica Pimenta and Regina Peon helped with many aspects of fieldwork and helped me to keep my feet on the ground during all those months of fieldwork by having to teach them. Roberto Ruiz introduced me to the spider monkeys in Chiapas, where Leonardo Heiblum and Ivana Sejenovich climbed the blue hills with me in search of monkeys. Finally, Mariana Gullco spent the winter in Philadelphia while I prepared this dissertation. It cannot have been much fun. Gracias, Lunis.

My studies at Penn were financed by a graduate scholarship from the National Council for Science and Technology

(CONACYT, Mexico). Fieldwork was supported by

grants from the Wildlife Conservation Society, the National Comission for the Knowledge and Use of Biodiversity (CONABIO, Mexico), the Mexican Fund for the Conservation of Nature (FMCN) and the Turner Foundation. v


ABSTRACT PATTERNS OF ASSOCIATION, FEEDING COMPETITION AND VOCAL COMMUNICATION IN SPIDER MONKEYS, Ateles geoffroyi Gabriel Ramos-Fernandez Dorothy L. Cheney The fission-fusion structure of animal groups may reduce direct feeding competition between group members but may require special interactions between group members to maintain social bonds. This study used behavioral observations of two habituated groups of black-handed spider monkeys (Ateles geoffroyi) in the northeastern Yucatan peninsula to explore these two aspects of their fission-fusion social structure. Association patterns between males show signs of active companionship that is probably based on affiliative social interactions. Females are found together in subgroups at rates that are indistinguishable from a random expectation, and the absence of social interactions between them suggests that they are simply converging at feeding spots, without any preference for particular female social companions. Males and females show signs of active avoidance, associating at lower rates than would be expected by chance. These association patterns may be influenced by long distance vi


vocal communication, as suggested by the results of simultaneous observations of two subgroups, as well as those of playback experiments using the species' most frequent vocalization, the whinny. Subgroups approach each other more often when they are within hearing range, and individuals in those subgroups vocalize more often, than when farther apart. Playback experiments showed that a close associate of the caller, who is not present in the same subgroup, is more likely to approach the speaker than another, less closely associated monkey. The size of subgroups shows no change between seasons with high fruit abundance and other seasons, and individuals do not travel more when in large subgroups compared to smaller subgroups. Together, these results suggest that social relationships in spider monkeys are maintained in part through vocal interactions between individuals in different subgroups. This in turn may influence the size and composition of subgroups in a manner that is unrelated to the abundance of food within a patch. Indirect feeding competition may still occur, probably not within subgroups but between all monkeys living within a given area.

vii


TABLE OF CONTENTS PP

ACKNOWLEDGMENTS

iii

ABSTRACT

vi

LIST OF TABLES

x

LIST OF ILLUSTRATIONS

xii

INTRODUCTION

1

CHAPTER 1: ASSOCIATION PATTERNS IN SPIDER MONKEYS: PASSIVE AGGREGATION OR ACTIVE COMPANIONSHIP AND AVOIDANCE? Summary Introduction Methods Results Discussion

12 13 16 22 39

CHAPTER 2: TESTING THE EFFECT OF SCRAMBLE COMPETITION ON THE SIZE OF SPIDER MONKEY SUBGROUPS Summary Introduction Methods Results Discussion

45 47 55 65 83

CHAPTER 3: WHINNIES BY SPIDER MONKEYS (Ateles geoffroyi): STAYING IN TOUCH WITH CLOSE ASSOCIATES? Summary Introduction Methods Results Discussion

93 95 102 116 138


CONCLUDING REMARKS

147

APPENDIX 1: list of all spider monkeys in the study groups

15 6

APPENDIX 2: list of tree species found in the study site BIBLIOGRAPHY

159 164

ix


LIST OF TABLES PP

Table 1-la: association indices between adult females, EU group

24

Table 1-lb: association indices between adult males, EU group

25

Table 1-lc: association indices between adult females and males, EU group

25

Table 1-ld: association indices between all adults, MX group

26

Table 1-2: summary of permutation test results

29

Table l-3a: number of greeting embraces between adult females, EU group

32

Table l-3b: number of greeting embraces between adult males, EU group

33

Table l-3c: number of greeting embraces between adult females and males, EU group 33 Table l-3d: number of greeting embraces between all adults, MX group

34

Table l-4a: grooming rates between adult females, EU group

35

Table l-4b: grooming rates between adult males, EU group

36

Table l-4c: grooming rates between adult females and males, EU group

36

Table l-4d: grooming rates between adults, MX group

37

Table 2-1: results of correlations between median subgroup size and daily range

79


Table 2-2: results of the correlation between subgroup size and distance traveled every hour by different individuals

82

Table 3-1: results of the analysis of variance on different acoustic parameters of the whinny

119

Table 3-2: Results of the discriminant function analysis using caller identity as a dependent variable

121

Table 3-3: results of the playback experiments

136


LIST OF IL LUSTRATIONS PP

Figure 1: photograph of a spider monkey subgroup

9

Figure 1-la: cluster analysis of association indices, EU group

27

Figure 1-lb: cluster analysis of association indices, MX group Figure 2-1: map of the study site

28 57

Figure 2-2: relative abundance of different tree species Figure 2-3: phenology profiles, 1998-1999

66

67

Figure 2-4a: frequency distribution of subgroup size, EU group

69

Figure 2-4b: frequency distribution of subgroup size, MX group

70

Figure 2-5a: monthly consumption of ramon fruit and subgroup size, EU group

72

Figure 2-5b: monthly consumption of ramon fruit and subgroup size, MX group

73

Figure 2-6: subgroup size and traveled distance, 0800 to 0900 hours, EU group

75

Figure 2-7: speed by sex of adult spider monkeys

77

Figure 2-8a: daily average subgroup size and daily travel distance, EU group

79

Figure 2-8b: daily average subgroup size and daily travel distance, MX group

80

Figure 3-1: spectrograms of whinnies from two different individuals

117


Figure 3-2: length of dyadic associations depending on initial whinny occurrence, EU group

123

Figure 3-3: length of dyadic associations depending on initial whinny occurrence, MX group

125

Figure 3-4: association length for each pair, MX group

126

Figure 3-5: association length for each pair, EU group

127

Figure 3-6: distance between subgroups and changes in distance

130

Figure 3-7a: individual whinny rate and distance to another subgroup, EU group

132

Figure 3-7b: individual whinny rate and distance to another subgroup, MX group

133


It is venturesome to think that a coordination of words (philosophies are nothing more than that) can resemble the universe very much.

It is also venturesome to think that

of all these illustrious coordinations, one of them -- at least in an infinitesimal way

—

does not resemble the

universe a bit more than the others.

Jorge Luis Borges "The Avatars of the Tortoise", 1952

x iv


INTRODUCTION


Animal group life poses interesting problems for biologists. As a consequence of living with others, animals must share space, resources and/or mates. They must also interact in various ways in order to achieve coordination in their daily activities, or to space themselves out and avoid potential conflicts. This dissertation addresses these two aspects of group life in spider monkeys

(Ateles

spp) .

The structure of animal groups can vary widely. On one end, there are loose aggregations of animals that simply use a common space, without developing long-term social relationships

(e.g. ungulate herds: Clutton-Brock et a l .

1982; schools of fish: Shaw 1962). On the other end, there are complex, highly structured groups with constant membership, where individuals of different age and sex, kin and nonkin, develop long-term bonds

(e.g. most primate and

cetacean species, Smuts et a l . 1987; Mann et a l . 2000)1.

1 Insect societies are not considered in this discussion, due to the fact that their social structure is based on a high degree of reproductive skew between close kin, so that the fitness of each individual depends on the fitness of the colony (rev. in Wilson 1971) .


There is a kind of social structure that appears to be particularly complex. Kummer (1968) coined the term fission-fusion for the group structure of Hamadryas baboons (Papio hamadryas). In this species, a troop of several hundred individuals using the same sleeping cliff splits in ever smaller units during the day, while baboons forage on seeds, leaves and fruits that appear widely scattered throughout the desert. The troop splits initially into two or three bands that begin their daily march in different directions. Along the way, bands split again into various clans of 20-60 baboons. Depending on the season, these clans can split into several one-male units (an adult male with several breeding females) that then forage on their own during the day, or can remain as part of the same clan. In this society, clan and band membership are constant, and most social interactions occur only within members of the same band (Stammbach 1987) . Similarly, members of one-male units never separate except for the occasional migration of an adult female or a change of male leadership. At the end of the day, when baboons return to their sleeping sites, one-male units join each other, clans join other clans and each band converges with others in the sleeping cliff where the whole troop spends the night. 3


In the social structure of Hamadryas baboons, there are distinct patterns of association between different group members. These patterns can range from the tight bonds between an adult male and the females in his unit, to the relationship between two females from different bands, who may sleep in the same cliff but never interact with one another in any other way. Variations on this social structure, albeit with less of a hierarchical organization in the lines along which fission and fusion occur, can be found in several species, including African elephants (Loxodonta africana, Poole et a l . 1988) spotted hyenas (Crocuta crocuta, Kruuk 1972),

bottlenose dolphins,

(Tursiops truncatus, Smolker et al. 1992), chimpanzees, (Pan spp, Goodall 1968) and spider monkeys

(Ateles spp, van

Roosmalen and Klein 1987; Symington 1990).

Of the species with fission-fusion social structure, spider monkeys and chimpanzees show remarkable similarities in their association patterns Hiraiwa-Hasegawa,

(Goodall 1968; Nishida and

1987; Symington 1990). In both species,

subgroups of one to several dozen individuals forage on fruits and leaves over large areas of tropical forest. 4


Subgroup membership, contrary to the Hamadryas baboon society, is not constant, and even the most closely bonded adult males are found in different subgroups on occasion. Although it is rare to find all group members together, the group is a clearly defined unit, whose members interact at higher rates between them than with members of other groups. Also, the group occupies a common range that the males defend from neighboring groups. Contrary to the pattern observed in the majority of primate species, organized in matrilines of closely related females, in spider monkeys and chimpanzees it is the males who form the closest bonds and stay in their natal groups. Females tend to be more solitary and normally migrate to other groups upon reaching adulthood. The convergence of social structure in two distantly related species living in similar habitats and feeding on similar resources, presents an interesting opportunity to study the relationship between environmental variables and social structure (Wrangham 1987; Symington 1990).

The fission-fusion social structure seems to be an ideal solution to the need for sharing resources with others in the group. By associating in temporary subgroups, 5


individuals of these species can forage over wide areas, without entering into direct conflicts over food with other group members. At the same time, fission-fusion presents a challenge to our understanding of the mechanisms by which groups are formed and maintained by individuals with different social relationships. Because the size and composition of subgroups are constantly changing, group members have developed social interactions that serve to mediate spacing and cohesion (e.g. Kummer 1995 in Hamadryas baboons; Mitani and Nishida 1993 in common chimpanzees,

Pan

troglodytes; McComb et al. 1999 in African elephants; Janik 2000 in bottlenose dolphins).

This dissertation is an attempt to shed light on these issues by studying the patterns of fission-fusion in two groups of free-living, habituated black-handed spider monkeys

(Ateles geoffroyi) living in the northeastern

Yucatan peninsula. Chapter 1 deals with the association patterns among individual spider monkeys. The goal of this chapter is to describe the social structure of spider monkeys in terms of subgroup membership and the social interactions between different individuals. The strategy of this chapter is to construct a random model of subgroup

6


membership, where monkeys simply join and leave each other without regard to the identities of others. This model is then compared to the observed patterns of association, testing the possibility that some pairs of monkeys actively search or avoid each other as association partners. Then, the nature of these nonrandom associations is explored further by comparing the social interactions between different pairs to their association rates.

Chapter 2 deals with the consequences of these association patterns in terms of competition for food. The goal of this chapter is to explore the possibility that by varying subgroup size, monkeys can track the natural variation in the distribution and abundance of their food resources in such a way as to maximize the energy gain of individuals. This chapter tests the hypothesis that feeding competition limits the size of subgroups because large subgroups need to travel more than smaller subgroups to find enough food for all members. This hypothesis is tested by comparing the size of subgroups in seasons with different abundance of food, as well as by analyzing the relationship between the distance traveled by individual

7


monkeys and the size of the subgroups in which they are found.

