Citation
The ecology, behavior and conservation of the black lion tamarins (Leontopithecus chrysopygus, Mikan, 1823)

Material Information

Title:
The ecology, behavior and conservation of the black lion tamarins (Leontopithecus chrysopygus, Mikan, 1823)
Creator:
Padua, Claudio Valladares, 1948-
Publication Date:
Language:
English
Physical Description:
xviii, 182 leaves : ill., photos ; 29 cm.

Subjects

Subjects / Keywords:
Black lion tamarin -- Brazil ( lcsh )
Tamarins -- Behavior ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 164-181).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Claudio Valladares Padua.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030180514 ( ALEPH )
30336039 ( OCLC )

Downloads

This item has the following downloads:


Full Text
THE ECOLOGY, BEHAVIOR AND CONSERVATION
OF THE BLACK LION TAMARINS
(Leontopithecus chrysopygus, MIKAN, 1823).
By
CLAUDIO VALLADARES-PADUA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1993




Copyright 1993
by
Claudio Valladares-Padua




To my beloved family, Suzana, Andr6, Filipe and Joana.




ACKNOWLEDGMENTS
I would like to express my gratitude to Dr. Kent Redford and Dr. John Robinson, my advisors and director and former director of the Program for Studies in Tropical Conservation of the University of Florida. Dr. Robinson was a source of inspiration and support during my master's studies and the first years of my Ph.D. work. His departure to another institution was a loss to many of my colleagues, to me, and to the University of Florida. I am fortunate to I have had the privilege of having a second advisor who is as inspirational as Dr. Robinson. Dr. Redford has become one of my most precious friends and his support has been more than I could state in words. Nevertheless, I would like to express my profound gratitude for his continuous and inexhaustible support. Our friendship will always be a reason to look back with joy to my years as a doctoral student, despite the stressful moments that come with completion of the project.
My appreciation goes to Drs. Eisenberg, Brockmann, Judd, and Sunquist, members of my committee, for their guidance and support. As a Latin American, I would like to express my appreciation to the Program for Studies in Tropical Conservation at the University of Florida. This program is undoubtedly the major North American program for the preparation of professionals in the area of tropical conservation. Professionals who graduated from this program have
iv




already decisively contributed to conservation in their own countries. To be part of this program is really a privilege.
I have come to admire and truly cherish Laury Cullen Jr. We
have shared wonderful moments together. He has always been a hard worker and a source of joy and enthusiasm. Several other people were responsible for this work. I would like to thank Sr. Jose, Homero, Preto, Zezinho, Serginho, Sr. Arnaldo, and all other staff of the Morro do Diabo Park. At the Instituto Florestal de Sao Paulo, I am deeply in debt to Marco Antonio Garrido, who has become a true friend, as well as Sandra Pacagnella, Helder Faria, Giselda, Marlene, Francisco Serio, Helio Ogawa, Francisco Kronka, Regis Guillaumon, and Genji Yamazoe. Their personal and institutional support was crucial for me to develop my work.
My sincere appreciation goes to Prof. Adelmar Coimbra-Filho, who helped me find my track in the primatological world. I will always be grateful for his support. Pissinati and my other colleagues at the Rio de Janeiro Primate Center have also always been supportive to my studies, and I thank them for this. Dr. Russell Mittermeier supported and encouraged this study, and helped me throughout my career in the conservation area.
I received support from the community of Teodoro Sampaio,
especially the CONDEPRO group, and I am thankful to them. To all the CESP personnel I offer my sincere thanks, especially to Antonio Audi, Francisco Guerra, Luis Fernando Galli, and Carlos de Lucca. The herbaria at the University of S&o Paulo at Piracicaba and at the Instituto Florestal de Sdo Paulo identified the botanical material for my study. I thank them for their input.
v




In the conservation world, I received unlimited support from Bill Konstant, Admiral Ibsen Camara, Faigal Simon, S6nia Rigueira, Lou Ann and Jim Dietz, Devra Kleiman, Beata and Bejamin Beck, Jeremy Mallinson, and Garry Eberhart. Their conservation efforts have made me feel that all the struggles in this field were worthwhile. Hope and Bob Stevens and Jon Ballou have been true sources of encouragement and inspiration. I admire them greatly.
Colleagues and friends at the University of Florida were an invaluable source of help. Monique Chacon, Pierre Buck Berner, Damian Rumiz, Jim Ellis, Alexandro Grajal, Gary Shaeff, Rosa and Joho Paulo Viana, Peter Crawshaw, Laurenz Pinder, and friends, colleagues, and staff from the Tropical Conservation and Development Program and the Department of Wildlife. At the Center for Latin American Studies, I would like to especially thank Terry M'Coy for letting me stay so long with my "extended family." I have made great friends at the Center whom I will never forget. Some families were also extremely supportive: the Fonsecas, the Johnsons, Limas, Principes, Maias, Robinsons, and the Redfords.
Suzana, Andre, Filipe, and Joana encouraged and supported me in ways impossible to describe. It must have been very difficult to tolerate my frequent moments of irritation, stress, associated with long periods of absence.
My studies at the University of Florida were funded by an Overseas Scholarship from the Brazilian Council of Science and Technology-CNPq. My field work in Sdo Paulo was funded by the Fanwood Foundation, the World Wildlife Fund-US, the Jersey Wildlife Preservation Trust, the Wildlife Preservation Trust International,
vi




the Conservation International, Whitley Animal Preservation Trust and the Instituto Florestal de Sdo Paulo. From the University of Florida I received support from the Program for Studies in Tropical Conservation, the Tropical Conservation and Development Program and the Center for Latin American Studies. The Internet Computer Network helped in transferring data and in my communication with people who are dear to me. This helped reduce the solitude I felt during my last months in the United States.
vii




TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS...................................................................I iv
LIST OF TABLES............................................................................x i
LIST OF FIGURES.......................................................................... xii
ABSTRACT ................................................................................. xvi
INTRODUCTION............................................................................. 1
The Black Lion Tamarin............................................................. 1
Aims of the Study..................................................................... 7
THE MORRO DO DIABO STATE PARK AS A HABITAT FOR THE
BLACK LION TAMARIN...............................................................1 0
Introduction ........................................................................... 10
The Morro do Diabo State Park................................................ 11
Methods................................................................................ 16
L. Chrysopygus Habitat in Morro do Diabo ..............................1 6
Types of Data Collected..................................................... 20
Characterizing the Four Habitats..........................................2 1
General Trends of Multivariate Analysis of Habitats............ 23
Results ................................................................................ 24
General Results ................................................................ 24
River area.................................................................... 25
Morro area .................................................................... 27
Little Bridge area........................................................... 27
Caldeir~o area ............................................................... 29
Results of the Multivariate Analysis of Habitats................. 30
Discussion............................................................................ 32
HABITAT DIFFERENCES AND THEIR EFFECTS OF FEEDING AND
FORAGING ECOLOGY................................................................ 34
Introduction ......................................................................... 34
Methods................................................................................ 36
Methods for Locating the Tamarins....................................... 36
v ii i




Methods for capturing and using radio-telemetry ............. 36
Home range maps and trail grids ...................................... 38
Methods for Recording and Analyzing Data ........................... 39
Composition and diversity of resources........................... 39
Phenology..................................................................... 40
Time budgets................................................................ 41
Movements and use of space ............................................ 44
Ranging behavior............................................................ 45
Feeding and foraging behavior ........................................ 46
Results ................................................................................ 48
Abundance and Distribution of Resources............................. 48
Variation with location................................................... 48
Phenology..................................................................... 51
Time Budgets..................................................................... 57
Movements and Use of Space............................................... 59
Height and substrate....................................................... 64
Home-range areas......................................................... 69
Intensity and diversity of quadrat use............................. 74
Sleeping sites .............................................................. 84
Diet................................................................................. 90
Plant food.................................................................... 90
Animal food................................................................... 93
Local variation ............................................................. 96
Seasonal variation ........................................................ 98
Variation in foraging ................................................... 105
Discussion .......................................................................... 107
Use of Time .................................................................... 108
Movements and Use of Space ............................................. 109
Diet................................................................................ 114
DISTRIBUTION, ABUNDANCE AND MINIMUM VIABLE POPULATION
OF THE BLACK LION TAMARINS ............................................... 117
Introduction........................................................................ 117
The Morro do Diabo State Park Sub-Populations ................. 11 8
Other Forest Fragments in the Pontal Region ...................... 11 9
The Fazenda Rio Claro ...................................................... 11 9
The Caetetus Ecological Reserve....................................... 121
Estimating Metapopulation Size ............................................ 121
The Morro do Diabo State Park .......................................... 122
Other Forest Fragments in the Pontal Region ...................... 124
Fazenda Rio Claro ............................................................ 127
The Caetetus Ecological Reserve....................................... 127
Results............................................................................... 127
i x




D iscussion ........................................................................................................... 12 8
METAPOPULATION EXTINCTION MODEL ........................................................... 134
Introduction ........................................................................................................ 134
V ortex O verview ............................................................................................... 136
M ethods ................................................................................................................. 136
Data Sources ....................................................................................................... 13 7
Environm ental variation .......................................................................... 14 0
C atastrophes ........... ...... ..... .......... 14 1
Inbreeding depression .............................................................................. 14 1
R esults .................................................................................................................. 14 2
D iscussion ........................................................................................................... 14 4
C onclusions ......................................................................................................... 14 9
CONCLUSIONS AND RECOMMENDATIONS ......................................................... 151
Past M anagem ent M easures .......................................................................... 153
T he Future of the Black Lion Tam arin ..................................................... 154
APPENDIX A LIST OF TREE SPECIES FOUND IN THE MORRO DO
D IABO PA R K ......................................................................................................... 159
APPENDIX B DATA SHEET FOR PROCESSING BLACK LION
TA M A R IN S ............................................................................................................ 163
REFERENC ES ............................................................................................................... 164
BIO G RAPH ICA L SKETC H ........................................................................................ 182
x




LIST OF TABLES
Table. page
2-1. Variables used to describe habitats. All variables were measured in all plots of the four study areas of black lion tamarins' home range in the Morro do Diabo State Park. Corresponding abbreviations for each variable are in parentheses........................................................................... 22
2-2. Mean habitat variables for the four study areas of the black lion tamarin home ranges in the Morro do Diabo State Park. Results with different letters indicate statistically significant differences............................................................................ 26
3-1. Characteristics of each study area quantifying availability of resources .......................................................................... 50
3-2. Ten species on which each study group spent the most time and percentage of total time spent in trees of each fruit species................................................................................. 52
3-3. Analysis of variance for monthly differences in six behaviors. Significance values are calculated from Type Ill partial sum of squares ............................................................9 1
3-4. Percentage of diet similarity among the four study groups as measured by the Renkonen index.......................................... 97
3-5. List of the most common plant items for each study group and the percentage of the total feeding records for each month................................................................................... 99
3-6. Rank list of the percentage of prey capture success for the four study groups................................................................. 106
4-1. Mean group size, number, age and sex of L. chrysopygus in the four study areas of the Morro do Diabo State Park............. 129
xi




4-2. Estimates of available habitat for each area where black lion tamarins are found, average home range size, group densities and total population sizes ...................................................... 131
5-1. Different population parameters used in the Vortex model ................................................................................. 138
5-2. Black lion tamarin carrying capacity based on the available known habitat in the original home range of the species ......... 139
xii




LIST OF FIGURES
Figure page
1-1. Photograph of a group of black lion tamarins
(Leontopithecus chrysopygus). Courtesy of The Jersey
Wildlife Preservation Trust ...................................................................... 2
2-1. The Morro do Diabo State Park ....................................................1 2
2-2. The annual rainfall and temperature means for the Morro
do Diabo region between 1979 and 1982 ............................................1 5
2-3. Morro do Diabo State Park vegetation map showing the 4
groups of black lion tamarins I censused. Adapted from Campos
and Heinsdijk (1970). Vegetation types ...............................................1 7
2-4. Model of quadrats used to collect habitat data ......................1 9
3-1. Fruit availability to black lion tamarins in the River Area
group measured by number of species carrying ripe fruit each
m o n th ....................................................................................................................... 5 3
3-2. Fruit availability to black lion tamarins in the Morro Area
group measured by number of species carrying ripe fruit each
m o n th ....................................................................................................................... 5 4
3-3. Fruit availability to black lion tamarins in the Little Bridge
Area group measured by number of species carrying ripe fruit
each m onth .................................................................................................... . 5 5
3-4. Fruit availability to black lion tamarins in the Caldeirdo
Area group measured by number of species carrying ripe fruit
each m onth .................................................................................................... . 5 6
3-5. Time budgets of each of the four study groups. (Summed
across the entire year) .............................................................................. 58
3-6. Movement of the River Area group during the study .......... 60
xii




3-7. Movements of the Morro Area group during the study ........6 1 3-8. Movements of the Little Bridge Area group during the study.................................................................................... 62
3-9. Movements of the Caldeir~o Area group during the study.... 63 3-10. Mean daily path length ( S.D.) for the River group during the one-year study.................................................................. 65
3-1 1. Mean daily path length ( S.D.) for the Morro group during the one year study .................................................................. 66
3-12. Mean daily path length ( S.D.) for the Little Bridge group during the one year study ....................................................... 67
3-13. Mean daily path length ( S.D.) for the Caldeira-o group during the one year study ....................................................... 68
3-14. The 1.13 km2 home range of the River group on a 50 mn x 50 mn grid .................................................................................. 70
3-15. The 1.20 km2 home range of the Morro group on a 50 mn x 50 m grid .................................................................................. 71
3-16. The 1.99 km2 home range of the Little Bridge group on a 50 m x50 mngrid........................................................................ 72
3-17. The 1.20 km2 home range of the Caldeirdo group on a 50 mn x 50 m grid ........................................................................... 73
3-18. Monthly mean use of quadrat by each study group. Means with same line are not significantly different (Wailer-Duncan K ratio).................................................................................... 75
3-19. Intensity of quadrat use by the River group..................... 76
3-20. Intensity of quadrat use by the Morro group.................... 77
3-21. Intensity of quadrat use by the Little Bridge group.......... 78
3-22. Intensity of quadrat use by the Caldeirdo group .............. 79
3-23. Frequency distribution of intensity of quadrat use by the River group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) ....... 80
x ii i




3-24. Frequency distribution of intensity of quadrat use by the Morro group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) .......8 1 3-25. Frequency distribution of intensity of quadrat use by the Little Bridge group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) ......................................................................... 82
3-26. Frequency distribution of intensity of quadrat use by the Caldeirao group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) ......................................................................... 83
3-27. Distribution of the duration of quadrat occupancy. Figure plots percentage order of quadrats against percentage of time of quadrat occupancy.................................................................. 85
3-28. Spatial distribution of tree dens in the River group home range.................................................................................... 86
3-29. Spatial distribution of tree dens in the Morro group home range.................................................................................... 87
3-30. Spatial distribution of tree dens in the Little Bridge group home range ........................................................................... 88
3-31. Spatial distribution of tree dens in the Caldeirdo group home range ........................................................................... 89
3-32. Overall use of different food items for the four study groups of L. chrysopygus......................................................... 95
3-33. Monthly variation in diet of the River group .................. 100
3-34. Monthly variation in diet of the Morro group.................. 101
3-35. Monthly variation in diet of the Little Bridge group .......102
3-36. Monthly variation in diet of the Caldeirdo group ............ 103
4-1. Satellite image of the Morro do Diabo State Park and other surveyed forest fragments in the Pontal region...................... 120
xiv




4-2. Cumulative use of quadrats per month by each of the four study groups of L. chrysopygus during one year of data collection ...................................................................................................... . 1 2 3
4-3. The map of the State of So Paulo with the main locations
where L. chrysopygus were surveyed or censused ........................1 25
5-1. Extinction probability over time for the six sub-populations and the metapopulation. Notice the change of scale in the m etapopulation figure .............................................................................. 1 4 3
5-2. The loss of heterozygosity over time for the six subpopulations and the metapopulation. Notice the change of scale in the m etapopulation figure .......................................................................1 45
xv




Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE ECOLOGY, BEHAVIOR AND CONSERVATION OF THE BLACK LION TAMARINS (Leontopithecus chrysopygus, MIKAN, 1823).
By
Claudio Valladares-Padua
May, 1993
Chairman: Dr. John G. Robinson Cochairman: Dr. Kent H. Redford Major Department: Wildlife and Range Sciences, School of Forest Resources and Conservation The black lion tamarin or golden-rumped lion tamarin
(Leontopithecus chrysopygus) is one of the most endangered species of New World primates. Though once more widespread, now the species has survived only in small forest fragments in the interior of the State of Sdo Paulo, Brazil. Two of these forest fragments are legally protected areas while the others are privately owned. Until recently the only known populations were the ones inhabiting the two protected areas. The total estimated population size for the species was 100 individuals, an estimate which caused the species to be classified as on the verge of extinction.
I compared four groups of black lion tamarins in the Morro do Diabo State Park in Brazil in order to investigate in what range of xvi




environmental conditions the species could survive. I selected four areas with different habitat characteristics and found that many aspects of the species biology were flexibile in responding to these habitat differences. The four study groups showed that they were able to adjust their use of space in the different habitats and to respond to habitat alterations through major dietary shifts. There were, however, no significant differences in their overall time budget. I concluded that black lion tamarins, like other members of the Callitrichidae, can live and reproduce in many different types of habitats. These findings suggest that the species has behavioral variation which may facilitate conservation management.
I estimated L. chrysopygus density to be 3.72 individuals per km2 for an average home range size of 1.38 km2. Based on this density, I estimated a metapopulation size of 1,004 individuals for the species. I combined demographic information with the data resulting from my ecology and behavior study, to predict the future outcome of black lion tamarin populations. For this purpose I used the Vortex model, a computer simulation model of population viability analysis. The results suggest that if treated individually, all black lion tamarin sub-populations except one, have more than a 50 percent chance of becoming extinct in the next 100 years. The exception was the population in the Morro do Diabo State Park where the species seems to have a greater chance of survival. Therefore, if all sub populations were managed as a metapopulation, which may include reintroduction, translocation and/or managed dispersal of individuals among its sub-populations, there would be a high probability for the long term survival of the species.
xvii




INTRODUCTION
The Black Lion Tamarin
The black lion tamarin or golden-rumped lion tamarin
(Leontopithecus chrysopygus Mikan, 1823) is one of the largest and most highly specialized members of the Callitrichidae family of New World primates (Hershkovitz, 1977) (Fig. 1-1). The species is endemic to the State of Sho Paulo, the most developed state in Brazil. Its habitat has been classified by Rizzini (1963, 1967) as the riparian forest that originally extended from the Atlantic coastal forest inland along the major rivers of the state. Little native habitat is left in the original range of these primates. Destruction of forest for lumber, agriculture and industrialization has reduced the State's original forest cover by about 95 percent (Serra-Filho et al., 1975; Victor, 1975; Ferri, 1980).
As a result of the demise of almost all its habitat, L.
chrysopygus was considered to be extinct from the beginning of the century. Then, in 1971, Coimbra-Filho rediscovered the species in the 35,000 ha Morro do Diabo State Park, in the far western part of the state (Coimbra-Filho, 1971). In 1976, Coimbra-Filho found a second population of black lion tamarin, this time in the 2,000 ha Caetetus Biological Reserve in the central part of the state (Coimbra-Filho, 1976). After the rediscovery of L. chrysopygus.
1




Figure 1-1. Group of black lion tamarins (Leontopithecus chrysopyqus). Courtesy of The Jersey Wildlife Preservation Trust.




AP
ko-




4
Coimbra-Filho and Mittermeier (1977) conducted a series of surveys to assess the conservation status of L. chrysopygus. Their estimates for the species population was in the range of 200 to 300 animals but the habitat of the species was continuing to be lost (reviewed in Coimbra-Filho, 1990). The need for conservation action for the species was evident and the response came in a series of measures both at the national and international levels. L. chrysopygus was placed on the Brazilian list of endangered species (Coimbra-Filho, 1971; Bernardes et al., 1990), in the IUCN Red Data Book (Thornbak and Jenkins, 1982), and listed in Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). In 1984, the International Union for the Conservation of Nature and Natural Resources (IUCN) placed L. chrysopygus prominently on a list of the world's twelve most endangered species (Murray and Oldfield, 1984).
The initial efforts for the conservation of black lion tamarins stimulated a series of studies on this species' biometry, morphology and physical characteristics (Coimbra-Filho, 1976; Hershkovitz, 1977, Kleiman, 1981; Kleiman et al., 1988). Black lion tamarins are small primates, the adults weighing an average of 600 g. The length of the head and body averages 261 mm and the length of the tail averages 370 mm. There is no sexual dimorphism. The color of black lion tamarin is predominantly black with a golden or reddish dot on the forehead, on the lower half of the forelimbs and on the proximal half of the dorsal surface of the tail.
If the general characteristics of the species are not subject to controversy, its taxonomic status is. Hershkovitz (1972) considered




5
it as a subspecies of Leontopithecus rosalia and many authors have followed this arrangement (e.g., Coimbra-Filho and Mittermeier, 1972, 1973, 1977; Kleiman, 1981, Kleiman et al., 1988). Forman et al. (1986), using the differentiation of alternate alleles at two polymorphic loci, argued in favor of this sub-specific designation. In 1981, Mittermeier and Coimbra-Filho revised their original taxonomic arrangement and elevated L. chrysopygus to a full species. This species designation was later supported by Rosenberger and Coimbra-Filho (1984) based on cranial-dental variability, by Snowdon et al. (1986) using the morphology of long call vocalizations, by Mittermeier et al. (1988) based on morphological characteristics, and by Natori and Hanihara (1989) based on morphological variation of the cranium. In a recent revision of the systematics of Brazilian primates, Coimbra-Filho (1990) maintained the full species classification of this taxon.
The genetics of the species was studied by Valladares-Padua (1987). An electrophoretic survey of blood enzymes from captive and free-ranging individuals showed that the species had no heterozygosity and it was genetically monomorphic at 25 loci (P=0.00; H=0.00) (Valladares-Padua, 1987). Results such as these are most frequently explained by one of the following interpretations:
(1) the low heterozygosity and polymorphism are a
measurement artifact in that electrophoretic assays only measure a small number of loci in the genome (Powell, 1983);




6
(2) the loss of genetic variability as a result of inbreeding depression in small populations (O'Brien, 1986; Rails and Ballou, 1983); or
(3) the lack of genetic diversity is a result of adaptation to
specific environmental conditions. Selection to this narrow range of conditions may result in genetic homogeneity. The species may loose its genetic variability if it occupies a stable niche for a long period of time (Beardmore, 1983; O'Brien, 1986).
It is difficult at this moment to know which of these best
explains the lack of genetic diversity in this species. ValladaresPadua (1987) reported that L. chrysopygus manifests other phenotypic signs of genetic depauperation, particularly those individuals in captivity. Some captive-born animals exhibited brown coloration, a problem correlated with the functioning of the melanin pigmentary system (Pissinatti, 1992). Such genetic diseases are caused by simple Mendelian recessive alleles revealed by consanguineous mating (Valladares-Padua, 1987).
Despite the work on biometry, taxonomy, and genetics, little information exists on the ecology and behavior of black lion tamarins in the wild. There are no published long-term studies on the species and the information available is based either on a small number of field observations or on anecdotal information. The existing natural history data are summarized in Coimbra-Filho (1970a; 1976) and Coimbra-Filho and Mittermeier (1973; 1977). More recent studies by Carvalho et al. (1989) and Carvalho and Carvalho (1989) reviewed some aspects of L. chrysopygus ecology and behavior, and later Keuroglian (1990) presented data on a very




7
short-term study of the ecology of the species. In addition to all these studies on organismic biology great concern continues to exist concerning the long-term survival of the species, a concern which can only be addressed by gaining data on the conservation biology of the black lion tamarins.
Aims of the Study
In a world increasingly affected by humans, where large tracts of relatively undisturbed habitats are being shredded and fragmented, one of the major challenges for conservation biology is ameliorating the long-term consequences of population fragmentation. This study was conceived as an answer to this challenge based on updating the concept of metapopulation developed by Levins (1969, 1970). To understand the conservation approach I am proposing, it is important to understand Levins' concept of "metapopulation", which he defined as an infinite number of sub or local populations of one species. Levins conceived of a species' metapopulation as a dynamic in which local extinctions of pestpredators was balanced by re-migration from other populations. Thus, in his view a metapopulation could be regarded as the net result of the establishment, survival, extinction and re-colonization of local populations.
This approach was adapted by conservation biologists, mainly by Gilpin (1987) and Hanski and Gilpin (1991). They proposed an adaptation of Levin's model to conservation biology using a finite number of sub-populations. Their models illustrated that the minimum viable size of a population was not solely dependent on its




8
size, but also on the patchiness of the existing habitats and on the movement of individuals between habitable patches. They included the extreme case where discontinuous habitats may result in the total impossibility of natural migration among local populations. Habitat or population fragmentation creates small isolated subpopulations, wich enhances their probability of extinction due to genetic, demographic and environmental forces acting within patches (Soul6, 1980; Rails and Ballou, 1983). Even if the subpopulations survive, isolation itself might cause genetic drift, leading to genetic divergence and consequent speciation (Wright, 1977; Franklin, 1980; Otte and Endler, 1989). Thus, in the cases where fragmentation precludes natural migration, metapopulation management will entail artificially moving animals from one patch to another. Managed animal migration must also take into consideration previous knowledge about the species so the animals can survive and reproduce in the new area (Foose, 1990).
Although Levins and Gilpin analyzed several aspects of
metapopulations, (eg. spatial distribution of habitat patches and extinction and recolonization dynamics) they did not directly consider the ecological and behavioral flexibility of a species as an important metapopulation dynamic variable. Lee (1991) suggested that investigations on how mammalian species respond to environmental change are of utmost importance in illuminating how animals might respond to fragmentation. In this study, I am particularly interested in behavioral flexibility related to environmental changes which includes maintenance, growth and reproduction.




9
The central questions of this dissertation are: do the behavior and ecology of groups of Leontopithecus chrysopygus differ significantly in different habitats, and if so, what are the implications of these differences for the conservation of the species? To answer this, in chapter 2, I compare four black lion tamarin habitats to determine if there are significant differences among these areas. Sixteen habitat variables were compared using multivariate statistical tests. In chapter 3, I investigate the effects of habitat differences in the feeding and foraging behavior among four groups of black lion tamarins. I collected quantitative and qualitative information on the variation in resource abundance and on the tamarins' use of time and space both as individuals and as groups. I examine in chapter 4, the size and density of each known sub-population of the black lion tamarin and estimate its metapopulation size. In chapter 5, I integrate the data resulting from all previous chapters by running a population viability computer simulation model (Vortex) (Lacy, 1990). With this program I examine, in light of black lion tamarin behavioral and ecological flexibility, the effects of deterministic forces as well as demographic, environmental and genetic stochastic events on the tamarin's metapopulation. Finally, in chapter 6, I summarize and discuss the dissertation's major results.




THE MORRO DO DIABO STATE PARK AS A HABITAT
- FOR THE BLACK LION TAMARIN
Introduction
The members of the family Callitrichidae seem to be quite
flexible in their general selection of habitat (Sussman and Kinzey, 1984). Moynihan (1970) reports that in Panama, Saguinus aeoffroyi uses both low, dense primary forest and abandoned, overgrown agricultural fields. Other authors describe at least three species of tamarins, S. mystax, S. fuscicollis and S. midas, living in primary and secondary forests (Castro and Soini, 1977; Mittermeier and van Roosmalen, 1981). Leontopithecus rosalia and Leontopithecus chrysomelas, two species congeneric with L. chrysopygus, show preference for primary forests (Rylands, 1982; Coimbra-Filho and Mittermeier, 1973; Kleiman et al., 1988).
The main question of this chapter was, therefore, to determine whether black lion tamarins showed this same trend in habitat flexibility and could live in different habitats. To answer this question, I selected four groups of black lion tamarins inside the Morro do Diabo State Park. The park is the only remaining place to attempt to answer this question because it is the last large continuous forest in the original range of the species, and because it is the only area that has a variety of forest types. The study sites
10




11
were selected after six months of evaluating habitats that seemed distinct from each other. The habitats used by the four study groups were quantitatively compared using means of sixteen preestablished variables (Table 2-1).
My null hypothesis was that there were no differences among the four habitats. To test this hypothesis, I used multiple discriminant function analysis (Ho = the common elements of the various mean vectors were identical to one another). I also used canonical discriminant analysis to create a habitat hypervolume (sensu Carey, 1981), which is a mathematical description of the species habitat range in multidimensional space. While the multiple discriminant function analysis confirms differences in habitat variables, the canonical discriminant analysis locates the differences and allows visual representation of these differences.
The Morro do Diabo State Park
The Morro do Diabo State Park is a protected area under the administration of the Forestry Institute of the State of S~o Paulo (Instituto Florestal de Sdo Paulo). It is located in the arrow-shaped region in the extreme western part of the State of S&o Paulo (220 30' S, 520 20' W), known as "Pontal do Paranapanema" (Fig. 2-1).
The park was created as a reserve in 1941 by the then
Governor of Sdo Paulo, Dr. Fernando Costa (Guillaumon et al., 1983). Of the original 290,000 ha park which was gazetted to protect regional fauna and flora, today only 34,156 ha survive (ValladaresPadua, 1987).




12
Brazil Sio Paul
N
V E
S
500 m Morro do Diabo
Paranapane Sat River
Figure 2-1. The Morro do Diabo State Park.




13
Most of the reduction in size occurred in the 1950s and 1960s. During this period the reserve was frequently invaded and illegally colonized by members of a wealthy elite, usually with the approval and connivance of important members of the state government (Leite, 1981). In the following decades the reserve experienced an increasing level of protection until 1986 when its status changed from reserve to park. Today, the park is well protected with legally demarcated, nondisputed boundaries. To the south there is the Rosana Dam Lake on the Paranapanema river, and pasture and farmland are to the east, west, north of the park.
The park's predominant relief forms are slightly undulating plains interspersed with fluvial plains and valleys and some sedimentary mesas. The Morro do Diabo is the highest of these mesas in the region. The altitude in the park ranges from 350 m above sea level at the top of the Morro do Diabo to 300 m above sea level at its lowest regions.
The park's predominant soil is dark red latosol sand phase (LEa) (Deshler, 1975; Ventura et al., 1965). A later study, done in 1979, by the Sdo Paulo Electricity Company (CESP) classified it as predominantly LEd, dark red latosol dystrophic plane relief phase (Guillaumon et al., 1983). No matter which of these classifications is more precise, the region in general has very poor soil (Setzer, 1949).
The regional climate is Cwa in the Koppen classification
(Guillaumon et al., 1983; Coimbra-Filho, 1976). The rainfall in the area of Morro do Diabo is seasonal; the area annually receives an average of 1,131 mm, of which 30 percent falls between April and




14
September (Fig. 2-2). This is half the mean for the Atlantic coastal forest. The low rainfall during the dry season in the Morro do Diabo region leaves the soil with little moisture and is responsible for the deciduous nature of the forest. The mean annual temperature is higher than in the coastal region.
Because it is part of the Atlantic rainforest, the forests of Morro do Diabo would be expected to match those described by Schimper (1903: 66) as "An evergreen, hygrophilous in character, at least 30 m high, rich in thick stemmed lianas and in woody as well as herbaceous epiphytes." But the Morro do Diabo area is also influenced by cold fronts which come from the south dropping the temperature below 00 C for one or two days each winter. These cold invasions influence the regional natural vegetation by reducing the regeneration rate of the trees (Hueck, 1978).
As a result, the best approximation of a correct classification of the park's forest would be upland semi-deciduous Atlantic Forest interspersed with some areas of "cerrad&o" (Baitello et al., 1988). The latter is described by Redford (1983: 126) as a "tall dense semideciduous xeromorphic savanna vegetation." The official denomination of the park's forest formation, according to the Instituto Brasileiro de Geografia e Estatstica--IBGE (Brazilian Geography and Statistics Institute) is the semi-humid tropical forest of the interior. It is characterized by being semi-deciduous with most of the emergent trees losing their leaves during the cool dry months between June and September (Hueck, 1978; C. Valladares-Padua pers. observ.).