Chapter 3 returns to the association patterns and asks, how are social relationships maintained in such a fluid social structure? In other words, what social interactions are needed to maintain the flexibility in subgroup size and the social bonds between individuals at the same time? The main goal of this chapter is to determine whether closely bonded individuals use vocal communication to maintain contact in the context of the observed pattern of subgroup fissions and fusions. Observations and playback experiments are used to test the hypothesis that through the use of vocalizations that can be heard by monkeys in nearby subgroups, signalers can inform their close associates about their presence and thereby influence their mutual distance and tendency to remain together.

Each of these chapters contains additional background on the topics of association patterns, feeding competition and vocal communication as they relate to the spider monkey's social structure. Each chapter also has a detailed section on methods used, with a brief description of the study


site. Also, each chapter interprets the results in light of available evidence from spider monkeys and other similar species.

Finally, in the section on Concluding Remarks, I attempt to put these results together and suggest additional questions and studies. Also discussed in this section are the implications of these results for the successful conservation of the spider monkey populations in the region. Figure 1: photographs of the study site and subjects

Figure 1 legend: on this page, China, Susana y Jose feed on a pixoy tree during the dry season (courtesy of David Taub) . 9


On this page, a view of the lake of Punta Laguna, from the edge (courtesy of Barbara Ayala) and from above (Gabriel Ramos-Fernandez).

V jA >

aSjji8»

*r "H> ,v« T«*/P>» -
e r» co e> o o o o o o e o> o o CO o o o o o e o o CO o m n r * . r » r - r » h » r - » r » r - » c o c o c o < o « o c o c o G o i o c o a < o a > 0 > o i 0 ) o > a > o > o > o ) 0 > c 3 > o ) a > o > a > 0 ) 0 > c » o > o i o > o > o > o > o > a ) a > o > a ) o > o > o ) 0 ) o > o > o > a > o > o > o > o > < n o > 0 )

o O CO Q

sss

Year Month

Figure 2-5 legend: bars represent the proportion of feeding samples for each month in which monkeys were feeding on ramon fruits (left vertical axis). Dots represent the median subgroup size for each month, with a line representing the range from the lower to the upper quartile (right vertical axis). Median subgroup size is negatively correlated with proportion of ramon fruit in diet (Pearson r2 = 0.18, N = 33, P < 0.05). P r e d i c t i o n 2: w h en the u t i l i z e d resou rce s abundant,

are h i g h l y

the r e l at iv e r a n gi ng cost s h o u l d be lower

If in a season with abundant resources, individuals do not need to travel far to obtain enough food, the cost to 73


each of them of adding subgroup members should be less than in seasons when they need to travel farther. In particular, RRC should be lower in seasons of high ramon fruit consumption than in other seasons.

For this analysis, RRC was calculated by measuring the distance traveled by subgroups of different size between 0800 and 0900 hours in various times of year (see Methods).

In neither of the study groups was there a significant correlation between the size of the subgroups and distance traveled (EU group, Pearson r2 = 0.07, N = 26, P < 0.19; MX group r2 = 0.04, N = 31, P < 0.28). In fact, when considering only those time periods when there was little or no consumption of ramon fruit, there was a negative correlation (Figure 2-6; RRC estimated by the slope divided by the intercept at subgroup of one = -0.25). There were too few samples to uncover any significant trend in the season of high ramon consumption (two data points are shown in Figure 2-6, showing the opposite tendency). Data in the MX group did not show any significant correlation between size of subgroups and distance traveled (data not shown).

74


In at least one of the study groups, RRC in the season of low ramon fruit consumption is clearly negative

Figure 2-6: subgroup size and traveled distance, 0800 to 0900 hours, E U group

1500

1250 -

1000

r

-

750 -

500 250 -

Subgroup size

Figure 2-6: correlation between subgroup size and distance traveled from 800 to 900 hours in the EU group, subgroups classified by season of ramon abundance (plus signs, ramon months; squares, non-ramon months). The correlation during non-ramon months is significant (Pearson r2 = 0.17, N = 24, P < 0.05; intercept = 621, intercept at subgroup size of one - 497.1, slope = -123.9). P r e d i c t i o n 3: ma les

travel faster than females

Males could be traveling in larger subgroups than females because of a difference in the costs of scramble


competition for each sex. If males travel faster than females, the effect of increasing subgroup size on the net per capita energy intake will be lower for males than for females. Females could then be maintaining their travel costs at the same level as the males1 by traveling in smaller subgroups. The prediction is that adult males should be faster than females.

This prediction was supported. Figure 2-7 shows the speed of adult females with dependent infants and male spider monkeys. Males are 1.3 times faster than females, on average. The difference between both sexes is significant (one-tailed Student's t-test, t = 1.99, df = 103, P < 0.05) .


Figure 2-7: speed, by sex of adult spider monkeys

7“

6“ 54"

2“

Sex

Figure 2-7 legend: Mean, standard errors and standard deviations shown as lines over the data points. Each point represents one measurement of an individual's speed. FI = females with dependent infants, M = males. General mean speed is shown as a horizontal line at 2. 7 km/hr. Prediction fe ma le s

3:

the relativ e r a n gi ng cost is h i g he r for

th an for m a l e s .

If scramble competition affects females more strongly than males, due to their different travel efficiencies, then for a given subgroup size females should be traveling more than males. Measuring the RRC for different individuals of each sex should reflect this pattern.

77


A data set obtained during 19 days in September-October 1999 allowed an estimation of the daily travel distance by 19 individuals

(8 females and 4 males in the EU group; 4

females and 3 males in the MX group). The subgroup size varied widely during the day. To account for this variation, two approaches were taken: one was to calculate the median subgroup size for each individual and correlate this with the distance traveled by him/her during the day. The other was to perform a correlation for each individual on the hourly measurements of subgroup size and distance traveled.

Figures 2-8a and 2-8b show the median subgroup size in which an individual was found and the distance traveled by him/her from 0800 to 1700 hours. None of the correlations is significant (Table 2-1). Values of RRC calculated from the estimated slopes and intercept at subgroup size of one are very low or close to zero.

78


Table 2-1: results of correlations between median subgroup size and daily range

Group and sex

r2

N

P

Slope

Intercept at subgroup size one

RRC

EU females EU males MX females MX males

0.01 0.01 0.00 0.32

8 4 4 3

n.s. n.s. n.s. n.s.

72.4 235 12.58 -1247

1650 1398 32.8 5460

0.04 0.17 0.38 -0.23

Table 2-1 legend: results of correlations for the data shown in figures 2-8a and 2-8b. RRC is the relative ranging cost, i.e. the slope of the correlation over the intercept at subgroup size of one. n.s. = non-significant correlation.

Figure 2-8a: daily average subgroup size and daily travel distance, EU group 4000

■eled ii

1* 3000

2000

5

1000

2

3

Median num ber o f adults in subgroup

79


Figure 2-8h: daily average subgroup size and daily travel distance, MX group 4000

~

3000 .

*o 1

2000

q

1000 .

2

3

Median num ber of adults in subgroup

Figure 2-8 legend: correlations between the median subgroup size in which an individual was found throughout the day and the total distance traveled by him/her during the day. Data are shown separately for females (diamonds) and males (squares) in the EU group (a) and in the MX group (b). Each point represents one individual followed for one compete day. The results of the correlations are shown in Table 2-1. Although not significant, the fitted line is shown on the graphs.

In the second approach to estimate RRC, hourly measurements of subgroup size and travel distance were correlated for each individual separately. Thus, for each individual, the data consisted of his/her subgroup size during each consecutive hour and the distance he/she 80


traveled during that hour, for a total of 10 hours. The results of these correlations in both study groups are shown in Table 2-2.

None of these correlations was significant, i.e. there was no significant relation between the subgroup size in which an individual was found every hour and the distance he/she traveled. The slope of most of the fitted lines was negative, except for one male and one female in the EU group and one female in MX. The intercepts at subgroup size of one show no apparent trends, except for the males in MX, whose lines intercept the subgroup size of one at a longer distance than females in the same group or both sexes in the EU group. Values of RRC calculated from these slopes and intercepts show either negative values or values close to zero.

81


Table 2-2: results of the correlation between subgroup size and distance traveled every hour b y different individuals

Sex

r2

N

p

Slope

Intercept at subgroup size one

RRC

RO EN PN AJ BT MO CU OC IS OF LU RI

m m m m f f f f f f f f

0.2 0.03 0.03 0.18 0.14 0.2 0.08 0.27 0.09 0.06 0.19 0

10 10 10 10 10 10 10 10 10 10 10 10

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

-99 -6 41.2 -32.7 -71.3 -15.3 68.1 -4 4.2 32.3 -53.6 117.2 3.68

560.7 118.2 29.2 285.9 270. 6 154.8 133.2 163.3 148.1 180.8 65.2 135

-0.18 -0.05 1.41 -0.11 -0.26 -0.10 0.51 -0.27 0.22 -0.30 1.80 0. 03

DA BE PA CE CH VE CL

m m m f f f f

0.11 0.24 0.24 0 0.08 0.07 0

10 10 10 10 10 10 10

n.s. n.s. n.s. n.s. n.s. n.s. n.s.

-78.7 -81.9 -188.1 0 -19.3 -22.3 7.9

417.5 406.2 762. 6 230 214.9 238.1 159

-0.19 I o NO O

ID

1 O o

-0.25 0. 00 -0.09 0. 05

Table 2-2 legend: results of the correlations between the hourly measurements of subgroup size and distance traveled by each individual during

one day. 10 hourly samples for each

individual were taken from 0800 to 1700 hours. RRC is the relative ranging cost, i.e. the slope of the correlation over the intercept at subgroup size of one. n.s. = non-significant correlation.


DISCUSSION

The results reported here do not support the hypothesis that scramble competition plays an important role in limiting the size of spider monkey subgroups. First, median monthly subgroup size did not increase, in either of the study groups, during those months when a very abundant resource was heavily consumed. The relative ranging cost (RRC), an estimation of the additional foraging effort that individuals in large subgroups made compared to those in smaller subgroups, yielded a negative figure for the season in which the abundant resource was not heavily consumed. There was some support for the prediction that differences between the sexes in the degree at which they are affected by scramble competition could be responsible for the differences in gregariousness: males traveled faster than females. However, there was no difference between the sexes in their estimates of RRC, which were generally negative or close to zero.

The apparent absence of scramble competition in these two study groups could be due to: 1) an artifact of the methods by which scramble competition was estimated; 2) the 83


possibility that RRC is not a good estimate of scramble competition in Ateles; 3) the possibility that scramble competition does not play an important role in this population and/or 4) the possibility that scramble competition does not play an important role in Ateles.

The assumption that spider monkeys in Punta Laguna are using an unlimited resource during the months that they feed heavily on ramon fruit seems justified. Trees of this species, at densities of more than 200 individuals per hectare, clearly do not impose on individual spider monkeys the need to travel far or decrease subgroup size in order to maintain a constant resource intake per capita. Such high densities of this species are common in remaining fragments of old-growth forest in the Yucatan peninsula and, because of their association with archeological sites, have been hypothesized to be remnants of forestry systems of the ancient Maya civilization (Gomez-Pompa and Kauz 1987). Whatever their age, ramon trees in Punta Laguna are big enough to provide abundant fruit at the patch scale and, because of their high density in the remaining medium forest, also at a larger, habitat scale. At least one of the two parameters, travel distance or subgroup size, 84


should have shown some change if scramble competition was at play.

The validity of the RRC measurements can be similarly justified. While none of the individual one-day follows yielded any significant correlation between subgroup size and daily traveled distance, the RRC of subgroups traveling from 0800 to 900 hours is clearly negative in one of the study groups. These data were taken in seasons when monkeys were not feeding heavily on ramon fruit, when, according to the hypothesis, RRC should have been greater than in other seasons. It is possible that it is the distance traveled in one day which is adjusted to the abundance of fruit and not the distance traveled in one hour. However, it is from 0800 to 0900 hours when spider monkeys in Punta Laguna travel farther and when the first foraging bout in a different site as the one used for sleeping is about to begin. An adjustment of subgroup size to the distance traveled would be expected during this time if scramble competition was at play.