15
30.000- -175.000
"" Temperature
2-150.000
25.000 .0. Rainfall
125 .000
20.000
(U -100.000 WL C
a 15.000 -.
0.. "- .
"' ... .75.000 '
"- 10.000 ,
50.000
5.000 25.000
0.000- I I I I I i I 0.000
C l ~ >1 c 0) c. > 0
ca 4) CO CL M ~ a) 0 a)
-,UL~ < U) 0 z 0
Months
Figure 2-2. The annual rainfall and temperature means for the Morro do Diabo region between 1979 and 1982.




16
Despite Morro do Diabo's physiognomic differences with the coastal forest of Brazil, its tree species and genera are similar to those described by Assumpg.o et al. (1982) for the State of S~o Paulo. At Morro do Diabo, I identified a total of 84 tree species in 31 families using a systematic sampling design (Appendix A). This is more than the list of Campos and Heinsdijk (1970) and a few less than the list produced by Baitello et al. (1988).
The vegetation survey conducted by Campos and Heinsdijk
(1970), for the park's first management plan, concluded that Morro do Diabo is composed of several distinct vegetation formations. They proposed a general classification of these formations in eight distinct types, six of which have some forest component (Fig. 2-3).
Methods
L. Chrysopygus Habitat in Morro do Diabo
Due to the comparative nature of my study, I selected four study sites that demonstrated the highest habitat variability detectable at the beginning of the field study. These were based on the vegetation map of Campos and Heinsdijk (1970) and my own personal observations of the general physiognomy of the area. After this initial qualitative classification, the next step was to quantify the study sites to demonstrate significant habitat differences among them. In order to quantify the vegetation characteristics in each of the four study sites, I set up trail transects forming 50 m x




17
N
Y YI IYY W
Y VYY Y
. T Y Brazil Sgo Paulo
T T
v v YY . T T
. . t et y 500m
T t s VEGETATION
Stt Type I
Y Type II
Y, I I
Type II
T" T tt
O Type IV
Type IV
Sg 3 S eType VII
10. gtt t ttittttttot
a t19 t z L'""' k.^. ^.. ttttt ttttttitt t it ttittit tt tt 9t ft tit tt t tt tt t ttttt tttt it
Met tt ; IP ttt tck rt ttit ttitttt
-)tttt t/ttitttt Paranapaneme River ttttr itt
T1 T
Figure 2-3. Morro do Diabo State Park vegetation map showing the 4 groups of black lion tamarins I censused. Adapted from Campos and Heinsdjik (1970). Vegetation types: 1. Densely stocked tall forest; 11. Moderately stocked tall forest; III. Poorly stocked forest; IV. Very impoverished forest; V. Heterogeneous vegetation; VI. Highly impacted forest fallow; VII. Abandoned pasture. L. chrysopygus groups: 1. River group; 2. Morro group; 3. Little Bridge group; 4. Caldeir&o group.




18
50 m quadrats. All transects were cut by machete to a width of approximately 1m along north-south and east-west compass bearings. I sampled 16 habitat variables in each of the 50 m x 50 m quadrats for a total of 35,776 observations. Among the measured variables, five were related to the composition and structure of trees with more than 10 cm of dbh (number of trees, number of tree species, diameter at breast height, mean tree height and canopy height). These were measured in 10 m x 10 m plots located in the northwestern corner of each 50 m x 50 m quadrat. The 11 variables not directly related to the tree composition of the forest were measured in a 1m x 10m rectangle aligned diagonally inside the 10 m x 10 m square (Fig. 2-4). In the four study areas, the habitat data matrix had 2,336 quadrats by 16 variables.
This method of data collection is called a "square grid"
systematic sampling design. It is widely used in forestry practices and it simplifies considerably field work. The problem in adopting this method is that it can cause imprecision if the systematic design coincides with periodicity in natural features like parallel ridges. In this case the results would not indicate variations in features such as valleys (de Vries, 1986). This did not seem to be the case in Morro do Diabo because, although I used systematic sampling, the distances between the samples were small (40 m) and the number of samples large (more than 2,300 quadrats). To avoid any risk of missing important habitat aspects and to prevent possible biases in the habitat sampling I used all 50 m x 50 m quadrats in a fine-grain approach. However, for the data analysis I




19
10 m x 10 m
North Western Quadrat
1 m x 10 m
Diagonal Rectangle Quadrat 50 m
50 m
Figure 2-4. Model of quadrats used to collect habitat data.




20
included only the quadrats used at least once by the primate groups I was studying. I reasoned that by restricting the analysis to the quadrats used by the primates I could obtain more accurate data on the habitat actually used by each group.
Types of Data C7ollected
Once I had selected the habitats to be compared and organized the network of quadrats in each study area, I collected quantitative and qualitative data in each of the quadrats. The protocol employed to measure the variables was a modified version of one used in many small mammal habitat studies (M'Closkey, 1976; August, 1983; Fonseca, 1989) (Table 2.1). In each l1in x l10m quadrat, I measured and identified all the trees with more than 10 cm dbh. These variables were related to number of trees and species composition, mean diameter at breast height and mean canopy and tree height (variables 1-5; Table 2-1). In the north-western corner of the quadrat I also measured the depth of the humus layer. In the 10 m x
1 m rectangle I counted and sorted in four equal size classes all trees with diameter less than 10 cm (variables 7-10; Table 2-1. The remaining variables in the 10 m x 10 m quadrats were recorded on a scale of 0 (minimum score for a parameter) to 10 (maximum score for a parameter). In this way I quantified epiphyte loads, vines/lianas loads and their degree of entanglement, ground bromeliads and canopy cover. I also used a "density board" to measure the degree of visibility in the-forest. I placed the board at the one meter end of the 1 in x 10 m rectangle and the observer at the other end (NRC, 1981; Wight, 1938).




21
Identification of tree species was done by my field assistant, Mr. Jose M. de Souza, who is a knowledgeable park ranger. Mr. Souza had participated in all previous primate studies and botanical inventories conducted at the Morro do Diabo State Park. He identified more than 90 percent of the trees in the study areas, showing consistency and accuracy in the identification of trees, and exceptional skills in following the animals. However, because the identifications relied mainly on bark and leaf characteristics, I marked for future identification any tree that could not be immediately identified as well as some common trees selected at random for checking its identification accuracy. When these species flowered or fruited, voucher specimens were collected and subsequently identified at the herbaria of the Forestry Institute of S5o Paulo or at the Forestry Department of the University of Sao Paulo. Mr. Jose correctly identified the common trees and of those for which he was not sure, his suggestions coincided 80 percent with those identified elsewhere.
Characterizing the Four Habitats
To characterize the habitats of each of the four groups, I used qualitative data to present a general description of the physical features, structure, and vegetation of each area. Quantitative data were used to test if each of the characteristics measured were significantly different and how they differed. Thus, I analyzed each of the habitat variables for the four areas by means of a series of independent one-way analyses of variance.




22
Table 2-1. Variables used to describe habitats. All variables were measured in all plots of the four study areas of black lion tamarins' home range in the Morro do Diabo state park. Corresponding abbreviations for each variable are in parentheses.
HABITAT VARIBLES DESCRIPTION
Number of Trees (NT) Number of trees in dbh sample
Number of Tree Species (NS) Number of tree species in dbh sample Mean diameter/ breast/height (dbh) Mean diameter (cm) of all trees with a dbh larger than 32 cm within a 1OX10 m quadrat
Mean Tree Height (TH) Mean tree height (m) of all trees in
dbh sample
Canopy Height (CH) Mean canopy height (m) of all trees in
dbh sample
Epithyte Density (ED) Density of vascular epithytes measured
on a sliding scale (0-2) in each 10X10 quadrat
Number of Trees 0.1-2.4 cm Number of trees 0.1-2.4 cm dbh in a
of dbh (T1) rectangle of 1OXl m in each quadrat
Number of Trees 2.5-4.9 cm Number of trees 2.5-4.9 cm dbh in a
of dbh (T2) rectangle of 1OXl m in each quadrat
Number of Trees 5.0-7.4 cm Number of trees 5.0-7.4 cm dbh in a
of dbh (T3) rectangle of 1OXl m in each quadrat
Number of Trees 7.5-10.0 cm Number of trees 7.5-10.0 cm dbh in a
of dbh (T4) rectangle of 1OXl m in each quadrat
Ground Bromeliads Percentage of ground bromeliads in each
density (GB) 1OX10 quadrat
Vine Density (VD) Density of vines expressed on a sliding
scale (0-2) in each 1OX10 m quadrat
Vine Entanglement (VE) Entanglement of the vines
measured on a sliding scale (0-2)
in each 1OX10 in quadrat
Humus (HU) Depth (cm) in the first soil layer
Percentage of Canopy Cover (CC) Percentage of canopy cover above each 1OX10 m quadrat
Understory Volume (UV) Number of areas of 30 cm visible on a
ruler od six divisions positioned
10 m distance in each 1OXl quadrat




23
General Trends of Multivariate Analysis of Habitats
The commonly recognized multidimensional nature of wildlife habitat has led to an increasing use of multivariate statistical techniques in studies of wildlife ecology (Capen, 1981). These techniques permit the use of a series of predictor habitat variables as an approximation of its multidimensional characteristic. The variables are then used in deriving a function that allows classification of the sample groups according to the probability of belonging to each group (Arita and Humphrey, 1988). In this study, I used the 16 variables measured in the four study habitats as the predictor variables. This allowed me to test the null hypothesis that the common elements in the four habitats (mean vectors for the variable predictors) were identical using a multiple discriminant function analysis (Manly, 1986). 1 evaluated the hypothesis of equal mean vectors with the Wilks' lambda test (Wilks, 1962). This test converts the value of the test result to a comparable F value and then compares it against a critical F-value at the 0.005 level to decide whether or not to reject the null hypothesis.
I also used canonical discriminant analysis as a confirmatory procedure, which allows differentiation among groups of observations (Owen and Chmielewski, 1985; Arita and Humphrey, 1988). This statistical technique is designed to find variable combinations that allow the maximum segregation among the groups. The canonical discriminant analysis also generates Mahalanobis distance, D2, which measures group dissimilarities: the larger the




24
value of D2 the less similar are the group mean vectors (Barcikowsky, 1983). I used the Mahalanobis distance technique to measure differences among the criterion groups' (habitat areas) centroids. The centroids are group mean values calculated in a single dimension or discriminant function (Tabachnick and Fidell, 1983).
To better visualize the relationships among areas, I plotted the centroids of each study area on a graph (Pielou, 1984; Huck et al., 1974). On the same graph I drew 95 percent confidence ellipses around each centroid (Owen and Chmielewsky, 1985). These comprise 95 percent of the observations for each group of tamarins studied in the multivariate space.
I performed all the statistical analyses using programs in
Statistical Analysis System on a personal computer (SAS Institute, Inc., 1985). Statistical significance was set at p< 0.05.
Results
General Results
Sites were different at highly significant levels (ANOVA, p<0.001) for all variables but one: epiphyte load. Epiphytes like bromeliads are the most important prey micro-habitats for the other lion tamarins (Peres, 1986). For L. chrysopygus this does not occur. The nonsignificant result for epiphytes can be explained by the fact that epiphytes are almost nonexistent in the Morro do Diabo (mean for the four areas = 0.01). In Table 2-2, I present the means for each variable by group and the general summary of these results.




25
Scheffd's multiple comparison test (Scheffd, 1959) was used to compare the means of variables for which significant differences existed. To illustrate the differences in means, I grouped variables into categories within which differences were not significant and assigned a letter code to each (Table 2-2). Differences across categories were significant while within each category they were not.
River area
This site was the most southerly of the areas I studied. It is an alluvial forest adjacent to the Paranapanema river, which influences its structure and composition. Occasional floods create a very conspicuous forest floor composed of many ridges and grooves sometimes filled with a layer of sediments carried in by the river. The area is swampy in many parts particularly close to the river.
The riparian forest at this site is the tallest forest I studied. As a general rule the canopy is even and closed with some emergent trees. It had the lowest mean number of trees per quadrat (mean number of trees = 4.45) but the tallest average height of canopy and largest average diameter at breast height (mean = 23.75 cm). The thick undergrowth consisted largely of seedling and sapling trees, shrubs, and young woody climbers and the density of these created a high understory volume. Vines were present but were not very dense. The River area is strikingly different from the other sites in its absence of ground bromeliads, which made data collection in this area almost a pleasure when compared with the other study areas.




Table 2-2. Mean habitat variables for the four study areas of the black lion tamarin home ranges in the Morro do Diabo state park. Results with different letters indicate statistically significant differences.
River Morro L. Bridge Caldeir=o Habitat Variables (N=262) (N=287) (N=377) (N=249)
Number of trees d 4.45 b 6.25 c 5.50 a 7.07
Number of Tree Species b 3.54 a 4.1 a 3.67 b 4.28
Mean DBH cm a 23.75 b 17.68 b 18.97 c 17.43
Mean Tree Height m a 8.69 c 7.68 b 7.64 b 7.60
Canopy Height m a 2.13 ab 2.01 b 1.99 c 1.76
Epiphyte Number a 0.06 a 0.12 a 0.10 a 0.10
Mean# trees of DBH .00-.80 cm ab 11.06 a 12.35 c 9.39 bc 10.11 Mean# trees of DBH .81-1.55 cm a 1.20 a 1.26 b 0.87 c 0.28 Mean# trees of DBH1.56-2.35 cm a 0.63 a 0.56 a 0.48 b 0.26 Mean# trees of DBH2.36-3.20 cm b 0.27 a 0.46 b 0.23 b 0.24 Ground Bromeliads Density a 0.00 c 0.46 a 2.30 b 1.54 Vine Density a 2.14. b 1.57 ab 1.82 c 1.11
Vine Entanglement a 1.00 a 0.85 a 0.89 b 0.59
Humus b 2.04 c 1.77 bc 1.82 a 2.32
Canopy Cover a 2.96 d 1.94 b 2.75c 2.30
Understory Volume a 3.25 a 3.06 b 2.71 a 3.14




27
When examining the 10 most common trees in the area, I found that they fell into 8 families, accounting for 67.8 percent of all species (Table 2-3). This gives the River area a tree composition quite distinct when compared to the Caldeir&o area at the other extreme of the park. These two areas shared only three of their 10 most common species (Table 2-3).
Morro area
The Morro area is located south of the east-west highway that bisects the park almost in front of Morro do Diabo, the hill after which the park is named. A 50 m strip of forest with trees of reduced height created an edge effect close to the highway. Outside of that strip of forest, the Morro trees had an average height higher than those at Little Bridge and Caldeirdo. The area has a large number of species per ha (mean = 4.1) and their average dbh was between the averages for the River and the Caldeirdo areas. The canopy cover is the most open among the four areas. Vines are present and dense but not as dense as in the River area. In the Morro area, the 10 most common species represented 78.7% of the total tree species found in the area (Table 2-3). Five of these 10 were in the Myrtaceae family and the rest were distributed among the Protaceae, Palmae, and Apocynaceae.
Little Bridge area
Northwest of the Morro area is the Little Bridge area, located on the north side of the highway. It has a xeromorphic aspect, characterized by many plants with adaptations to dry habitats, such




Table 2-3. The 10 most common tree species in the four study areas.
River Morro
Species Family % Species Family %
Lonchocarpus leucanthus Leguminosae F. 13.64 Eugenia uvalha Myrtaceae 20.00
Croton floribundas Euphorbiaceae 10.30 Psidium sp. 1 Myrtaceae 19.30
Chrysophyllum gonocarpum Sapotaceae 7.40 Nectandra salina Lauraceae 12.50
Jaracatia spinosa Caricaceae 7.30 Myrcia sp. Myrtaceae 10.80
Chrysophyllum sp. Sapotaceae 5.90 Enidlicheria paniculata Lauraceae 6.60
Gallesia gorazema Phytolaccaceae 5.50 Eugenia sp. Myrtaceae 5.60
Nectandra salina Lauraceae 5.00 Syagrus romanzoffiana PaIrnae 4.80
Campomanesia sp. Myrtaceae 4.50 Sebastiana serrata Euphorbiaceae 3.10
Psidium sp. 1 Myrtaceae 4.30 Myrciaria sp Myrtaceae 2.80
Aspidosperma polyneuron Apocynaceae 4.00 Eugenia sp. Myrtaceae 2.60
Other 46.32 Other 22.97
N.)
00
Little Bridge Caldeirho
Species Family % Species Family %
Psidium sp. Myrtaceae 31.80 Psidium sp. Myrtaceae 27.20
Eugenia uvalha Myrtaceae 11.00 Eugenia uvalha Myrtaceae 21.80
Myrceugenia ovata Myrtaceae 9.20 Nectandra salina Lauraceae 6.20
Nectandra salina Lauraceae 7.80 Syagrus romanzoffiana Palmae 4.40
Enidlicheria paniculata Lauraceae 4.50 Enidlicheria paniculata Lauraceae 3.40
Eugenia sp. Myrtaceae 3.20 Myrciaria sp. Myrtaceae 2.90
Roupalia brasiliensis Proteaceae 3.10 Helietta longifoliata Rutaceae 2.60
Syagrus romanzoffiana Palmae 2.90 Myrcia sp. Myrtaceae 2.60
Aspidosperma polyneuron Apocynaceae 2.70 Aspidosperma polyneuron Apocynaceae 2.30
Myrciaria sp. Myrtaceae 2.50 Roupala brasiliensis Proteaceae 2.30
Other 30.91 Other 33.50




29
as ground bromeliads and cacti. This site has the highest density of ground bromeliads (2.30 mean percentage per quadrat; Table 2.2).
It has the third lowest mean number of trees but on average contains the same number of tree species as the River and Morro area (mean number of species per quadrat = 3.67). Trees here are similar in height to the Morro and Caldeirdo areas. The canopy cover is dense but broken by the presence of some small islands of "cerrado" (savanna type) especially in the western part of the area. The Little Bridge site has a high percentage of Myrtaceous trees (4 out of the 10 most common species), representing 54.5 percent of the trees in the area (Table 2-3). The understory vegetation at the Little Bridge area is not very thick. However, a large number of ground bromeliads evenly cover the area, together with the second highest number of vines and lianas and the presence of many palm trees gives the Little Bridge site a physiognomy of a typical "cerrad.o" (rough savanna type).
Caldeirdo area
Located in the northwestern region of the Morro do Diabo park, Caldeir~o is the area with the highest mean number of trees but trees with the smallest average height and the lowest dbh (mean height = 7.6 m and mean dbh = 17.43). In this area the 10 most common species made up 88.1 percent of its total number of individual trees. The Myrtaceae is also heavily represented, with six species and 61.1 percent of individual trees (Table 2-3). Despite the high mean number of species in the area, the dominance of the Myrtaceae family, associated with dense understory cover, gives the




30
Caldeirdo a very homogeneous physiognomy. The Caldeirao site has the lowest vine entanglement and density of the study areas contributing to its homogeneous appearance. The uniform aspect of the area is exacerbated even more by the low and even canopy height with its few emergent trees.
Results of the Multivariate Analysis of Habitats
The result of Wilks' lambda test to evaluate the hypothesis of the equal mean vectors for the four study areas was 44,806, which is comparable to an F ratio of 22.8 with 45/3440 degrees of freedom. The probability of obtaining an F ratio this large by chance is less than 0.0001. This result permits rejection of the null hypothesis, which predicted habitat uniformity and strongly suggests that habitat as measured by 16 variables differs significantly among the study sites.
The results of the canonical analysis showed that the greatest habitat difference is between the River and Caldeirdo areas. The other two areas do not differ significantly from one another or from the other two areas. The differences between the four study areas obtained from the discriminant equations are best visualized in a projection of the centroids for each group in the first two canonical variable spaces. The centroids are the points that correspond to the two mean discriminant scores of each group (distance and height). In the same graph I present the Mahalanobis distances between the centroids of each area (Fig. 2-5).
I also constructed a 95 percent confidence interval ellipse for each area. I used this as a graphical representation of the scatters




31
2
~Little oL 1.9 271ie
0-.
C:
U 1
0 0
M
-2
-2 -1 0 12
Canonical Function 1
Figure 2-5. Projection of the four habitat areas in space of the first two canonical variables. The ellipses represent 95% of the confidence interval for the habitat of each group. The lines and numbers connecting the centers (centroids) of the ellipses are the Mahalanobis distances.




32
of scores of each group in a bivariate plot of the first two canonical variates. Based on my observations of the habitat of black lion tamarin inside and outside Morro do Diabo, I am assuming that the four distinct habitats types at the park cover most of the range of habitat types occupied by black lion tamarins. Under this assumption, the four study groups' 16 variables dimensional space, obtained in combining the four groups ellipses (Fig. 2-5), would be statistically analogous to the hypergeometric model of the ecological niche proposed by Huchinson (1965). Assuming no difference in food availability the resulting space if compared with other forest fragments will facilitate the future location of reintroduction and translocation areas for the management of the species.
Discussion
The results of this aspect of the study suggest that there are significant habitat differences among the areas used by the four black lion tamarin groups I studied in the Morro do Diabo State Park. These differences are in accordance with the previous botanical and forest inventories conducted in the Morro do Diabo State Park, which showed a meaningful variability in the forest composition in different areas of the park (Campos and Heinsdijk, 1970; Baitello et al., 1988). The Scheffe test results indicate which habitat variables contribute most to these differences (Table 2-2). Among the 16 variables measured, only 1, epiphyte number, showed no significant difference among all the four areas. All other variables were significantly different among two or more of the areas. At this




33
point, I do not have enough data to conclude if any one of the four study areas is either an optimal or at least a preferred habitat for black lion tamarins but they certainly look dramatically different. These differences are clearly supported by the statistical analysis. Therefore, it seems safe to conclude that black lion tamarins, can live in many different types of forests. These results show that many of the suppositions about tamarins preferring either secondary forest, edge habitats or mature primary forests are not completely accurate (Rylands, 1982; Coimbra-Filho and Mittermeier, 1977; Sussman and Kinzey, 1984). The results of this study also give enough evidence to conclude that black lion tamarin is a generalist that is able to live and reproduce across a range of different forest types. This conclusion certainly has an important impact on the conservation management of the species, as I will demonstrate in the following chapters.




HABITAT DIFFERENCES AND THEIR EFFECTS ON FEEDING AND FORAGING ECOLOGY
I ntrodu ct ion
In the last 25 years it has became apparent that social behavior in primates is influenced by habitat, which produces differences even between populations of the same species (Crook and Gartlan, 1966; Eisenberg et al., 1972; Chivers, 1986). This may be due to the fact that primates and other relatively long-lived animals are exposed to change on different time scales than organisms living short periods of time. As a consequence of their long life, some primate species naturally experience seasonal and annual variation in the availability of resources during their lifetime and this may allow them to learn from individual experience about ranges of variation and outcomes of responses (Clark, 1991). Throughout life, the change in demographic and ecological conditions give individuals the ability to promote their survival and reproductive potential depending on that store of knowledge and many times on the knowledge of their group members (Pereira and Altmann, 1985). Furthermore, this constant exposure of primates to frequent novel and rapidly changing conditions may then select for "flexibility" (Fagen, 1982). Flexibility is the attribute that allows animals to vary short-term behavior in the face of
34




35
change and new information, largely through processes of learning (Clark, 1991). Ecological and behavioral flexibility is then of vital importance for the survival of a species in a world of frequent environmental shifts.
My major goal in this dissertation is first, to document intraspecific ecological variation in social groups of black lion tamarins and then try to relate it to ecological changes. In the last chapter, I demonstrated that the four study groups of black lion tamarins have significantly different habitats. In this chapter I concentrate on a detailed intra-specific study habitat use of the four study groups of L. chrysopygus in the Morro do Diabo State Park. I use the intraspecific comparative method to describe the strategies these four groups of tamarins employ in responding to the different patterns of seasonal and local habitat variation. The understanding of the patterns of social behavior and of habitat use within and between species is important for unraveling ecological aspects crucial for designing conservation management initiatives for a given species (Chivers, 1986).
The major question of this chapter is therefore related to black lion tamarin habitat use: Does the behavior and ecology of groups of Leontopithecus chrysopygus differ significantly in different habitats? To answer this question I describe and compare differences in phenological patterns, time budgets, use of space, diet, feeding and, foraging success among the four study groups.




36
Methods
I started my field work at the Morro do Diabo State Park in
February of 1988. Since the main goal of this study was to compare groups of L. chrysopygus living under different conditions, I selected areas based on their distinct forest cover physiognomy (see chapter 2). I captured the first study group in June of 1988 and by the end of August I had captured six groups in different areas of the park. From these six groups I selected the four target groups for my study.
Methods for Locating the Tamarins
Methods for capturing and using radio-telemetry
On the basis of my previous experience with this species (Valladares-Padua, 1987) I was aware that L. chrysopygus was harder to trap than other primates of the Callitrichidae. Golden lion tamarins for example, can easily be caught in Tomahawk or other traps by baiting them with bananas (Peres, 1986). There are several difficulties in capturing black lion tamarins. They are not easily baited (I experimented with more then 50 types of bait items without success) and they have a large home range for an animal of their size, which makes them difficult to locate even with the help of radio-telemetry devices.
The only capture procedure that was effective was to follow the animals to their den tree where they were caught. Even in this case, experienced field personnel were required to locate the animals and perform the trapping procedure safely (Coimbra-Filho, 1977). In order to locate a group of L. chrysopygus for the first




37
time, I randomly searched the forest in locations where the animals had been previously observed by the park staff. At the beginning of my field work I usually detected them first by sound and then by sight, but by the end of the first year of field work, I was fairly successful in attracting them with the use of play-backs. I have since adopted this method to locate the tamarins (Muckehirn, 1967).
After detecting a group of black lion tamarins, I followed them until dusk when they sheltered in a tree hole (den). If the position of the den allowed a field assistant to climb to it, a platform would be placed near the den in such a way that neither the field personnel, nor the animals would be at risk. If the den was in a position that offered unsafe capture conditions, I would follow the animals for another day until they entered a tree cavity that seemed fit for capturing. The following day at dawn, I would come back to capture the animals.
To capture the animals, I checked the correct position and
shape of the den and closed all other possible exits with pieces of iron mesh. I then measured the den with a scaled flexible stick and filled it with burlap bags. The bags compressed the animals into the bottom of the den. Finally, a hole large enough to fit a human hand was opened at the top of the compressed chamber with the use of a small chain saw. The animals were hand caught one by one and slipped into burlap bags. After the capture a small wooden door was installed to close the saw hole. This door was useful because the lion tamarins occasionally re-used the same dens. If I had to recapture animals I would wait until they entered one of these modified dens. I used this method more than 30 times to capture




38
and/or recapture the animal and excessive stress was not evident on any of these occasions.
Once located, at least two animals in each group were radio
collared. Radios were especially helpful in finding the groups at the beginning of data-collection days or when I lost the whole group. Otherwise, radios were not usually needed as habituated groups were relatively easy to follow.
The first time a group of lion tamarins was captured, the group was processed at the field headquarters in the park's administration area. This procedure consisted of anesthetizing the animals to obtain individual weights and measurements. For this purpose I modified a protocol used for the golden lion tamarin project (Appendix B). Each animal was also tattooed with an individual code. Additionally, either a radio transmitter (Telonics, Inc. and Wildlife Materials, Inc.) or a bead-chain collar with different colors coded for individual recognition were placed on each tamarin. The animals were released the following morning where they had been captured.
Home range maps and trail grids
From the day I first captured a group until the day I started to collect data on a systematic basis, all radio-collared animals were followed at least once a week with two major objectives. First, I marked location points on a map of the area which allowed me to get an estimate of their home range. Second, I wanted to begin habituating the groups to my presence. This phase lasted from August to December of 1988. During the first days of November of 1988, 1 had finally marked enough points on maps to be able to get a




39
first approximation of the size and location of their home ranges. I set up transects throughout the four study areas using a 50 m x 50 m trail grid. This process lasted until June of 1989.
Trails were opened at all areas in which the tamarins had been seen. However, I later found out that the first six months of observation did not include the entire home range of the tamarins, and many expansions of the original grid system had to be made as the systematic data collection progressed.
To identify the tamarin's positions and movements in their range, I labeled a tree in each corner of each 50 m x 50 m quadrat with coded aluminum tags. This system allowed me to recognize with precision and ease the group location when collecting behavioral and ecological data.
In July of 1989, 1 finally finished the grid system, had all the quadrats labeled, and had the animals habituated to my presence. I considered a group to be habituated when all its members would feed normally at a two-meter distance from the observer (Richard, 1978). I had spent eighteen months of work and opened more than 270 km of trails before I could finally start the systematic collection of data on the four study groups.
Methods for Recording and Analyzing Data
Composition and diversity of resources
To measure resource abundance, I developed an index based on local vegetation in each of the four study group's home ranges. I used the same systematic grid sampling technique I used to measure habitat variables (Chapter 2) to estimate the available plant




40
resources for the primates. In each of the 10 m x 10 m plots inside the quadrats, in an area roughly corresponding to the home range of each group, I counted and identified all trees over 10 cm dbh. This included a total of 17,582 surveyed trees. To examine differences between home ranges, I used the biometry and taxonomy of the surveyed trees to calculate a series of means and percentages that allowed for a structural as well as taxonomic description of the four different home ranges.
Phenology
Transects for phenological data collection were established in each of the four study areas. All transects for this purpose followed existing east-to-west trails but were spaced 300 m apart to cover most of the home range of the study groups. Due to transect sizes and the constraints of time imposed by a multi-group comparative study, I used sample swaths every 50 m (NRC, 1981). In these swaths I tallied all trees that occurred within 5 m of the trail (10 m x 10 m) with dbh greater than 10 cm. The diameters were measured to the nearest cm. Where buttresses interfered with the measurement, the section of the trunk between them was measured, summed and added to estimate the portion of the trunk occupied by the buttresses. I ignored the presence of lianas in my tree measurements.
Between July, 1989 and June, 1990, I spent the first week of each month collecting phenological data. Since I did not know then which fruits the tamarins ate, I collected data of all tallied trees. I assessed the phenological state of each tree monthly (n = 12 months




41
for all areas). I checked for the presence and abundance of fruits, flowers and buds. I recorded these abundance based on the percentage of the canopy where they occurred (0-100%). For each study area, I computed monthly means of the proportions of fruit trees which had fruits of all stages of ripeness. By recording the abundance of fruits each month, I compared the estimated availability of food for each group throughout the year. I also conducted a regression analysis to test whether rainfall explained a significant amount of the variability in fruiting.
Time budgets
I started systematic behavioral data collection on the first group of black lion tamarins in August, 1989. The other groups I started later, in subsequent months. For each of the four target groups I recorded data on all animals that seemed to be older than one year of age for at least two full days per month for the period of twelve months. I assigned age classes based on a combination of their weight and body size and on the basis of information from known-age captive animals. I tried to make the observations continuous by watching each group during daylight hours for 48 hours. Whenever data collection could not be done continuously, I compensated by collecting the same amount of missing hours on the next possible day.
Usually data collection started as soon as the first animal left the den tree--this generally occurred between 6:02 h and 7:43 h.I stopped collecting data when the last individual of the group entered their den, which occurred between 16:16 h and 17:31 h.