It is possible that the intensity of scramble competition in spider monkeys is not captured by an index 85


such as RRC. Isbell (1991) postulated that, when foods are dispersed, scramble competition should not occur in species that can simply increase their foraging swath. It is possible that when feeding heavily on ramon fruit, spider monkeys simply increase their dispersion. The difficulties in defining a subgroup in fission-fusion primate species have been discussed by Chapman et al. (1993). One explanation for the apparent absence of scramble competition is that, for the spider monkeys in Punta Laguna, subgroups have been wrongly defined. In simultaneous observations, subgroups in the same group have been seen to maintain parallel travel paths at distances of 100-300m (this study, Chapter 3). It may be that subgroups coordinate their position by avoiding or approaching each other, communicating with long-distance calls (see Chapter 3). If this were the case, in order to understand the influence that food abundance and distribution has on grouping patterns in this species, the definition of subgroup would need to be extended to encompass all those individuals that could potentially influence each other's foraging paths.

86


The distance traveled between food patches may be affected by other factors unrelated to food. Ultimately, an increase in the distance traveled by larger subgroups compared to smaller ones is the result of the decisions of individual monkeys. It is unknown how animals come to "agree" on the time to leave a food patch and move elsewhere

(rev. in Boinski and Garber 2000). A negative

value of relative ranging cost could arise if larger subgroups took more time to start moving than smaller subgroups, regardless of the resource abundance and only as a result of the time they take to reach a collective decision to leave. This would cause larger subgroups to move slower and therefore travel less distance than smaller subgroups.

The temporal scale at which RRC was measured in this study could be inappropriate. In a study of Ateles paniscus in Manu, Peru, dealing specifically with feeding competition, Symington (1988) found a significant positive correlation between the size of subgroups and the distance they traveled in a day. Specifically, this relationship was between the monthly averages of subgroup size and daily travel distance. This estimate of scramble competition is 87


on a much larger temporal scale than the one obtained here, in which data points for distance traveled by subgroups of different size consisted of days or even hours. It may be that the relative ranging cost is not a good measure of scramble competition over short temporal scales, at least for fission-fusion species in which subgroup size changes several times during the day.

The possibility that the population of spider monkeys in Punta Laguna is an atypical one also needs to be considered. As mentioned above, the area of remaining medium forest is much smaller than other studies have reported, and the density of spider monkeys within this fragment is higher than reported for any other population (Ramos-Fernandez, unpublished results). Outside this remaining fragment of medium forest, the vegetation consists of 30 to 40-year old successional forest that spider monkeys utilize on around 30% of their foraging trips. This area appears to have been disturbed by a fire that occurred in the dry season following hurricane Behula in 1967, the effects of which may have been particularly strong for large arboreal species

(Lopez-Portillo 1990,

reports such an effect of a fire after hurricane Gilbert in 88


1988). If true, this would imply that the spider monkey population may have decreased abruptly and that, as the successional forest provides increasingly more food, it is currently not limited by resources. In support of this interpretation, the per capita birth rate found in four years of study (0.48 +/- 0.12 SD births per year, calculated on 4 females with more than one offspring born during the study period) are higher than those reported in any other study to date (Milton, 1981 Ateles geoffroyi in Panama = 0.38 +/- 0.03;

Symington, 1990 in Ateles paniscus

in Peru = 0.35 +/- 0.06).

Finally, it is possible that scramble competition is not an important determinant of subgroup size in spider monkeys. Either subgroup size or traveling distance can be adjusted to a particular abundance of food. If the density of resources increases, subgroups could become larger and travel the same distance or stay the same size and travel less. The crucial variable that mediates between these two is foraging efficiency (i.e. the amount of food consumed per unit of time or distance). For at least part of the time, spider monkeys could simply be increasing their foraging efficiency by eating more food when in large 89


subgroups. Many of this species' foods occur in slowly depletable patches. This is suggested by the fact that spider monkeys may return to feed on the same patch on several days (see also Chapman 1989) . Other species in their diet are found in patches that are quickly depletable (Symington 1988/ Chapman 1988) and it may be only during those times that scramble competition is important.

Still less clear is the influence that scramble competition could have on the establishment and maintenance of social relationships. Wrangham (2000) predicted that males in fission-fusion societies would be faster than females and that, because of suffering less scramble competition, they would be able to forage in larger subgroups than females. Although in the present study the difference in speed was as predicted, there was no evidence that males do in fact suffer lower levels of scramble competition.

A similar situation is found in chimpanzees: even though males are faster than females, there is only weak evidence that large subgroups range more than small ones. Wrangham and Smuts (1980) found a 15-28% increase in the travel 90


distance per hour when in parties than alone and Wrangham (2000) found a small but non-significant correlation between the distance traveled per hour for parties of different size. Wrangham (2000) interpreted this as evidence that chimpanzees may sometimes tolerate increased travel costs incurred by the presence of others. Whether this tolerance of travel costs is due to reasons other than food is currently unknown.

In conclusion, this study highlights the shortcomings of both the theory of scramble competition and our understanding of the way in which individual decisions can influence properties of groups as basic as their size and movements. It may be that only a deeper understanding of the decisions made by individual animals will allow a more general explanation of the consequences of their group life.


Ill Wh innies by spider monkeys (A t e l e s geoffroyl) : staying in touch with close associates?


SUMMARY

Results of a three-year long field study of spider monkeys (Ateles geoffroyi) in Punta Laguna, Mexico are used to explore the information content of the species' most frequent vocalization, the whinny. Results are consistent with the most parsimonious interpretation: that whinnies contain information about individual identity and location. This information potentially allows listeners to maintain contact with individuals in their own subgroup or learn about the composition of subgroups nearby. Acoustic analyses confirmed previous findings of consistent acoustic variability among the whinnies of different individuals. Observations on the use of the whinny were made in single subgroups as well as in two subgroups simultaneously. Within a subgroup, whinnies appeared to influence the length of time two associates stayed in the same subgroup. In one of the study groups, dyadic associations appeared to last longer when one of the two individuals had emitted a whinny near the beginning of the association than when neither of them had. Results from the second study group show the opposite: associations lasted longer when none of the dyad's members had emitted a whinny. In betweensubgroup observations, it was found that two different 93


subgroups that were within the active space of the whinny approached each other more often than subgroups that were farther apart. Individual adults in these subgroups also emitted more whinnies when they were within hearing range of another subgroup than when farther apart. A paired playback design was used to determine whether whinnies could influence the behavior of close associates as opposed to non-associated individuals. Although non-associates were as likely as close associates to respond vocally to playbacks of whinnies, only the close associate ever approached the speaker. Together, these results suggest that whinnies could be used by spider monkeys to achieve flexibility in spacing while maintaining specific social relationships.


I N T R O DUCTION There is a general class of animal signals, normallyreferred to as "contact" calls, that appear to play a role in maintaining the spatial coherence of the social unit. This unit can be a pair

(e.g. Butterfield 1970 in zebra

finches Poephila guttata; Smith 1972 in Carolina chickadees Parus carolinensis) or a large group (e.g. Green 1975 in Japanese macaques Macaca fuscata; Robinson 1982 in wedgecapped capuchins Cebus nigrivittatus;

Dittus 1988 in toque

macaques Macaca sinica; Boinski 1991 in squirrel monkeys Saimiri oerstedii; Wilkinson and Boughman 1998 in greater spear-nosed bats Phyllostomus hastatus;

McComb et al. 2000

in African elephants, Loxodonta africana). A common feature of these signals is that they may be heard by some group members that are not in visual contact with the caller. This situation is frequent in large groups that range widely, where individuals in a long progression may be far from others

(e.g. Cheney et al. 1996 in chacma baboons,

Papio cynocephalus) or in fission-fusion societies, where group members may spend several days at long distances from each other (Goodall 1968 and Wrangham 1977 in chimpanzees Pan troglodytes; Smolker et a l . 1992 in bottlenose dolphins Tursiops truncatus; van Roosmalen and Klein 1987 Symington 95


1990, Chapman 1990 in spider monkeys, Ateles spp; Holekamp et al. 1997 in spotted hyenas, Crocuta crocuta).

While these signals seem to serve a general contact function, it is difficult to specify their information content, a necessary step in determining how individuals who use them achieve spatial coordination. Various different messages could produce the same response from recipients. A call that contains information about an imminent danger (e.g. predator alarm calls, Seyfarth et al. 1980)

can produce a decrease in the group's spread, but so

might any other signal that informed others about food or the presence of a neighboring group. Moreover, group members may give calls simply to maintain contact with or a certain distance from others, in which case the calls might elicit no obvious responses in listeners

(e.g. dolphin

signature whistles, Smolker et a l . 1993; baboon barks, Cheney et al. 1996; elephant contact calls, Poole et a l . 1988). A final difficulty in studying the information content of these calls is the fact that each of several recipients hearing the call may respond differently, depending on his/her particular social relationship with the caller. 96


Spider monkeys live in fission fusion societies, in groups of 15 to 40 members that occupy an area of 30 to 200 hectares

(van Roosmalen and Klein, 1987; Symington 1987;

Chapter 2). Subgroups of 1 to 20 individuals split and come together several times during the day, resulting in some individuals spending several days at distances of up to 2 km from others in their group. There are distinct patterns of association between different group members:

males tend

to associate more often and for longer periods of time than females, and within each sex there are associations of different strength (Chapter 1; Symington, 1987; Chapman 1990).

The most frequent call of the species is the whinny. Eisenberg (1976, pp. 39) first described it as a "position indicator" that "appears to involve contact maintenance and can involve group movements." Whinnies are given in widely different social contexts and callers can be resting, feeding or travelling at the time of calling. Other group members, in the same or in a different subgroup, sometimes respond with another whinny within seconds of hearing the first. Chapman and Weary (1990) found that the acoustic


structure of this vocalization varied consistently among individuals, so that it could potentially convey information on the identity of the caller. They also found that the acoustic structure of whinnies from some individuals tended to resemble that of their parents' whinnies.

Teixidor and Byrne

(1997, 1999) performed two sets of

playback experiments in Santa Rosa, Costa Rica, to test two hypotheses about the information content of the whinny. The first was that whinnies conveyed information on whether the caller was from the same or another group, so that upon hearing whinnies from strange conspecifics, spider monkeys would respond more aggressively than when hearing whinnies from group members. The results were not conclusive, as the subjects responded vocally and by approaching the speaker in both cases. The second set of experiments was designed to test the idea that whinnies conveyed information about the activity of the caller. Subjects were played back whinnies that were originally emitted in a feeding or a traveling context. Again, the response of the subjects to each whinny class was not significantly different, as they

98


responded vocally and by approaching the speaker in both conditions

(Teixidor and Byrne 1999).

Individual identity appears to be the most consistent source of acoustic variation among whinnies

(Chapman and

Weary 1990; Teixidor and Byrne 1999). The inconsistent results in the second playback experiment cited above could be the result of recipients responding selectively to different individuals with whom they hold particular social relationships, without regard to the original context in which the whinnies are emitted. In spider monkeys, social relationships among different group members include active companionship as well as avoidance (Chapter 1). These relationships play a large role in determining subgroup size and composition. By carrying information on individual identity, whinny exchanges could be used by individuals who are traveling in the same subgroup to remain together in an environment where visibility is impeded by dense vegetation. Similarly, for individuals who are not part of the same subgroup, the whinny could inform listeners about the presence and composition of another subgroup as the two come near each other. Through the vocal replies elicited from others, the original caller could obtain information 99


on the composition of other subgroups. Individuals in a spider monkey subgroup might then use whinnies as a means to decide whether they stay where they are or move elsewhere, depending on the presence of particular others.

This study was designed to test the hypothesis that the whinny serves a contact function by informing close associates about the caller's position and identity. Four different methods were used to test predictions derived from this hypothesis. First, an acoustic analysis of several whinnies was used to confirm previous findings documenting individual variation in the acoustic features of whinnies

(Chapman and Weary 1990; Teixidor and Byrne

1999). Second, within-subgroup observations were used to test the prediction that close associates in the same subgroup stay together for longer periods of time when they are hearing each other's whinnies than when they are not. Third, between-subgroup observations were used to test the prediction that subgroups within the active space of the whinny influence each other's movement patterns through the use of this vocalization. Lastly, playback experiments are used to test the prediction that the listeners' responses

100


to a whinny depend on their social relationship with the caller.

101


METHODS

S tu dy site and animals

Data come from a three years' study of two free-ranging, habituated groups of black-handed spider monkeys

(Ateles

geoffroyi) living around the lake of Punta Laguna, in Yucatan, Mexico (20°38! N, 87°38' W, 14 m altitude). Their habitat consists of one 60-hectare fragment of semi evergreen medium forest, with trees no taller than 25 meters, and a much larger area of forest in different stages of secondary succession, with trees no taller than 15 m. Trails have been cut throughout the 60 hectares of forest that these two communities use regularly, and trees and other landmarks have been used to make accurate maps of this area (see Figure 2-1 in Chapter 2). Visibility conditions for observers on the ground are very good, as monkeys use the canopy at heights from 5 to 25 m.

One group was habituated to human presence before the study began and the other was habituated during 1997. Individuals were identified by facial marks and other unique features. Adults were defined by their darker faces and fully descended testes in the case of males. All 8 adults

(6 females, 2 males) in the MX group (easternmost) 102


could be reliably identified by the end of 1996. All 19 adults

(13 females, 6 males) in the EU group (westernmost)

could be reliably identified by the end of 1997. Between January 1997 and December 1999, data were gathered by myself and two field assistants. Data reported here include two years

(January 1998-December 1999, 712 total hours) for

the EU group and three years (January 1997-December 1999, 1184 total hours) for the MX group.

D e f i n i t i o n of s u bg ro up

Data for the within- and between-subgroup observations were derived from 20-minute scan samples obtained from one or two subgroups simultaneously. For each sample, the identity and location of all independently locomoting individuals was noted, as well as their activity and position with respect to landmark trees or paths. A subgroup was defined using a chain rule of 30m; that is, all individuals within 30m or less of every other were considered as part of the same subgroup and therefore in association for a particular instantaneous sample. The cutoff distance for the chain rule was originally derived by choosing one adult monkey and noting its distance to all other individuals within a 200 m radius. This procedure was repeated 5 times and a cutoff was selected as the shortest 103


distance at which the distribution of the number of individuals with respect to distance showed a steep decline.

D e f i n i t i o n of v o c a l i z a t i o n s

All vocalizations heard were classified by ear into 9 different types. This initial classification was generally consistent across observers, although some calls could fall into two of these categories. To supplement this classification, spectrograms of several exemplars of each type were produced (see below). A whinny was defined as a frequency modulated call, with 2-10 repeating elements and lasting from 0.3 to 1.5 seconds, containing abrupt changes in fundamental frequencies, so that the call had both grunt and whistle qualities simultaneously (see Figure 1). This feature distinguished whinnies from "high whinnies", which did not contain the grunt-like quality and were of lower amplitude than the whinny. This impression was confirmed in our simultaneous observations, when the maximum distance at which the two vocalizations could be heard was determined to be 200 m for the whinny and 100 m for the high whinny (see Amplitude measurements below).

104


D e f i n i t i o n of close associates

For the playback experiments reported here, a dyad was defined as close if its two members had a high association index (see Chapter 1) and if the pair had been observed either in a grooming interaction or a coalition during the course of the study. Non-close associates were defined by having a low association index (see Chapter 1) and by having been observed interacting aggressively at least once during the course of the study.

Part I . A c o u s t i c analyses

Calls were recorded opportunistically using a portable Sony Walkman Professional analog tape recorder and a Sennheiser K3U-ME88 directional microphone. After each call, the identity of the caller, his/her activity, and the date and time of day were recorded on the same tape. All calls were recorded from a distance between 5 and 30 m. Recordings were digitized in the field using a portable PC computer with an ESS ES1688 audio drive, running Wave SE-II (monoaural recording, 16 bit; sampling rate, 44 kHz; 10 kHz low pass and 60 Hz high pass filters).

The amplitude of whinnies was estimated by the following procedure: a whinny was recorded noting the distance to the 105


microphone, the exact location of the caller on the tree, as well as the recording level. Afterwards, from this same distance, a continuous pure tone of 2 kHz was played back at an absolute SPL of 100 dB at 1m (measured with a RadioShack Sound Level Meter 33-2050, weighting set on C) and was recorded with the tape recorder set at the same recording level. Sound intensity of these recordings, measured as peak-to-peak voltage, was measured at the maximum intensity point, in the case of the recorded calls, or at 10 arbitrary points in the case of the pure tone, using an oscillometer-like display in Spectra Plus. The ratio of maximum intensity of calls to average intensity of the pure tone was converted to dB and this was taken as the difference in maximum RMS SPL between the calls and the tone at a reference distance of 1 m. An approximate amplitude of 84 dB at 1 m for natural whinny occurrences was thus arrived at.

All acoustic analyses were conducted by Friedekrike Range, from the Psychology Department, University of Pennsylvania. Recorded whinnies with a high signal/noise ratio were selected for acoustic analysis. This sample consisted of 53 whinnies from 7 adults, 4 adult females 106


from the MX group and two adult females and one male from the EU group. The acoustic parameters for analysis were chosen based on the results of previous studies on the same vocalization (Chapman and Weary 1990; Teixidor and Byrne 1999). The following parameters were analyzed: call duration, (the time in milliseconds from the onset of the call to the end); number of elements, (the number of falling and rising elements, i.e. any continuous frequency band, bordered in time by silence or a completely different frequency band, seen in the spectrogram as arches or inverse arches or broad frequency bands); average length (average duration of all falling and rising elements); maximum frequency (the highest fundamental frequency in any element); minimum frequency (the lowest fundamental frequency in any element); frequency range (the minimum frequency subtracted from the maximum); duration of ending element (the time in milliseconds from onset to ending of the last element of the call, which occasionally showed a steep increase of frequency in the beginning that ended in a rather long, frequency modulated tone; see example in Figure 1).

107


These acoustic parameters were measured using a Sun workstation running WAVES software for signal processing and analysis. All acoustic parameters were measured by the same individual

(F. Range).

A separate Kruskall-Wallis one-way analysis of variance by ranks for each of these 7 parameters was used to test for significant variation between whinnies from different individuals. Once ,the sources of significant variation had been identified, a direct discriminant function analysis was used to test whether variation in these parameters could serve to identify the caller correctly. This analysis identifies boundaries between groups of objects

(in this

case individuals), being defined in terms of those variable characteristics

(acoustic parameters) that distinguish or

discriminate the objects in the respective groups. Because predictor variables such as duration and frequency had different scales of measurement, all variables were transformed to z-scores prior to the analysis. Following Teixidor and Byrne (1999), the standard (direct) procedure was used, in which "all predictors enter the equation at once and each predictor is assigned only the unique association it has with groups. Variance shared among 108


predictors contributes to the total relationship, but not to any one predictor"

(Tabachnick and Fidell 1989). Missing

data (n=l) were replaced with the mean of the individual's other calls (Tabachnick and Fidell 1989).

Statistical analyses were performed with the SPSS (v. 7.5.1.) statistical software package. Results were considered significant when P < 0.05 (a trend was reported for 0.05 < P < 0.1). Part II.

Be h av io ra l obs ervations w i th i n subgroups

In order to explore the relationship between the occurrence of whinnies and the length of the association between closely bonded individuals, dyads with more than 30 and 100 hours of observation in the EU and MX group, respectively, were chosen for analysis. These dyads consisted of 5 male and 11 female dyads in the EU group and 3 male and 8 female dyads in MX, for a total of 27 dyads. Series of two or more consecutive 20-minute scan samples in which these dyads were in the same subgroup were selected from the database of all scan samples.

These association series included both complete and censored series. Complete series started at the time the 109


two individuals were seen in the same subgroup for the first time and ended when one of the two was found in a subgroup without the other. Series that began at dawn

(i.e.

if both individuals were together before 0700 hours) and those that ended at dusk (i.e. if both individuals continued together after 1800 hours) were also considered complete. The mean duration of complete series was 4 8.9 (+/- 35.2 SD, N = 85) and 64.3 (+/- 47.6 SD, N = 198) minutes in the EU and MX groups, respectively. The analysis also considered censored series, which ended when the subgroup containing the two individuals was lost or the observers left the subgroup. The mean duration of censored series was 67.5 (+/- 57.1 SD, N = 66) and 84.1

(+/- 65.5

SD, N = 198) minutes in the EU and MX groups, respectively.

These association series were classified into two categories: a) those series in which at least one of the dyad's members had emitted at least one whinny during the first 20 minutes

(whinny series) and b) those series in

which neither of the dyad's members had emitted any whinny in the beginning of the association (no-whinny series) .

110


These two categories of association series were compared in different ways. First, all whinny series were compared to all no-whinny series, without regard to the specific dyad, using a separate survival analysis for each study group (see below). Second, whinny series and no-whinny series were analyzed as matched pairs with respect to the specific dyad involved using a Wilcoxon signed-ranks test. Finally, whinny series were compared to no-whinny series within dyads, using a separate survival analysis for each dyad.

A Kaplan-Meier product-limit survival analysis 1981)

(Miller,

was performed to estimate the mean time that these

associations lasted. This analysis is used commonly in life history studies to estimate survival probabilities given a set of data that may or may not consist of complete observations until death; that is, it allows for censored data to be included in the analysis. The outputs are product-limit survival estimates for each data category that can be compared using non-parametric tests. In this case, these estimates were the fraction of all association series that remained after different time lags and their mean length, for two categories: whinny and no-whinny series. A log-rank chi-square non-parametric approximation 111


was used to compare the mean length of series from both categories. Pa rt III.

B e h a vi or al o bs e r v a t i o n s b e t w e e n s u bgr oup s

During 1998 and 1999, three observers occasionally followed two different subgroups from the same group at the same time. A total of 285 and 128 hours of observation for the EU and MX groups, respectively, was completed in this manner. An instantaneous scan sample was taken every 20 minutes, at the same time in both subgroups, noting the location of the subgroup with respect to identified landmarks, as well as the identity and activity of all independently locomoting individuals. In addition, all instances of vocalizations were registered, noting the calling individual and his/her activity. Observers remained in contact through the use of radios.

The maximum distance at which a whinny given in one subgroup could be heard by the observer following the other subgroup was 200 m. A conservative estimate of the maximum distance at which monkeys themselves could hear each other's whinnies was therefore chosen to be 300 m. It was assumed that those subgroups at larger distances than this could not hear each other's whinnies and therefore 112


constituted a null set of data for the hypothesis that the whinny influenced the relative movement of subgroups. Simultaneous samples were then divided into those in which subgroups were near (30 to 100 m), at intermediate distance (100-300 m) and far (more than 300 m ) .

In order to determine whether subgroups could be influencing each other's movements, the changes in distance between the two were analyzed with respect to their initial distance. Simultaneous samples were divided into those that, compared to the previous sample 20 minutes before, constituted either 1) an approach; 2) no change in distance from the previous sample, or 3) a retreat. Approaches and retreats were defined as changes in distance longer than 30 m.

In order to determine whether whinnies could influence the relative movements of subgroups, the occurrence of whinnies by any member of the two subgroups was analyzed during the 20 minutes prior to the sample where the relative distance between the two subgroups was established.

113


Part IV.

Pla yb ac k exp er ime nts

Playback experiments were performed on both groups during the second half of 1999. Stimuli were whinnies from 12 individuals

(1-2 call exemplars per individual)

that had

been recorded as described above (see Acoustic analysis) and transferred to compact discs.

In these playback experiments, two individuals who were currently in the same subgroup and separated by 5 to 50 m, were played back the whinny of a third individual who was currently in another subgroup. One of the subjects was a close associate of the individual whose call was to be played back and the other was not. The absence of the caller was verified by continuing the scan samples on the same subgroup for at least one hour after the experiment.

Experiments were carried out during the morning observation period (0700 - 1100 hours) or the afternoon (1500-1900 hours). As soon as the subgroup's composition was noted, a Nagra DSM speaker-amplifier was lifted into a tree at a minimum height of 5 m (average 5.5) and at a horizontal distance of 50 m from both subjects

(range, 40-

60m; estimated visually), pointed in a direction between the two. Two observers began a focal observation on each of 114


the two subjects. After 5 minutes, one whinny was played back at an absolute sound pressure level of 84 dB at lm (see Acoustic analysis above). Focal observations continued for 10 more minutes. Observers noted any vocalization emitted by the subject and all changes in his/her position and activity. In addition, all instances of vocalizations by others in the subgroup were registered. No individual was tested more than once on a given day.