42
When deciding on systematic rules for recording behavior, two levels of decision must be made: sampling rules to specify which subjects to choose and when, and recording rules to specify how the behavior is recorded. The methods I used were focal animal as sampling rule, and instantaneous or point sampling as recording rule, as described by Martin and Bateson (1986). With focal animal sampling the observer records all instances of individual animal behavior and with instantaneous or point sampling as the recording rule the observer divides an observation session into short intervals and in the end of each sample interval the animal behavior is recorded. Focal animals were chosen randomly using numbers generated by a Hewlett-Packard calculator. Each observation session on a focal individual lasted ten minutes, with a ten minute interval between each session. These breaks allowed me to rest and to locate the next focal animal. During the ten minute sessions, I used sample intervals of one minute and sampled at the end of these intervals. To choose the size of the sample interval, I relied on a combination of common sense and trial and error methods. While habituating the groups, I did a pilot project in which I tried to find the interval that would balance theoretical accuracy against ease and reliability of measurement as suggested by Martin and Bateson (1986).
Minute by minute data on the subject animal were collected in 12 categories. In many cases the categories contained two or more subdivisions. The categories and their subdivisions are listed below:
1) Time of day at which recording was made;
2) Quadrat occupied by the subjects (quadrats were identified
by their corner labels);




43
3) Identification of the subject animal (all animals were
individually marked with collars);
4) Subject's height above the ground in meters;
5) Subject's action divided into discrete categories;
5.1) Inactive--an individual was stationary and not performing
any other activity including resting;
5.2) Moving--an individual was locomoting;
5.3) Feeding--an individual was chewing fruit, exudate or prey;
5.4) Foraging--an individual was searching for fruit, prey, or
manipulating within a particular micro-habitat searching
for animal prey contained;
5.5) Not visible--an individual was out of sight;
5.6) Social--an individual was interacting with conspecifics
of the same or a different group;
6) Substrate--I recorded whether the animal was on the
ground, in understory cover, or on trees, vines or lianas;
7) Object/substrate identity--whether it was on the trunk,
branches or twigs (for substrata less than 5 mm of
diameter);
8) Capture success;
9) Height in the tree;
10) Position in the tree;
11) Nearest neighbor;
12) Nearest neighbor distance.
During the focal period, I also recorded all occurrences of rare
behaviors or events involving the focal animal like mounting or
copulating. Finally, I kept an "ad libitum" record book to describe
unusual events or any other information that I thought important but
that did not fit into the behavior recording forms. I could usually
observe the whole group at the same time, as intra-group
cohesiveness is high in L. chrysopvgus. Maximum distances between
animals of the same group rarely exceeded 25 m. This
characteristic facilitated observation and data collection.
By using a combination of focal sampling and instantaneous or
point sampling I obtained a score of the proportion of all sample
points on which the behavior pattern was occurring for each animal




44
in each study group. Despite being aware that instantaneous sampling does not give true frequencies of behavior, I present the results as frequencies because the intervals I used were short. A series of studies have demonstrated that instantaneous sampling gives an approximation of the proportion of time spent in each behavior sample (Dunbar, 1976; Leger, 1977; Rhine and Flanigon, 1978).
Based on these methods, a total of 6,436 occurrences of
behavior were recorded for the River group, 6,440 for the Caldeirao group, 6,543 for the Morro group and 7,042 for the Little Bridge group. These occurrences sum up to a total of 24,461 records for the whole study period, and were derived from a total contact time of 815 hours on 96 sampling days (8.49 h/day).
As for the intra-specific comparative analysis of behavior actions, I used two methods of analysis of variance (ANOVA, MANOVA) to simultaneously evaluate the effects of suites of independent variables (groups, and months of the year) on activity budgets (Neter et al., 1985; Kinnaird, 1990). The assumptions of the analysis of variance were met by an arc-sin transformation of the proportional data (Sokal and Rohlf, 1981).
Movements and use of space
There are many approaches to defining the size of a homerange area (Harvey and Barbour, 1965; Ludlow, 1986; Rudran, 1978; NRC, 1981). The number of quadrats entered by each group at the time the curve reaches an asymptote is considered in some studies as the home range for the species (Rylands, 1982; Peres, 1986).




45
However, I estimated the home range size of black lion tamarins in a more conservative way, including all the quadrats that were within the perimeter of the quadrats they used. In each home range I also recorded all the den trees used by each group during the year.
To examine the intensity of use of black lion tamarin range, I ranked each quadrat according to the number of times it was used by the group. To examine differences among use of quadrats by each group, I compared monthly mean use using analysis of variance. If the differences were significant I used the Waller-Duncan K ratio to compare the means for which significant differences existed. I also tested the distribution of sightings throughout each group's range against a Poisson distribution to determine if use of space was random.
I tested den distribution in the home range space using the relationship between the mean and variance of the number of dens per sampling unit (Ludwig and Reynolds, 1988). By this method when the variance is equal to the mean the distribution has a random pattern, when is bigger has a clumped pattern and when is smaller a uniform pattern. The choice of a scale is important and patterns can be detected in one scale and not at another (Pielou, 1979). 1 used a 64 quadrats (8 x 8) as sample unit since with smaller scale I could not detect any pattern.
Ranging behavior
To record animal movements I used the same 50 m x 50 m grid system that was used to measure habitat differences. The aluminum tags placed in each quadrat corner allowed the observer to check his




46
position while following the group. Group locations were recorded at each instantaneous sampling and all locations were plotted on graph paper maps. I obtained daily path lengths by direct measurement of these maps with a manual odometer. In the analysis of rates of movements, I considered the straight-line distances between the central point of each quadrat. This produces a relatively conservative distance estimate, because animals rarely move in a linear or uni-dimensional way (Peres, 1986). Only complete day samples were used in this analysis.
Feeding and foraging behavior
Many different methods are available to quantify primate diets (for review see Clutton-Brock, 1977). Due to the comparative nature of my study I measured the diet composition of black lion tamarins by recording the proportion of feeding and foraging time spent on different food items (Clutton-Brock, 1975; Richard, 1978). This method seemed to be very appropriate for this type of study since it is easily repeatable and is not subject to major biases due to differences in recording methodology (Clutton-Brock, 1975; Struhsaker, 1975). I used five pre-established behavioral categories to cover the many different components of food acquisition: feeding on animals; feeding on fruits; feeding on exudates; foraging on animals and manipulative foraging. These categories are almost all self explanatory with the exception of manipulative foraging which involves searching for food through movements such as probing woody crevices, grabbing, biting and/or turning over decomposing materials.




47
Plant items in all feeding records were identified to genus or species level. The tamarins' feeding plants are relatively large, and easy to identify. Whenever I observed a tamarin eating a new fruit, I collected a sample of it and fixed it in 70 percent alcohol. I labeled and boxed all e aten exudates and other miscellaneous items. This small collection of dietary specimens was used to help in the identification of black lion tamarin diet composition.
Recording of animal prey was not as easy. Black lion tamarins prey mostly on relatively small animals and ingest them in a rapid manner so I was rarely able to recognize the taxonomic group. Whenever possible I collected animal prey fragments dropped by the primates and conserved them in 10 percent formaldehyde. By the end of my study period I had a total of 22 fauna components of the tamarin's diet. Unfortunately, the museum to which I sent them to be identified has lost them. Nevertheless, the prey animals I listed were those I could identify.
I measured capture success through direct observation. I also described the item and the substrate from where it was taken. As suggested by Robinson (1986) for animal prey, a "capture" included discovery of an item but not necessarily its immediate ingestion. This is a more reliable approach to the measurement of capture success. With this method I avoided the bias of a prey being successfully captured during the point sampling but only ingested in the intervals between samplings. Furthermore, with this method I avoided misinterpretation of the feeding item taken during the point sampling.




48
I compared feeding behavior within and between groups. A chisquare test compared the proportions of the three major diet items eaten by each group every month. Since fruits constituted the bulk of their diet, I used two methods to distinguish the diet of the four groups in time and space. The first was the percentage of feeding records of the most consumed fruit items by month and the second was the Renkonen index of percentage similarity (Renkonen, 1938). This method despite its simplicity is one of the best quantitative similarity coefficients available (Krebs, 1989). The index range from 0 (no similarity) to 100 (complete similarity) (Wolda, 1981).
I used analysis of variance methods (ANOVA, MANOVA) to simultaneously evaluate the effects of suites of independent variables (group and month of the year) on feeding and foraging activity budgets (Neter et al., 1985; Kinnaird, 1990).
Results
Abundance and Distribution of Resources
Variation with location
The amount of data collected on plant resource abundance was slightly different at each site due to differences in home range sizes. For the River area, for example, I recorded 3,231 individual trees in a total sample area of 6.65 ha. The Morro area had 4,907 trees in 7.49 ha, the Little Bridge had 5,034 trees in 9.4 ha and finally the Caldeir.o area had 4,320 trees in 6.89 ha (Table 3-1).
I found that some areas were quantitatively and qualitatively different from others. The River area not only had the smaller




49
number of trees, but also the lowest occurrence of feeding trees per ha (303 frt/ha where frt = trees of spp from which the tamarins ate). The highest occurrence of feeding trees per ha was in the Morro area (534 frt/ha) followed by the Caldeirdo and the Little Bridge areas with 506 frt/ha and 442 frt/ha respectively.
Despite its small number of trees, the River area had the
largest diversity of trees with 64 species in 30 families while the Caldeirdo had the smallest diversity with 44 species in 21 families. The other two areas ranged between these two.
The average tree height and dbh were also found to be higher in the River area (8.42 m and 74.59 cm respectively) when compared to the height and dbh of trees in the other three areas with 7.66 m and 54.74 cm for the Caldeirdo area, 7.58 m and 59.56 cm for the Little Bridge area, and 7.12 m and 55.12 cm for the Morro area. (see Table 3-1).
These comparisons were useful to furnish information on the black lion tamarins' environments as their sources of food and shelter. The River area for example, has a forest composition quite different from that of the other study areas. The other three areas have a smaller number of tree species and a larger number of shorter and thinner trees. In the River area, the ten most common species belonged to seven families and only one of these species was on the list of the ten most eaten fruits by L. chrysopygus there. The Caldeir&o area had its ten most common species from only three families, of which five species were from the Myrtaceae. Six of these tree species were among the ten most eaten fruits by black lion tamarins.




50
Table 3-1. Characteristics of each study area quantifying availability of resources.
Table 3-1. General characteristics of each study area considering availability of resources
AREA RiVBR CALDEIRAO MORRO L. BRIDGE
Number of Quadrats 665 689 749 940
Hectares 7 7 7 9
Number of Families 30 21 27 26
Number of Species 64 44 64 55
Number of Trees 3231 4320 4907 5034
Number of Feeding Species 32 24 33 24
Number of Feeding Trees 2017 3487 4000 4156
Average Height (in) 8 8 7 8
Feeding Trees per Hectare 303 506 534 442
Average D.B.H. 24 17 18 19




51
The Morro and the Little Bridge areas had their ten most
common species from six families. In the Morro area, five of these common species were on the list of the most eaten species, while the Little Bridge group only ate fruits from four of the ten most common species. Additionally, the ten most common species of each of the four study areas are reasonably different (Table 3-2). The River and Caldeirao areas share only two of the ten most commonly eaten fruit species and these two were in very different ranking order of abundance. The other two areas are more similar to each other although they also vary from the former areas (Table 3-2).
Phenology
To examine the annual variation in fruits available to the four groups of black lion tamarins, I plotted the number of species that had ripe fruits during the twelve months of my study (Fig. 3-1 to 34). This fruit availability was poorly correlated with rainfall: In only the Little Bridge area was there a significant correlation between the average of rainfall per month and the number of tree species with fruits (rs = .69 p<.005). The Morro and the Caldeirao areas had low positive correlation (rs of .19 ns and .33 ns) respectively. The River area was the only area with a negative correlation, although this correlation was low (rs = -.22 ns).
Phenological data reflected the same general patterns of
differences in number and taxonomy of trees among the four areas.




52
Table 3-2. Ten species on which each study group spent the most
time and percentage of total time spent in trees of each fruit
species.
Food species Time
feeding (%)
RIVER GROUP
Myrceugenia ovata 15.5
Syagrus romanzoffiana 10.2
Campomanesia sp. 8.9
Helietta Iongifoliata 7.6
Cabralea canjerana 5.9
Myrcia sp. no.2 4.8
Celtes spinosa 4.1
Vochysia tucanorum 3.3
Zygocactus sp 2.8
Psidium sp. no. 2 2.5
CALDEIRAO GROUP
Myrcia sp. no. 2 30.0
Psidium sp. no. 1 13.9
Syagrus romanzoffiana 10.4
Xylopia brasiliensis 6.7
Eugenia sp. no. 2 6.3
Eugenia uvalha 5.7
Myrciaria sp. no. 1 3.1
Trichilia pallida 2.9
Myrceugenia ovata 2.9
Sebastiana serrata 2.0
MORROGROUP
Eugenia uvalha 10.9
Syagrus romanzoffiana 10.9
Xylopia brasiliensis 10.7
Myrcia sp. no. 1 10.3
Eugenia sp. no. 2 5.8
Ficus enormis 5.3
Vochysia tucanorum 3.8
Myrceugenia ovata 3.8
Psidium sp. no. 1 3.8
Myrcia sp. no. 2 3.1
LITTLE BRIDGE GROUP
Eugenia sp. no. 2 22.3
Syagrus romanzoffiana 19.4
Myrcia sp. no. 1 14.6
Campomanesia sp. 5.6
Cereus sp. 3.9
Terminalia sp. 3.7
Myrceugenia ovata 3.7
Psidium sp. no. 1 3.4
Helietta Iongifoliata 3.1
Anadenanthera falcata 2.8




53
20
18 t 5 16 .2- 14
-C
12 .T 10
8
0.
S6
E
-, 4
z
2
J F M A M J J A S 0 N D Months
Figure 3-1. Fruit availability to black lion tamarins in the River
Area group measured by number of species carrying ripe fruit each
month.




54
2018 16D 14
12
-10-
0
U) 8 D6
E
z 4 -',
2
0 I I I I II
J F M A M J J A S 0 N D Months
Figure 3-2. Fruit availability to black lion tamarins in the Morro
Area group measured by number of species carrying ripe fruit each
month.




55
20
218 .16
. 14
12
U,
- 100 8
6
E 4 +'
Z 2
0
F M A M J J A S 0 N D Months
Figure 3-3. Fruit availability to black lion tamarins in the Little Bridge Area group measured by number of species carrying ripe fruit each month.




56
20
~ 18 16
C)
14
124
2
0 l I l
J F M A M J J A S O N D Months
Figure 3-4. Fruit availability to black lion tamarins in the Caldeirdo Area group measured by number of species carrying ripe fruit each month.




57
More than 50 percent of the fruiting trees in the Morro, the Little Bridge and the Caldeirdo areas were members of the Myrtaceae (average = 62.3 percent; range = 54.2 percent to 76.2 percent). In the River area, only 18.5 percent of the trees that fruited belonged to the Myrtaceae. In Morro do Diabo, trees from the Myrtaceae usually produce fruits in the rainy season, from October to February (pers. observ.).
Time Budgets
In the next sections of this chapter I will address a
comparative investigation on how the four studied social groups of black lion tamarin use their habitat. Primates customarily do not use their habitat in a uniform way and in order to compare the ecology of the four study groups of black lion tamarins it is important to examine their activity budgets and patterns (Chivers, 1986).
Overall black lion tamarins spent the majority of their time (53-70 percent) in inactivity (defined as stationary and resting categories combined); social behavior on the other hand, was engaged in for only a short amount of time (0.2-0.5 percent) (Fig. 35). The four groups showed rather similar time budgets. The MANOVA failed to show significant effects of all the independent variables on group membership. The Wilks lambda test was calculated to be 0.38.




58
10090
80- 0 RIVER
70- 0 CALDEIRAO
S60
HMMORRO S50
- L. BRIDGE
_ 4030
20
10
EAT FOR MOV NVIS INAC SOC
Behavior Action
Figure 3-5. Time budgets of each of the four study groups. (Summed across the entire year).




59
This is equivalent to an F ratio of 1.15 with 33 degrees of freedom. The probability of obtaining an F ratio this large by chance is less than 0.28. The ANOVAS analysis also failed to show significant effects except for inactivity (F= .64, .39, .01, 1.91 and .10 for feeding, foraging, moving, inactive and social, respectively, df=4, p>0,005) (Fig. 3-32). These results indicate that despite the differences in habitat by area, statistical tests failed to show differences among the four groups in their activity budgets for the six examined behavior categories. However, there was a significant difference in monthly variation in the time budgets of the four study groups of black lion tamarins' (Table 3-3). 1 use the term local variation to express the differences observed among the study groups at Morro do Diabo. I have not analyzed time budgets based on individuals or their age or sex classes. However, according to my observations if there were differences, these were not striking.
Movements and Use of Space
Home range use can be described as falling into 4 categories: uniform coverage, focused on shifting patches, concentrated at the center or at the periphery, or around the periphery of the area (Terborgh, 1983). Each pattern reflects the influence of a dominant factor in the life of the species: for some species that factor is the structure of the habitat, particularly the vegetation (Peres, 1986). For other species it can be social interactions. In this section I examine how do the four study groups-of black lion tamarin's use of space differ as a function of habitat?




60
FFIEEDDCC'BBAAZ XV[IU!IT IS 1 R 10Q P OINA L JIA !H;G!F!E D 1C11BA'
-101 __ _ _ __ _ _ _ -10
____ ____ ___-8
-61 -6-1 1 1> -413
-1
22 1 ____ 5__ 2
26!
27'7 29 9___13 1 13___1411 32 1 15
33 71111
34' 17I
1___ AA8IT ~ PONML. GF11
19 19
Figur 3-.Mvmn2fth0ie ragopuigtesuy




61
1 2 i 3 4 51 6 7 8 19 11l1 12 1314161617 18 19 2012112 223 2 4252 62 728291 301313 3213 33 4 363 613 7 Hill!H
GGIcr
AF A61.
N 4
A ~~~ Z _2 r1 xL6 011 311 679122_''7'2 ~'~ t233 23143,63
Fiur 3 Voeet fteMroAe ru uigtesuy




62
UT 8! OQP ON M IL J I H GF ED C A 8 X11CDiEF G~H1 JJ LL.MM;NN 0OP QO
47i 4___ 47
462____ 46
451 41
44, 44
421 r4 _ _42
41: 1 _ 41
40'-4 39, '600 ] :3______ 39
38 _____ 38
37 ______37
36 3 6 ____35 -___4__35
34 3 4
_________ 33
229
28> 2 8
27 2 7
i 26
_____________20
Figure~~~~~~~~~~~~~~ 3-9Mvmnso h iteBig ragopdrn h
study17




63
A B C ID !E FI7G I777I L ;MN 10 !P 0 R S !T !U !V !X Z AAIBB CCD DO
52 52
51 51
50! 50
491_________ 49
48 4
47 47
46 46
45 45
44 44
41 ___401
39i
367
341
325
31
301 30
29 29
28i2 27 2
26'
25112 24 12
23i
22! 2
221
2022 21
171 21
1519 14:
137 1216
115
83! __3X1
A2 B12 p { AB CD
Fiue310oeet1o h adi~ ragru uigtesuy




64
The means of the daily path length for the 77 full days of the
data collection ranged from 1,362 m for the River group to 2,088 for the Little Bridge group. The rate of movements measured as mean distances traveled by each group, each month is presented in Fig. 310 to 3-13.
Cumulative group movements for all full day observations in each group are presented in Fig. 3-6 to 3-9. For estimating group movements, I also assumed that the lion tamarins moved in straight lines from one point to the next.
Height and substrate
My data indicate that the use of forest heights by the four groups of black lion tamarins did not vary greatly. In all sites they spent on average a considerable amount of time slightly above the middle level of the forest. The average height used by the River group was
8.5 m (S.D. 2.7) while the other groups were between 7 and 8 m (mean height = the Morro group, 7.1 m, S.D. 2.0; the Little Bridge group, 7.6 m, S.D. 2.2 and the Caldeirdo group, 7.7 m, S.D. 2.0). Black lion tamarins rarely go to the ground. During my entire study I only once saw an individual from the River group go to the ground for a few seconds while chasing a grasshopper. Although I have not conducted a full analysis of feeding or sleeping average heights, preliminary results indicate that these activities occurred at all levels and at least in one case, black lion tamarins spent a night below the ground level in a den that was deep in the tree roots.




5000
475045004250? 40003 3750E
E 3500 "O
S3250
O
0
S3000 2750 S2500
6 2250 0
S 2000 -T 1
ccO T S175001500 -Ig ) 1250- l
> O 0 O
1000- O O T
750 13
500- 1
250
I I I I I I I I
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Months
Figure 3-10. Mean daily path length ( S.D.) for the River group during the one
year study.




5000 4750 4500 4250 S4000 S3750 E 3500- .l
3250
0
oT
S3000 8 2750
0
. 2500 O 2250
- 2000 T T o
S1750- El 0
1500 T 1
CD E
500 250
I I I I I I IIIIII
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Months
Figure 3-11. Mean daily path length (+ S.D.) for the Morro group during the one
year study.




5000
4750450042504000(D
E 3750-a 35000) T
0 32503000- 0
o
2750
2500 0
2250- O]
2000 0
S150
cc 1500 l
1250- 0 .
1000 750
500
250
I I I I I I I IIII
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Months
Figure 3-12. Mean daily path length ( S.D.) for the Little Bridge group during
the one year study.




5000
4750
- 4500 S4250E 4000 )- 3750o 3500e 3250- T3 El
S3000
. 2750S25000
2250
C 2000- T I
S1750CD)
< 15001250- T
1000- E O
750
500250
I I I I I I I I I I I I
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Months
Figure 3-13. Mean daily path length ( S.D.) for the Caldeirao group during the
one year study.




69
Home-range areas
The cumulative percentage of quadrats entered by each of the four study groups as a function of observation time will be presented in chapter 4. All the cumulative number of quadrats entered by each group had reached an asymptote by the end of the first year of data collection with the possible exception of the Little Bridge group. The total number of quadrats entered by each group at the time the curve reaches an asymptote is considered in some studies as the home range for the species (Rylands, 1982; Peres, 1986). However, I estimated the home range size of black lion tamarins in a more inclusive way, including all the quadrats that were within the perimeter of the quadrats they used (Rudran, 1978). I sighted black lion tamarins of the River group in 294 quadrats, the Morro group in 342 quadrats, the Little Bridge group in 447 quadrats, and the Caldeir~o group in 318 quadrats (Fig. 3-19 to 3-22). The inclusion of peripheral quadrats increased the home range sizes from .73 km2 to 1.13 km2 for the River group, from .85 km2 to 1.20 km2 for the Morro group, from 1.12 km2 to 1.99 km2 for the Little Bridge group and from .79 km2 to 1.20 km2 for the Caldeirdo group (Fig. 3-14 to 317). Overall, the mean home range size for L. chrysovgua in Morro do Diabo using the first method or only the quadrats they were seen to enter would be 0.87 km2- If one includes quadrats within their range boundaries, the mean home range size at Morro do Diabo was
1.38 km2.




70
FF 1EDDJCCIBB AAZ X V U IT IS R Q P 0ON M IL J 1t H G *F tE iD iC B iAI
-11 -10
___9 ____ -9
-8 ________ -8
-71_______ -7
-61_ _ __ _ -6
-41----------------------------------------------------4
-31
-2 -2
1 3
44
8 8
91
121 12
13 ]----____13
141 14__15! 115 161 1 16----------------I -__
191i 19
201 ______ 20
211 ______ 21
22I 22
23 1j 23
2511 1
261 1111 11 1 1 26
271 2~j 2
281 28
291 1-------------------------------------29
30'-------------------------------------------------------------------321 32
341 34
FHD~CB A UTRQPI ON M L J I IH G !F IE D C!B iA
Figure 3-14. The 1.13 km2 home range of the River group on a 50 m x 50 m grid.




71
12 3:415 6781 911011112 13 14116116171181912012122232425262728291303113233134i3536!37 HH !HH
FF I__cc ___ _i cc
AA: A___ IA
_______ z
X -4
___ v
U r U - ----'-~--*----
B: 11 _N N
M M
Li L
< I H
G~
I, il
23 4 5S 78911123451118901222 2 2627 282913013132:33i34;3S1361371
Figure 3-15. The 1.20 km2 home range of the Morro group on a 50 m x 50 m grid.




72
U T iS R 10OPON IMIL J I HG Jr F ED C B JA AABBCMEIFG-~l JJLIMN'10PI 481 48
4 7:_____ 47
46i 46
451 45
441 44
431143
421 42
411 41
40 40
391 ~1~,39
38, 38
371 37
36C 36
35,i______ 35
34, 3 4
331 33
321 32
311 31
30, 1 - - - 30
291 I29
281 ,28 271 127
26 26
25 25
24,2 231 1 723
221- -2
211 21
201 20
191 1 9
- --1- --7_ 17
16_______ 1 6
1 51 _____ 15
141 ______ 14
13 13
12________ 12
lo0 10
61I 6
51
'lit
VUTSR0P LJI H F E 0 C8 A A A jBCDM FFHHI JJ1LL!M1Ni*4!PPIQ1
Figure 3-16. The 1.99 Ikm2 home range of the Little Bridge group on a
50 m x 50 m grid.




73
A B C D E F G H J L MN OP OR ST U V X ZAABBICCIDDI
52 52
51 51
50 50
49 49
481 48
47 |_ 47
461 46
45 45
44i i 44
431 43
42! 42
41 41
40 40
391 39
381 38
37 37
36 36
35 35
34 34
33 33
32 32
31 31
301 30
29 29
281 28
271 27
26! 26
251 25
24 24
231 23
22 22
21 -------21
201 20
19; 19
181 18
17- ----17
161 -----16
15 15
141 1 F I i 14
4 14 131 ,13
12 12
11 11
10 10
5' 6
9 9
8 I 8
7 7
6 6
5 5
41 4
3 3
2 2
1 t 1 I 1
ABC EF HIJ MN PO ST VX AAIBB CCDD
A IB IC 'D F H J N 0 P O R S T U V Z IDD
Figure 3-17. The 1.20 km2 home range of the Caldeirho group on a 50 m x 50 m grid.




74
Intensity and diversity of quadrat use
A clear variation existed in the monthly use of quadrats
between the River group and the other three groups (Fig. 3-18). 1 found the four group means to be significantly different from each other (F=4.52 df = 3 Pr>F=O.0076). Although the difference was significant, the ANOVA test did not locate where the means differed. For that purpose I used the Wailer-Duncan K-ratio T test as a multiple comparison test. The results showed that the River group used fewer quadrats than the other three groups, which is consistent with the fact that their home range size was the smallest of the study groups (Fig. 3-18).
Likewise, the use of space by each group was not distributed evenly. The River group concentrated much of their time along the perimeter of their range. The Morro and the Little Bridge groups used their space more evenly but still showed some tendency for a greater use of the edges of their ranges. The pattern shown by the Caldeir~o group was distinctly different, showing intense use of certain parts of the periphery of their range, but concentrating mainly in a core area in its center. The patterns for their home range use are illustrated in Fig. 3-19 to 3-22.
1 tested the distribution of sightings throughout each group's range against a Poisson distribution model (Ludwig and Reynolds, 1988) to determine if their use of space was at random. I found the sightings significantly deviated from the expected Poisson distributions for all four groups.




75
60
50- """ ''''''''
. . . . . . . ..ii ii iii .. .. . . . . .. . . . .
:..... ,.. .+ ... ,,'-......-............' ..... ..... .....
'.... ... . . oo. ... .. .. .. ..
20 -~ ~ ... ...............':::""''""'''""''" ...............:::::::::::::::
...... ... ...... ....,.,... .,.... .,...,,...... ......,: .: :. ::: :::::.
...............i!!~i~~i ....:::::::::::::"' " " '. . . .
4 0-~~~~oooo O ,, ..-.......% .o... ...........,.. .. .. .. ..
.- ............ -....... ..:.:::::. :::::.:. : ::: .. ....... ... ...:::: :::::::::::: ::::::
1 0 ..................." " ""::::::::::::::: ...............:::: ::::::::::::::::::
...... .............. :.......... .:.:.: ...............'::' :::::::::::::::::
... .. .... ........ ...,................. ... ... ....:.: .. ..::: :...::...::...::..
S Oo oo.Oo....,o. ...... *.......................... . . .
.......... ...... .......:.:...:.:..., ............:::::::::: .::::::::......
. . . . . . . .. .... ... . . .
............ L.Bridg.Group3.Rive
.. .. . .. .. .. .. .. .. .. .. ..,r.. .. .. .. .
..........1.. nthly......... of.quadrat.by.each.study.group.
Means...... wit ......... ...... .... no.ig iic n ly dff r nt ( ale -D n a
. . . . . . . . . .. . .. .




76
FIF EEDDI B8 AAIZIX V U iT !S R 0 P !0 N !M IL J I H i G F E !D iC B !A
_:10 INUMER UTILIZATION ______ _________-10
68 ______2 -8
71 -7
6 -6
-41 >15-4 31 -3
2 -2
31 3
4 4
6 6
8, 8
1 0 1_0
1021 10
1 31 11 3 1 41 14
15 8
1 9~ 18
201 20
211 ____21
23 2
241 24
25i 25
26 28
29 29
311 31
32i 32
33i %\_3 3
ICC-. B-AA4 IX V IUT IROP 10- A___ IL~ Ba AA1 IM LG !F IC a'
Figure 3-19. Intensity of quadrat use by the River group.




77
1 21 3 41 51 6 7 81 9110111 12 13 14 15 16 17 18119120121122123124125 261 2 8 29 30 31132133134,351,36137, HH I HH
GG NUMBER UTIUZATION GG
EE 2 EE
DD 3 D
cc 4 CC
S SB
R R
Hv
A A4
T i i MP, T
~L
2 4 5 6 9 01 21 41 81 81 02 22 2 2 62 93 132133 34)35 36137!
Figure 3-20. Intensity of auadrat use by the Morro arouo.




78
UT S R 0 !P0 IN IM L J A H G !F ED iC B I A *AA BB CDD IE 46 1 '46
44i 3 44
43 41 43
42: 51 42
40 k 40
39i 3 9
3813
3 7' 37
361 36
35 35
341 A ___________1_34
33 i________33
32 1X\32
311 31
30i 3 0
291 '29
281 28
27; 27
26 126
251 25
24 0 24
23'2 3
U2 T22Ro LJIHGFE D ABcID~~H~I .. LMI~0W
Fiur 3-1 Inest fquda2s1y h iteBrdegop




79
A ;B C !D IE :F IG H :II J L A IN 10 P Q ,R !S !T U V X AA'B !CC DD
5 2 52
51 !NUMBER UTLiz~llONi 51
50i____________ 50
491112 49
48 3148
471_____4___ 47
S46
45
44 44
431 43
42: 42
41: 41
401 X 40
39i 39
381 38
37T____________ 37
36'_____ 36
3 5;1-\X 35
34!_____ 34
33IIII _____ 33
32,___ 32
311 31
30! 30
29 29-24 4
231 23__22' 22________21.2 204 2
18' 1
17 1
161 1
151 15___ ___________141 1
131 1
12 1
10 10
91 2
8 8 1
6 161 __
5'
41 4
2 [ I2
A B !C ID E IFIG HI aJ L NST u ly Z7 AAlBB'CCDIM[
Figure 3-22. Intensity of quadrat use by the Caldeirao group.




80
50
45
40
35
30- 0
20
LL........
15 0iiiiiii
2 0 ...... .......
0 ....... :
0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ Number of times quadrats were used
Figure 3-23. Frequency distribution of intensity of quadrat use by the River group. Diamonds indicate the values expected if quadrat
occupancy were random (following a Poisson distribution).




81
50
45 40 35
0
30
:25 0
LL ...
20 : : .:.: :**:*
......::::::::ii~i .......
159 .......
Num... .......
Figure3-24.Frequnumbeisriofties oinetofquadrats wereuse
the Morro group. Diamonds indicate the values expected if quadrat
occupancy were random (following a Poisson distribution).




82
50
45
40 .
0
35 30
0
25
0
LL 20
15 10
0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ Number of times quadrats were used
Figure 3-25. Frequency distribution of intensity of quadrat use by the Little Bridge group. Diamonds indicate the values expected if
quadrat occupancy were random (following a Poisson distribution).




Full Text

PAGE 1

THE ECOLOGY, BEHAVIOR AND CONSERVATION OF THE BLACK LION TAMARINS (Leontopithecus chrysopygus, MIKAN, 1823). By CLAUDIO VALLADARES PADUA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1993

PAGE 2

Copyright 1993 by Claudio Valladares-Padua

PAGE 3

To my beloved family, Suzana, Andre, Filipe and Joana.