A trial was suspended or discarded if: either of the two subjects moved more than 10 m during the focal observation so that the distance from each subject to the speaker was not the same; any individual in the subgroup emitted a whinny during the 5 minutes previous to the playback; or the caller joined the subgroup within 60 minutes following the playback.

115


RESULTS

Pa rt I. A c o u s t i c analyses

Previous studies had shown that whinnies contained consistent acoustic variation across individuals

(Chapman

and Weary 1990; Teixidor and Byrne 1999). To confirm these results for the groups and vocalizations under study, recordings of whinnies from 6 adult females and one adult male were analyzed by Friederike Range in two different ways. First, separate analyses of variance were conducted on seven acoustic parameters from each caller's whinnies. Second, a discriminant function analysis was used to determine whether the variation in these parameters was sufficient to distinguish the whinnies from each individual. Three whinny exemplars from two adult females are shown in Figure 3-1. Figure 3-1 (opposite page): spectrograms of whinnies from two different individuals

116


Call duration I wl7 i

liSTT

'T

10000

5000

[ *\ -

'■%.

i

« 0

J..M

l..|,ct.^,liJkl^LUb.UJJ

b.

l-L 1— 1 -I.L.L

UULI

hl-fUM-iLU. U

Ending element

10000

w* Iw.f

5000 wwwv«t*£

Ill


Figure 3-1: Spectrograms of three whinny exemplars. Spectrogram (a) is from adult female BI, (b-c) are from adult female CL. Horizontal axis in seconds, vertical in hertz. Shown are also definitions of the different elements used for the acoustic analysis: RA, regular arch; IVA, inverse regular arch. Note different time scale in in spectrogram (a) .

Kruskal-Wallis one-way analyses of variance were conducted for each of the seven acoustic features listed in Table 1. Of the seven acoustic parameters considered, 5 showed statistically significant variation between individuals. The number of elements per call and the average length of elements were not significantly different among individuals, although the latter fell just short of statistical significance (H = 11.24, df = 6, P < 0.08; Table 3-1).

118


Table 3-1: results of the analyses of variance on different acoustic parameters of the whinny Acoustic parameter

H (df=6)

P

Call duration

13.844

0.5; Figure 35) . Figure 3-4: association length for each pair, M X group

200

120

□ no W in 20'

EjW in 20'

Pair

Figure 4 legend: means (+/- SE) from the product-limit analysis of the association length of different pairs in the MX group when at least one of them has emitted a whinny in the first 20 minutes and when none have emitted any whinny during the same time. The 5 rightmost group of bars come from male-male pairs, including two pairs sampled at different times of the study (DA-BE and PA-BE, whose relationship changed due to BE's development into adulthood during 1998). All others are female pairs. No bar in the whinny category implies there are no associations included in the sample; no error bar implies only one association was included. (***) = The female 126


pairs CH-CL and FL-VE show a significant difference in the within-pair comparison of association length (see text). Figure 3-5: association length for each p a i r , EU group

140 120 I

100

sz

o 0) c

c o

4->

□ no W in 20' □ W in 20'

60 .

CO

‘O o < /) >
300

Distance range (m)

Figure 3-la legend: average rate of whinny emission for adults in the EU group. Each line represents one individual's whinny rate when another subgroup was at each of the three distance ranges. Solid lines represent adult females, dotted lines adult males.

132


Figure 3-7b: individual whinny rate and distance to another subgroup, MX group

7

6

5

4

3

2

1

0 30-100

100-300

>300

Distance range (m)

Figure 3-7b legend: average rate of whinny emission for adults in the the MX group. Each line represents one individual's whinny rate when another subgroup was at each of the three distance ranges. Solid lines represent adult females, dotted lines adult males.

These results suggest that two subgroups that are within hearing range of each other can potentially influence each other's movement patterns. Individuals in these subgroups also emit more whinnies than when their subgroups are far apart. The possible causal relation between the whinny and

133


these changes in distance was tested through the use of playback experiments.

Part IV.

P l a yb ac k Ex pe ri me nt s

The observations above suggested that whinnies could affect the behavior of individuals within the same subgroup and in other, nearby subgroups. If whinnies functioned to maintain contact between individuals in the same or different subgroups, close associates of the signaler might be expected to respond more strongly to a whinny than an individual who was less closely associated. It was predicted that, upon hearing a whinny from an individual that was not present in the same subgroup, the responses of a close associate of the caller and another, non-closely associated individual would be qualitatively different. Possible responses might include a vocal response or an approach toward the speaker.

A total of 20 paired playback trials was performed in both study groups. All playbacks involved the whinny from a monkey that was not present in the subgroup and two subjects: a close associate of the calling monkey and another, less closely associated individual. Table 3-3 shows the individual that was predicted to respond based on 134


his/her association with the caller and the actual subject who responded. Three different responses to the playback are listed in Table 3-3: whinnies emitted immediately after the playback, approaches to the speaker and additional whinnies during the 10 minutes following the playback.

In 5/20 (25%) trials, the predicted individual answered the playback within 10 seconds of the call; in 4/20 (20%) trials, the other subject answered. Both the predicted and the non-predicted subjects answered in 6/20

(30% trials). A

third, different individual that was not being followed answered in 1/20 (5%) trials. No monkey answered in the remaining 4/20

(20%) trials (Sign test on the null

hypothesis that both subjects were equally likely to answer, one tailed P = 0.5).

135


Table 3-3: results of the playback experiments Subjects

.

Trial No.

Date

Group

4 5 7 9a 10 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

8/14/99 10/11/99 10/20/99 10/27/99 11/1/99 11/9/99 11/11/99 11/16/99 11/18/99 11/19/99 11/22/99 11/23/99 11/29/99 11/29/99 12/3/99 12/8/99 12/8/99 12/9/99 12/14/99 12/14/99

MX EU EU MX MX EU EU EU EU EU EU EU MX EU MX MX EU MX MX MX

Gives whinny in Approaches response? speaker?

Close Other associate CL DN OF DA PA IS EN RO PN RO Rl PN VE EN VE CH GO CH FL CE

CH MO LU PA BE MO BA OF BT AJ OC BT CE OF CH CL FB CE PA CH

Gives additional whinnies?

Caller and Other Close Other Close Close Other Voc No. associate associate associate FI10 Lu1 Mo21 Ve39 Da13 Ba4 Ro18 Pn2 Go20 En4 Mo21 Ro18 CI65 Go19 FI10 Ce16 En4 CI22 CI22 FI10

yes yes

yes yes yes yes yes yes yes yes

yes

yes

yes yes yes yes yes yes yes

yes yes yes yes

yes yes yes

yes yes

yes yes yes

yes

Table 1 legend: results of the playback trials with respect to three responses by the close associate and another, unrelated individual: emitting a whinny immediately in response to the played back whinny, approaching the speaker and emitting any additional whinnies during the 10 minutes following the playback. An empty space in the results columns indicates no response by the subject in the column. In addition to the identity of subjects and callers, the key to the vocalization used in each playback is shown in the middle column. See methods for experimental procedure.

In most trials, neither subject approached the speaker. However, when one did, it was only the close associate who approached the speaker in 7/20 (35%) trials. The probability that only one of the two subjects approached 136


the speaker on seven occasions by chance alone is significantly small (Sign test on the null hypothesis that both subjects were equally likely to approach the speaker, one-tailed P < 0.01). In one of these (trial #16), both the subject (RO, an adult male) and another male who was not the subject in the experiment began traveling in the direction of the speaker after playback of the whinny by RO's closely associated male, PN. The two males went past the speaker and continued moving and vocalizing in the same direction.

Finally, in 4/20 (20%) trials one of the two individuals gave additional whinnies during the 10 minutes following the playback. In this case it was also the close associate and not the other, unrelated subject who showed this response. The probability that it was only one of the two subjects who gave additional whinnies on four occasions by chance alone is not significantly small, although it approaches statistical significance (Sign test on the null hypothesis that both subjects were equally likely to give additional whinnies after playback, one-tailed P = 0.06).

137


DISCUSSION

The results of this study support the hypothesis that the whinny serves a contact function by providing information to listeners about the identity and location of the caller. For monkeys in the same subgroup, the results were inconclusive. While in one of the study groups adult pairs were more likely to stay together when they had heard each other's whinnies than when not, in the other group the opposite was true. The evidence for the between-subgroup contact function of whinnies based on information on identity and location is more clear. First, subgroups within the active space of the whinny approached more often than subgroups that were farther apart. Second, individuals emitted more whinnies when another subgroup was within hearing range than when it was farther apart. Finally, the results of the playback experiments suggest that one important determinant of whether listeners will join a caller from a different subgroup is the social relationship they have with him/her.

Be t w e e n s u b gr ou p c o m m un ic at io n

Playback experiments were designed in such a way that the calling individual whose whinny was to be played back 138


was not present in the subgroup of the two subjects. This situation is analogous to that which spider monkeys would encounter if an individual with whom they have not been associated lately is found nearby. In these cases, listeners have little information about the caller's presence, because they had not been associated with him/her recently. This could explain the different behavior of close associates compared to other, less closely associated subjects in response to the playback: while the former may be motivated to join the caller, the latter may simply announce their own presence with a whinny in response to the newcomer's whinny. In spider monkeys, fusions of two subgroups are normally accompanied by whinny exchanges between individuals in both subgroups and are followed by greeting embraces between closely associated monkeys that had, until recently, been in different subgroups

(Eisenberg

1976; this study, Chapter 1). This suggests that whinnies are a means by which group members can spend time apart from each other and still recognize each other at a distance.

There are interesting parallels between the results of this study and those obtained in studies of another 139


fission-fusion species, the common chimpanzee. Mitani and Nishida (1993) found that a chimpanzee male is more likely to give a pant hoot when his allies or associates are near (defined as those individuals seen within an hour before or after the call) than when they were absent (defined as those individuals not seen during the entire day) or in the same subgroup as the caller. Mitani

(1996) suggested that

in this way chimpanzees recruit the company and support of allies and close associates. The fact that spider monkeys in subgroups that were within the active space of the whinny did in fact vocalize more than subgroups that were farther apart from each other suggests that whinnies could be serving a similar function in a similar social system. The fact that close associates were more likely to approach the speaker in the playback experiments also supports the functional analogy between pant hoots and whinnies. In the case of the spider monkey, however, the vocalization is given equally by males and females and is heard over shorter distances than the chimpanzees' pant hoots.

Based on an analysis of calling subgroups and their subsequent change in size, Chapman and Lefebvre (1990) suggested that whinnies were used to attract other group 140


members to recently discovered fruiting trees, and to adjust the foraging subgroup size to an optimal size given the opposing influences of feeding competition and predation. As a result, the whinny of spider monkeys has been cited as an example of a food call used to recruit others to a fruiting tree (Hauser 1996; Heinrich and Marzluff 1991). However, while it is clear that calling subgroups in Chapman and Lefebvre's

(1990) study were

joined by others more frequently than subgroups that remained silent, only 17% of calling subgroups were actually ever

joined (Chapman and Lefebvre 1990) .

In contrast to Chapman and Lefebvre's

(1990), in the

present study the identity of callers and listeners, as well as their social relationships, were taken into account. Responses to the whinny appear highly variable, depending at least in part on the relationship between callers and listeners. When receiving information about the caller's identity, listeners can choose whether to join a caller or not. As a result, a whinny will sometimes result in a listener joining the caller, but in others it will result in the opposite, the listener retreating from the caller. This would explain the small proportion of calling 141


subgroups actually joined by other monkeys observed by Chapman and Lefebvre

(1990), and it would be compatible

with the positive relationship between calling frequency and subgroup size increases found in the same study.