PAGE 4

ACKNOWLEDGMENTS I would like to express my gratitude to Dr. Kent Redford and Dr. John Robinson, my advisors and director and former director of the Program for Studies in Tropical Conservation of the University of Florida. Dr. Robinson was a source of inspiration and support during my master's studies and the first years of my Ph.D. work. His departure to another institution was a loss to many of my colleagues, to me, and to the University of Florida. I am fortunate to I have had the privilege of having a second advisor who is as inspirational as Dr. Robinson. Dr. Redford has become one of my most precious friends and his support has been more than I could state in words. Nevertheless, I would like to express my profound gratitude for his continuous and inexhaustible support. Our friendship will always be a reason to look back with joy to my years as a doctoral student, despite the stressful moments that come with completion of the project. My appreciation goes to Ors. Eisenberg, Brockmann, Judd, and Sunquist, members of my committee, for their guidance and support. As a Latin American, I would like to express my appreciation to the Program for Studies in Tropical Conservation at the University of Florida. This program is undoubtedly the major North American program for the preparation of professionals in the area of tropical conservation. Professionals who graduated from this program have iv

PAGE 5

already decisively contributed to conservation in their own countries. To be part of this program is really a privilege. I have come to admire and truly cherish Laury Cullen Jr. We have shared wonderful moments together. He has always been a hard worker and a source of joy and enthusiasm. Several other people were responsible for this work. I would like to thank Sr. Jose, Homero, Preto, Zezinho, Serginho, Sr. Arnaldo, and all other staff of the Morro do Diabo Park. At the lnstituto Florestal de Sao Paulo, I am deeply in debt to Marco Antonio Garrido, who has become a true friend, as well as Sandra Pacagnella, Helder Faria, Giselda Marlene, Francisco Serio, Helio Ogawa, Francisco Kronka, Regis Guillaumon, and Genji Yamazoe. Their personal and institutional support was crucial for me to develop my work. My sincere appreciation goes to Prof. Adel mar Coimbra-Filho, who helped me find my track in the primatological world. will always be grateful for his support. Pissinati and my other colleagues at the Rio de Janeiro Primate Center have also always been supportive to my studies, and I thank them for this. Dr. Russell Mittermeier supported and encouraged this study, and helped me throughout my career in the conservation area. I received support from the community of Teodoro Sampaio, especially the CONDEPRO group, and I am thankful to them. To all the CESP personnel I offer my sincere thanks, especially to Antonio Audi, Francisco Guerra, Luis Fernando Galli, and Carlos de Lucca. The herbaria at the University of Sao Paulo at Piracicaba and at the lnstituto Florestal de Sao Paulo identified the botanical material for my study. I thank them for their input. V

PAGE 6

In the conservation world, I received unlimited support from Bill Konstant, Admiral Ibsen Camara, Fai9al Simon, Sonia Rigueira Lou Ann and Jim Dietz, Devra Kleiman, Beata and Bejamin Beck Jeremy Mallinson, and Garry Eberhart. Their conservation efforts have made me feel that all the struggles in this field were worthwhile. Hope and Bob Stevens and Jon Ballou have been true sources of encouragement and inspiration. I admire them greatly Colleagues and friends at the University of Florida were an invaluable source of help. Monique Chacon, Pierre Buck Berner Damian Rumiz, Jim Ellis, Alexandro Grajal Gary Shaeff, Rosa and Joao Paulo Viana Peter Crawshaw, Laurenz Pinder and friends, colleagues, and staff from the Tropical Conservation and Development Program and the Department of Wildlife At the Center for Latin American Studies, I would like to especially thank Terry M'Coy for letting me stay so long with my "extended family. I have made great friends at the Center whom I will never forget. Some families were also extremely supportive: the Fonsecas, the Johnsons, Limas, Principes Maias, Robinsons and the Redfords. Suzana, Andre, Filipe and Joana encouraged and supported me in ways impossible to describe. It must have been very difficult to tolerate my frequent moments of irritation, stress associated with long periods of absence. My studies at the University of Florida were funded by an Overseas Scholarship from the Brazilian Council of Science and Technology-CNPq. My field work in Sao Paulo was funded by the Fanwood Foundation the World Wildlife Fund-US the Jersey Wildlife Preservation Trust, the Wildlife Preservation Trust International vi

PAGE 7

the Conservation International, Whitley Animal Preservation Trust and the Institute Florestal de Sao Paulo. From the University of Florida I received support from the Program for Studies in Tropical Conservation, the Tropical Conservation and Development Program and the Center for Latin American Studies. The Internet Computer Network helped in transferring data and in my communication with people who are dear to me. This helped reduce the solitude I felt during my last months in the United States vii

PAGE 8

TABLE OF CONTENTS ACKNOWLEDGEMENTS ............................................. .. .. ................. .. ................... ..... i V LIST OF TABLES ............................. ........... .... ..................... .. ........... .. .. .......... .. .... .. x i LIST OF FIGURES ........... ... ............................. ........... ....... ............. ... ................. ..... x ii ABSTRACT .... ... .... .. ..... .. .. ................ .......... .......... .. ... .. .. .... ......... ... .. .......... ..... ... XV i INTRODUCTION .... ... ...... .................. .... .... .. ............ ........ .... ...... ... ..... .. ... ............. ... 1 The Black Lion Tamarin ...... .. ..... .............. ..................... ..... .. ....................... ... 1 Aims of the Study .................... .. .. ................. ... ...... ... ... .. ..... .... .. .............. .. ..... . 7 THE MORRO DO DIABO STATE PARK AS A HABITAT FOR THE BLACK LION TAMARIN ............. .... .. ............... ... ....... ... ..................... ... ... ... ...... 1 0 Introduction .. .. ........................ .. ... ... ................................ ... .. .. ... .................. .. ..... 1 0 The Morro do Diabo State Park ........ ...... ..... .................... ... ... .... .. ...... .. .. .. .. 1 1 Methods .. .... .. ....... ....... ... .... .. .................... .. ....... .. .......................... ... .............. 1 6 L. Chrysopygus Habitat in Morro do Diabo ............... ...... ..... .. ...... ...... 1 6 Types of Data Collected ............................................ .. ................... .... 2 0 Characterizing the Four Habitats .. ......... .. ......................... .. .......... .. .. .. .. 2 1 General Trends of Multivariate Analysis of Habitats .......... ........ 2 3 Results .................................................................................................... ...... .. ....... 2 4 General Results ... .. .................... .. ..................... ............. .... .. .. ............... ... .. .. 2 4 River area .... .... ..... .. .. ... .. .. ........................... ... .. .. .. .. ... ............. .... ...... .. 2 5 Morro area ........................ ...... .... .... .... ..... .... .. ............... ...... ...... .. ..... .... 2 7 Little Bridge area .................. .. ..................................... ...... .............. .. .. 2 7 Caldeirao area ....................................................... ........... ... ... ... ... ........ 2 9 Results of the Multivariate Analysis of Habitats ........ .. ........... 3 O Discussion ..................................... .. ..... .. ..... .. .. ... ............................................ ..... .. 3 2 HABITAT DIFFERENCES AND THEIR EFFECTS OF FEEDING AND FORAGING ECOLOGY ...... .. .. .. ..... .... ........ ... ... .. ............... ...... ...... .. ... ............. 3 4 lntroduction .... .. ........ .. ................ .... .. ............................................ .. .. .... ... ......... 34 Methods ........................... ............. ........................................................ ... ... ... .... 3 6 Methods for Locating the Tamarins .................................... .. ........ ... ..... 3 6 Viii

PAGE 9

Methods for capturing and using radio-telemetry ... ............ .... 3 6 Home range maps and trail grids ................. ... ................................. 3 8 Methods for Recording and Analyzing Data ................................. .. .... 3 9 Composition and diversity of resources .. .. .......... .............. ......... 3 9 Phenology ................................................................................................... 4 0 Time budgets .................................. .. ...................................................... 4 1 Movements and use of space ........... ....................... ..... ....... .. .. .. ... .. .. 4 4 Ranging behavior ..................................................................................... 4 5 Feeding and foraging behavior ............ ..... ...... ............................. .. 4 6 Results ......................................................................... ....................... .. ................. 4 8 Abundance and Distribution of Resources ......................................... 4 8 Variation with location .......................................... .. .......... .. ............... 4 8 Phenology ............................ .. ... ............................................................ ..... 5 1 Time Budgets ............ .... ....................................... .. ............................. ........... 5 7 Movements and Use of Space ...................... ..... ......... ... .................... ... ... 5 9 Height and substrate ........................................................ .... .............. .. 6 4 Home-range areas .................................................................................. 6 9 Intensity and diversity of quadrat use ..... ........................ ........... 7 4 Sleeping sites ................... .. ............ .......... .... .... ........................... .. ........ 8 4 Diet ..................................................................................................................... 9 0 Plant food ............................................................ ... .................... ........ .. .... 9 O Animal food ..................... ............. ........ ... ............ ............... ................. .. 9 3 Local variation .......................................... ...................... .. ............. .... ... 9 6 Seasonal variation .......... ....................................... .............................. 9 8 Variation in foraging .......................................................................... 1 05 Discussion ........................................................................................................... 107 Use of Time ... ................ ................................................................................ 1 0 8 Movements and Use of Space ................................................................ 1 O 9 Diet ................ ........ ............................ .... ....... ... ............ .... ............................... 114 DISTRIBUTION, ABUNDANCE AND MINIMUM VIABLE POPULATION OF THE BLACK LION TAMARINS ..... ... .......... .... ........... .......... ........... ........... 11 7 lntroduction ............................................................................................ ..... .... 117 The Morro do Diabo State Park Sub-Populations .......................... 1 1 8 Other Forest Fragments in the Pontal Region ....................... .... ..... 11 9 The Fazenda Rio Claro .............................................. ............... ............. 1 1 9 The Caetetus Ecological Reserve ..................................................... .. 1 2 1 Estimating Metapopulation Size ......................................................... ...... 1 21 The Morro do Diabo State Park ............................................................. 1 2 2 Other Forest Fragments in the Pontal Region ....... .... ... ................. 1 2 4 Fazenda Rio Claro .......................................... ..................................... .. .. .. 127 The Caetetus Ecological Reserve ........................................................ 1 2 7 Results .................... ... ................................... ....................................................... 127 ix

PAGE 10

Discussion ........................................................................................................... 128 METAPOPULATION EXTINCTION MODEL ........................................................... 1 34 lntroduction ........................................................................................................ 134 Vortex Overview ............................................................................................... 136 Methods ................................................................................................................. 1 3 6 Data Sources ........................................................................................ .............. 1 3 7 Environmental variation ......................... : ................................................ 1 40 Catastrophes ................................................................................................ 1 41 Inbreeding depression .............................................................................. 1 41 Results .................................................................................................................. 1 42 Discussion ........................................................................................................... 1 44 Conclusions ......................................................................................................... 149 CONCLUSIONS AND RECOMMENDATIONS .................................. ........ ............. 1 5 1 Past Management Measures .......................................................................... 1 5 3 The Future of the Black Lion Tamarin ..................................................... 154 APPENDIX A LIST OF TREE SPECIES FOUND IN THE MORRO DO DIABO PARK ......................................................................................................... 1 5 9 APPENDIX B DATA SHEET FOR PROCESSING BLACK LION TAMARINS ............................................................................................................ 163 REFERENCES ............................................................................................................... 1 6 4 BIOGRAPHICAL SKETCH ........................................................................................ 1 82 X

PAGE 11

LIST OF TABLES Table Q..g,gg 2-1. Variables used to describe habitats. All variables were measured in all plots of the four study areas of black lion tamarins' home range in the Morro do Diabo State Park Corresponding abbreviations for each variable are in parentheses ... .......................................................... ... ............. ............ ......... .. .... 2 2 2-2. Mean habitat variables for the four study areas of the black lion tamarin home ranges in the Morro do Diabo State Park Results with different letters indicate statistically significant differences .................................................................................................. .. ...... 2 6 3-1 Characteristics of each study area quantifying availability of resources ... .. ........ .......................................... .. ............................ .......... ......... 5 0 3-2. Ten species on which each study group spent the most time and percentage of total time spent in trees of each fruit species ...................................................... ...... ........................................................ 52 3-3. Analysis of variance for monthly differences in six behaviors. Significance values are calculated from Type Ill partial sum of squares ................................................. ..... .................. .......... ... 9 1 3-4. Percentage of diet similarity among the four study groups as measured by the Renkonen index ........................................................... 9 7 3-5. List of the most common plant items for each study group and the percentage of the total feeding records for each month ....................................................................................................................... 99 3-6. Rank list of the percentage of prey capture success for the four study groups ............. ...... .......................................................................... 1 06 4-1 Mean group size, number, age and sex of L. chrysopygus in the four study areas of the Morro do Diabo State Park ................... 129 xi

PAGE 12

4-2 Estimates of available habitat for each area where black lion tamarins are found, average home range size group densities and total population sizes ..... ...... ..................... .. ....... ............ ... ... .. .. ... 1 3 1 5-1. Different population parameters used in the Vortex model ....................... ... .. ........................................... .. .......... .. .. ....... ...... ..... ..... 1 38 5-2 Black lion tamarin carrying capacity based on the available known habitat in the orig i nal home range of the species ... ... .... 139 xii

PAGE 13

LIST OF FIGURES Figure 1-1. Photograph of a group of black lion tamarins (Leontopithecus chrysopygus). Courtesy of The Jersey Wildl i fe Preservation Trust ........ .. .. ... .. .. .... .... ...... ...... .. ..... .. .... ... ..... .. .... 2 2-1. The Morro do Diabo State Park ... ...... ...... ...................... .. .. .... .... .... .. .. 1 2 2-2. The annual rainfall and temperature means for the Morro do Diabo reg i on between 1979 and 1982 ..... .. .......... .. ........ .... .. .......... .... .. 1 5 2-3. Morro do Diabo State Park vegetation map showing the 4 groups of black lion tamarins I censused Adapted from Campos and Heinsdijk (1970). Vegetation types .... ...... .............. .... ...... .. .... ...... ..... 1 7 2-4. Model of qua drats used to collect habitat data .. .... .. ............ .... 1 9 3-1. Fruit availability to black lion tamarins in the River Area group measured by number of species carrying ripe fruit each month ........................................... .. ...... .. ........ .. ............. .. ................. .... .. .... .... ... 53 3-2. Fruit availability to black lion tamarins in the Morro Area group measured by number of species carrying ripe fruit each month ............. .. ....... ............ .................................................. .. ... .. ....................... .. 54 3-3. Fruit availability to black lion tamarins in the Little Bridge Area group measured by number of species carrying ripe fruit each month .. ..... .. ...... .. ................................................................. .. ...... .. ............ .. 5 5 3-4. Fruit availability to black lion tamarins in the Caldeirao Area group measured by number of species carrying ripe fruit each month ....................................... .. .... .. ................................. .. ............ .. .... .. ..... 5 6 3-5. Time budgets of each of the four study groups. (Summed across the entire year) ........................................................... .... .................. .. 5 8 3-6. Movement of the River Area group during the study ......... .. .... 6 O xii

PAGE 14

3-7. Movements of the Morro Area group during the study ............ 6 1 3-8. Movements of the Little Bridge Area group during the study .................... .... ............................................................................... ............ ..... 62 3-9. Movements of the Caldeirao Area group during the study .... 6 3 3-10. Mean daily path length S.D.) for the River group during the one-year study .................................................................... ... .................. ... 6 5 3-11. Mean daily path length S.D.) for the Morro group during the one year study .............................................................................................. 6 6 3-12. Mean daily path length S.D.) for the Little Bridge group during the one year study ............................................................................... 6 7 3-13. Mean daily path length S.D.) for the Caldeirao group during the one year study ............................................................................... 6 8 3-14. The 1.13 km2 home range of the River group on a 50 m x 50 m grid ...................................................... .. .............................................................. 7 0 3-15. The 1 20 km2 home range of the Morro group on a 50 m x 50 m grid ...................................................................................................................... 7 1 3-16. The 1.99 km2 home range of the Little Bridge group on a 50 m x 50 m grid ................................................................................................. ... ... 7 2 3-17. The 1.20 km2 home range of the Caldeirao group on a 50 m x 50 m grid ............................................................................................................ 7 3 3-18. Monthly mean use of quadrat by each study group. Means with same line are not significantly different (Waller-Duncan K ratio) ........................................................................................................................ 7 5 3-19. Intensity of quad rat use by the River group .............................. 7 6 3-20. Intensity of quadrat use by the Morro group ....... .. .................... 7 7 3-21. Intensity of quadrat use by the Little Bridge group .............. 7 8 3-22. Intensity of quad rat use by the Caldeirao group ..................... 7 9 3-23. Frequency distribution of intensity of quadrat use by the River group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) ........... 8 0 xiii

PAGE 15

3-24. Frequency distribution of intensity of quadrat use by the Morro group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) ... ........ 8 1 3-25. Frequency distribution of intensity of quadrat use by the Little Bridge group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution__) .............. .. ...................................................... ...... ........ .... .. .. .... .. .... .. 8 2 3-26. Frequency distribution of intensity of quadrat use by the Caldeirao group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution) .. ........ .. .. .. .......... .............. .. .... .. ...... .................. .. .. .... ........... .... .. .. ..... 83 3-27. Distribution of the duration of quadrat occupancy. Figure plots percentage order of quadrats against percentage of time of quadrat occupancy ... ... ....... .......... ... .. .. ....... .... .... .. .... ... ... .. ... ............ .. .. ... ... 8 5 3-28. Spatial distribution of tree dens in the River group home range .................... .... ........ .. .... .. ...... .. ...... .. .... .... ....... .. .......... .... .......... ... .. ............ ..... 8 6 3-29. Spat i al distribution of tree dens in the Morro group home range ...................... .. ........ .... ........ .... ............ .. .... .. ..... .. ...... ..... .. ...... .. ........ ......... .. ... 8 7 3-30. Spatial distribution of tree dens in the Little Bridge group home range ..... .............. .. ........................................................... .. .. .. .. ...... .... .. ....... 8 8 3-31. Spatial distribution of tree dens in the Caldeirao group home range .... .. .. ... .. .......... .. .......................... .. .......... .................. .. ........ ..... ... ..... .. 89 3-32. Overall use of different food items for the four study groups of L. chrysopygus ... .. .. ............................ .. .... .. .. .. ..... .. .. .... .................... 9 5 3-33. Monthly variation in diet of the River group ....... .. .................. 1 00 3-34. Monthly variation in diet of the Morro group .......................... 101 3-35. Monthly variation in diet of the Little Bridge group ... .. .. .. .. 1 02 3-36. Monthly variation in diet of the Caldeirao group .................. 1 03 4-1 Satellite image of the Morro d o Diabo State Park and other surveyed forest fragments in the Pontal region .......................... .. .... 120 xiv

PAGE 16

4-2. Cumulative use of quadrats per month by each of the four study groups of L. chrysopygus during one year of data collection ........................... .................................... ............. ... .............. ......... .. 1 2 3 4-3. The map of the State of Sao Paulo with the main locations where L. chrysopygus were suNeyed or censused .......... .. ......... .. .... 125 5-1. Extincti9n probability over time for the six sub-populations and the metapopulation Notice the change of scale in the metapopulation figure .. ............... ... .. .. .. ........ .. ........ .. ... .. .... .... ........ ... .. .. .. .. .. 1 4 3 5-2 The loss of heterozygosity over time for the six sub populations and the metapopulation. Notice the change of scale in the metapopulation figure ........ .. .............. .... ................ ...... .............. .. ......... 1 4 5 xv

PAGE 17

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE ECOLOGY, BEHAVIOR AND CONSERVATION OF THE BLACK LION T AMARI NS (Leontopithecus chrysopygus MIKAN 1823) By Claudio Valladares-Padua May, 1993 Chairman: Dr. John G. Robinson Cochairman: Dr. Kent H. Redford Major Department: Wildlife and Range Sciences, School of Forest Resources and Conservation The black lion tamarin or golden-rumped lion tamarin (Leontopithecus chrysopygus) is one of the most endangered species of New World primates. Though once more widespread, now the species has survived only in small forest fragments in the interior of the State of Sao Paulo, Brazil. Two of these forest fragments are legally protected areas while the others are privately owned. Until recently the only known populations were the ones inhabiting the two protected areas. The total estimated population size for the species was 100 individuals, an estimate which caused the species to be classified as on the verge of extinction. I compared four groups of black lion tamarins in the Morro do Diabo State Park in Brazil in order to investigate in what range of xvi

PAGE 18

environmental conditions the species could survive. selected four areas with different habitat characteristics and found that many aspects of the species biology were flexibile in responding to these habitat differences. The four study groups showed that they were able to adjust their use of space in the different habitats and to respond to habitat alterations through major dietary shifts. There were, however, no significant differences in their overall time budget. I concluded that black lion tamarins, like other members of the Callitrichidae, can live and reproduce in many different types of habitats. These findings suggest that the species has behavioral variation which may facilitate conservation management. I estimated L. chrysopygus density to be 3.72 individuals per km2 for an average home range size of 1.38 km2. Based on this density, I estimated a metapopulation size of 1 004 individuals for the species. I combined demographic information with the data resulting from my ecology and behavior study, to predict the future outcome of black lion tamarin populations. For this purpose I used the Vortex model, a computer simulation model of population viability analysis. The results suggest that if treated individually, all black lion tamarin sub-populations except one, have more than a 50 percent chance of becoming extinct in the next 100 years. The exception was the population in the Morro do Diabo State Park where the species seems to have a greater chance of survival. Therefore, if all sub populations were managed as a metapopulation, which may include reintroduction, translocation and/or managed dispersal of individuals among its sub-populations, there would be a high probability for the long term survival of the species. xvii

PAGE 19

INTRODUCTION The Black Lion Tamarin The black lion tamarin or golden-rumped l i on tamarin (Leontopithecus chrysopygus Mikan, 1823) is one of the largest and most highly specialized members of the Callitrichidae family of New World primates (Hershkovitz 1977) (Fig. 1-1 ). The species i s endemic to the State of Sao Paulo the most developed state i n Brazil. Its habitat has been classified by Rizzini (1963 1967) as the riparian forest that originally extended from the Atlantic coastal forest inland along the major rivers of the state. Little native habitat is left in the original range of these primates Destruction of forest for lumber agriculture and industrialization has reduced the State s original forest cover by about 95 percent (Serra-Filho et al. 1975; Victor 1975; Ferri 1980). As a result of the demise of almost all its habitat L. chrysopygus was considered to be extinct from the beginning of the century. Then, in 1971, Coimbra-Filho rediscovered the species in the 35,000 ha Morro do Diabo State Park, in the far western part of the state (Coimbra-Filho, 1971 ). In 1976, Coimbra-Filho found a second population of black lion tamarin, this time in the 2,000 ha Caetetus Biological Reserve in the central part of the state (Coimbra-Filho, 1976). After the rediscovery of L chrysopygus, 1

PAGE 20

Figure 1-1. Group of black lion tamarins (Leontopithecus chrysopygus). Courtesy of The Jersey Wildlife Preservation Trust.

PAGE 22

4 Coimbra-Filho and Mittermeier (1977) conducted a series of surveys to assess the conservation status of L. chrysopygus. Their estimates for the species population was in the range of 200 to 300 animals but the habitat of the species was continuing to be lost (reviewed in Coimbra-Filho, 1990). The need for conservation action for the species was evident and the response came in a series of measures both at the national and international levels. L. chrysopygus was placed on the Brazilian list of endangered species (Coimbra-Filho 1971; Bernardes et al. 1990) in the IUCN Red Data Book (Thornbak and Jenkins 1982) and listed in Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) In 1984 the International Union for the Conservation of Nature and Natural Resources (IUCN) placed 1.. chrysopygus prominently on a list of the world s twelve most endangered species (Murray and Oldfield, 1984). The initial efforts for the conservation of black lion tamarins stimulated a series of studies on this species' biometry, morphology and phys i cal characteristics (Coimbra-Filho 1976 ; Hershkovitz 1977 Kleiman, 1981; Kleiman et al., 1988). Black lion tamarins are small primates, the adults weighing an average of 600 g. The length of the head and body averages 261 mm and the length of the tail averages 370 mm. There is no sexual dimorphism. The color of black lion tamarin i s predominantly black with a golden or reddish dot on the forehead on the lower half of the forelimbs and on the proximal half of the dorsal surface of the tail. If the general characteristics of the species are not subject to controversy, its taxonomic status is. Hershkovitz ( 1972) considered

PAGE 23

5 it as a subspecies of Leontopithecus rosalia and many authors have followed this arrangement (e.g., Coimbra-Filho and Mittermeier, 1972, 1973, 1977; Kleiman, 1981 Kleiman et al., 1988). Forman et al. (1986), using the differentiation of alternate alleles at two polymorphic loci, argued in favor of this sub-specific designation. In 1981, Mittermeier and Coimbra-Filho revised their original taxonomic arrangement and elevated L. chrysopygus to a full species. This species designation was later supported by Rosenberger and Coimbra-Filho (1984) based on cranial-dental variability, by Snowdon et al. (1986) using the morphology of long call vocalizations, by Mittermeier et al. ( 1988) based on morphological characteristics, and by Natori and Hanihara ( 1989) based on morphological variation of the cranium. In a recent revision of the systematics of Brazilian primates, Coimbra-Filho (1990) maintained the full species classification of this taxon. The genetics of the species was studied by Valladares-Padua (1987) An electrophoretic survey of blood enzymes from captive and free-ranging individuals showed that the species had no heterozygosity and it was genetically monomorphic at 25 loci (P=0 00; H=0.00) (Valladares-Padua 1987). Results such as these are most frequently explained by one of the following interpretations: ( 1) the low heterozygosity and polymorphism are a measurement artifact in that electrophoretic assays only measure a small number of loci in the genome (Powell, 1983);

PAGE 24

6 (2) the loss of genetic variability as a result of inbreeding depression in small populations (O'Brien, 1986; Ralls and Ballou, 1983); or (3) the lack of genetic diversity is a result of adaptation to specific environmental conditions. Selection to this narrow range of conditions may result in genetic homogeneity. The species may loose its genetic variability if it occupies a stable niche for a long period of time (Beardmore, 1983; O'Brien, 1986). It is difficult at this moment to know which of these best explains the lack of genetic diversity in this species. Valladares Padua ( 1987) reported that L chrysopygus manifests other phenotypic signs of genetic depauperation, particularly those individuals in captivity. Some captive-born animals exhibited brown coloration, a problem correlated with the functioning of the melanin pigmentary system (Pissinatti, 1992). Such genetic diseases are caused by simple Mendelian recessive alleles revealed by consanguineous mating (Valladares-Padua, 1987). Despite the work on biometry, taxonomy, and genetics, little information exists on the ecology and behavior of black lion tamarins in the wild. There are no published long-term studies on the species and the information available is based either on a small number of field observations or on anecdotal information. The existing natural history data are summarized in Coimbra-Filho ( 1970a; 1976) and Coimbra-Filho and Mittermeier ( 1973; 1977). More recent studies by Carvalho et al. ( 1989) and Carvalho and Carvalho (1989) reviewed some aspects of L. chrysopygus ecology and behavior, and later Keuroglian ( 1990) presented data on a very

PAGE 25

7 short-term study of the ecology of the species. In addition to al I these studies on organismic biology great concern continues to exist concerning the long-term survival of the species, a concern which can only be addressed by gaining data on the conservation biology of the black lion tamarins. Aims of the Study In a world increasingly affected by humans, where large tracts of relatively undisturbed habitats are being shredded and fragmented, one of the major challenges for conservation biology is ameliorating the long-term consequences of population fragmentation This study was conceived as an answer to this challenge based on updating the concept of metapopulation developed by Levins (1969, 1970). To understand the conservation approach am proposing, it is important to understand Levins' concept of "metapopulation", which he defined as an infinite number of sub or local populations of one species. Levins conceived of a species' metapopulation as a dynamic in which local extinctions of pest predators was balanced by re-migration from other populations. Thus, in his view a metapopulation could be regarded as the net result of the establishment, survival, extinction and re-colonization of local populations. This approach was adapted by conservation biologists, mainly by Gilpin (1987) and Hanski and Gilpin (1991 ). They proposed an adaptation of Levin's model to conservation biology using a finite number of sub-populations. Their models illustrated that the minimum viable size of a population was not solely dependent on its

PAGE 26

8 size but also on the patchiness of the exist i ng habitats and on the movement of individuals between habitable patches. They included the extreme case where discontinuous habitats may result in the total impossibility of natural migration among local populations. Habitat or population fragmentation creates small isolated sub populations, wich enhances their probability of extinction due to genetic demographic and environmental forces acting within patches ( Soule, 1980; Ralls and Ballou, 1983). Even if the sub populat i ons survive, isolation itself might cause genetic drift leading to genetic divergence and consequent speciation ( Wright 1977 ; Franklin 1980; Otte and Endler 1989). Thus i n the cases where fragmentation precludes natural migration metapopulation management will entail artificially moving animals from one patch to another Managed animal migration must also take into consideration previous knowledge about the species so the animals can survive and reproduce in the new area (Foose 1990). Although Levins and Gilpin analyzed several aspects of metapopulations (eg spatial distribution of habitat patches and extinction and recolonization dynamics) they did not directly consider the ecological and behavioral flexibility of a species as an important metapopulation dynamic variable. Lee (1991) suggested that investigations on how mammalian species respond to environmental change are of utmost importance in illuminating how animals might respond to fragmentation. In this study, I am particularly interested in behavioral flexibility related to environmental changes which includes maintenance, growth and reproduction.

PAGE 27

9 The central questions of this dissertation are: do the behavior and ecology of groups of Leontopithecus ch rysopygus differ significantly in different habitats, and if so, what are the implications of these differences for the conservation of the species? To answer this, in chapter 2, compare four black lion tamarin habitats to determine if there are significant differences among these areas. Sixteen habitat variables were compared using multivariate statistical tests. In chapter 3, I investigate the effects of habitat differences in the feeding and foraging behavior among four groups of black lion tamarins. I collected quantitative and qualitative information on the variation in resource abundance and on the tamarins' use of time and space both as individuals and as groups. I examine in chapter 4, the size and density of each known sub-population of the black lion tamarin and estimate its metapopulation size. In chapter 5, I integrate the data resulting from all previous chapters by running a population viability computer simulation model (Vortex) (Lacy, 1990). With this program I examine, in light of black lion tamarin behavioral and ecological flexibility, the effects of deterministic forces as well as demographic, environmental and genetic stochastic events on the tamarin's metapopulation. Finally, in chapter 6, I summarize and discuss the dissertation's major results.

PAGE 28

THE MORRO DO DIABO STATE PARK AS A HABITAT FOR THE BLACK LION TAMARIN Introduction The members of the family Callitrichidae seem to be quite flexible in their general selection of habitat (Sussman and Kinzey, 1984). Moynihan (1970) reports that in Panama, Saguinus geoffroyi uses both low dense primary forest and abandoned, overgrown agricultural fields. Other authors describe at least three species of tamarins, S. mystax, .s_ fuscicollis and S. midas, living in primary and secondary forests (Castro and Soini, 1977; Mittermeier and van Roosmalen, 1981 ). Leontopithecus rosalia and Leontopithecus chrysomelas, two species congeneric with L. chrysopygus, show preference for primary forests (Rylands, 1982; Coimbra-Filho and Mittermeier 1973 ; Kleiman et al., 1988). The main question of this chapter was, therefore, to determine whether black lion tamarins showed this same trend in habitat flexibility and could live in different habitats To answer this question, I selected four groups of black lion tamarins inside the Morro do Diabo State Park. The park is the only remaining place to attempt to answer this question because it is the last large continuous forest in the original range of the species, and because it is the only area that has a variety of forest types. The study sites 1 0

PAGE 29

1 1 were selected after six months of evaluating habitats that seemed distinct from each other. The habitats used by the four study groups were quantitatively compared using means of sixteen pre established variables (Table 2-1 ). My null hypothesis was that there were no differences among the four habitats. To test this hypothesis I used multiple discriminant function analysis (Ho = the common elements of the various mean vectors were identical to one another). I also used canonical discriminant analysis to create a habitat hypervolume ( sensu Carey, 1981), which is a mathematical description of the species habitat range in multidimensional space. While the multiple discriminant function analysis confirms differences in habitat variables, the canonical discriminant analysis locates the differences and allows visual representation of these differences The Morro do Diabo State Park The Morro do Diabo State Park is a protected area under the administration of the Forestry Institute of the State of Sao Paulo (lnstituto Florestal de Sao Paulo). It is located in the arrow-shaped region in the extreme western part of the State of Sao Paulo (22 30' S, 52 20' W), known as "Pontal do Paranapanema" (Fig. 2-1) The park was created as a reserve in 1941 by the then Governor of Sao Paulo, Dr. Fernando Costa (Guillaumon et al., 1983). Of the original 290,000 ha park which was gazetted to protect regional fauna and flora, today only 34,156 ha survive (Valladares Padua, 1987).

PAGE 30

1 2 L Br-azi1 Sio Pau1o N _, +, s r--i 500 m Morro do Diabo ---..,:::=--Highway Road Railroad Poronop8nem8 River Figure 2-1. The Morro do Diabo State Park.

PAGE 31

1 3 Most of the reduction in size occurred in the 1950s and 1960s. During this period the reserve was frequently i nvaded and illegally colonized by members of a wealthy elite, usually with the approval and connivance of important members of the state government (Leite, 1981 ). In the following decades the reserve experienced an increasing level of protection until 1986 when its status changed from reserve to park. Today, the park is well protected with legally demarcated nondisputed boundaries. To the south there is the Rosana Dam Lake on the Paranapanema river and pasture and farmland are to the east, west, north of the park. The park's predominant relief forms are slightly undulating plains interspersed with fluvial plains and valleys and some sedimentary mesas. The Morro do Diabo is the highest of these mesas in the region. The altitude in the park ranges from 350 m above sea level at the top of the Morro do Diabo to 300 m above sea level at its lowest regions. The park s predominant soil is dark red latosol sand phase (LEa) (Deshler 1975 ; Ventura et al. 1965) A later study, done in 1979, by the Sao Paulo Electricity Company (CESP) classified it as predominantly LEd, dark red latosol dystrophic plane relief phase (Guillaumon et al., 1983). No matter which of these classifications is more precise, the region in general has very poor soil (Setzer, 1949). The regional climate is Cwa in the Koppen classification (Guillaumon et al. 1983; Coimbra-Filho, 1976). The rainfall in the area of Morro do Diabo is seasonal; the area annually receives an average of 1,131 mm, of which 30 percent falls between April and

PAGE 32

1 4 September (Fig. 2-2). This is half the mean for the Atlantic coastal forest. The low rainfall during the dry season in the Morro do Diabo region leaves the soil with little moisture and is responsible for the deciduous nature of the forest. The mean annual temperature is higher than in !he coastal region. Because it is part of the Atlantic rainforest, the forests of Morro do Diabo would be expected to match those described by Schimper (1903: 66) as "An evergreen, hygrophilous in character, at least 30 m high, rich in thick stemmed lianas and in woody as well as herbaceous epiphytes." But the Morro do Diabo area is also influenced by cold fronts which come from the south dropping the temperature below 0 C for one or two days each winter. These cold invasions influence the regional natural vegetation by reducing the regeneration rate of the trees (Hueck, 1978). As a result, the best approximation of a correct classification of the park's forest would be upland semi-deciduous Atlantic Forest interspersed with some areas of "cerradao" (Baitello et al., 1988). The latter is described by Redford (1983: 126) as a "tall dense semi deciduous xeromorphic savanna vegetation." The official denomination of the park's forest formation, according to the lnstituto Brasileiro de Geografia e Estatfstica--lBGE (Brazilian Geography and Statistics Institute) is the semi-humid tropical forest of the interior. It is characterized by being semi-deciduous with most of the emergent trees losing their leaves during the cool dry months between June and Septemb~H (Hueck, 1978; C. Valladares-Padua pers. observ ).

PAGE 33

1 5 3 0. 000 <>. .. 1 75 000 --0T e mp er a tu re 2 5 000 \, 15 0 000 .... . O R ai n fall o Q) 125 000 ... 2 0 000 :::l ca \ 1 00 000 ... Q) 15 000 ~ Cl. E .. ~.(). .... 75 000 Q) \ \. .... 1 0.000 ~ 50 000 .---o 5 .000 ~ 25 000 0 000 0 000 C .Cl ;;; .... >, C "S Cl a. u > (.) C1l Q) a. C1l ::::J -, ::::J Q) 0 Q) -, LL
PAGE 34

1 6 Despite Morro do Diabo's physiognomic differences with the coastal forest of Brazil, its tree species and genera are similar to those described by Assumpgao et al. (1982) for the State of Sao Paulo. At Morro do Diabo I identified a total of 84 tree species in 31 families using a systematic sampling design (Appendix A) This is more than the list of Campos and Heinsdijk (1970) and a few less than the list produced by Baitello et al. (1988) The vegetation survey conducted by Campos and Heinsdijk ( 1970), for the park's first management plan concluded that Morro do Diabo is composed of several distinct vegetation formations. They proposed a general classification of these formations in eight distinct types, six of which have some forest component (Fig 2-3). Methods L. Chrysopygus Habitat in Morro do Diabo Due to the comparative nature of my study, I selected four study sites that demonstrated the highest habitat variability detectable at the beginning of the field study. These were based on the vegetation map of Campos and Heinsdijk (1970) and my own personal observations of the general physiognomy of the area. After this initial qualitative classification, the next step was to quantify the study sites to demonstrate significant habitat differences among them In order to quantify the vegetation characteristics in each of the four study sites, I set up trail transects forming 50 m x

PAGE 35

1 7 N +, s Brazil Sao Pau1o H 500 m V EGETATION ttt Type I ttt V V V Type 11 V V Y :r1' 1' Type 111 :r1' 1' Group Q Type IV Ty pe V Tfdpe V I ....... Type V I I **~ Poronopanemo River Figure 2-3. Morro do Diabo State Park vegetation map showing the 4 groups of black lion tamarins I censused. Adapted from Campos and Heinsdjik (1970). Vegetation types: I. Densely stocked tall forest; II. Moderately stocked tall forest; Ill. Poorly stocked forest; IV. Very impoverished forest; V. Heterogeneous vegetation; VI. Highly impacted forest fallow; VII. Abandoned pasture. L. chrysopygus groups: 1. River group; 2. Morro group; 3. Little Bridge group; 4. Caldeirao group.

PAGE 36

1 8 50 m quadrats. All transects were cut by machete to a width of approximately 1 m along north-south and east-west compass bearings. I sampled 16 habitat variables in each of the 50 m x 50 m quadrats for a total of 35,776 observations. Among the measured variables five were related to the composition and structure of trees with more than 10 cm of dbh (number of trees number of tree species, diameter at breast height, mean tree height and canopy height). These were measured in 1 O m x 1 O m plots located in the northwestern corner of each 50 m x 50 m quadrat. The 11 variables not directly related to the tree composition of the forest were measured in a 1 m x 1 Om rectangle aligned diagonally inside the 10 m x 10 m square (Fig. 2-4 ). In the four study areas, the habitat data matrix had 2,336 quadrats by 16 variables This method of data collection is called a "square grid" systematic sampling design. It is widely used in forestry practices and it simplifies considerably field work. The problem in adopting this method is that it can cause imprecision if the systematic design coincides with periodicity in natural features like parallel ridges. In this case the results would not indicate variations in features such as valleys (de Vries, 1986). This did not seem to be the case in Morro do Diabo because, although I used systematic sampling, the distances between the samples were small (40 m) and the number of samples large (more than 2,300 quadrats). To avoid any risk of missing important habitat aspects and to prevent possible biases in the habitat sampling I used all 50 m x 50 m quadrats in a fine-grain approach However, for the data analysis

PAGE 37

1 9 10mx10m North Western Ouadrat 1 m x 1 0 m ~--+--.0 iagona I Rectangle Quadrat 50 m 50 m Figure 2-4. Model of quadrats used to collect habitat data.

PAGE 38

20 included only the quadrats used at least once by the primate groups was studying. I reasoned that by restricting the analysis to the quadrats used by the primates I could obtain more accurate data on the habitat actually used by each group. Types of Data Collected Once I had selected the habitats to be compared and organized the network of quadrats in each study area, I collected quantitative and qualitative data in each of the quadrats. The protocol employed to measure the variables was a modified version of one used in many small mammal habitat studies (M'Closkey, 1976; August, 1983; Fonseca, 1989) (Table 2.1 ) In each 1 Om x 1 Om quadrat, I measured and identified all the trees with more than 1 0 cm dbh. These variables were related to number of trees and species composition, mean diameter at breast height and mean canopy and tree height (variables 1-5; Table 2-1 ). In the north-western corner of the quadrat I also measured the depth of the humus layer. In the 10 m x 1 m rectangle I counted and sorted in four equal size classes all trees with diameter less than 10 cm (variables 7-10; Table 2-1. The remaining variables in the 10 m x 1 0 m quad rats were recorded on a scale of 0 (minimum score for a parameter) to 10 (maximum score for a parameter). In this way I quantified epiphyte loads, vines/lianas loads and their degree of entanglement, ground bromeliads and canopy cover. I also used a "density board" to measure the degree of visibility in the forest. I placed the board at the one meter end of the 1 m x 1 0 m rectangle and the observer at the other end (NRC, 1981; Wight, 1938).

PAGE 39

21 Identification of tree species was done by my field assistant, Mr. Jose M. de Souza, who is a knowledgeable park ranger. Mr. Souza had participated in all previous primate studies and botanical inventories conducted at the Morro do Diabo State Park. He identified more !han 90 percent of the trees in the study areas showing consistency and accuracy in the identification of trees and exceptional skills in following the animals. However, because the identifications relied mainly on bark and leaf characteristics, marked for future identification any tree that could not be immediately identified as well as some common trees selected at random for checking its identification accuracy. When these species flowered or fruited, voucher specimens were collected and subsequently identified at the herbaria of the Forestry Institute of Sao Paulo or at the Forestry Department of the University of Sao Paulo. Mr. Jose correctly identified the common trees and of those for which he was not sure, his suggestions coincided 80 percent with those identified elsewhere. Characterizing the Four Habitats To characterize the habitats of each of the four groups, I used qualitative data to present a general description of the physical features, structure, and vegetation of each area. Quantitative data were used to test if each of the characteristics measured were significantly different and how they differed. Thus, I analyzed each of the habitat variables for the four areas by means of a series of independent one-way analyses of variance.

PAGE 40

22 Table 2-1. Variables used to describe habitats. All variables were measured in all plots of the four study areas of black lion tamarins home range in the Morro do Diabo state park. Corresponding abbreviations for each variable are in parentheses. HABITAT V ARIBLES Number of Trees ( NT) Number of Tree Species ( NS) Mean diameter/ breast/height ( dbh ) Mean T r ee Height ( TH ) Canopy Height ( CH ) Epithyte Density ( ED ) Number of Trees 0.1 2.4 cm of dbh (T1 ) Number of Trees 2 5-4 9 cm of dbh ( T2) Number of Trees 5.0-7.4 cm of dbh ( T3) Number of Trees 7 5-10.0 cm of dbh ( T4) Ground Bromeliads density ( GB) Vine Density ( VD) Vine Entanglement (VE ) Humus (HU) Percentage of Canopy Cover ( CC) Understory Volume ( UV) DESCRIPTION Number of trees i n dbh sample Number of tree species in dbh samp l e Mean diameter ( cm ) of a l l trees w i th a dbh larger than 32 cm w i thin a 1 OX10 m quadrat Mean tree height ( m ) of all trees i n dbh sample Mean canopy height ( m ) of all trees i n dbh sample Density of vascular epithytes measured o n a sliding scale ( 0-2 ) i n each 10X 1 0 quadrat Number of trees 0.1-2.4 cm dbh i n a rectangle of 1 OX 1 m in each quadrat Number of trees 2.5-4 9 cm dbh in a rectangle of 1 OX 1 m in each quadrat Number of trees 5.0-7.4 cm dbh i n a rectangle of 1 OX 1 m in each quadrat Number of trees 7 5-10 0 cm dbh i n a rectangle of 1 OX 1 m in each quadrat Percentage of ground bromeliads in each 1OX10 quadrat Density of vines expressed on a sliding scale (0-2) in each 1OX10 m quadrat Entanglement of the vines measured on a sliding scale (0-2) in each 1OX10 m quadrat Depth (cm) in the first soil layer Percentage of canopy cover above each 10X10 m quadrat Number of areas of 30 cm visible on a ruler od six divisions positioned 10 m distance i n each 1 OX 1 quadrat

PAGE 41

23 General Trends of Multivariate Analysis of Habitats The commonly recognized multidimensional nature of wildlife habitat has led to an increasing use of multivariate statistical techniques in studies of wildlife ecology (Capen, 1981 ). These techniques permit the use of a series of predictor habitat variables as an approximation of its multidimensional characteristic. The variables are then used in deriving a function that allows classification of the sample groups according to the probability of belonging to each group (Arita and Humphrey, 1988) In this study used the 16 variables measured in the four study habitats as the predictor variables. This allowed me to test the null hypothesis that the common elements in the four habitats (mean vectors for the variable predictors) were identical using a multiple discriminant function analysis (Manly, 1986). I evaluated the hypothesis of equal mean vectors with the Wilks' lambda test (Wilks, 1962). This test converts the value of the test result to a comparable F value and then compares it against a critical F-value at the 0.005 level to decide whether or not to reject the null hypothesis. also used canonical discriminant analysis as a confirmatory procedure, which allows differentiation among groups of observations (Owen and Chmielewski, 1985; Arita and Humphrey, 1988). This statistical technique is designed to find variable combinations that allow the maximum segregation among the groups The canonical discriminant analysis also generates Mahalanobis distance, D 2 which measures group dissimilarities: the larger the

PAGE 42

24 value of 02 the less similar are the group mean vectors (Barcikowsky, 1983). I used the Mahalanobis distance technique to measure differences among the criterion groups' (habitat areas) centroids. The centroids are group mean values calculated in a single dimensio n or discriminant function (Tabac h nick and Fidell 1983). To better visualize the relationships among areas, I plotted the centroids of each study area on a graph (Pielou 1984; Huck et al., 1974). On the same graph I drew 95 percent confidence ellipses around each centroid (Owen and Chmielewsky 1985). These comprise 95 percent of the observations for each group of tamarins studied in the multivariate space. I performed all the statistical analyses using programs in Statistical Analysis System on a personal computer (SAS Institute Inc 1985). Statistical significance was set at p< 0 05. Results General Results Sites were different at highly significant levels (ANOVA, p<0 001) for all variables but one: epiphyte load. Epiphytes like bromeliads are the most important prey micro-habitats for the other lion tamarins (Peres 1986). For L. chrysopygus this does not occur. The nonsignificant result for epiphytes can be explained by the fact that epiphytes are almost nonexistent in the Morro do Diabo (mean for the four areas = 0.01 ). I n Table 2-2 I present the means for each variable by group and the general summary of these results.

PAGE 43

25 Scheffe's multiple comparison test (Scheffe, 1959) was used to compare the means of variables for which significant differences existed. To illustrate the differences in means I grouped variables into categories within which differences were not significant and assigned a lette_r code to each (Table 2-2). Differences across categories were significant while within each category they were not River area This site was the most southerly of the areas I studied. It is an alluvial forest adjacent to the Paranapanema river which influences its structure and composition. Occasional floods create a very conspicuous forest floor composed of many ridges and grooves sometimes filled with a layer of sediments carried in by the river. The area is swampy in many parts particularly close to the river. The riparian forest at this site is the tallest forest I studied. As a general rule the canopy is even and closed with some emergent trees. It had the lowest mean number of trees per quadrat (mean number of trees = 4.45) but the tallest average height of canopy and largest average diameter at breast height (mean = 23.75 cm). The thick undergrowth consisted largely of seedling and sapling trees shrubs, and young woody climbers and the density of these created a high understory volume. Vines were present but were not very dense. The River area is strikingly different from the other sites in its absence of ground bromeliads, which made data collection in this area almost a pleasure when compared with the other study areas.

PAGE 44

Table 2-2. Mean habitat variables for the four study areas of the black lion tamarin home ranges in the Morro do Diabo state park. Results with different letters indicate statistically significant differences. River Morro L. Bridge Habitat Variables (N = 262) (N = 287) (N = 3 77) Number of trees d 4.45 b 6 25 C 5.50 Number of Tree Species b 3.54 a 4 1 a 3.67 Mean DBH cm a 23 75 b 17 68 b 18.97 Mean Tree Height m a 8 69 C 7.68 b 7.64 Canopy Height m a 2.13 ab 2.01 b 1.99 Epiphyte Number a 0 06 a 0.12 a 0 10 Mean# trees of DBH 00-.80 cm ab 11.06 a 12 35 C 9 39 Mean# trees of DBH .81-1 .55 cm a 1.20 a 1 26 b 0 87 Mean# trees of DBH1 .56-2.35 cm a 0.63 a 0 56 a 0 48 Mean# trees of DBH2.36-3 20 cm b 0.27 a 0.46 b 0 23 Ground Bromeliads Density a 0 00 C 0.46 a 2.30 Vine Density a 2.14 b 1.57 ab 1 82 Vine Entanglement a 1.00 a 0 85 a 0 89 Humus b 2.04 C 1. 77 be 1.82 Canopy Cover a 2.96 d 1.94 b 2 75 Understory Volume a 3.25 a 3 06 b 2 71 a b C b C a be C b b b C b a C a Caldelrao (N = 24 9 ) 7 07 4 28 17.43 7.60 1 .76 0.10 1 0. 11 0.28 0.26 0 24 1 54 1 11 0 59 2 32 2 30 3.14 I\) O')

PAGE 45

27 When examining the 10 most common trees in the area I found that they fell into 8 families, accounting for 67 .8 percent of all species (Table 2-3). This gives the River area a tree composition quite distinct when compared to the Caldeirao area at the other extreme of the park. These two areas shared only three of their 10 most common species (Table 2-3). Morro area The Morro area is located south of the east-west highway that bisects the park almost in front of Morro do Diabo the hill after which the park is named. A 50 m strip of forest with trees of reduced height created an edge effect close to the highway. Outside of that strip of forest, the Morro trees had an average height higher than those at Little Bridge and Caldeirao. The area has a large number of species per ha (mean = 4.1) and their average dbh was between the averages for the River and the Caldeirao areas. The canopy cover is the most open among the four areas. Vines are present and dense but not as dense as in the River area. In the Morro area, the 10 most common species represented 78. 7% of the total tree species found in the area (Table 2-3). Five of these 10 were in the Myrtaceae family and the rest were distributed among the Protaceae, Palmae, and Apocynaceae. Little Bridge area Northwest of the Morro area is the Little Bridge area, located on the north side of the highway. It has a xeromorphic aspect, characterized by many plants with adaptations to dry habitats such

PAGE 46

Table 2-3. The 10 most common tree species in the four study areas. River Morro Species Family % Species Family Lonchocarpus /eucanthus Leguminosae F. 13 64 Eugenia uvalha Myrtaceae Croton floribundus Euphorbiaceae 10 30 Psidium sp. 1 Myrtaceae Chrysophyllum gonocarpum Sapotaceae 7 40 Nectandra saligna Lauraceae Jaracatia spinosa Caricaceae 7.30 Myrcia sp Myrtaceae Chrysophyllum sp Sapotaceae 5 90 Enidlicheria paniculata Lauraceae Gal/esia gorazema Phytolaccaceae 5.50 Eugen ia sp Myrtaceae Nectandra saligna Lauraceae 5 00 Syagrus romanzoffiana Palmae Campomanesia sp Myrtaceae 4.50 Sebastiana serrata Euphorbiaceae Psidium sp 1 Myrtaceae 4.30 Myrciaria sp Myrtaceae Aspidosperma polyneuron Apocynaceae 4.00 Eugenia sp. Myrtaceae Other 46 32 Other Little Bridge Caldeirao Species Family % Species Family Psidium sp Myrtaceae 31 80 Psidium sp Myrtaceae Eugenia uvalha Myrtaceae 11 00 Eugenia uvalha Myrtaceae Myrceugenia ovata Myrtaceae 9 20 Nectandra saligna Lauraceae Nectandra saligna Laurace ae 7 80 Syagrus romanzoffiana Palmae Enidlicheria paniculata Lauraceae 4 50 Enidlicheria paniculata Lauraceae Eugenia sp. Myrtaceae 3 20 Myrciaria sp Myrtaceae Roupalia brasiliensis Proteaceae 3. 10 Helietta longifoliata Rutaceae Syagrus romanzoffiana Palmae 2 90 Myrcia sp Myrtaceae Aspidosperma po/yneuron Apocynaceae 2 70 Aspidosperma polyneuron Apocynaceae Myrciaria sp. Myrtaceae 2 50 Roup a la br asi/iensis Proteaceae Other 30 91 Oth er % 20 00 19 30 12 50 10.80 6 60 5 60 4 80 3 10 2 80 2.60 22.97 % 27 20 21 80 6 20 4.40 3.40 2.90 2 60 2.60 2 30 2 30 33 50 I\) CX>

PAGE 47

29 as ground bromeliads and cacti. This site has the highest density of ground bromeliads (2.30 mean percentage per quadrat; Table 2.2) It has the third lowest mean number of trees but on average contains the same number of tree species as the River and Morro area (mean number of species per quadrat = 3.67). Trees here are similar in height to the Morro and Caldeirao areas The canopy cover is dense but broken by the presence of some small islands of "cerrado (savanna type) especially in the western part of the area. The Little Bridge site has a high percentage of Myrtaceous trees (4 out of the 1 O most common species) representing 54 5 percent of the trees in the area (Table 2-3). The understory vegetation at the Little Bridge area is not very thick However a large number o f ground bromeliads evenly cover the area together with the second highest number of vines and lianas and the presence of many palm trees gives the Little Bridge site a physiognomy of a typical "cerradao" (rough savanna type) Caldeirao area Located in the northwestern region of the Morro do Diabo park Caldeirao is the area with the highest mean number of trees but trees with the smallest average height and the lowest dbh (mean height = 7.6 m and mean dbh = 17.43). In this area the 10 most common species made up 88.1 percent of its total number of individual trees. The Myrtaceae is also heav i ly represented with six species and 61 1 percent of individual trees (Table 2-3). Despite the high mean number of species in the area, the dominance of the Myrtaceae family, associated with dense understory cover gives the

PAGE 48

30 Caldeirao a very homogeneous physiognomy. The Caldeirao site has the lowest vine entanglement and density of the study areas contributing to its homogeneous appearance. The uniform aspect of the area is exacerbated even more by the low and even canopy height with its few emergent trees. Results of the Multivariate Analysis of Habitats The result of Wilks' lambda test to evaluate the hypothesis of the equal mean vectors for the four study areas was 44 806 which is comparable to an F ratio of 22.8 with 45/3440 degrees of freedom. The probability of obtaining an F ratio this large by chance is less than 0.0001. This result permits rejection of the null hypothesis, which predicted habitat uniformity and strongly suggests that habitat as measured by 16 variables differs significantly among the study sites. The results of the canonical analysis showed that the greatest habitat difference is between the River and Caldeirao areas. The other two areas do not differ significantly from one another or from the other two areas. The differences between the four study areas obtained from the discriminant equations are best visualized in a projection of the centroids for each group in the first two canonical variable spaces. The centroids are the points that correspond to the two mean discriminant scores of each group (distance and height). In the same graph I present the Mahalanobis distances between the centroids of each area (Fig. 2-5). I also constructed a 95 percent confidence interval ellipse for each area. I used this as a graphical representation of the scatters

PAGE 49

N C 0 __, u C ::J LI.. (0 u C 0 C (0 u 31 2 Little 0 -1 2 -------.------.......--------r------1 2 -1 0 Canonical Function 2 Figure 2-5. Projection of the four habitat areas in space of the first two canonical variables. The ellipses represent 95% of the confidence interval for the habitat of each group. The lines and numbers connecting the centers (centroids) of the ellipses are the Mahalanobis distances.

PAGE 50

32 of scores of each group in a bivariate plot of the first two canonical variates. Based on my observations of the habitat of black lion tamarin inside and outside Morro do Diabo, I am assuming that the four distinct habitats types at the park cover most of the range of habitat types o~cupied by black lion tamarins. Under this assumption, the four study groups' 16 variables dimensional space, obtained in combining the four groups ellipses (Fig. 2-5), would be statistically analogous to the hypergeometric model of the ecological niche proposed by Huchinson ( 1965). Assuming no difference in food availability the resulting space if compared with other forest fragments will facilitate the future location of reintroduction and translocation areas for the management of the species. Discussion The results of this aspect of the study suggest that there are significant habitat differences among the areas used by the four black lion tamarin groups I studied in the Morro do Diabo State Park. These differences are in accordance with the previous botanical and forest inventories conducted in the Morro do Diabo State Park, which showed a meaningful variability in the forest composition in different areas of the park (Campos and Heinsdijk, 1970; Baitello et al., 1988). The Scheffe test results indicate which habitat variables contribute most to these differences (Table 2-2). Among the 16 variables measured, only 1, epiphyte number, showed no significant difference among all the four areas. All other variables were significantly different among two or more of the areas. At this

PAGE 51

33 point, do not have enough data to conclude if any one of the four study areas is either an optimal or at least a preferred habitat for black lion tamarins but they certainly look dramatically different. These differences are clearly supported by the statistical analysis. Therefore it seems safe to conclude that black lion tamarins can live in many different types of forests These results show that many of the suppositions about tamarins preferring either secondary forest edge habitats or mature primary forests are not completely accurate (Rylands 1982; Coimbra-Filho and Mittermeier 1977 ; Sussman and Kinzey 1984). The results of this study also give enough evidence to conclude that black lion tamarin is a generalist that is able to live and reproduce across a range of different forest types. This conclusion certainly has an important impact on the conservation management of the species as I will demonstrate in the following chapters.

PAGE 52

HABITAT DIFFERENCES AND THEIR EFFECTS ON FEEDING AND FORAGING ECOLOGY Introduction In the last 25 years it has became apparent that social behavior in primates i s i nfluenced by habitat which produces differences even between populations of the same species ( Crook and Gartlan 1966 ; Eisenberg et al. 1972; Chivers 1986). This may be due to the fact that primates and other relatively long lived animals are exposed to change on different time scales than organisms living short periods of time. As a consequence of their long life, some primate species naturally experience seasonal and annual variation in the availability of resources during their lifetime and this may allow them to l earn from individual experience about ranges of variation and outcomes of responses (Clark 1991 ). Throughout life the change in demographic and ecological conditions give individuals the ability to promote their survival and reproductive potential depending on that store of knowledge and many times on the knowledge of their group members (Pereira and Altmann, 1985). Furthermore, this constant exposure of primates to frequent novel and rapidly changing conditions may then select for flexibility" (Fagen, 1982) Flexibility is the attribute that allows animals to vary short-term behavior in the face of 34

PAGE 53

35 change and new information largely through processes of learning (Clark, 1991 ). Ecological and behavioral flexibility is then of vital importance for the survival of a species in a world of frequent environmental shifts. My major goal in this dissertation is first, to document intra specific ecological variation in social groups of black lion tamarins and then try to relate it to ecological changes. In the last chapter I demonstrated that the four study groups of black lion tamarins have significantly different habitats In this chapter I concentrate on a detailed intra-specific study habitat use of the four study groups of L. chrysopygus in the Morro do Diabo State Park. I use the intra specific comparative method to describe the strategies these four groups of tamarins employ in responding to the different patterns of seasonal and local habitat variation. The understanding of the patterns of social behavior and of habitat use within and between species is important for unraveling ecological aspects crucial for designing conservation management initiatives for a given species (Chivers, 1986). The major question of this chapter is therefore related to black lion tamarin habitat use: Does the behavior and ecology of groups of Leontopithecus chrysopygus differ significantly in different habitats? To answer this question I describe and compare differences in phenological patterns, time budgets, use of space, diet, feeding and foraging success among the four study groups

PAGE 54

36 Methods I started my field work at the Morro do Diabo State Park in February of 1988. Since the main goal of this study was to compare groups of L. chrysopygus living under different conditions I selected areas based on their distinct forest cover physiognomy (see chapter 2). I captured the first study group in June of 1988 and by the end of August I had captured six groups in different areas of the park. From these six groups I selected the four target groups for my study Methods for Locating the Tamarins Methods for capturing and using radio-telemetry On the basis of my previous experience with this species (Valladares-Padua, 1987) I was aware that L. chrysopygus was harder to trap than other primates of the Callitrichidae. Golden lion tamarins for example, can easily be caught in Tomahawk or other traps by baiting them with bananas (Peres, 1986). There are several difficulties in capturing black lion tamarins. They are not easily baited (I experimented with more then 50 types of bait items without success) and they have a large home range for an animal of their size which makes them difficult to locate even with the help of radio-telemetry devices. The only capture procedure that was effective was to follow the animals to their den tree where they were caught. Even in this case, experienced field personnel were required to locate the animals and perform the trapping procedure safely (Coimbra-Filho, 1977). In order to locate a group of L. chrysopygus for the first

PAGE 55

37 time, I randomly searched the forest in locations where the animals had been previously observed by the park staff. At the beginning of my field work I usually detected them first by sound and then by sight, but by the end of the first year of field work I was fairly successful in attracting them with the use of play-backs. I have since adopted this method to locate the tamarins (Muckehirn 1967). After detecting a group of black lion tamarins, I followed them until dusk when they sheltered in a tree hole ( den). If the position of the den allowed a field assistant to climb to it, a platform would be placed near the den in such a way that neither the field personnel, nor the animals would be at risk. If the den was in a position that offered unsafe capture conditions I would follow the animals for another day until they entered a tree cavity that seemed fit for capturing. The following day at dawn, I would come back to capture the animals. To capture the animals, I checked the correct position and shape of the den and closed all other possible exits with pieces of iron mesh. I then measured the den with a scaled flexible stick and filled it with burlap bags. The bags compressed the animals into the bottom of the den. Finally, a hole large enough to fit a human hand was opened at the top of the compressed chamber with the use of a small chain saw. The animals were hand caught one by one and slipped into burlap bags. After the capture a small wooden door was installed to close the saw hole. This door was useful because the lion tamarins occasionally re-used the same dens. If I had to recapture animals I would wait until they entered one of these modified dens. I used this method more than 30 times to capture

PAGE 56

38 and/or recapture the animal and excessive stress was not evident on any of these occasions. Once located, at least two animals in each group were radio collared. Radios were especially helpful in finding the groups at the beginning of da ta-collection days or when I lost the whole group. Otherwise, radios were not usually needed as habituated groups were relatively easy to follow. The first time a group of lion tamarins was captured, the group was processed at the field headquarters in the park's administration area. This procedure consisted of anesthetizing the animals to obtain individual weights and measurements. For this purpose modified a protocol used for the golden lion tamarin project (Appendix B). Each animal was also tattooed with an individual code. Additionally, either a radio transmitter (Telonics, Inc. and Wildlife Materials, Inc.) or a bead-chain collar with different colors coded for individual recognition were placed on each tamarin. The animals were released the following morning where they had been captured. Home range maps and trail grids From the day I first captured a group until the day I started to collect data on a systematic basis, all radio-collared animals were followed at least once a week with two major objectives. First, I marked location points on a map of the area which allowed me to get an estimate of their home range. Second, I wanted to begin habituating the groups to my presence. This phase lasted from August to December of 1988. During the first days of November of 1988, I had finally marked enough points on maps to be able to get a

PAGE 57

39 first approximation of the size and location of their home ranges. set up transects throughout the four study areas using a 50 m x 50 m trail grid. This process lasted until June of 1989 Trails were opened at all areas in which the tamarins had been seen However I later found out that the first six months of observation did not include the entire home range of the tamarins and many expansions of the original grid system had to be made as the systematic data collection progressed To identify the tamarin's positions and movements in their range I labeled a tree in each corner of each 50 m x 50 m quadrat with coded aluminum tags. This system allowed me to recognize with precision and ease the group location when collecting behavioral and ecological data. In July of 1989 I finally finished the grid system had all the quadrats labeled and had the animals habituated to my presence I considered a group to be habituated when all its members would feed normally at a two-meter distance from the observer (Richard, 1978). I had spent eighteen months of work and opened more than 270 km of trails before I could finally start the systematic collection of data on the four study groups. Methods for Recording and Analyzing Data Composition and diversity of resources To measure resource abundance, I developed an index based on local vegetation in each of the four study group's home ranges. I used the same systematic grid sampling technique I used to measure habitat variables (Chapter 2) to estimate the available plant