A similar argument can explain the responses of spider monkeys to the playbacks performed by Teixidor and Byrne (1999). These experiments indicated that scanning behavior was more frequent after a "feeding whinny" than after a "traveling whinny". However, subjects were no more likely to respond vocally or approach the speaker after either of these whinny types. Because their acoustic analysis revealed that individual identity was the main source of information contained in the whinnies, Teixidor and Byrne (1999) interpreted these results as the outcome of a hierarchical perception process where context of emission would be recognized only after individual identity had been taken into account.

Although the context in which the recorded whinnies were emitted was not taken into account, in this study monkeys were more likely to approach the speaker if the playback whinny was from a close associate than from another 142


individual. This is consistent with Teixidor and Byrne's (1990) interpretation of their own results, highlighting the importance of individual identity over the possible additional referents these calls might have.

Within-subgroup communication

A strong test of a within-subgroup contact function for the whinny is more difficult than testing whether it can function to inform monkeys in other subgroups. This is because, within their own subgroup, callers may simply be informing others of something that, in a sense, they already know: that the caller is still present. Under this hypothesis, the expected response of monkeys within the caller's subgroup would be to stay where they are and do little else except for vocalizing in response.

Results from the two groups under study yielded contradictory results. In the larger group (EU), whinnies in the beginning of a pair's association predicted longer association times. This contrasts with the finding in the smaller group (MX), in which a whinny in the beginning of an association predicted shorter association times between pairs. There are important differences between these two groups in their social relationships and use of space that 143


may explain these results. Because there were more female pairs available for the analysis and each female pair had more association series measured, the females' contribution to the observed results is stronger than the males'. Although the number of adults is much smaller in the MX group (6 female and 2 male adults as of 1997, compared to 13 female and 6 male adults in the EU group), subgroups are larger in the MX group. Individual core areas are larger in the EU group but daily ranges are shorter (B. Ayala, in preparation). While in the MX group females do not seem to form close social bonds and in some cases appear to avoid each other, in the EU group some pairs of females do maintain close social relationships

(Chapter 1). It is

possible that females in the MX group use the whinny as a within-subgroup spacing call, while females in the EU group may use it to maintain close contact.

Individuals frequently responded to whinnies from other individuals within their own subgroup. While there is no pattern to the identity of the answering individuals (Ramos-Fernandez, unpublished results), vocally acknowledging receipt of a contact call has been suggested as one of the primary functions of an analogous call in 144


another species with a similar social system: bottlenose dolphins exchange signature whistles during many of their interactions

(rev. in Smolker 2000) . In some of these,

individual dolphins match each other's signature whistles, even when at long distances from each other (Janik 2000) . Whether spider monkeys also address particular others through their calls is currently unknown. However, the results of a set of playback experiments conducted in captivity by Masataka (1986) suggest that they could potentially do so. Both bottlenose dolphins and spider monkeys appear to face similar needs to remain in touch while foraging in loosely aggregated social units that belong to a larger group.

Acoustic

struct ure of whinnies

A word of caution is in order with regard to the comparison of results from different studies of spider monkey's whinnies. Because during the observational phase of this study whinnies were found to grade into "high whinnies" and "R calls", an attempt was made to maintain consistent criteria for classification of whinnies. This was mainly the presence of a "grunt" like quality in the sound at the same time as the frequency modulation occurred at a higher frequency. Other researchers

(Chapman and Weary 145


1990; Teixidor and Byrne 1999) have not provided enough detail on their definition of a whinny to justify a strict comparison of the three studies' results. It is possible that Teixidor and Byrne (1999) included vocalizations in their analysis of the whinny that in this study would fall into the "high whinny" category and would therefore not have been analyzed.

Other vocalizations of the species need to be studied in more detail. Although less frequent than whinnies, R calls and screams travel farther and could play a large role in the interactions between monkeys in different subgroups within the same group, as well as in the relationships with other neighboring groups.

Conclusion

The results of this study suggest that by broadcasting information on their presence to others who may in turn announce theirs, spider monkeys may be able to spend time away from their close associates while remaining in contact through vocal communication. These interactions may be crucial for maintaining social relationships within a fluid social structure.

146


CONCLUDING REMARKS

147


The previous chapters have shed light on the mechanisms underlying the social structure of spider monkeys. They also have highlighted some shortcomings of the theories used to understand the how ecological factors influence grouping patterns.

Chapter 1 demonstrated that subgroups of spider monkeys cannot be characterized simply as the result of random aggregation. Rather, individuals with close social relationships tend to be found together more often than others with looser bonds, and some pairs, especially mixedsex pairs, show signs of avoidance. Close bonds are observed mostly among males, while females generally associate at rates that cannot be distinguished from those expected by the random model. High association rates are observed between individuals that also groom and embrace each other often. The absence of affiliative social interactions between many female pairs suggests that they could simply be converging at feeding spots, without any preference for particular social companions.

Chapter 2 found no evidence to support the idea that the variability in association patterns enables spider monkeys 148


to track the current availability of resources. None of the results agreed with the predictions of the hypothesis that feeding competition in the form of increased travel distance for larger subgroups is a limit on subgroup size. Subgroups did not increase in size during seasons of high food abundance and individuals in large subgroups did not travel farther than those in smaller ones. It may be that the composition, and not the size of subgroups, is a more important factor affecting whether a monkey joins a subgroup or whether monkeys in a subgroup separate from each other. Scramble competition may still occur, not within subgroups, but between all individuals living in a common area.

The results from chapter 3 complemented those from chapter 1 by highlighting the importance of social relationships in determining the composition of subgroups. This chapter also suggested that vocal communication plays an important role in modifying the composition of subgroups. Monkeys in different subgroups appeared to coordinate their travel distance, but only when they were within hearing range. Individuals in these subgroups also vocalized more than when another subgroup was far apart. 149


The results of the playback experiments showed that close associates are more likely to approach a previously absent caller than other, less closely associated individuals. Together, the results from this chapter suggest that spider monkeys could be using vocal communication to maintain and establish contact with their close associates. This contact mechanism would then have a direct influence on the relative movements, size and composition of different subgroups.

Suggestions

for fu rt h e r studies

The results from chapter 3 left many questions unanswered: for example, what is the composition of the subgroups that approach each other, compared to that of subgroups that do not? Do the subgroups that approach each other contain closely associated individuals? Do subgroups that do not contain closely bonded individuals avoid each other? This finding would provide strong support for the hypothesis that monkeys can know the composition of nearby subgroups through the use of vocalizations and that they use this information to decide whether to join them or maintain themselves at a distance from them.

150


It seems clear from chapter 1 that male spider monkeys have close bonds with each other and females do not. But if the primary function of whinnies is to maintain contact with close associates, why do females vocalize at all? The rates of vocalization by some adult females are two or three times higher than those of males

(Ramos-Fernandez,

unpublished results). It is possible that females use whinnies to maintain themselves apart from each other, thereby avoiding the potential conflict arising from feeding with many individuals. The results of the withinsubgroup analysis in chapter 3 support this claim, for only one of the study groups: pairs of females were more likely to stay in the same subgroup when neither of the two had given a whinny during the beginning of their association than when they had. However, this hypothesis assumes that competition for food occurs within subgroups, which the results from chapter 2 do not support.

C o n s e r v a t i o n issues

It seems necessary that the results of a study on an endangered species

(Rylands et al. 1995) be applicable to

the successful conservation of both the species and its habitat. Even when this was not the original purpose of this study, some of the results reported here, as well as 151


those of other studies that I participated in during my fieldwork, have already been used for this purpose.

The results discussed below constitute the basic information used for the design of a protected area, the Otoch Ma'ax Yetel Koh sanctuary for spider monkeys

("House

of the Monkey and the Puma" in Yucatec Maya) This sanctuary is managed by the local communities, Yucatec Mayas with a minimum income and formal education level, who nevertheless initiated the conservation efforts that almost led to the declaration of the area as federally protected by the end of 2000.

The most important result of this study for conservation purposes is that spider monkeys in Punta Laguna utilize the secondary forest. By foraging in this type of forest at least on 30% of their daily foraging trips, monkeys in the area could be obtaining additional food that the 60 remaining hectares of undisturbed forest where they live cannot provide them. As discussed in chapter 2, this secondary forest may be the result of a fire that occurred more than 30 years ago. It is possible that as the secondary forest regenerates, it provides increasingly more 152


fruit that spider monkeys can feed upon and, as a result, the population might be in expansion. High per capita birth rates and the apparent absence of feeding competition within subgroups support this conclusion.

It is helpful to compare this situation to what occurs in other areas of the tropics where the forest is converted to cattle ranches (Estrada and Cortes-Estrada 1988). In these areas, small fragments of old-growth forest remain isolated within large extensions of cleared land that cannot be utilized by spider monkeys. In the Yucatan peninsula, cattle ranching is not a profitable activity because grass cannot be grown easily in the limestone floor (Challenger 1998). Therefore, human disturbances to forested areas consist mainly of mosaics of forest of different ages produced by slash-and-burn agriculture. Local people that practice this form of agriculture set aside areas of forest in different stages of secondary succession so the soil can regain its nutrient content. Among these, it is the oldest of these areas that spider monkeys use in Punta Laguna.

153


The main application of this study for the successful preservation of the population of spider monkeys has been the demarcation of an area of approximately 5000 hectares as the necessary habitat for the population to survive. Within this area, there are two lakes surrounded by approximately 1000 hectares of undisturbed medium forest, while the rest is mainly 30-50 year old secondary forest.

The local communities who own the land in the area of Punta Laguna are motivated to preserve the spider monkeys, due principally to the affluence of tourists from nearby Cancun and Playa del Carmen, who visit the area to observe the spider monkeys in their natural habitat. They have agreed on setting aside the undisturbed forest and to maintain a corridor of secondary forest between the two lakes. The rest of the protected area will be used for slash-and-burn agriculture and apiculture. Depending on the number of people that practice agriculture, these activities may prove to be compatible with the conservation of the spider monkey population, provided that enough regenerating secondary forest is left for spider monkeys to feed upon. The number of people living in rural areas has been declining steadily in the Yucatan peninsula for the 154


past 20 years, due mainly to the affluence of people into fast-growing cities like Cancun (http://www.inegi.gob.mx). This is certainly a sign of hope for the spider monkeys in Punta Laguna.

The results discussed above are currently being used as the guidelines for a study with a wider scope. A regional census on the primates of the Yucatan peninsula began in June, 1999 (Serio-Silva and Ramos-Fernandez, unpublished). The main goal of this study is to identify those populations that, based on their size and the condition of their habitat, are

still likely to survive if their

habitat is properly managed. The criteria of minimum area of undisturbed forest and minimum population size for a population to be included in this recommendation were based on the results from the population in Punta Laguna.

155


A P P E N D I X ONE:

LIST OF ALL SPIDER MO NK EYS STUDY GROUPS

IN THE

Below is a list of the study subjects in both groups as of December 1999, together with their sex, age class for the majority of the study period and additional observations like date of birth, for the case of young infants, or date or emigration or death, for the case of older monkeys. They are grouped by kin relationships, some of which are inferred (such as those between subadults and their presumed mothers) and others were observed (such as all juveniles and infants, whose mother is the first adult female above them).

MX group

Name_________ Sex

Age Class

Key______ Observations

Bigotona Benito Pepen

F M M

adult juvenile infant

BI BE PE

Cecilia Sofia Celia Pilar

F F F F

adult subadult juvenile infant

CE SO Cl PI

China New female Susana Alicia Jose

F F F F M

adult infant subadult juvenile infant

CH SU AL JO

dissapeared 8/99 born 7/96 disappeared 3/98 born 3/98 born 12/98 disappeared 4/97 disappeared 1/99 156


MX group (cont.) Name

Sex

Age Class

Key

Observations

Claudia Karina Mica Lamat

F F F F


CL MS MI LT

born 4/14-17/98

Flor Concha Licho

F F M


FL CO LI

born 4/97

Veronica Damian Archi Kaban

F M M F


VE DA AR KB

born 4/98

Pancho

M

adult

PA

New

F

subadult

Sex

Age Class

Key

Barbara SinCola New male New

F F M F

adult subadult infant infant

BA SC Camilo

Beti Licha New female Ocelote

F F F F

adult juvenile infant adult

BT LA OC

Lucimira

F

infant

LM

Cubana Fidel

F M

adult juvenile

CU FI

Daniela

F

subadult

DN

Isabel Chichan Ana Margarita

F F F F


IS CN AN MG

Attempted to join group 8/98

EU group

Name

Observations

born 12/97 born 5/99, died 11/99

born presumed of born

12/98 daughter BT 1/98

157


EU group (cont.) Name

Sex

Age Class

Key

Observations

Jessee Adam Zane

F M M


JE AD ZA

born 2/98

Lupe Bronco

F M

adult juvenile

LU BR

Maribel New male

F M

adult infant

MB

Morticia Mirta Chupacabras

F F M


MO MT CC

Ofelia New

F

adult infant

OF

Rigoberta New female

F F

adult infant

RI

Wendy Peter

F F

adult juvenile

WE PT

Yanis Silvia Calavera New

F F F


YA SI CA

Julieta Coqueta Claudio

F F M


JU CQ CD

Roberto Gonzalo Alejandro Enrique Floyd Fabio Pan

M M M M M M M

adult adult adult adult adult adult adult

RO GO AJ EN FY FB PN

born 11/98 born 1/98

born 1/99

disappeared 4/99

158


A P P E N D I X TWO: LIST OF TREE SPECIES FOUND IN V E G E T A T I O N ANALYSIS Below is a list of the trees with a diameter at breast height larger than 10 cm found in the vegetation census (see Chapter 2). Included in this census were the two main vegetation types that spider monkeys utilized: undisturbed medium forest and 30-50 year old successional forest. Species are grouped by family, shown with both Maya and scientific names. Also indicated is whether the species was only found in the medium forest ("m"; 32 species), only in the successional forest ("s"; 23 species) or in both

("ms";

35 species). The last column shows whether the species is endemic to the Yucatan peninsula ("E"). "NI" indicates 8 species that were not identified with scientific name and one that was not identified with Maya nor scientific name.

The scientific name and family classification comes from Flores and Carvajal

(1994) and Sosa et a l . (1985).

159


List of tree species found in vegetation analysis Family

Maya name

Scientific name

Veg.

End.

Anacardiaceae c h i 'abal k'ulirn c h e ' chechem k'inil ju 1jub

Spondias purpurea L. Astronium graveolens Jacq. Metopium brownei (Jacq.) Urb. Spondias lutea L. Spondias mombin L.

m ms ms ms ms

anona guanabana saramuyo pox e 'lemuy

Annona Annona Annona Annona Malmea Fr.

m m m m m

sakchechem N i k t e 'ch'oom

Cameraria latifolia L. Plumeria obtusa L. var Sericifolia (C. Wright ex Griseb. ) Woodson

m s

pochote

Ceiba aesculifolia (Kunth.) Britten & E.G. Baker Ceiba pentandra (L..) Gaertn.

s

Annonaceae cherimola Mill. muricata L. squamosa L. reticulata L. depressa (Baill.)

R.E.

Apocyanaceae

Bombacaceae

ceiba

m

Boraginaceae bohom bek

Cordia ailiodora (Ruiz & Pav.) Oken Ehretia tinifolia L.

s m

Burseraceae chakah pom

Bursera simaruba (L.) Sarg. Protium copal (Shldtl. & Cham.) E n g l .

ms ms

kitamche1 Top lajun

Caesalpinia gaumeri Greenm. Caesalpinia yucatanensis Greenm. Caesalpinia mollis (Kunth.) Spreng.

ms ms

guarumbo

Cecropia peltata L.

ms

c h 'ooy

Cochlospermum vitifolium Willd. ex Spreng.

s

almendro

Terminalia catappa

m

silil

Dyospyros cuneata Standi.

Caeasalpinaceae

c hakte'

m

Cecropiacea Cochlospermaceae

Combretaceae L.

Ebenaceae s

160


List of tree species Family

(cont.)

Maya name

Scientific name

Veg. End.

Euphorbiaceae taanche1 e k 1ulub sakits chakche'

Croton fragilis H. B. et S Drypetes lateriflora (Sw.) Krug & Urb Euphorbia gaumerii millsp. Manihot spp.

ms ms

Dalbergia glabra (Mill.) Standi. Lonchocarpus rugosus Benth. Lonchocarpus violaceus (Jacq) DC. Lonchocarpus yucatanensis Pittier. Lonchocarpus parviflorus Benth. Piscidia piscipula (L.) Sarg. Swartzia cubensis (Britton & Wilson) Standi, var cubensis

m

ms ms

Fabaceae sitsmuk k 'anasin ba a l c h e ' x u 'u l

ya'ax xu'ul ha'bin katalox

s ms ms s ms m

Flacourtiaceae ximche tamay

Caesaria nitida Jacq. Zuelania guidonia (Sw.) Britton & Millsp.

m ms

tats 1i 1

Hippocratea celastroides H.B. et S .

ms

h o o c h 'o c h e '

Nectandra aff. sanguinea (Rottb.) Roland

ms

laal muuch

Gronovia scandens L.

m

h o 'ol

Hampea integerrima Slecht.

s

cedro

Cedrela mexicana M.

Hippocrateaceae

Lauraceae

Loasaceae Malvaceae Meliaceae ms

161



(cont.)

Maya name

Scientific name

Veg.

End.

Mimosaceae kaatsim subin maybaca ya'ax - e k 1 pich tsalam chukum chi'may

Acacia gaumeri S.F. Blake Acacia globulifera Saff. Bauhinia divaricata L. Chloroleucon mangense (Jacq) Enterolobium cyclocarpum (Jacq) Griseb. Lysiloma bahamensis Benth. Pithecellobium dulce (Roxb.) Benth. Pithecellobium albicans (Kunth) Benth.

m s

m •s ms ms s

Moraceae ramon koopo' alamo

Brosimum alicastrum Sw. Ficus cotinifolia Kunth. Ficus spp.

sak loob

Eugenia mayana Standi.

x a 1an

Sabal yapa Wright ex Beccari

ms ms

m

Myrtaceae Palmae ms

Polygonaceae xtohyub

Coccoloba acapulcensis Standi. Coccoloba barbadensis Jacq. boob t s 'i t s 'i lche' Gymnopodium floribundum Rolfe

m

pimienta c h e 1 Colubrina greggii S. Watson var yucatanensis M.C.Johnston Ziziphus jujuba (L.) Lam. ciruela

m

Rhamnaceae

Rubiaceae t a s t a 1ab

Guettarda combsii Urb.

ms

naranja xikche1

Citrus sinensis (L.) Osbeck. Zanthoxylum fagara (L.) Sarg.

m m

Rutaceae

Sapindaceae wayum koox

Exothea diphylla (Standi.) Lundell waaya Talisia olivaeformis (Kunth. Radik. k'aan chunuub Thouinia paucidentata Radik.

m ms

162



(cont.)

Maya name

Scientific name

Veg.

End.

Sapotaceae chikeeh sapotillo zapote caracolillo k 'a n a s t e '

Chrysophylum mexicanum Brandegee. Manilkara spp. Manilkara zapota (L.) van Royen Sideroxylon Capiri (M. DC.) Pittier Sideroxylon foetidissimum Jacq. subsp. gaumeri (Pittier) T.D. Penn

ms m ms m ms

Simaroubace beel siinik che' p a 1s akche'

Alvaradoa amorphoides Liebm.

pixoy

Guazuma ulmifolia Lam.

k 'askat holol

Luehea speciosa Willd. Trichospermum mexicanum Baill.

y a 'axnik t a t a k 'c h e '

Vitex gaumeri Greenm. Citharexylum schottii Greenm.

Simarouba glauca DC.

Sterculiaceae ms

Tiliaceae (DC.)

s ms

Verbenaceae ms

m

NI aceite c h e ' botox morax sakwisiche tanakash t z a y a 'ac toon tzimin wayum'che' spp with no maya name

m ms m

m m s

m s m

163


BIBLIOGRAPHY

164


INTRODUCTION

Clutton-Brock, T.H., Guiness F.E., and Albon, S.D. 1982. Red deer: behavioral ecology of two sexes. University of Chicago Press. Goodall J. 1968. The behaviour of free-living chimpanzees in the Gombe stream Reserve. Animal Behaviour Monographs 1, 165-311. Janik V.M. 2000. Whistle matching in wild bottlenose dolphins truncatus) . Science 289, 1355-1357.

(Tursiops

Kruuk H. 1972. The spotted hyena. A study of predation and social behavior. University of Chicago Press. Kummer H. 1968. Social organization of hamadryas baboons. Chicago Press.

University of

Kummer H. 1995. In quest of the sacred baboon: a scientist's journey. Princeton University Press. Mann J., Connor R.C., Tyack P.L. and Whitehead H. (Eds.) Cetacean societies: field studies of dolphins and whales. University of Chicago Press. McComb K . , Moss C., Sayialel S. and Baker L. 2000. Unusually extensive networks of vocal recognition in African elephants. Animal Behaviour 59, 1103-1109. Mitani JC and Nishida T. 1993. Contexts and social correlates of long distance calling by male chimpanzees. Animal Behaviour 45:735-746. Nishida T. and Hiraiwa-Hasegawa M. 1987. Chimpanzees and b o n o b o s : cooperative relationships among males. In: Smuts B.B., Cheney D.L., Seyfarth R.M., Wrangham R.W. and Struhsaker T.T.(Eds). Primate societies . University of Chicago Press Poole J.H., Payne K., Langbauer W.R.Jr. and Moss C. 1988. The social contexts of some very low frequency calls of African elephants. Behavioral Ecology and Sociobiology 22, 385-92. van Roosmalen M.G.M. and Klein, L.L. 1987. The spider monkeys, Genus Ateles. In: Ecology and Behavior of Neotropical Primates . RA Mittermeier and AB Rylands (eds). World Wide Fund, Washington. Shaw E. 1962. The schooling of fishes. 138.

Scientific American 206,

128-

165


I NT RO DU C TI ON

(CONT.)

Smolker R, Richards AF., Connor RC and Pepper JW. 1992. Sex differences in patterns of association among Indian Ocean bottlenose dolphins. Behaviour 123, 38-29. Smuts B.B., Cheney D.L., Seyfarth R.M., Wrangham R.W. and Struhsaker T.T. 1987. Primate societies. University of Chicago Press. Stammbach E. 1987. Desert, forest and montane baboons: multilevel societies. In: Smuts B.B., Cheney D.L., Seyfarth R.M., Wrangham R.W. and Struhsaker T.T.(Eds). Primate societies. University of Chicago P ress. Symington, M.M. 1990. Fission-fusion social organization in Ateles and Pan. International Journal of Primatolology, 11, 47-61. Wilson E.O.

1971.

The insect societies. Harvard University Press.

Wrangham R.W. 1987. Evolution of social structure. In: Smuts B.B., Cheney D.L., Seyfarth R.M., Wrangham R.W. and Struhsaker T.T.(Eds). Primate societies . University of Chicago Press

166


C HA PTER 1 Bedjer L., Fletcher D. and Brager S. 1998. A method for testing association patterns of social animals. Animal Behaviour, 56, 725.

719-

Clutton-Brock T.H., Guiness F.E., and Albon S.D. 1982. Red deer: behavioral ecology of two sexes. University of Chicago Press. Chapman C.A. 1990. Association patterns of spider monkeys: the influence of ecology and sex on social organization. Behavioural Ecology and Sociobiology, 26:409-414. Connor R.C., Wells R.S., Mann J. and Read A.J. 2000. The bottlenose dolphin: social relationships in a fission-fusion society. In: Cetacean societies: field studies of dolphins and whales. Mann J., Connor R.C., Tyack P.L. and Whitehead H. (Eds.) University of Chicago Press. Goodall J. 1986. The chimpanzees of Gombe: patterns of behavior. Harvard University Press. Jaccard P. 1908. Novelles recherches sur la distribution florale. Soc. Vaud. Sci. Nat., 44, 223-270. Kummer H. 1968. Social organization of hamadryas baboons. Chicago Press.

Bull.

University of

Symington M.M. 1987. Ecological correlates of party size in the black spider monkey, Ateles paniscus chamek. Ph.D. thesis, Princeton University. Symington M.M. 1988. Food competition and foraging party size in the black spider monkey (Ateles paniscus chamek ). Behaviour, 105, 177134 . Symington M.M. 1990. Fission-fusion social organization in Ateles and Pan. International Journal of Primatolology, 11, 47-61. Whitehead H. 1999. Testing association patterns of social animals. Animal Behaviour, 51, 26-29. Wrangham R.W. 1977. On the evolution of ape social systems. Science Information, 18, 334-368.