PAGE 58

40 resources for the primates. In each of the 10 m x 10 m plots inside the quadrats in an area roughly corresponding to the home range of each group I counted and identified all trees over 1 0 cm dbh. This included a total of 17 582 surveyed trees. To examine differences between home ranges I used the biometry and taxonomy of the surveyed trees to calculate a series of means and percentages that allowed for a structural as well as taxonomic description of the four different home ranges Phenology Transects for phenological data collection were established in each of the four study areas. All transects for this purpose followed existing east-to-west trails but were spaced 300 m apart to cover most of the home range of the study groups. Due to transect sizes and the constraints of time imposed by a multi-group comparative study, I used sample swaths every 50 m (NRC, 1981 ). In these swaths I tallied all trees that occurred within 5 m of the trail (10 m x 1 0 m) with dbh greater than 10 cm. The diameters were measured to the nearest cm. Where buttresses interfered with the measurement, the section of the trunk between them was measured summed and added to estimate the portion of the trunk occupied by the buttresses. I ignored the presence of lianas in my tree measurements. Between July 1989 and June, 1990 I spent the first week of each month collecting phenological data. Since I did not know then which fruits the tamarins ate, I collected data of all tallied trees. assessed the phenological state of each tree monthly (n = 12 months

PAGE 59

41 for all areas). I checked for the presence and abundance of fruits flowers and buds I recorded these abundance based on the percentage of the canopy where they occurred (0-100%). For each study area I computed monthly means of the proportions of fruit trees which had fruits of all stages of ripeness By recording the abundance of fruits each month I compared the estimated availability of food for each group throughout the year. I also conducted a regression analysis to test whether rainfall explained a significant amount of the variability in fruiting Time budgets I started systematic behavioral data collection on the f irst group of black lion tamarins in August 1989. The other groups I started later, in subsequent months For each of the four target groups I recorded data on all animals that seemed to be older than one year of age for at least two full days per month for the period of twelve months I assigned age classes based on a combination of their weight and body size and on the basis of information from known-age captive animals. I tried to make the observations continuous by watching each group during daylight hours for 48 hours. Whenever data collection could not be done cont i nuously compensated by collecting the same amount of missing hours on the next possible day. Usually data collection started as soon as the first animal left the den tree--this generally occurred between 6:02 h and 7:43 h stopped collecting data when the last individual of the group entered their den which occurred between 16 : 16 h and 17 : 31 h

PAGE 60

42 When deciding on systematic rules for recording behavior, two levels of decision must be made: sampling rules to specify which subjects to choose and when, and recording rules to specify how the behavior is recorded. The methods I used were focal animal as sampling rule, and instantaneous or point sampling as recording rule as described by Martin and Bateson (1986). With focal animal sampling the observer records all instances of individual animal behavior and with instantaneous or point sampling as the recording rule the observer divides an observation session into short intervals and in the end of each sample interval the animal behavior is recorded. Focal animals were chosen randomly using numbers generated by a Hewlett-Packard calculator. Each observation session on a focal individual lasted ten minutes, with a ten minute interval between each session. These breaks allowed me to rest and to locate the next focal animal. During the ten minute sessions used sample intervals of one minute and sampled at the end of these intervals. To choose the size of the sample interval, I relied on a combination of common sense and trial and error methods. While habituating the groups, I did a pilot project in which I tried to find the interval that would balance theoretical accuracy against ease and reliability of measurement as suggested by Martin and Bateson (1986). Minute by minute data on the subject animal were collected in 12 categories. In many cases the categories contained two or more subdivisions. The categories and their subdivisions are listed below: 1) Time of day at which recording was made; 2) Quadrat occupied by the subjects (quadrats were identified by their corner labels);

PAGE 61

43 3) Identification of the subject animal (all animals were individually marked with collars); 4) Subject's height above the ground in meters; 5) Subject's action divided into discrete categories; 5.1) lnactive--an individual was stationary and not performing any other activity including resting; 5.2) Moving--an individual was locomoting; 5.3) Feeding--an individual was chewing fruit, exudate or prey ; 5.4) Foraging--an individual was searching for fruit prey or manipulating within a particular micro-habitat searching for animal prey contained; 5.5) Not visible--an individual was out of sight ; 5.6) Social--an individual was interacting with conspecifics of the same or a different group ; 6) Substrate--! recorded whether the animal was on the ground, in understory cover or on trees, vines or lianas ; 7) Object/substrate identity--whether it was on the trunk branches or twigs (for substrata less than 5 mm of diameter) ; 8) Capture success ; 9) Height in the tree; 10) Position in the tree; 11) Nearest neighbor; 12) Nearest neighbor distance. During the focal period, I also recorded all occurrences of rare behaviors or events involving the focal animal like mount i ng or copulating Finally, I kept an ad libitum" record book to descr i be unusual events or any other information that I thought important but that did not fit into the behavior recording forms. I could usually observe the whole group at the same time, as intra-group cohesiveness is high in L. chrysopygus. Maximum distances between animals of the same group rarely exceeded 25 m. This characteristic facilitated observation and data collection By using a combination of focal sampling and instantaneous or point sampling I obtained a score of the proportion of all sample points on which the behavior pattern was occurring for each animal

PAGE 62

44 in each study group. Despite being aware that instantaneous sampling does not give true frequencies of behavior, I present the results as frequencies because the intervals I used were short A series of studies have demonstrated that instantaneous sampling gives an approximation of the proportion of time spent in each behavior sample (Dunbar 1976; Leger 1977 ; Rhine and Flanigan 1978). Based on these methods, a total of 6,436 occurrences of behavior were recorded for the River group 6,440 for the Caldeira.a group 6 543 for the Morro group and 7 042 for the Little Bridge group. These occurrences sum up to a total of 24,461 records f or the whole study period and were derived from a total contact time of 815 hours on 96 sampling days (8.49 h/day) As for the intra-specific comparative analysis of behav i or actions, I used two methods of analysis of variance (ANOVA MANOVA) to simultaneously evaluate the effects of suites of independent variables (groups, and months of the year) on activity budgets (Neter et al., 1985; Kinnaird 1990). The assumptions of the analysis of variance were met by an arc-sin transformation of the proportional data (Sokal and Rohlf, 1981 ). Movements and use of space There are many approaches to defining the size of a home range area (Harvey and Barbour, 1965 ; Ludlow 1986 ; Rudran 1978; NRC, 1981 ). The number of quadrats entered by each group at the time the curve reaches an asymptote is considered in some studies as the home range for the species (Rylands, 1982; Peres 1986).

PAGE 63

45 However, I estimated the home range size of black lion tamarins in a more conservative way, including all the quadrats that were within the perimeter of the quadrats they used. In each home range I also recorded all the den trees used by each group during the year. To examine the intensity of use of black lion tamarin range, ranked each quadrat according to the number of times it was used by the group. To examine differences among use of quadrats by each group, I compared monthly mean use using analysis of variance. If the differences were significant I used the Waller-Duncan K ratio to compare the means for which significant differences existed. I also tested the distribution of sightings throughout each group's range against a Poisson distribution to determine if use of space was random. I tested den distribution in the home range space using the relationship between the mean and variance of the number of dens per sampling unit (Ludwig and Reynolds, 1988). By this method when the variance is equal to the mean the distribution has a random pattern, when is bigger has a clumped pattern and when is smaller a uniform pattern. The choice of a scale is important and patterns can be detected in one scale and not at another (Pielou, 1979). I used a 64 quadrats (8 x 8) as sample unit since with smaller scale I could not detect any pattern. Ranging behavior To record animal movements I used the same 50 m x 50 m grid system that was used to measure habitat differences. The aluminum tags placed in each quadrat corner allowed the observer to check his

PAGE 64

46 position while following the group. Group locations were recorded at each instantaneous sampling and all locations were plotted on graph paper maps. I obtained daily path lengths by direct measurement of t h ese maps with a manual odometer. In the analysis of rates of movements, I considered the straight-line distances between t he central point of each quadrat. This produces a relatively conservative distance estimate, because animals rarely move in a linear or uni-dimensional way (Peres, 1986). Only complete day samples were used in this analysis. Feeding and foraging behavior Many different methods are available to quantify primate diets (for review see Glutton-Brock, 1977). Due to the comparative nature of my study I measured the diet composition of black lion tamarins by recording the proportion of feeding and foraging time spent on different food items (Glutton-Brock, 1975; Richard, 1978). This method seemed to be very appropriate for this type of study since it is easily repeatable and is not subject to major biases due to differences in recording methodology (Glutton-Brock, 1975; Struhsaker, 1975). I used five pre-established behavioral categories to cover the many different components of food acquisition: feeding on anima l s; feeding on fruits; feeding on exudates; foraging on animals and manipulative foraging. These categories are almost all self explanatory with the exception of manipulative foraging which involves searching for food through movements such as probing woody crevices, grabbing, biting and/or turning over decomposing materials.

PAGE 65

47 Plant items in all feeding records were identified to genus or species l evel. The tamarins' feeding plants are relatively large, and easy to i dentify. Whenever I observed a tamarin eating a new fruit I collected a sample of it and fixed it in 70 percent alcohol. labeled and boxed all ea t en exudates and other miscellaneous items. This small collection of dietary specimens was used to help in the identification of black lion tamarin diet composition. Recording of animal prey was not as easy. Black l i on tamarins prey mostly on relatively small animals and ingest them in a rapid manner so I was rarely able to recognize the taxonomic group. Whenever possible I collected animal prey fragments dropped by the primates and conserved them in 1 O percent formaldehyde By the end of my st u dy period I had a total of 22 fauna components of the tamarin s diet. Unfortunately, the museum to which I sent them to be identified has lost them Nevertheless, the prey animals I listed were those I could identify. I measured capture success through direct observation also described the item and the substrate from where it was taken. As suggested by Robinson (1986) for animal prey, a "capture included discovery of an i t em but not necessarily its immediate ingestion This is a more reliable approach to the measurement of capture success. With this method I avoided the bias of a prey being successfully captured during the point sampling but only ingested in the intervals between samplings. Furthermore with this method I avoided misinterpretation of the feedin g item taken during the point sampling.

PAGE 66

48 I compared feeding behavior within and between groups. A chi square test compared the proportions of the three major diet items eaten by each group every month. Since fruits constituted the bulk of their diet I used two methods to distinguish the diet of the four groups in time and space The first was the percentage of feeding records of the most consumed fruit items by month and the second was the Renkonen index of percentage similarity (Renkonen 1938). This method despite i ts simplicity is one of the best quantitative similarity coefficients available (Krebs, 1989). The index range from O (no similarity) to 100 ( complete similarity) (Wolda 1981). I used analysis of variance methods (ANOVA MANOVA) to simultaneously evaluate the effects of suites of i ndependent variables (group and month of the year) on feeding and foraging activity budgets (Neter et al. 1985; Kinnaird 1990). Results Abundance and Distribution of Resources Variation with location The amount of data collected on plant resource abundance was slightly different at each site due to differences in home range sizes. For the River area for example, I recorded 3 231 individual trees in a total sample area of 6.65 ha. The Morro area had 4 907 trees in 7.49 ha, the Little Bridge had 5,034 trees in 9.4 ha and finally the Caldeirao area had 4,320 trees in 6 89 ha (Table 3-1 ). I found that some areas were quantitatively and qualitatively different from others. The River area not only had the smaller

PAGE 67

49 number of trees, but also the lowest occurrence of feeding trees per ha (303 frt/ha where frt = trees of spp from which the tamarins ate). The highest occurrence of feeding trees per ha was in the Morro area (534 frt/ha) followed by the Caldeirao and the Little Bridge areas with 506 f rt/ha and 442 f rt/ha respectively. Despite its small number of trees the River area had the largest diversity of trees with 64 species in 30 families while the Caldeirao had the smallest diversity with 44 species in 21 families. The other two areas ranged between these two. The average tree height and dbh were also found to be higher in the River area (8.42 m and 74.59 cm respectively) when compared to the height and dbh of trees in the other three areas with 7.66 m and 54.74 cm for the Caldeirao area, 7.58 m and 59.56 cm for the Little Bridge area, and 7.12 m and 55.12 cm for the Morro area. (see Table 3-1). These comparisons were useful to furnish information on the black lion tamarins' environments as their sources of food and shelter. The River area for example, has a forest composition quite different from that of the other study areas. The other three areas have a smaller number of tree species and a larger number of shorter and thinner trees. In the River area, the ten most common species belonged to seven families and only one of these species was on the list of the ten most eaten fruits by L. chrysopygus there. The Caldeirao area had its ten most common species from only three families, of which five species were from the Myrtaceae. Six of these tree species were among the ten most eaten fruits by black lion tamarins.

PAGE 68

50 Table 3-1. Characteristics of each study area quantifying availability of resources. Table 3-1. General characteristics of each study area considering availability of resources AREA RIVER CALDEIRAO tvmRO Number of Quadrats 665 689 749 Hectares 7 7 7 Number of Families 30 21 27 Number of Species 64 44 64 Number of Trees 3231 4320 4907 Number of Feeding Species 32 24 33 Number of Feeding Trees 2017 3487 4000 Average Height (m) 8 8 7 Feeding Trees per Hectare 303 506 534 Average D.B.H 24 1 7 1 8 L. BRIDGE 940 9 26 55 5034 24 4156 8 442 1 9

PAGE 69

51 The Morro and the Little Bridge areas had their ten most common species from six families. In the Morro area, five of these common species were on the list of the most eaten species, while the Little Bridge group only ate fruits from four of the ten most common species. Additionally, the ten most common species of each of the four study areas are reasonably different (Table 3-2). The River and Caldeirao areas share only two of the ten most commonly eaten fruit species and these two were in very different ranking order of abundance. The other two areas are more similar to each other although they also vary from the former areas (Table 3-2). Phenology To examine the annual variation in fruits available to the four groups of black lion tamarins, I plotted the number of species that had ripe fruits during the twelve months of my study (Fig. 3-1 to 34 ). This fruit availability was poorly correlated with rainfall: In only the Little Bridge area was there a significant correlation between the average of rainfall per month and the number of tree species with fruits (rs = .69 p<.005). The Morro and the Caldeirao areas had low positive correlation (rs of .19 ns and .33 ns) respectively. The River area was the only area with a negative correlation, although this correlation was low (rs = -.22 ns). Phenological data reflected the same general patterns of differences in number and taxonomy of trees among the four areas.

PAGE 70

52 Table 3-2. Ten species on which each study group spent the most time and percentage of total time spent in trees of each fruit species. Food species RIVER GROUP Myrceugenia ovata Syagrus romanzoffiana Campomanesia sp. Helietta longifoliata Cabralea canjerana Myrcia sp no.2 Ce/tes spinosa Vochysia tucanorum Zygocactus sp Psidium sp. no 2 CALDEIRAO GROUP Myrcia sp. no 2 Psidium sp. no. 1 Syagrus romanzoffiana Xylopia brasiliensis Eugenia sp. no 2 Eugenia uvalha Myrciaria sp. no. 1 Trichilia pa/Iida Myrceugenia ovata Sebastiana serrata ~ROGROUP Eugenia uvalha Syagrus romanzoffiana Xy/opia brasiliensis Myrcia sp. no. 1 Eugenia sp. no. 2 Ficus enormis Vochysia tucanorum Myrceugenia ovata Psidium sp. no. 1 Myrcia sp. no. 2 LITTLE BRIDGE GROUP Eugenia sp. no. 2 Syagrus romanzoffiana Myrcia sp. no. 1 Campomanesia sp. Cereus sp. Terminalia sp. Myrceugenia ovata Psidium sp. no. 1 Helietta longifoliata Anadenanthera falcata Time feeding ( % ) 15 .5 10 .2 8.9 7 6 5.9 4 8 4 1 3.3 2.8 2 5 30.0 13 9 10 .4 6.7 6 3 5.7 3. 1 2.9 2.9 2.0 10 9 10.9 10 7 10 .3 5.8 5.3 3.8 3.8 3.8 3. 1 22.3 19 .4 14.6 5.6 3.9 3.7 3.7 3.4 3. 1 2.8

PAGE 71

CJ) -:'. ::i ... Q.) Cl. .:: ..c. ~ CJ) ~ (.J Q.) Cl. CJ) 0 ... Q.) ..a E ::i z 53 20 T 1 8 1 6 1 4 1 2 1 0 8 6 4 2 0 I t J F M A M J J A s 0 N D Months Figure 3-1. Fruit availability to black lion tamarins in the River Area group measured by number of species carrying ripe fruit each month.

PAGE 72

20 1 8 ~ 1 6 2 ~14 ;:: ..c 1 2 : IJl (l) 1 0 u (l) a. IJl 8 0 .... 6 (l) ..c E ::::, 4 z 2 0 54 ...,.. -l-i--:I I \ I \ \ \ \ J F M A M J J A s 0 N D Months Figure 3-2. Fruit availability to black lion tamarins in the Morro Area group measured by number of species carrying ripe fruit each month.

PAGE 73

20 ::::, 1 8 + .... ~16 T ;:: ..c 1 4 -~ 1 2 T (/) Q) I u 1 0 + Q) a. I (/) 8 I 0 l .... 6 Q) ..Cl + E 4 ::::, z 2 0 J F M A M 55 n J J Months A s 0 \ \_ N D Figure 3-3. Fruit availability to black lion tamarins in the Little Bridge Area group measured by number of species carrying ripe fruit each month.

PAGE 74

20 T 2 18 + ] 16 t Q) I g1 4 l 12 1 I Cl) ~ 1 0 t I 8 0 6 ,._ Q) .0 E 4 ::::, z 2 .L I 0 I J F 56 M A M J J A s 0 N D Months Figure 3-4. Fruit availability to black lion tamarins in the Caldeirao Area group measured by number of species carrying ripe fruit each month.

PAGE 75

57 More than 50 percent of the fruiting trees in the Morro, the Little Bridge and the Caldeirao areas were members of the Myrtaceae (average = 62.3 percent; range = 54.2 percent to 76 2 percent). In the River area only 18.5 percent of the trees that fruited belonged to the Myrtaceae. In Morro do Diabo, trees from the Myrtaceae usually produce fruits in the rainy season, from October to February (pers. observ.) Time Budgets In the next sections of this chapter I will address a comparative investigation on how the four studied social groups of black lion tamarin use their habitat. Primates customarily do not use their habitat in a uniform way and in order to compare the ecology of the four study groups of black lion tamarins it is important to examine their activity budgets and patterns (Chivers, 1986) Overall black lion tamarins spent the majority of their time (53-70 percent) in inactivity ( defined as stationary and resting categories combined); social behavior on the other hand, was engaged in for only a short amount of time (0.2-0.5 percent) (Fig. 35). The four groups showed rather similar time budgets. The MANOV A failed to show significant effects of all the independent variables on group membership. The Wilks lambda test was calculated to be 0.38.

PAGE 76

58 1 00 9 0 8 0 RIVER 7 0 IE! CALDEIRAO . Ql 6 0 Ol II tv'ORR) C Ql 5 0 () .... L. BRIDGE Ql a.. 40 3 0 2 0 10 0 EAT Fal MOV NVIS INAC Behavior Action Figure 3-5. Time budgets of each of the four study groups. (Summed across the entire year).

PAGE 77

59 This is equivalent to an F ratio of 1.15 with 33 degrees of freedom. The probability of obtaining an F ratio this large by chance is less than 0.28. The ANOVAS analysis also failed to show significant effects except for inactivity (F= .64, .39, .01, 1.91 and .1 O for feeding, foraging, moving, inactive and social, respectively, df=4, p>0 005) (Fig. 3-32). These results indicate that despite the differences in habitat by area statistical tests failed to show differences among the four groups in their activity budgets for the six examined behavior categories. However, there was a significant difference in monthly variation in the time budgets of the four study groups of black lion tamarins' (Table 3-3). I use the term local variation to express the differences observed among the study groups at Morro do Diabo. I have not analyzed time budgets based on individuals or their age or sex classes. However, according to my observations if there were differences, these were not striking. Movements and Use of Space Home range use can be described as falling into 4 categories: uniform coverage, focused on shifting patches, concentrated at the center or at the periphery, or around the periphery of the area (Terborgh, 1983). Each pattern reflects the influence of a dominant factor in the life of the species: for some species that factor is the structure of the habitat, particularly the vegetation (Peres, 1986). For other species it can be social interactions. In this section I examine how do the four study groups of black lion tamarin's use of space differ as a function of habitat?

PAGE 78

60 I FF EE i DO OC l BBI A/AZ X I V I U T s R a I P i O I N I M i l J i i I H i G i F I E I D I C I B i A -11 I I I I I I I I I i : I -11 I I I I -1 o I i I I I I I i I I -1 0 -9 1 i I i I l i i i l I I I -9 I -8 I I I i I i I I i I : I I 8 7 1 I : I I I rh i i i I I -7 -6 1 I I I I I r,--lcI 6 I j .-, -5 i I I i I I I i I i -5 I ...... I I -4 i I I i ) ""'VT -+.,.... ~ I I I I 4 -3 I I I I. r I I I I -3 2 I ... i I I I -2 .i. -1 I 7"'I y ,,, I ~ ... 1 I I : 1 1 i I ~ A. iI / ~ 1 2 1 1 L~ 1 ,. Y i I I r-rr~ I I I 2 3 1 I I L,i-, 1 I i I I I _:; I I 3 4 I I A [ I I I i I I J ----,, 4 S i I ,I I ... I i I '/ I I 5 6 1 i i : .: I -~ I i j : I I .J 6 1 : I i i 1'. I I A. I L I L:...J I 7 I I : 8 1 : I I I I l'-t, ,, J I I "-! : I I 8 9 1 : I I I '-J 1 r I I 9 ,I I r7' 1 0 1 i I :' .... I I r, I i r-i-' I 1 ""'11" i i 1 0 11 1 i [\ I -J i i ,) : 1 1 I I I I I 1 2 1 I i I I -iJ .,. .h_ l ,I I I i 1 2 I I 1 3 1 I tt!. \ I "II:"' I' 'lii,,L : I I I I 1 3 I 1 4 1 I I --I I I i 14 I I I 1 5 i i T H ... I I I I i I 1 5 1 6 1 i I I l. ... --;., I I I I I I I : I 16 17 I I I I -=, .... i I I i I i 17 1 8 I I .... ,~ l"'I I -> I I I I I 18 1 9 I i ,,, 'I.. .J ....,--4!""T -I I I I I 1 9 20 1 ; I I i I ,I I I I I I i : 20 I I 21 I I i I I I I I I I 21 2 2 1 I I I I I i I I I ; I 22 I 23 1 I i I I : .,,. 1 I : 1 --, I I I I 23 I I 24 1 : I I I I I L'.... I 1 I i I I I 24 I : 25 1 ; I I L' I I I i 25 I I I I i 26 I I i l 1J/~ I I :"Il l I I : 26 27 i I I I I .... I I 27 I 28 I I I I ........ I 28 29 i I i I ....... 1 1 i 29 30 I I I I i I I L I 30 31 : I I I ... ... I 31 32 I : I I I ""' i--, 32 33 i I I l'-U 33 34 I I I I 34 FF EE DC ex: BB A/AZ X I V u T I S R I O p 0 N I M I L J I H G I F E D C I B I A I Figure 3-6. Movement of the River Area group during the study.

PAGE 79

61 1 2 1 3 i 4 1 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 20 2 1 2 2 2 3 24 25 26 2 7 2 8 29 3 0 1 3 1 l 32 1 33 34 : 35 1 36 1 37 1 HH I ... I "'I I HH a:; I ,c. -t:: ... "I I I ex; FF I I I ~ I FF El: I I I I m El: I 00 I '\.. I : i I 00 cc I cc BB --~ )/ "'! J I BB AA { l ,r '\ ,. ,, '\. I A~ z \ I" l ., 7' I z X I I X V I I I I "'C'"" i V u i I J. "' .. i !"'i.. 1 : u T I I \ ...,. ,,,. '\. : I ~ T s : \ l. I..~ s R I I 1 L.. I J I I R a I i \ -I"\ I' 'l""v' I a p LI I I I\ Ji"\ I 1 P 0 I I I ,v ... LI,. I +--f I \ : o N I .. ... \ I N I M I --1' ,., .. _,,. ,_ '1 I M L ,,. ...I ----r !J I M" L J I I- r;;:, ..I I,"" """ I .. i J I 07 ..... .. C. .r, I 4 ...:..1 I I H ; .... 'II,,... ""' .,,. -_., ._ ._._...j I H G ~. ,...,,. ... f"'I I..., I I I G F : r ... -::li ]; I : I F E .,, L.:!11. I ... 7' I'1 I I E 0 I I I I o C .. i I C B I I I B A I I i I I A 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 14 1 5 1 6 1 7 1 8 1 9 2 0 21 2 2 2 3 2 4 2 5 2 6 27 28 29 3 0 31 3 2 3 3 34 35 l 3 6 l 37 1 Figure 3-7. Movements of the Morro Area group during the study

PAGE 80

62 u 1 T I s R I Q 1 P I O I N M I L : J I H G F E O C ; B A I AA BB I CC 1 00 I EE FF 00 I HH II JJ I LL MM I NN OO I PP I OO i 48 + I I I I I I I I 1 48 4 7 1 I i I i I I I I I I i i 4 7 46 1 I I : I I i I I I 46 45 1 I I i I I 7 7 4 5 44 1 I i I I I I I I I 1 44 43 1 I I i i I, I i I 1 ~ 1 : i i 1 43 42 1 : ~ l.. i I I : l 42 41 I I I I I i I ) I i IA ., f'-..L... 7 i I 41 40 1 I I I 1 ., I : -1-, i 7 I I 1 40 3 9 1 I I ; V C J I I i I 7 ; l I 3 9 3 8 1 I : I I ,.._.. I l "' J I I I 7 3 8 37 I I I i L. : 3 7 36 I I I <'f.....__ I I I 3 6 3 S t I I ..., ~ I / I 35 34 I / I -~ I I I i ~ I 3 4 3 3 1 ~ i / "'I I I..., ,' 3 3 3 2 : I I I ( ....,., i ""'! I ,j : i 3 2 31 I I : I I I I I.) I I 1.J .....,.... i 3 1 3 o : I I I I i I ri I I i I i 3 0 29 : /! : l i .-;,, I I I : 2 g 2 8 1 I I ;J I I ~ i I r...... 1 ; ~ I I I 2 8 27 i ,.., I r I I I i"" _;..a..., : ...._ I I I 2 7 2 6 1 I I i I 1 ----,-a ~ I i"I I I H-, I 26 2 5 i ,.. I I .__ I i I ......... 1 ,--r-, 1 I i 2 5 24 1 ..)/ I I .,.~ I I I I I N I I i 2 4 23 1 I J II,. i "'I" ... .,... ,o(._ I I I I I 23 22 1 J _..., ~r ., I i I I I I I l 22 21 I /'I \ J i j.J i I I I I I I 21 20 1 \.I 1.L 1 ...,. i I l 20 a l 1 : ~ I ca. I I I : 1 9 1& l "114...r '-. -;::.i I \ ,J' ''[_.. ..... I I I I I i 1 8 17 1 ~, I "" i I I I I ( "' 1 7 i i i 16 1 < l,J / I I '1 l-,---' I I I 1 6 1 5 1 N X r l I .A. i L I I .. r 1 5 14 I I i ~ ~1 ( i'-+J "\ n. '""!:I : t / I I l'X L.; -"C i 7 14 1 3 i I ( I / I ) i I I I n .1 1 3 1 2 1 I ,, /J/1 i.J.. I ... V I I J..-"'I I I I i V i I I :Ai.,' I 7'l_ I 1 2 11 1 I I ,.,~ ... __ ,-1 .... I /I I I ,. I 11 10 1 I L r I ... 3,..,.._ 1 .... .., I i/U I. I r 1 I I 10 9 1 I I r-tI r,.. 1 : I 1"1 I VI/ l }'l '~ I 9 8 I i l I ~ L I 1-.. ( 1./1 J I 1 V" I 8 7 I I ,it. ..... I ~ I i',.. ...... I I/ 'I I 7 6 I I :, I r,I u i N 6 5 I ,,.. -!-' ,._ l'r......, V .... ... : I I' 5 4 I I I ~ !-i.. ( 4i I I 4 3 1 I -~ rl..J 1 ... i I I I i I 3 I 2 I i i,----i I I i I I I I I I 2 1 1 I .,_ I -i--r I I I I 1 l u I r s R a p 0 N M L I J I H G F E l o I C I B I A AA I BB cc 00 I EE I FF OO I HH I II i JJ 7 LL l MM I NN oo pp aa 1 Figure 3-8. study. Movements of the Little Bridge Area group during the

PAGE 81

I A j s C D I E F G H I [ J 52 I 1 I I 51 1 50 1 49 1 I 48 1 47 1 46 [ I I 45 1 I I 44 1 I I 43 1 42 1 I I 41 1 I 63 L i M I N 1 0 P Q I R I I l "-r, I '--h I : "I I i I I l I l s i T i u i v I x z I AA I BB 1 cc 1 00 I : I I I 52 i I i I 51 : I 50 i 49 I 48 I I I 47 i I I 46 I I I i : I 45 I i I I I 44 I I 43 42 I I 41 I I I I I 40 40 39 1 i J,,---I I I : I 39 38 1 37 1 36 1 35 1 34 : 33 32 : 31 [ 30 1 29 1 28 1 27 26 25 1 24 1 23 I : 22 1 I 21 [ 1 2o i I 1 a l 17 j 15 1 14 1 12 1 I 11 I 10 1 9 I a l I 1 I I 6 1 I 5 i I 2 I i 1 I I I A I s I I I I I I I I I I I i I I I I I : 1 I i ; I "' _" i I I I~ ... I I I J~ !X J.-. -, I I t~ I I I J I "" I 1 -b, i .._......_ I I < 1 ., I I I I I I I I I --~ I .. -~ I -:: i i I I I ( 7' I I 1,.1 I : \r\ I I : "1 .. I I \. \ \. I i i i I I I I i ; I I I I I /l : rf-/ I j 38 37 I 36 ; 35 34 33 I 32 31 30 29 28 27 I 26 I 25 : 24 1 23 22 I 21 20 I 19 I 1 a I 1 7 16 15 I 14 I 13 -....1 1 .7 I ""-:: J I ... ..1 : I 12 ( l \ i I I i : i I 11 I 1 I 10 I I 1 I 9 I I i 1 8 I I 7 I I 6 I I 1 I 5 I I I 4 I I I 3 I I I 2 I : I 1 C I D E F I G H I I J L i M N O P Q R S T u v x i z AA I BB cc : oo l Figure 3-9. Movements of the Caldeirao Area group during the study.

PAGE 82

64 The means of the daily path length for the 77 full days of the data collection ranged from 1,362 m for the River group to 2,088 for the Little Bridge group. The rate of movements measured as mean distances traveled by each group, each month is presented in Fig. 310 to 3-13. Cumulative group movements for all full day observations in each group are presented in Fig. 3-6 to 3-9. For estimating group movements, I also assumed that the lion tamarins moved in straight lines from one point to the next. Height and substrate My data indicate that the use of forest heights by the four groups of black lion tamarins did not vary greatly. In all sites they spent on average a considerable amount of time slightly above the middle level of the forest. The average height used by the River group was 8 5 m (S.D. 2.7) while the other groups were between 7 and 8 m (mean height = the Morro group, 7.1 m, S.D. 2.0; the Little Bridge group, 7.6 m, S.D. 2.2 and the Caldeirao group, 7.7 m, S.D. 2.0). Black lion tamarins rarely go to the ground. During my entire study I only once saw an individual from the River group go to the ground for a few seconds while chasing a grasshopper. Although I have not conducted a full analysis of feeding or sleeping average heights, preliminary results indicate that these activities occurred at all levels and at least in one case, black lion tamarins spent a night below the ground level in a den that was deep in the tree roots.