Social

167


C HAP TE R 2 Alexander R.D. 1974. The evolution of social behavior. Annual Review of Ecology and Systematics 5, 325-383. Boinski S. and Gerber PA (Eds). 2000. On the Move: travel in groups. University of Chicago Press.

how and why animals

Bradbury J.W. and Vehrencamp S.L. 1976. Social organization and foraging in emallonurid bats. II. A model for the determination of group size. Behavioural Ecology and Sociobiology 1, 383-404. Chapman C.A., White F.J. and Wrangham R.W. 1993. Defining subgroup size in fission-fusion societies. Folia Primatologica 61, 31-34 Chapman C.A. 1988. Patch use and patch depletion by the spider and howling monkeys of Santa Rosa National Park, Costa Rica. Behaviour 105, 88-116. Chapman C.A. 1989. Spider monkey sleeping sites: Implications for primate group structure. American Journal of Primatology 18, 53-60. Chapman C.A. 1990. Association patterns of spider monkeys: the influence of ecology and sex on social organization. Behavioural Ecology and Sociobiology 26, 409-414. Chapman C.A and Chapman L.J. 2000. Determinants of group size in primates: the importance of travel costs. In: S Boinski and PA Gerber (Eds). On the Move: how and why animals travel in groups. University of Chicago P r ess. Flores J.S and Carvajal I.E. 1994. Tipos de vegetacion en le peninsula de Yucatan. Fasciculo 3 de: Etnoflora Yucatanense. Universidad Autonoma de Yucatan. Gomez-Pompa A. and Kauz A. 1987. The conservation of resources by traditional cultures in the tropics. World Wilderness Congress, Estes Park, Colorado. September, 1987. Higgins K.F., Oldemeyer J.L., Jenkins K.J., Clambey G.K. and Harlow R.F. 1994. Vegetation sampling and measurement. In: Research and Management techniques for wildlife and habitats . T.A. Bookhout (Ed). Bethseda, MA: The Wildlife Society. Isbell L.A. 1991. Contest and scramble competition: patterns of female aggression and ranging behavior among primates. Behavioural Ecology and Sociobiology 2, 143-154. Isbell L.A. 1998. Rank differences in ecological behavior: a comparative study of patas monkeys (Erythrocebus p a t a s ) and vervets (Cercopithecus aethiops) . International Journal of Primatology 20, 257-272.

168


C H AP TE R 2

(CONT.)

Janson C.H. and Goldsmith M. 1995 Predicting group size in primates: foraging costs and predation risks. Behavioral Ecology 6, 326-336. Janson C.H. 2000. Primate Socio-Ecology: Evolutionary Antropology 9, 73-86.

The End of a Golden Age.

Klein L.L. and Klein D.J. 1977. Feeding behavior of the Colombian spider monkey, pp 153-181 in: Clutton-Brock, T.H. (Ed.) Primate Ecology. Academic Press, London. Lomnicki A. 1988. Population ecology of individuals . Princeton University Press. Lopez-Portillo J., Keyes M.R, Gonzalez A, Cabrera E.C and Sanchez 0. 1990. Los incendios de Quintana Roo: catastrofe ecologica o evento periodico? Ciencia y Desarrollo , 16, 43-54. Milton K. 1981. Estimates of reproductive parameters for free-ranging Ateles geoffroyi. Primates 22, 574-579. van Roosmalen M.G.M. and Klein, L.L. 1987. The spider monkeys, Genus Ateles. In: Ecology and Behavior of Neotropical Primates . RA Mittermeier and AB Rylands (eds). World Wide Fund, Washington. van Schaik C.P. 1989. The ecology of social relationships amongst female primates. In: V Standen, RA Foley (eds) Comparative socioecology. The behavioural ecology of humans and other mammals. Blackwell, Oxford. Sosa V., Flores J.S., Rico-Gray V., Lira R. and Ortiz J.J. 1985. Lista floristica y sinonomia m a y a . Fasciculo 1 de; Etnoflora Yucatanense. INIREB. Sterck E.H.M., Watts D.P. and van Schaik C.P. 1997. The evolution of female social relationships in nonhuman primates. Behavioural Ecolology and Sociobiology 41, 291-309. Symington, M.M. 1987. Ecological and social correlates of party size in the black spider monkey, Ateles paniscus chamek. Ph D. thesis, Princenton University. Symington M.M. 1988. Food competition and foraging party size in the black spider monkey (Ateles paniscus chamek). Behaviour 105: 117-134. Wrangham R.W. and Smuts B.B. 1980. Sex differences in the behavioural ecology of chimpanzees in the Gombe National Park, Tanzania. Journal of Reprodution and Fertility (Suppl) . 28:23-31. Wrangham R.W., Gittleman J.L. and Chapman C.A. 1993. Constraints on group size in primates and carnivores: population density and dayrange as assays of exploitation competition. Behavioural Ecology and Sociobiology 32: 199-209.

169


CHAP TE R 2

(CONT.)

Wrangham R.W. 2000. Why are male chimpanzees more gregarious than mothers? A scramble competition hypothesis. In: Primate Males: casues and consequences of variation in group composition. Kappeler P. (Ed). Cambridge University Press.

170


CHAP TE R 3 Boinski S. 1991. The coordination of spatial position: a field study of the vocal behaviour of adult female squirrel monkeys. Animal Behaviour 41, 89-102. 1970. The pair bond in the zebra finch. In: Social behaviour in birds and mammals. Ed by JH Crook. New York: Academic

Butterfield P.A. Press.

Chapman, C.A. 1990. Association patterns of spider monkeys: the influence of ecology and sex on social organization. Behavioural Ecology and Sociobiology 26:409-414. Chapman, C.A and Lefebvre L. 1990. Manipulating foraging group size: spider monkey food calls at fruiting trees. Animal Behaviour 39:89196. Chapman C.A. and Weary D.M. 1990. Variability in spider monkey's vocalizations may provide basis for individual recognition. American Journal of Primatology 22:279-284. Chapman C. A . , Wrangham R.W. and Chapman L.J. 1995. Ecological constraints on group size: an analysis of spider monkey and chimpanzee subgroups. Behavioural Ecology and Sociobiology 36:59-70. Cheney D.L., Seyfarth R.M. and Palombit R. 1996. The function and mechanisms underlying baboon 'contact' barks. Animal Behaviour 52, 507-518. Dittus W.P.J. 1988. An analysis of toque macaque cohesion calls from an ecological perspective. In: Primate vocal communication, ed. Todt et al. Springer-Verlag. Eisenberg J.F. 1976. Communication mechanisms and social integration in the black spider monkey, Ateles fusciceps robustus, and related species. Smithsonian Contributions to Zoology 213. Goodall J. 1968. The behaviour of free-living chimpanzees in the Gombe stream Reserve. Animal Behaviour Monographs 1, 165-311. Green S. 1975. Variation of vocal pattern with social situation in the Japanese monkey (Macaca fuscata): a field study. In: Primate Behavior. LA Rosenblum (ed). New York: Academic Press. Hauser M.D.

1996. The Evolution of Communication. MIT Press.

Heinrich B. and Marzluff J.M. 1991. Do common ravens yell because they want to attract others? Behavioural Ecology and Sociobiology 28:1321. Holekamp K.E., Cooper S.M., Katona C.I., Berry N.A., Frank L.G. and Smale L. 1997. Patterns of association among female spotted hyenas (Crocuta crocuta). J. Mammal 78, 55-64. 171


CHAPTER 3

(CONT.)

Janik V.M. 2000. Whistle matching in wild bottlenose dolphins truncatus). Science 289, 1355-1357.

(Tursiops

Kachigan S.K. 1991. Multivariate statistical analysis: a conceptual introduction. 2d ed. Radius Press, New York. Masataka N. 1986. Rudimentary representational vocal signalling of fellow group members in spider monkeys. Behaviour 96, 4 9-61. McComb K, Moss C, Sayialel S and Baker L. 2000. Unusually extensive networks of vocal recognition in African elephants. Animal Behaviour 59, 1103-1109. Miller R.G.

1981.

Survival analysis . Wiley, New York.

Mitani J.C. 1996. Comparative studies of African ape vocal behavior. In: Great Ape Societies, ed. by WC McGrew, LF Marchant and T Nishida. Cambridge: Cambridge University Press. Mitani J.C. and Nishida T. 1993. Contexts and social correlates of long distance calling by male chimpanzees. Animal Behaviour 45:735-746. Poole J.H., Payne K . , Langbauer W.R.Jr. and Moss C. 1988. The social contexts of some very low frequency calls of African elephants. Behavioural Ecology and Sociobiology 22, 385-92. Robinson J.G. 1982. Vocal systems regulating within-group spacing. Primate Communication, de. CT Snowdon, CH Brown and M Petersen. Cambridge: Cambridge University Press.

In:

van Roosmalen M.G.M. and Klein L.L. 1987. The spider monkeys, Genus Ateles. In: Ecology and Behavior of Neotropical Primates . Ed. by Mittermeier R.A. and Rylands A.B. World Wide Fund, Washington. Seyfarth R.M., Cheney D.L. and Marler P. 1980. Vervet monkey alarm calls: semantic communication in a free-ranging primate. Animal Behaviour 28, 1070-1094. Smith W.J. P r ess.

1977. The behavior of communicating.

Harvard University

Smolker R . , Richards A.F., Connor R.C. and Pepper J.W. 1992. Sex differences in patterns of association among Indian Ocean bottlenose dolphins. Behaviour 123, 38-29. Smolker R. 2000. Keeping in touch at sea: group movement in dolphins and whales. In: On the Move: how and why animals travel in groups. Ed. by Boinsky S. and Garber P. A. University of Chicago Press.

172


C H A PT ER 3

(CONT.)

Symington M.M. 1987. Ecological and social correlates of party size in the black spider monkey, Ateles paniscus chamek. Ph D. thesis, Princeton University. Symington M.M. 1990. Fission-Fusion Social Organization in Ateles and Pan. International Journal of Primatology 11(1):47 — 61. Tabachnik B.G. and Fidell L.S. Harper and Row, New York.

1989. Using multivariate statistics.

Teixidor P. and Byrne R.W. 1997. Can spider monkeys (Ateles geoffroyi) discriminate vocalizations of familiar individuals and strangers? Folia Primatol 68:254-264. Teixidor P. and Byrne R.W. 1999. The 'whinny' of spider monkeys: individual recognition before situational meaning. Behaviour 136, 279-308. Wilkinson G.S and Boughman J.W. 1998. Social calls coordinate foraging in greater spear-nosed bats. Animal Behaviour 55, 337-350. Wrangham R.W. 1977. Feeding behaviour of chimpanzees in Gombe National Park Tanzania. In: Primate Ecology. Ed. by Clutton-Brock T.H. Academic Press (London).

173


CON CLUSION Challenger A. 1998. Utilizacion y conservacion de los ecosistemas terrestres de Mexico: pasado, presente y futuro. CONABIO, UNAM, Sierra Madre. Estrada, A. and Cortes-Estrada R. 1988. Tropical Rain forest conversion and perspectives in the conservation of wild primates (Alouatta and Ateles) in Mexico. American Journal of Primatology 14, 315-27. Flores J.S and Carvajal I.E. 1994. Tipos de vegetacion en le peninsula de Yucatan. Fasciculo 3 de: Etnoflora Yucatanense. Universidad Autonoma de Yucatan. Mexican Government's Instituto Nacional de Estadistica, Informatica. http://www.inegi.gob.mx

Geografia e

Rylands, A. B., Mittermeier, R. A. and Rodriguez Luna, E. et al. 1995. A species list for the New World Primates (Platyrrhini): distribution by country, endemism, and conservation status according to the MaceLande system. Neotropical Primates 3(suppl), 113-60. Sosa V . , Flores J.S., Rico-Gray V., Lira R. and Ortiz J.J. 1985. Lista floristica y sinonomia maya. Fasciculo 1 de: Etnoflora Yucatanense. I NIREB.

174