PAGE 83

5000 475045004250en 4000... Q) 3750Q) E 3500-0 Q) 3250> 0 3000Q) 2750CJ c:: Cll 2500(JJ i3 225022000ca 0 1750Q) Ol 1500Cll ... Q) 1250> <( 1000750500250D I Jan T D l I Feb T D ..L I Mar I D 1 I Apr l D l I May T D l I Jun D I Jul Months T D l I Aug T D 1 I Sep T ..L I Oct T T l l. I I Nov Dec Figure 3-10. Mean daily path length S.D.) for the River group during the one year study 0) 01

PAGE 84

5000 4 7 50 4500 4250 en 4000 .... Q) 3750 Q) g 3500 l -0 Q) 3250 > 0 3000 Q) 2 7 50 (.) C co 2500 l en 0 2250 2000 I T O') co T O') 0 T 17 50 l Q) T l O> 1500 co l T .... Q) 1 250 > T l <( 1000 1 l 7 50 1 500 250 Jan Feb Mar Apr May Jun Jul A u g Sep Oct Nov Dec Months Figure 3-11 Mean daily path length ( S D ) for the Morro group during the one year study

PAGE 85

5000 4 7504500 en 4250 ... 4000Q) Q) 3 750E l:l 3500 Q) l > 3250 0 3000 D Q) l u c:: 2750 T D T en l l Cl) 2500l D 0 T l D 2250 D T D ..L en 20001 l D en 0 D D -..J Q) 1750l D l C'l l T en 1500T ... Q) > 1250D D <( .l 1 10007 50 500250I I I I I I I I I I I I Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months Figure 3-12 Mean daily path length S.D.) for the Little Bridge group during the one year study.

PAGE 86

5000 4750 en 4500... Q) 4250Q) s 4000"Cl 3750 Q) > 0 3500 Q) 3250u C: 3000ca Cl) 2750i:5 2500 ca 22500 Q) 2000CJ) ca ... 1750Q) > <{ 15001 2501000750500250l l l j l l l l T l l l. T l I I I I I I I I I I I I Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Months Figure 3-13. Mean daily path length S.D.) for the Caldeirao group during the one year study. 0) (X)

PAGE 87

69 Home-range areas The cumulative percentage of quadrats entered by each of the four study groups as a function of observation time will be presented in chapter 4. All the cumulative number of quadrats entered by each group had reached an asymptote by the end of the first year of data collection with the possible exception of the Little Bridge group. The total number of quadrats entered by each group at the time the curve reaches an asymptote is considered in some studies as the home range for the species (Rylands, 1982; Peres, 1986) However, estimated the home range size of black lion tamarins in a more inclusive way, including all the quadrats that were within the perimeter of the quadrats they used (Rudran, 1978). I sighted black lion tamarins of the River group in 294 quadrats, the Morro group in 342 quadrats, the Little Bridge group in 447 quadrats, and the Caldeirao group in 318 quadrats (Fig. 3-19 to 3-22). The inclusion of peripheral quad rats increased the home range sizes from 73 km 2 to 1.13 km2 for the River group, from .85 km2 to 1.20 km2 for the Morro group, from 1.12 km2 to 1.99 km2 for the Little Bridge group and from .79 km2 to 1 20 km2 for the Caldeirao group (Fig. 3-14 to 317). Overall, the mean home range size for L. chrysopygus in Morro do Diabo using the first method or only the quadrats they were seen to enter would be 0.87 km2. If one includes quadrats within their range boundaries, the mean home range size at Morro do Diabo was 1.38 km2.

PAGE 88

70 I FF I EE DD I CC BB AA Z X V u T s R a p 0 N M L J I H G f F I E I D I C i B I A -11 I I i I i : l -1 1 -10 I I I : I I 10 9 1 I I I I I 9 8 1 I I I 8 7 1 I I I I 7 6 1 I I 6 S i I i I I I -5 4 1 I I I 4 -3 i I 3 2 1 I I I I -2 1 I I -1 1 I i I I I : 1 2 1 I 2 3 I 3 4 1 i I I I I I : 4 5 I I I 5 6 1 I 6 7 1 i ; I I 7 I S i I i I I I 8 9 1 i \ I I : 9 1 0 I I 1 0 11 I I I i : I 11 1 2 i I I I I I 1 2 1 3 I I I I I 1 3 1 4 I : I I 1 4 I 1 5 I I I I i 1 5 I I 1 6 1 i I I I 1 6 11 1 i I I I I 1 7 18 I I 1 8 1 9 I 1 I : 19 2 0 1 I i ; : I 20 2 1 1 I [ I I I 21 I 22 I I 22 23 I I 23 24 24 25 I 25 26 i I 26 27 I I I I 27 28 28 2 9 1 I 29 30 I I 30 31 I l 31 3 2 I I 32 33 I I I 33 34 I I ; i 34 FF EE DD cc BB AA Z X V u T s R a p 0 N M L J I H G I F I E I D C B A Figure 3-14. The 1 .13 km 2 home range of the River group on a 50 m x 50 m grid.

PAGE 89

71 1 1 2 1 3 1 4 5 6 7 8 II 1 0 11 1 2 1 3 14 1 5 1 6 1 7 1 8 111 20 2 1 22 2 3 24 2 5 2 6 2 7 2 8 211 3 0 I 3 1 I 32 I 33 I 34 I 35 l 3 6 3 7 i HH I I I I I HH ex. I I ex; FF I : FF I I EE EE I I DO I i I D D cc : cc BB I I 1 BB AA I i I I Al z z X I : X V I l I I V u I I u T I I I T s : : I s R I I I I i R a I a p I I p 0 i I : I I 0 N : I I N M I I i M L I I I I : L J i I I : I I J I I : I : I i i H : I I i i H G i I I G F : I i I F E fl i : J E D I : i D C I I I C B l B A : I I A 1 1 2 a l 4 1 5 6 7 8 II 1 0 11 1 2 13 1 4 1 5 1 6 17 18 111 2 0 21 2 2 2 3 2 4 2 5 2 6 2 7 28 211 1 3 0 I 3 1 I 3 2 : 3 3 1 3 4 1 3 5 36 37 Figure 3-15 The 1.20 km 2 home range of the Morro group on a 50 m x 50 m grid.

PAGE 90

72 u i T s 1 R a p 0 N M L J I H G F E D C B A A~ BB cc DD EE FF 00 HH I I I JJ I LL '-M I NN I CO I PP I QQ I 48 i I i I I 1 4 8 4 7 1 i 4 7 46 I I I I I 4 6 4 S I ; I I i I i 1 4 5 44 I I I 1 44 43 I I I I : 4 3 42 I I I I I I 42 4 1 1 I I ; 41 40 ; I 40 I 3 9 1 I I I I 3 9 3 8 38 3 7 1 i I I 37 3 6 i I 36 3 s l I 35 34 1 34 33 I I I 33 3 2 I : I 32 31 1 I I 31 30 i I 30 29 I I I : I 2 9 2 8 I I I i I 2 8 27 I I I 27 26 I 2 6 25 I i I I I I 1 2 5 24 I I I 1 24 23 I ; : I I 1 2 3 22 I I I i 2 2 21 I I I I I I 2 1 20 I I I I I 20 19 I i I : I 1 1 9 1 8 I I I I I I I 1 8 1 7 I ; I I I 1 7 16 I I I I 1 6 1 s I I I 1 5 14 I I 14 I 13 I I : 1 3 1 2 I ; : I 1 2 I 11 I I I I I 1 1 1 0 i I I : I I I 1 0 9 I I I I 9 8 I I I 8 7 I I : I 7 6 I I I I 6 S I I 5 4 i I I I 4 3 I I I I 3 2 I ; i 2 1 I I I I 1 u I T s R a p 0 N M L J I H G F E D C B A A, BB cc DD EE FF 00 HH I I I J J I LL i '-M l NN l co l PP l aa i Figure 3-16. The 1.99 km 2 home range of the Little Bridge group on a: 50 m x 50 m grid.

PAGE 91

73 I A 8 l e I D E F G H I J L M N 0 p a R s T u V X z I AA I BB I CC I DD I 52 : i I i 52 51 I I I i I I 5 1 50 : 5 0 49 1 I I 49 48 i I I 48 47 I I j I I 4 7 46 i I 46 45 I I i I I I 45 I 44 i I l 44 I I 43 1 I ; i I I I I 43 42 I I I i I 42 41 I i I I i 41 40 I I i 40 39 I I 39 38 I : 38 37 i i I I i I 37 I 36 1 I I I I I I I I I 36 35 I 35 34 1 i I I 3 4 33 I I I 33 I 32 I I 32 I I 3 1 1 i I I 31 I 3 0 I I I 30 I i 29 i I I 29 2 8 I i I I i I 2 8 27 I i I I I 27 I 26 I : I : I I 26 25 I I I I i I I i I i 2 5 I I 24 I I I I i I 2 4 I i 23 I I I I : 2 3 22 I I I 22 21 I I I I 2 1 20 I I I I 20 19 I I I I 19 18 I I 18 17 I i 17 I 16 I I I I 16 I 15 i i 15 14 i I 14 1 3 1 I i I 13 12 I 12 I 11 1 11 10 I : 10 9 I 9 8 I 8 7 i 7 6 6 5 I 5 4 4 3 I 3 2 I I '. 2 I 1 i I I 1 A 8 C D E F G H I J L M N 0 p a R s T u V X z AA BB I CC DD Figure 3-17. The 1 20 km2 home range of the Caldeirao group on a 50 m x 50 m grid.

PAGE 92

74 Intensity and diversity of guadrat use A clear variation existed in the monthly use of quadrats between the River group and the other three groups (Fig. 3-18). found the four group means to be significantly different from each other (F=4.52 df = 3 Pr>F=0.0076). Although the difference was significant, the ANOVA test did not locate where the means differed. For that purpose I used the Waller-Duncan K-ratio T test as a multiple comparison test. The results showed that the River group used fewer quadrats than the other three groups, which is consistent with the fact that their home range size was the smallest of the study groups (Fig. 3-18) Likewise the use of space by each group was not distributed evenly The River group concentrated much of their time along the perimeter of their range. The Morro and the Little Bridge groups used their space more evenly but still showed some tendency for a greater use of the edges of their ranges The pattern shown by the Caldeirao group was distinctly different showing intense use of certain parts of the periphery of their range, but concentrating mainly in a core area in its center. The patterns for their home range use are illustrated in Fig. 3-19 to 3-22. I tested the distribution of sightings throughout each group's range against a Poisson distribution model (Ludwig and Reynolds, 1988) to determine if their use of space was at random I found the sightings significantly deviated from the expected Poisson distributions for all four groups.

PAGE 93

60 so40(/) cii .... "'C ro 30:::, a 0 1 z 20100 ..... ......... ... ... ....... . . . . .... ... ........ : ::::::: ::::: ::: :: ::: :::: :: ::: I Caldeirao 75 11111111 111 :::::::::::::::::::: :: :::::::: /\U// H/I I L.Bridge Groups I Group 3 ... .. ... .. .. .. ........ ... .. :::::::::::::::: ::: ::::::::::: : ; : i: i: ~: ~: ~: ~: ~: ~: i: ~: ~: : ~: : : :::::::::::: :: : : ::::: ::: :::: -: -::-: -:: -:-:-:-: :-:-:-:: T River Figure 3-18. Means with K ratio). Monthly mean use of quadrat by each study group same line are not significantly different (Waller-Duncan

PAGE 94

76 E l o J c a I A I -11 -1 0 9 1 l -9 8 I 8 -7 1 -7 6 6 -S i -5 -4 1 I i l -4 3 1 -3 -2 1 -2 -1 -1 1 1 2 : I i 2 ----1 3 1 3 4 4 5 5 6 6 7 7 8 1 8 9 1 9 1 0 1 0 1 1 1 11 1 2 1 2 1 3 I 1 3 1 4 1 5 I 1 5 1 6 1 6 1 7 I I 1 7 1 8 i 1 8 1 9 1 1 9 20 I I l I 20 21 I 21 2 2 1 I i 22 23 I I 23 24 I 24 25 25 2 6 1 26 27 27 28 28 2 9 1 30 31 32 3 3 3 3 3 4 34 FF I EE DD I CC BB AA Z X V U T S R Q P O N M L J I H G 1 F E I D C 8 A Figure 3-19. Intensity of quadrat use by the River group.

PAGE 95

T s R a L H G F E D C B A 77 g 10 111213 1415 1& 1118 a 20 21 22 23 24 25 26 21 28 2g 30 J 31 32 33 34 1 35 1 36 1 37 1 Figure 3-20. Intensity of quadrat use by the Morro group.

PAGE 96

78 0 N M L i J I H 1 G F I E O C B A AA I BB CC I O O EE J J I LL MM I NN I OO R' aa l I I i 4 8 1 4 7 1 4 5 45 1 4 4 4 3 ; 42 41 4 0 39 38 3 7 37 3 6 1 3 5 3 5 3 5 3 4 1 I 3 4 3 3 3 3 3 2 3 2 31 I 3 1 JO I I 3 o 29 2 9 28 1 2 8 2 7 1 2 7 26 1 2 6 ; 2 5 24 1 2 3 : 22 1 2 1 i I 4 1--+--+--.--4 3 -----f,,,,. .... .,;+--,--. -,--,--+--.----,--t---: 3 2 2 OO I H H II JJ LL MM NN l oo R' l aa l Figure 3-21. Intensity of quadrat use by the Little Bridge group.

PAGE 97

79 A 1 e 1 c i o l e F I G H 1 J L I M N o P : a I R 1 s I T l u v I x l z I AA I BB 1 cx: 00 1 52 I 52 I 50 I 49 I 48 4 7 l I 46 45 44 43 1 43 42 I I 42 41 I 41 40 1 I : ~ I : I I I r--i I I : 40 39 39 38 1 38 37 1 I I I I 3 7 36 1 3 6 35 1 35 34 1 3 4 33 33 32 32 31 1 31 30 3 0 29 I I 2 9 28 i I I I I I 28 27 1 27 26 1 I 26 25 i I I 25 24 1 I 24 23 1 2 3 22 2 2 21 i I 21 20 1 20 19 1 10 1 17 1 I I I 17 -+---,----+--+--, ~ ---1 ---1 6 15 I I 1 15 14 I 14 1 3 13 12 i 12 1 1 1 11 10 1 10 9 9 8 8 7 7 6 6 5 5 4 4 3 3 2 2 A l e C D E F G H I J L TU V X Z I AA I BB I CX:00 1 Figure 3-22. Intensity of quadrat use by the Caldeirao group.

PAGE 98

>, (.) C Q) ::l O" Q) ,_ LL 80 50-.------------------------------. 4 5 4 0 35 30 25 2 0 1 5 1 0 5 0 0 .... .. 0 0 0 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ Number of times quadrats were used Figure 3-23. Frequency distribution of intensity of quadrat use by the River group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution)

PAGE 99

81 50-,-----------------------------, 45 40 35 30 >(..) C: 25 CT Q) .... LL 20 1 5 1 0 5 ....... <> <> <> .. ... ....... o ...U ;;;;. ;;i;; ;;;;, LL -:.::; ~ ;;;l ---1; 2.;; -~ ~ LJl;, :4 :.:.::: :J..._~LJl::.:. :::: ~ : ~::::; :'.:'. i : ___..=Q. ......... ....c:::::::o:::::::i..,_J,;;, ~ :,;:. ;;.;;i LJ 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70 79 Number of times quadrats were used 80+ Figure 3-24. Frequency distribution of intensity of quadrat use by the Morro group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution).

PAGE 100

>, (.) C: Q) :::::, 0Q) .... u. 82 50 45400 3530250 201 5 10-:-: :'!": -:-: 5 0 I k ::::::::: ::I 1:::: : :0: :::: i .A. I I I I I T T T 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ Number of times quadrats were used Figure 3-25. Frequency distribution of intensity of quadrat use by the Little Bridge group. Diamonds indicate the values expected if quadrat occupancy were random (following a Poisson distribution).

PAGE 101

>, (.) C 83 so~-------------------------------, 454035<> 30. ... .. 25<> 0" u... 2015<> 105tl]t:: 1 :::::: 0.: :) 1 . : I 0 ......_._ -,-, ....._ ........ ..., j--.___ ......... ...,. j--.__ ........ _,,;,-. .... _...__.,.., ........ ____. ... .... ... .... .... : 1 _____.1 .. ..., -;-~ -..._l_.,l __ ~ ~~ -.. 1.___._ ....
PAGE 102

84 This can be better visualized in Fig. 3-23 to 3-26 The four groups used their home-range space very unevenly. The cumulative proportion of time spent in a quadrat plotted against the percentage of different quadrats used by the tamarins showed that the River, the Caldeirao, and the Morro groups spent over 50 percent of their time in less than 20 percent of the quadrats in their home ranges (Fig. 3-27). The Little Bridge group which had the largest home range size, distributed its time less evenly than the other three groups. This group spent over 50 percent of its time in 5 percent of its range quadrats (Fig. 3-27). Sleeping sites The average number of tree dens used as sleeping sites among the four study groups of black lion tamarins at Morro do Diabo was 17 (ranging from 16 to 19) (Fig. 3-28 to 3-31). L. chrysopygus rarely returned to a tree den used on a previous night although one study has documented frequent return at another site (Valladares Padua and Cullen Jr. 1992). During the study period, only once did a whole group sleep outside a tree den. On two occasions I observed one or two animals sleeping outside the tree where its group was spending the night. In both cases agonistic behaviors had been observed during the preceding day, and I interpreted this observation to indicate social conflict within the group. Also the dens are located in the or close to the areas of intensive use by L. chrysopygus.

PAGE 103

Q) E i= 0 100 75 ---0--River O Caldeirao 50 ----0---Morro ----6---L. Bridge 25 o~----------------------.---------r-----------, 0 25 50 75 100 % Ouadrats Figure 3-27. Distribution of the duration of quadrat accupancy Figure plots percentage order of quad rats against percentage of time of quad rat occupancy. 00 CJ1

PAGE 104

86 FF EE DD cc BB AA Z X V u T s R a p 0 N M L J I H G F E D C B -11 i I -1 0 1 s Den Trees I 9 8 I I I 7 l I I -6 1 -5 -4 1 I -3 1 2 1 5 I -1 I I I I I I 1 I I I s I I 2 1 s i I i 3 1 S s I ; I 4 1 I i I I I s l I I I i i : 6 1 I I I i I I s 7 1 i I i & j I I I I 9 1 I I s 10 I I I I 11 1 2 1 s I 1 3 1 4 s s 1 S I I s I I I I i 16 1 i I I 1 7 1 8 I s s 1 9 1 I I 20 1 I I I 21 22 I I 23 1 24 s 25 1 s I 26 27 28 I 29 30 31 32 33 34 FF EE DD cc BB AA Z X V u T s R a p 0 N M L J I H G F E D C B Figure 3-28. Spatial distribution of tree dens in the River group home range. A 11 1 0 -9 -8 -7 -6 5 4 -3 2 -1 1 2 3 4 5 6 7 8 9 1 0 11 1 2 1 3 1 4 1 5 1 6 17 1 8 1 9 I 20 21 22 23 I 24 25 26 27 28 s 29 30 31 32 33 34 A

PAGE 105

87 1 2 1 3 1 4 5 6 7 8 1 g I 1 0 11 12 13 1 14 1 5 1 6 17 1 8 u2o j 21 22 232425 26 27 28 1 29 30 31 3 2 1 3 3 3 4 1 3 5 36 HH i i I CD s Den Trees I FF I I I EE I I DD cc I I BB s AA s z i I s s X s s 5 V I u I s T I I s ; R I I Q I p I 0 I I I N I s s I M I l L s J I I i s H s s s G I ; s : F s E s D C B I I A 1 2 3 4 5 6 7 8 9 I 1 o 11 1 2 13 14 1 5 1 6 1 7 18 1 9 20 21 22 23 24 1 25 26 27 28 29 30 31 32 33 34 35 Figure 3-29. Spatial distribution of tree dens in the Morro group home range. s 36 3 7 HH CD FF EE DD cc BB AA z X V u T s R Q p 0 N M L J I H G F E D C B A 37

PAGE 106

88 u T I S R a p 0 N M L J j l H G F I E D C 1 8 I A AA 88 cc DD EE FF G'., HH I II J J I LL MM I NN I OO I PP 1 00 1 4 8 I i I : I : 47 s De n T re es I I I I 4 6 1 I I I I 4 5 I I I I I I I 4 4 1 I I I I 43 i I I I I 42 I I I I I I 41 I I I I I i I I 40 I I I I : 3 9 I I I 38 I I : I I I I 3 7 I I I I I ; 3 6 1 i I I : i I I I I I i 3 S I I i I i I I i I 3 4 1 I I I I I I : I 3 3 1 I I I I I I I i I I 3 2 1 I I I s I I I I I : 3 1 I i I I s I I I I : I 3 o l i I I I i I I i : I 29 i I I 28 s I I I i I I 2 7 I s ; i I I 28 i I 25 I i 24 I i I 2 3 I I I I i I I 22 I I I I I I 2 1 I I I I I I I 20 I I I I I I I I 19 i : s I 1 8 I s s I I I I 1 7 I s I 1 6 I I I 1 5 I I i i S 14 I S I 1 3 I I I 1 2 I s I 11 I s I 1 0 i i I g I 8 I s s I 7 I 6 s I I s s 4 I 3 I i I 2 I I 1 u T s R a p 0 N M L J I H G F E D C 8 A I AA 88 cc DD EE FF G'., HH II J J I LL I MM I NN I OO I PP I QQ Figure 3-30. Spatial distribution of tree dens in the Little Bridge group home range. 4 8 47 46 4 5 4 4 43 4 2 i 4 1 40 39 3 8 I 3 7 3 6 : 3 5 3 4 I 3 J 3 2 I 31 3 0 2 g i 2 8 1 27 I 2 6 25 24 23 2 2 2 1 2 0 1 g 1 8 1 7 1 6 1 S 1 4 1 3 1 2 11 1 0 g 8 7 6 s 4 3 2 1

PAGE 107

89 A B : c D E F I G H I I J L I M I N i Q I p 1 0 1 R I s r : u V i X Z I AA : BB i CC : oo 52 i 1 I I I i i I I I 52 51 i S i D e n Tree s I I I i I 51 so i i I I I I 50 49 1 I I I I I I 49 48 1 I i 48 4 7 1 I I I I I 47 46 i I : I : I 46 45 1 I I I 45 44 1 s I 44 43 1 43 42 I i 42 4 1 1 I I s 41 40 1 39 1 I I 38 1 I I 37 1 I I 36 35 34 33 1 32 1 31 30 29 2 8 1 2 7' 26 1 25 1 I I 24 1 I I 22 1 21 1 : I i 20 1 19 1 i i 1 8 1 I 17 1 I I I : I 16 1 15 1 i 1 4 i 1 3 1 I 1 2 I i 11 1 I I 10 1 I 1 I I 9 1 : I I 8 1 i I I I 7 1 i i I : : I I I s I I I I i i I I I I I I I I I I I i 1 e l e l o I E I F G H I I s i I I i I I I I i : s I I I I i i I ; I i I ; i I : I s s s s s I I s : s I I I I S : i I S I I I I j : I S I I i S I I I i I I S I I s i I I I l I I I I i I I i I I i I I 1 S i I I I I I I s : I i I I i J I L I M N o I P a I R : s 1 r ; u I v : x l z 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 I 22 : 21 20 19 18 17 16 1 5 14 13 12 11 I 10 I i 9 i 8 I I 7 I I 6 I s 4 I i 3 I I 2 1 1 AA I BB : cc 1 00 Figure 3-31 Spatial distribution of tree dens in the Caldeirao group home range.

PAGE 108

90 The results of the variance-to-mean relationship analyses showed in fact that the distribution of dens were clumped for the four study areas (1 07/1.41 for the River area 1 35/2 86 for the Morro area, 1.00/1.2 for the Little Bridge area and 1.58/2.08 for the Caldeira.a area). The importance of dens for lion tamarins are often neglected in the literature although it seems to have an important role in the ecology of this primates. This genus is the only among the Callitrichidae family that frequently use dens as sleeping sites Saguinus, Cebuela and Callithrix usually prefer tree branches or tangles of vines or twigs and only occasionally sleeps in tree dens ( Snowdon and Soini 1988 ; Stevenson and Rylands 1 988). A central issue in the ability of animals to tolerate environmental change lies in the species capacity to locate consume and process adequate food I examined the variation in proportion of time spent foraging and feeding across months in the four groups The ANOVA analysis failed to show an effect of group but not of months on the proportion of time spent in feeding and foraging (Table 3-3) Most important for answering the question were the differences in each group s diet: Plant food The diet of L. chrysopygus is composed of three primary types of food: fruits, animals, and plant exudates. Of the 2,146 feeding records collected for the four study groups, 1,686 (78 5 percent)

PAGE 109

91 Table 3-3. Analysis of variance for monthly differences in six behaviors Significance values are calculated from Type Ill partial sum of squares Source df Sum of Mean Squares Squares F p Social 11 11.47 1.04 3 3 0 003 Inactive 1 1 1309.10 11 9 1.83 0.085 Resting 1 1 6867.71 624.33 6.71 0.0001 Moving 11 611.51 55.59 3.8 0.001 Foraging 1 1 77.73 7.06 6.73 0.0001 Feeding 1 1 259.59 23.59 3 2 0.004

PAGE 110

92 were fruits. Animal matter make up 13.5 percent and plant exudates 7 8 percent. Black lion tamarins consumed at least 53 species of fruits belonging to 24 different families. Among families 65 percent were from the family Myrtaceae the most popular fruit source for black lion tamarins both i n number of feeding records as well as in number of species involved. The fruits from the Myrtaceae were small and fleshy and tamarins ate them when they were ripe. The second most frequently eaten fruit was the Palmae (13 percent) with only one species accounting for this total Syag rus romanzoffiana. The tamarins ate the sticky soft outer part of these fruits. They also ate the fleshy aggregation of berries of Annona, which have larger and harder fruits The tamarins also ate some young fruits of the harder legumes. Additionally L. chrysopygus relied on the fruits of a series of vines lianas and at least two epiphytes of the Cactaceae Because of the difficulty of identifying vine and lianas, I lumped all these species into one category called vines. These represent six percent of the total records of the fruits the tamarins ate. Epiphytes represent less than one percent of black lion tamarin fruit diet and were composed of Rhipsalis fruits (Cactaceae) and the inflorescence of the Araceae Philodendron sp (Araceae) I do not have observations of black lion tamarins eating nectar in Morro do Diabo although I recorded these animals feeding on the flower's nectar of M abea fistulifera in the Caetetus reserve in 1985

PAGE 111

93 used exudate as a general term to refer to saps, gums and resins (for definitions see Bearder and Martin, 1980). Black lion tamarins ate exudates from only eight species of trees belonging to seven families. The most important sources of exudates were two species of Rutaceae Helietta longifoliata and Esenbekcia sp. (around 20 percent of the exudate records). Vines and lianas represent a minor source of gum (less than one percent). Despite the difficulties in identification, I was able to identify at least two of the most frequently eaten vines, which were the thorny Acacia polyphylla (Leguminosae) and Xomelia sp. (Rubiaceae). The percentage of fruits eaten was negatively correlated with movements for the Caldeirao group (r 5 =-0.38 ns). It was however positively correlated in the other three groups (River r 5 =0.32 ns; Morro r 5 =0 1 ns and Little Bridge r 5 =0.2 ns). Animal food Almost all the existing literature on the feeding ecology of the tamarins describes them as relying heavily on insects (Coimbra Filho, 1985; Ramirez, 1985; Rylands, 1982; Peres, 1986; Sussman and Kinzey, 1984). The faunivorous part of the black lion tamarin's diet is composed of small vertebrates and insects. Among the latter, I observed tamarins eating both mobile and non-mobile insect prey. Seventy-two percent of the animals eaten by L. chrysopygus were non-mobile prey versus 27 percent, which were mobile. Among the non-mobile prey, I frequently observed them eating cryptic adult cockroaches and larvae of Lepidoptera and Coleoptera. Mobile prey included mainly orthopterans insects like grasshoppers, stick

PAGE 112

94 insects and katydids. Tamarins also consumed snails, caterpillars, spiders, and vertebrates such as tree frogs (Hyla sp.), small lizards, fledgling birds, and even some snakes. Black lion tamarins used two principal foraging strategies. The first was usually employed in capturing mobile prey and animals tended to visually search and then make rapid grabs. The second was a method used to capture non-mobile prey, and consisted of tactile probing of natural tree cavities, cracked tree bark and other similar foraging substrata. At other sites, important foraging micro habitats for Leontopithecus are the underside and the inside of bromeliad rosettes (Peres, 1986; Rylands, 1982; Coimbra-Filho and Mittermeier, 1973). However, at Morro do Diabo park, there were no bromeliads. Ad-libitum data from my study show that black lion tamarins occasionally consumed unusual items like abandoned bee hives or spider webs. These were observed opportunistically. The four study groups showed significant differences in their diet (animals, plants and exudates) (X2=139.9 df = 6 p<.0001 ) All groups ate more fruits than animals or exudates, but the relative proportions varied (Fig. 3-32). The Caldeirao group ate more fruits (84.5 percent) while the River group relied the least on fruits for its survival (71.5 percent). The Morro and the Caldeirao groups ate little or no exudates, but they consumed proportionally more animals and fruits than the other groups. The River and the Little Bridge groups ate less animals but consumed more exudates. Therefore, there was considerable variation in the consumption of animal prey in L. chrysopygus diets.

PAGE 113

100 90 80 p 70 e 60 r C 50 e n 40 t 3-0 a g 20 e 1 0 0 ----------------River Caldeirao Morro L Bridge Figure 3-32. Overall use of different food items for the four study groups of L. chrysopygus. co 01

PAGE 114

96 Local variation The four study groups at Morro do Diabo differed in the food items they ate. The species of tree that are included in their diets vary quite significantly. Table 3-2 presents the ten tree species on which each group spent most of their time feeding. Only one tree species was common in the diet of all the groups: the palm Syagrus romanzoffiana. Among feeding trees, the strongest difference was between the groups located in the north and south extremes of the park: the River and the Caldeirao groups The animals in these areas had only two feeding tree species in common: the palm Syagrus romanzoffiana, which was the second most important tree and the myrtaceous tree Myrceugenia ovata, which was the most abundant in the River area and the ninth most frequent in the Caldeirao area list. Overall diets of black lion tamarins from the River group were the most different when compared to the other three groups. The Morro and the Little Bridge groups showed the most similar diets and both also had dietary similarities to the Caldeirao group. The results of the Renkonen index of similarity reinforce these difference in food items between the groups showing an increasing gradient of differences in the percentage similarity for the four study group diets going from the River area to the Caldeirao area (Table 3-4). The extreme differences are again between the River and Caldeirao areas that presented only 26 percent of similarity in their diet. Considering that the Renkonen index goes from O or no similarity to 100 or complete similarity I conclude that there is a low similarity in the diet of the four groups

PAGE 115

97 Table 3-4 Percentage of diet similarity among the four study groups as measured by the Renkonen index. AREA Morro L. Bridge Caldeirao River 36.2 28.6 26.3 Morro 53.9 58.3 L. Bridge 51 6

PAGE 116

98 Seasonal variatio n The plant comp o nent of each group's diet changed almost completely through the year (Table 3-5). This change however is more related to the quality and distribution of feeding trees than the abundance of fruits. There was not a strong correlation between the availability of fruit and the percentage of time spent eating fruit (River rs=.002 ns Morro rs= 287 ns Little Bridge rs=.007 ns and Caldeirao, r 5 =.488 ns). The River and the Caldeirao groups foraged and fed on the same fruit species during only four months. On three of these occasions the groups shared two members of the Myrtaceae Myrciar i a ovata and Psidium sp. This occurred when the rainy season was beginning almost at the end of the year. In the dry season they shared one species, Syagrus romanzoffiana. The other two groups were in an intermediary position, sharing many diet species with the River and the Caldeirao groups throughout the year. The palm S. romanzoffiana, is an important species for the four groups of black lion tamarins in the dry season. They ate the sticky fleshy outside of the ripe fruits. At the onset of the rainy season, the Xylopia brasiliensis (Annonaceae) represented a major source of food for the Caldeirao group and to a lesser extent for the two central groups, the Morro and the Little Bridge. During the same period of the year, the River group relied mainly on fruits and gum of vines of the Rutaceae Helietta longifoliata (Rutaceae) Seasonal variation in tamarin diets was not restricted to monthly differences of plant species. The relative proportions of animals, plants and exudates in the tamarins' diet changed throughout the year (Fig. 3-33 to 3-36). The River group compensated for a decline

PAGE 117

Table 3-5 List of the two most common plant items for each group and the percentage of the total feeding records for each month SAMPLE RIVER M:ffO L. BRIDGE CALDEIRAO PERm FOOD ITEMS % FOOD ITEMS % FOOD ITEMS % FOOD ITEMS % Jan 1 Myrcia sp. 2 31 Eugenia sp. 2 48 Eugenia sp 2 87 Eugenia sp 2 75 Jan 2 Ce/tis spinosa 1 6 Eugenia sp 1 33 Eugenia uva/ha 6 Eugenia uvalha 25 Fev 1 Vines 29 Myrcia sp 2 42 Eugenia sp 2 62 Eugenia uvalha 55 Fev 2 Cedrela fissilis 24 Eugenia uvalha 38 Cereus sp. 24 Xylop i a brasiliensis 1 8 Mar 1 Esenbeckia sp 44 Xylopia brasil i ensis 22 Psidium sp 2 25 Xylop i a brasiliensis 34 Mar2 Helietta longifoliata 33 Cereus sp 1 6 Syagrus romanzoffiana 1 3 Eugenia sp 2 32 Apr 1 Vines 32 Xylop i a brasi/iensis 53 Syagrus romanzoffiana 25 Trichi/ia pa/Iida 47 Apr 2 Helietta longifol i ata 1 8 Cassia ferruginea 1 3 Psidium sp 2 1 3 Xylopia brasiliensis 1 7 May 1 Helietta longifoliata 100 Syagrus r o manzoffiana 64 Syagrus romanzoffiana 70 Syagrus romanzoffiana 38 May 2 Xy/opia brasi/iensis 8 Esenbeckia sp 1 6 Psid i um sp 2 29 Jun 1 Helietta longifoliata 50 Ficus sp 52 Xylopia brasiliensis 3 1 Myrcia sp 1 7 1 Jun 2 Vines 1 8 Syagrus romanzoffiana 2 1 Helietta /ongifoliata 1 7 Syagrus romanzoffiana 1 0 co co Jul 1 Helietta longifoliata 40 Myrcia sp 83 Myrcia sp 88 Myrcia sp 1 97 Jul 2 Syagrus romanzoffiana 27 Xylopia brasiliensis 8 V i nes 5 Nectandra sa/igna 2 Aug 1 Syagrus romanzoffiana 32 Eugenia uvalha 50 Syagrus romanzoffiana 28 Syagrus romanzoffiana 33 Aug 2 Vines 25 Eugenia sp 2 2 1 Helietta /ongifo/iata 22 Myrcia sp 1 25 Sep 1 Cabra/ea canjerana 55 Erisma sp 71 Eugenia sp 2 70 Myrcia sp 1 47 Sep 2 Erisma sp 32 Cabra/ea canjerana 1 8 Esenbeckia sp 20 Syagrus romanzoffiana 1 4 Oct 1 Myrceugenia ovata 91 Myrceugenia ovata 60 Myrceugenia ovata 50 Myrceugenia ovata 48 Oct 2 Psidium sp 2 4 Erisma sp 20 Psidium sp 2 50 Psidium sp 2 43 Nov 1 Myrceugen i a ovata 88 Psidium sp. 2 2 1 Myrceugenia ovata 100 Psidium sp .. 2 41 Nov 2 Chrysophyllum gonocarpum 2 Myrciaria sp .. 22 Myrceugenia o v ata 1 8 Dec 1 Campomanesia sp .. 52 Philodendron sp 28 Campoman e sia sp 4 1 Psidium sp 2 43 Dec 2 Psidium sp 1 1 5 Savia dyct i ocarpa 28 Anaden a nthera falcat a 1 8 Sebastiania serrata 1 6

PAGE 118

100 ? I ? m 90 80 70 C Q) co 60 ICl) C Q) 50 ... Q) __.. .0 0 E 4 0 ::::, 0 z 0 30 20 1 0 0 J F M A M J J A s 0 N D M on th s Figure 3-33. Monthly variation in diet of the River group

PAGE 119

100 l f 11 90 r ~nim~~ l 80 -----~ C 70 Q) <1l f60 en C Q) 50 ..... Q) ..Cl E 40 ::::, 0 z 0 30 20 1 0 0 -~---.-.-~ ....... 1 ---~ ~--1---~ ___ ... ,_, ___ J F M A M J J A s 0 N D Months Figure 3-34. Monthly variation in diet of the Morro group.

PAGE 120

100 -90 80 -------F ~i~ a ~ -~ ------70 C Q) .)/! cu 60 Ien C 2 50 l 20 1 0 0 J F M A M J J A s 0 N D Months Figure 3 35. Monthly variation in diet of the Little Bridge group.

PAGE 121

100 u -----90 ~~~ als l 80 C 70 Q) ca I60 en C Fruits Q) 50 ,.._ Q) .0 0 E 40 w ::, z 0 30 20 1 0 0 J F M A M J J A s 0 N D Months Figure 3-36. Monthly variation in diet of the Caldeirao group.

PAGE 122

104 in the availability of fruits in the rainy season by an increase in the consumption of animals and exudates. During the month of November, when fruits started to be scarce, stick-insect abundance increased noticeably. The diet of the River group then shifted to include animal prey. In January when the River group's diet reduced even more the percentage of fruits taken and showed a tremendous increase in the percent of eaten exudates. During the month of April fruit production peaked with a correspondent increase in the amount of fruits eaten and a corresponding decline in the amount of exudates in the tamarins' diets. From May until July, the River group ate almost only fruits and exudates of H. longifoliata. From July until November this group became almost completely frugivorous again The Little Bridge group also relied on animal prey and exudates to complement their diet in the rainy season. The time of these occurrences was however slightly different from that of the River group. From October until February they supplemented their fruit diet exclusively with animals. From March until July part of their animal supplement was replaced by exudates. In August exudates were replaced by animals and in September it worked the other way around, with no animals and only exudates in their diets Exudates were a minor component of the Caldeirao group's diet throughout the year. During the dry season this group relied heavily on fruits and during the wet season they consumed less fruits and more animals Exudates were only eaten, and then slightly, during the months of April and November. The Morro group ate more exudates than the Caldeirao group, but less than the other two groups. Animals were important

PAGE 123

105 secondary components of L chrysopygus diets except during the months of May July and November. Exudates were eaten from March until July. In September, when the dry season ends, and in December right in the middle of the rainy season there were two other peaks of exudates in the Morro group s diets. In order to compare plant and animal sources of food, I lumped exudates and fruits into one category. I could then compare plant items to animal items eaten by each group per season. Using this comparison he River and the Caldeira.a groups spent more time eating plants in the dry season than in the rainy season. On the other hand they spent twice as much time eating animals in the rainy season than in the dry season. The Morro and the Little Bridge groups also spent more time eating plants in the dry season than in the rainy season but they spent the same percentage of their time feeding on animals. Variation in foraging Foraging is another important component in the investigation of black lion tamarins' diets. To exploit foods in a tropical forest primates must have some method of tracking these resources through space and time (Milton 1980). In this section I will examine how social groups of black lion tamarins forage for food. Since there was an intra-specific variation in the quantity and quality of the tamarin's diet, the next question to be examined is whether foraging efficiency is better in some habitats than in others. To measure this efficiency I used the overall capture

PAGE 124

106 Table 3-6 Rank list of the percentage of prey capture success for the four study groups. GROJP CAPTURE SUCCESS PREY% RANK River 13. 7 2 Q Morro 18.6 1 Q L. Bridge 13.3 3 Q Caldeirao 11 8 4 Q

PAGE 125

107 success of each group. considered capture success as any capture that resulted in feeding. The percentages and ranked capture successes for each group are presented in Table 3-6 The River and the Morro groups had greater capture successes than the Caldeirao and the Little Bridge groups. There was a strong negative relationship between the percentage of foraging and movements for the Caldeirao group (r 5 =0.7 p>0.005). The other three groups had a weak correlation (Spearman rank correlation: r 5 =0.32 for the River group, r 5 =0.056 for the Morro group and rs=0.189 for the Little Bridge group) Discussion The need for comparative studies of primates in different habitats has been recently recognized due to the unabated destruction of tropical rain forest (Johns, 1991; Marsh et al., 1987). The collection of adequate and precise data related to habitat use in different habitats is thus gaining importance in the primatological field for the conservation and management of threatened species in the wild as well as in various forms of captivity (Box, 1991 ). The importance of this study is that it gathers a variety of biological information on L. chrysopygus which can also be used in management plans for the species. I documented the variation of four groups of black lion tamarins ecology, behavior based on their use of different habitats at the Morro do Diabo park. Overall the four study groups responded differently to each habitat. The greatest differences appeared in the use of space and feeding ecology among the four groups. Although their use of time showed no significant

PAGE 126

108 categorical differences, the many strategies they used to respond to habitat differences were distinct among each of the study groups, and of black lion tamarins in relation to other New World primate species. Use of Time The complex interaction of the different habitat components makes it difficult to explain how the time budgets of black lion tamarins would not be influenced by these habitat differences. The results of this study show similar overall use of time by black lion tamarins despite the fact that they differ significantly on a monthly basis. Black lion tamarins spend very little time feeding and foraging, and in contrast to other lion tamarins, L. chrysopygus are inactive most of their time. Their use of time also differs from other members of the Callitrichidae family (Terborgh 1983). This population of black lion tamarin is the only one among the four lion tamarins that live in a semi-deciduous forest of the interior. This environmental condition may explain the high level of inactivity of these primates when compared with other Callitrichidae Leontopithecus has the morphological adaptations of a predator. They have the largest body size among the callitrichids, their teeth are relatively large and the canines are better developed (Hershkovitz 1977). The time budget of omnivorous primates is usually influenced by the food supply (e.g., Terborgh, 1983; Robinson, 1986). The black lion tamarin's use of time shows that they depend heavily on fruits.

PAGE 127

109 None of the study groups used their range randomly They seemed to forage for prey as well as for exudates opportunistically while moving between fruit trees where they gathered the bulk of their diet. Leontopithecus chrysopygus show a high variation i n their seasonal use of fruits and an intense monthly use of a small number of available fruits (Waser 1977 ; Kinnaird, 1990; Robinson 1986). These fruits might provide the tamarins with their necessary energetic requirements without them having to spend a long time in locomotion, foraging and feeding. These results are quite different from the golden lion tamarins or for the golden headed lion tamarins, which spend most of their time moving and foraging (Peres, 1986 ; Rylands, 1982). Although Rylands (1982) was unable to record the amount of time spent by golden-headed-lion tamarin foraging and eating prey, Peres (1986) stated that searching for prey in golden lion tamarins was time consuming and crowded the time budget for all other activities. This study's four groups of black lion tamarins never spent more then three percent of their time foraging for prey. Movements and Use of Space Range size of the study groups of black lion tamarins varied among groups. No matter whether I considered the number of quadrats used by each group or all the quadrats included within the group's range the size of the areas used among the four groups varied considerably The Little Bridge group occupied the largest site, which was 80 percent larger than the smallest site, that of the River group. As a species their mean range size is among the

PAGE 128

11 0 largest of most primates (see Sussman and Kinzey, 1984; Glutton Brock and Harvey, 1977) and is by far the largest among the other Callitrichidae Conversely they have one of the lowest densities among the members of their family (for review see Goldizen 1987). Their daily average range seems to be correlated with the size of their group The Little Bridge group ranged 80 percent larger than the River group. A large range with a small density calls for an increase i n daily range length. Despite the cost-benefit limitations of an increased daily range length against the large size of their range I would expect black lion tamarins to present a larger range length than that described for other Callitrichidae. The results of this study however show that the average of black lion tamarin daily range length is not very large and i s comparatively equivalent in size to those described in the literature for the genus the Saguinus and Callimico (Dawson, 1976; Terborgh, 1983; Pook and Pook, 1981). It seems that black lion tamarins have enough fam i liarity with their large home range to relocate their use of space within their range depending on fruit availability (Zack and Falls, 1976; Robinson, 1986). Additionally I found very little territorial defense. This could be explained by a relaxation in territorial defense as found by Dawson (1976) for some groups of Panamanian tamarins. However, the inter-group encounters observed showed all the characteristic behaviors of a very territorial animal and I never observed any group inside another group's range. The only cases of clear territorial invasion I observed occurred not by groups but generally by duos of males when dispersing from or into a group. Leontopithecus chrysopygus have

PAGE 129

1 1 1 sternal and circungenital glands that they use for scent marking. Results from studies in captivity showed that scent marking are used by lion tamarins for territorial marking (Mack and Kleiman 1978). The intensity and diversity of habitat use varies considerably among the four study groups. The same is true for the Callitrichidae but at the inter-specific level. A number of species seems to prefer secondary forest and edge habitat such as Cebuella pygmmaea, and many species of the genus Saguinus (Castro and Soini 1977 ; Dawson 1976 ; Mittermeier and Coimbra-Filho 1977) The genus Leontopithecus on t he contrary was described as either using exclusively mature primary forest or at least having patterns of habitat selection i nversely related to habitat disturbance (Rylands 1982 Peres 1986). Neither distribution of dens nor range use by the four study groups were uniform. Different groups used different parts of their range with different intensities. The River group more intensely used the periphery of its range with clear open areas in its central part, the Caldeirao group intensely used a core area in the central part of its range. Black lion tamarins have multiple constraints on activity and isolation fields such as the effects of two or more limiting resources (sleeping sites and food) as suggested by Altmann (1974) for different groups of baboons. Their infrequent use of the same den tree and their intense use of the areas surrounding dens suggests that black lion tamarins follow the strategy of an animal that is familiar with its area of use (Waser and Wiley, 1979). Depending on the characteristics of their habitats

PAGE 130

11 2 the four study groups used different combinations of a strategy to economize the effects of more then one limiting resource. Going from den to den they moved more or less straight ahead during any one day over a defined route. The "knowledge' of the resources available in each area might be an explanation for the observed differences on intensity and diversity of habitat use by the four groups (Robinson, 1986). This singular strategy would be similar to the one described for the blue monkey (Cercopithecus mitis) by Rudran (1978) in that black lion tamarins have a diverse diet that includes non-mobile insects Waser and Wiley ( 1979) suggest that a primate that takes non-mobile prey which have a low abundance and a low renewable rate, should avoid repeating routes. Garber (1988) describing nectar feeding by Saguinus mystax and Saguinus fuscicollis, showed that groups of these species were able to decide which tree to use based on minimizing distances traveled between trees. The author suggest that his study groups probably retained a cognitive map of the spatial distribution of certain tree species within their home range. A number of studies (Coimbra-Filho, 1977; Rylands, 1982 and Peres, 1986), suggest that lion tamarins would have very little flexibility in their use of habitat as they are adapted to primary forests where plant food sources are widely scattered. The same authors suggested that tamarins would require a mature forest habitat because of their use of holes (dens) as sleeping sites, which seem to be absent in secondary growth forest. An additional or alternative reason suggested by Rylands (1982) for the tamarins' use of high quality forest would be their animal-prey foraging. Lion

PAGE 131

113 tamarins would not be able to exploit non-mobile prey fauna in second growth forest because they are absent in these forests (Winnet-Murray, 1986 ; Peres, 1986). Rylands (1982) added that primary forest was the preferred habitat for lion tamarins because of their use of the epiphytic bromeliad as an important foraging site Such plants are not present in secondary forest growth The results of my study do not confirm these predictions and show flexibility in forest use by black lion tamarins. The habitats used by the four study groups at Morro do Diabo have characteristics analogous to both mature and second growth forest. The area inhabited by the River group could be compared to a seasonally flooded lowland mature forest. The Caldeirao area on the other hand showed more similarities to a secondary forest due to its botanical homogeneity, low stature and density in the understory cover. Black lion tamarins also appear flexible in relation to differences in resource availability. They use alternatives for foraging for non-mobile prey. As the Morro do Diabo habitats have no epiphytic bromeliads these are replaced differently i n larger or smaller quantities by other sources of prey depending on the habitat. As an example groups rely more on fruits in the River area but the Morro group presents the highest success in foraging for prey. The use of tree dens is also considered an ecological constraint for tamarins living in secondary forests, but this does not seem to be a limiting factor in black lion tamarin habitat use at Morro do Diabo. In all four study sites at Morro do Diabo, I was able to observe the tamarins using an average of 17 tree dens per area.

PAGE 132

114 Finally, in order to examine the flexibility of L. chrysopygus habitat use it is necessary to examine their use of plant and other diet components. The diets of marmosets and tamarins are still a puzzle. Wh i le Sussman and Kinzey ( 1984) in their ecological review of the Callitrichidae stated that the diet of tamarins is predominantly insects plus fruits the studies of lion tamarins conducted by Rylands (1982) and by Peres (1986) showed exactly the opposite. The bulk of L. chrysomelas and L. rosalia's diets are of plant matter followed by anima l prey. Black lion tamarin feed on three primary types of food, as do most Callitrichids, but with a significant variation through space (in different habitats) and through time (monthly). In reference to the variation through space, habitat differences primarily affected the percent of plant versus animal items in the composition of the tamarin's diet. Richard ( 1978) found similar results for Propithecus verreauxi in Madagascar and suggested that this may have been due to differing traditions between groups, or by local differences in the availability and distribution of the various food species I have demonstrated that there were quantitative and qualitative differences in the availability of fruit species in the habitat of the four study groups of black lion tamarins but only the qualitative differences affected the tamarins differences in diet. These differences however are not mutually exclusive with Richard's ( 1978) different-tradition hypothesis. Black lion tamarins are territorial animals and my data

PAGE 133

11 5 suggest that they form new groups by the dispersal of animals among groups (Valladares-Padua 1992). The dispersal, sometimes as much as five kilometers away from their original areas requires some flexibility in using the newly adopted habitat (Valladares Padua 1992). The seasonal changes were different in the study areas as well as were the consequences of these changes on the tamarins. The four groups depended on fruit most of the year but the River group and the Little Bridge groups tended to compensate fruit decrease with an increase of animals in their diet. The other two groups the Morro and the Caldeirao are more similar to each other in all seasons of the year. The above results show that black lion tamarin's diet seems to be a balance between fruits, animal prey and exudates. Fruit is by far the most important dietary component throughout the year as well as on a monthly basis and animal prey and exudates augment the diet in different proportions depending on the time and habitat. Undoubtedly animal prey is an essential diet component of black lion tamarins as it is to other Callitrichids (Peres, 1986). But L chrysopygus seem to have mechanisms to compensate for the reduced amount of animal protein by increasing their use of plant matter. Exudates that constitute an important proportion of the diet of severa l tamarins like Saguinus oedipus or S. fuscicollis, are relatively important to some groups of. black lion tamarins but almost irrelevant for others (Garber 1984 ; Terborgh, 1983). While at times black lion tamarins seem to be opportunistic exudate

PAGE 134

11 6 feeders, they can sometimes rely on exudates for up to 50 percent of their diet. This dependency may last for as much as six months like the River and the Little Bridge groups. When fruits are scarce exudates are a common source of food for tamarins and marmosets (Ferrari 1989; Rylands, 1982). L. chrysomelas and many other Callitrichids feed on the gum of Parkia pendula. For black lion tamarins the important sources of exudate are from only tree species such as the Combretaceae Terminalia sp (Combretaceae) and the Helietha longifoliata (Rutaceae). Exudate feeding confirms the variability in feeding behavior among the four study groups at the Morro do Diabo park. The generalization I obtained from this intra-specific study on habitat use by black lion tamarins is that different groups of this species have the ability to live in a wide range of environmental conditions. This study also provided information on behavioral and ecological differences between Leontopithecus chrysopygus and other species of primates.

PAGE 135

DISTRIBUTION ABUNDANCE AND MINIMUM VIABLE POPULATION OF THE BLACK LION T AMARI NS Introduction One of the greatest chal l enges in dealing with the conservation of an endangered species is estimating how many individuals still ex i st and how many are necessary if the species i s to survive over the long term. The estimate of the number of individuals is especially difficult when habitats are fragmented, and the species is isolated in different sub-populations. Only by studying all these sub-populations is it possible to estimate a species metapopulation structure. Small populations, even when not declining, might already be in jeopardy because of the genetic and demographic problems intrinsic to the dynamics of its small size (Ballou 1990). Inbreeding creates a threshold for finite populations below which the population begins to decline in performance and fertility. According to Soule (1980) an effective population (Ne, where Ne are roughly the individuals that reproduce) of 50 individuals is this threshold population size or minimum viable population. In cases of extreme reductions of sub-populations within a metapopulation. these sub-populations might go below the threshold line and require a special integrated management plan to assure their surviva l. Gilpin (1987) added the important role played by space, especially 117

PAGE 136

118 the patchiness of habitat, in characterizing a minimum viable population of a species. That is why when estimating the number of individuals needed for long term survival of a species, one has to take a number of factors into consideration: the species' population size; the fragmentation of its habitat; the protection of the existing habitat and the management of the population as a whole. Leontopithecus chrysopygus can serve as an example of a species with a small total metapopulation. Initial estimates of no more than one or two hundred animals were provided by Coimbra Filho (1970b) and Coimbra-Filho and Mittermeier (1977). An international committee for the conservation of the species recommended immediate conservation actions for its survival (Seal et al., 1990). The first and most important of these was a more accurate estimate of the population number of the species. Following this suggestion, in 1990 and 1991 I conducted a survey of the species in forest fragments within the area of its original distribution. In this chapter, I present the results of this survey and discuss possible conservation measures based on these estimates of black lion tamarin abundance and distribution. To determine the metapopulation size, I used fewer sources of data. The Morro do Diabo State Park Sub-Populations To assess L. chrysopygus metapopulation sizes I began the intensive study described in the previous chapters. As Brockelman and Ali (1987) suggested, no one should attempt to sample primate densities without understanding species characteristics, such as their habitat, group structure, activity patterns, and vocalizations.

PAGE 137

119 With my long term study at the Morro do Diabo State Park I gathered information on their average home range and group size which then allowed me to calculate a range density in different habitats. I used the density estimate from the study groups at the Morro do Diabo State Park and extrapolated to the remaining park forested area as well as to some of the areas described below in which I rediscovered the species. Other Forest Fragments in the Pontal Region To find new populations of black lion tamarins I conducted surveys in nine forest f r agments located in the western part of the distribution of the species which is also known as the Pontal region (Fig. 4-1 ). These surveys were designed to obtain information on the presence or absence of the species and on the amount of habitat available in each forest fragment. However, the estimate of the total number of black lion tamarins for these areas resulted from extrapolating the density results obtained for the species at the Morro do Diabo park to each of these areas. The F azenda Rio Claro The numbers of black lion tamarins at Fazenda Rio Claro were based on direct density estimates and not on extrapolations. This area has special habitat characteristics with a production forest of Pinus sp. interspersed with natural forest. expected that this would produce different densities than in the Morro do Diabo. Exhaustive surveys gave me the confidence to use the actual number of counted individuals.

PAGE 138

120 Figure 4-1. Satellite image of the Morro do Diabo State Park and other surveyed forest fragments in the Pontal region.

PAGE 139

121 The Caetetus Ecological Reserve I did not have an opportunity to carefully examine the situation at the Caetetus Ecological Reserve. The data sources for this area were either taken from a black lion tamarin's genetic study I conducted in tflis reserve in 1986 or from information compiled from the work of other researchers, supplemented with information gathered from employees of the Reserve (Valladares-Padua, 1987; Keuroglian, 1990; Passos, 1992). Estimating Metapopulation Size To estimate the metapopulation size for L. chrysopygus, I assumed as a rule, that all the forest isolates where I found black lion tamarins were at carrying capacity. My estimates were based on the following observations: (a) I observed distinct groups in almost all forested parts of the Morro do Diabo State Park, no matter how different the forests seemed; (b) I observed many encounters of all my study groups with neighboring groups at Morro do Diabo; (c) I located at least one group in each of the inhabited forest isolates of the Pontal do Paranapanema region; ( d) I was able to physically count a large sample of the L. chrysopygus at the Fazenda Rio Claro; (e) I located two groups of black lion tamarins in the Caetetus Biological Reserve when developing the genetic study and the

PAGE 140

122 Reserve's wardens' information is that black lion tamarins density is high in the area (Valladares-Padua, 1987). The Morro do Diabo State Park I obtained group and individual densities per km 2 using the method described in NRC (1981) for long-term monitoring of specific groups. According to this method, when territories of neighboring groups do not overlap, which is the case with black lion tamarins, home range size and group size are the information needed for estimating the species density. Group density is obtained by dividing the average home range size by the number of monitored groups. Individual densities are obtained dividing average home range size by the mean group size of monitored groups. The time over which data need to be collected to confirm home range size depends on the habitat and the size of the species home range (animals with smaller home range will need shorter time) (NRC, 1981 ). To determine how long my observations would be, I calculated the number of new quadrats (0.25 ha) entered each month and plotted the cumulative monthly use of quadrats by each study group of lion tamarins. Ideally, data should be collected until one is confident that the whole range is known or the cumulative curve of range use against study time reaches a plateau (Rudran, 1978; Peres, 1986; Robinson, 1986). In my study, the curve started to reach an asymptote for three of the four study groups after one year of data collection and a total number of records of 6436 for the River area,

PAGE 141

123 6 00 f/1 --0R i ver G r oup 11' 5 0 0 '"" "C Calde1rao Group 11' ::I 400 0 Group 3 0 300 ---6---L. Brid ge Group '"" Q,) E 200 ::I z 1 0 0 0 C ai >::::, a. u > '...) l1l co a. ro ::::: :.:J (IJ 0 (IJ ..., ,, <{ --: ..., (J) :> z 0 <{ Months Figure 4-2. Cumulative use of quadrats per month by each of the four study groups of L. chrysopygus during one year of data collection.

PAGE 142

124 6543 for the Morro area, 7042 for the L. Bridge area, and 6440 for the Caldeirao area (Fig. 4-2). However, the Little Bridge group was still showing a slight increase in the number of new quadrats entered at this time. On the basis of the other groups' results, I assumed the curve for this group was about to level off and stopped the data collection at this point. To obtain the average home range size of black lion tamarins in the Morro do Diabo park, I used data from my long term study of the ecology and behavior of the species. To calculate the home range size for each group, I encircled the grid of quadrats occupied by the groups and counted them (see chapter 3, Fig 3-14 to Fig. 3-17) I included all the quadrats inside the home range even if they were never used by the group I also calculated the mean group size from data of the four Morro do Diabo State Park study groups. I estimated the black lion tamarin population for the remaining forest area of the park based on individual density data derived from these target groups and multiplying it by the total forest habitat available at Morro do Diabo. Other Forest Fragments in the Pontal Region In the forest fragments of the Pontal region my objective was not to obtain the population density but to record the presence or absence of black lion tamarin in the fragments. To conduct surveys in these areas I used the following methods: followed old hunting trails between 7:00 and 12:00 h and between 14:00 h and 17:30 h. did not make any observation between 12:00 h and 14:00 h, when primate activity is very low (Rylands, 1982)

PAGE 143

125 -45 11 10 2371 8 9 2 6 Pontal Region Areas with recently found sub populations Protected Areas: of black lion tamarins : 3 Fazenda Tucano 1 Morro do Diabo State Park 4 Fazenda Rosanella 2 Caetetus 5 Fazenda Ponte Branca Ecological Station 6 Fazenda Rio Claro 7 Fazenda Ribeirao Bonito N s Areas surveyed with no black lion tamarins 8 Fazenda Santa Rita 9 Fazenda Mosquito 10 Fazenda Vista Bonita 11 Fazenda Santa Zelia 12 Fazenda Nova Pontal Figure 4-3. The map of the State of Sao Paulo with the main locations where .L.. chrysopygus were surveyed or censused.

PAGE 144

126 I mapped and marked survey trails every 50 m. Transects were walked daily for at least 15 days in each area. This involved walking slowly (1.0 to 1 5 km/h) as suggested by Emmons (1984) stopping at least every 50 m (NRC, 1981 ). When a group was sighted it was followed, captured radio collared and the sex and weight (as an approximation of age) recorded as described by Valladares-Padua (1987) The groups were located and observed for the rest of the day at least once a week for a minimum of six months. I mapped each of these group locations and made ad libitum observations on the ecology and behavior I surveyed nine areas in the Pontal region (Fig 4-3 ) The tamarins were found in only four of these areas: The Fazenda Ponte Branca which had 800 ha of forest; Fazenda Tucano and Fazenda Rosanella, which together had 800 ha of continuous forest ; and finally, the 107 ha Fazenda Ribeirao Bonito was the smallest area surveyed that had tamarins. Unfortunately this small patch of forest was logged while the research was in progress. One group of tamarins with three individuals had to be rescued and transferred to the Sao Paulo Zoo did not find tamarins in the other five areas I surveyed. The largest of them Fazenda Mosquito, has almost 2 000 ha of forest in good condition and belongs to the King Ranch of Texas, USA. Both the Fazenda Vista Bonita and the Fazenda Santa Zelia have 250 ha; the first is well protected by its owner and the forest seems quite undisturbed, the latter had signs of human action, like fire and hunting. Another farm, the Fazenda Santa Rita (159 ha), has similar

PAGE 145

127 traces of human interventions. The last farm surveyed was the Fazenda Nova Pontal (169 ha) and it had forest in good condition. Fazenda Rio Claro The survey method used at the Fazenda Rio Claro was different from the one used on the other study sites Since suitable habitat consisted mainly of gallery forests, I used a variant of the sweep sample method in this area (Whitesides et al., 1988). A total of 270 trails covering 71 km were used for this study A group of at least three well-trained field workers walked simultaneously in a straight line, one on the trail and the others 100 m apart in a parallel line through the gallery forest. Using this method it was easy to see the animals either as they escaped from the observers or when they got to the boundaries of their territory and returned to its center The animals were counted twice at intervals of approximately one month between each sweep. The addition of all observed animals resulted in the minimum number of groups dwelling in the Fazenda's natural forest. The Caetetus Ecological Reserve Although there have been several studies on black lion tamarins at the Caetetus reserve (Valladares-Padua, 1987; Keuroglian 1990; Passos 1992) they do not provide an estimate of the species population size there. These studies nevertheless, clearly show the recent presence of L. chrysopygus in the site.

PAGE 146

128 Results The estimated density of black lion tamarins in the Morro do Diabo park was 0.73 groups per km2 or 3.72 individuals per km2. This was calculated from the average home range size of 1.38 km 2 The estimated mean group size of black lion tamarins in the Morro do Diabo Park was 4.75 individuals (S.D. 1.78) (Table 4-1 ). Based on the mean home range and group size of the four groups surveyed at Morro do Diabo, I estimated that the park has about 821 individuals. The other surveyed areas inhabited by the tamarins (excluding the Fazenda Rio Claro), comprise a total area of 4 000 ha. If I use the same density estimate as those of Morro do Diabo for these areas they would have a total of 139 individuals. The number of .L.. chrysopygus censused (counted) in Fazenda Rio Claro was 44 individuals. When the density figures for the black lion tamarins were applied to the combined areas of Morro do Diabo, Caetetus Ponte Branca and Rosanella the total number of individuals was 959. Adding the 44 individuals reported for Fazenda Rio Claro (Valladares-Padua and Cullen Jr. unpublished report) yields a metapopulation size of 1 004 individuals distributed across six forest fragments (Table 4-2). Discussion The survival of an endangered species depends on the survival of a population at a level that maintains its vigor and its potential for evolutionary adaptation. Such a population is termed a minimum viable population (Soule, 1987) However, if I treat different

PAGE 147

129 Table 4-1. Mean group size, number, age and sex of L. chrysopygus in the four study areas of the Morro do Diabo State Park Group Total Adult Adult Juvenile Juvenile Per Area Male Female Male Female River 3 1 1 0 1 Morro 7 4 1 1 1 Little Bridge 6 2 2 1 1 Caldeirao 3 2 1 0 0 Total 1 9 9 5 2 2 Average Group Size = 19/4 = 4.75 Individuals

PAGE 148

130 populations of a species (sub-populations) as a single managed population I can certainly improve its long term survival by maintaining at the same time at least the same vigor and evolutionary characteristics. Such a population is a minimum viable metapopulation. There is no single magic number that constitutes a minimum viable metapopulation (MVM) for all species or for any one species at all times Rather a MVM is a dynamic process that depends on the known sub-populations, on their ecological characteristics, and on the objectives of existing or planned management programs for the taxon of concern. An accurate census for any species is the first step in the establishment of a MVM goal. The estimated number resulting from this study, of more then 1,000 black lion tamarins in the wild, substantially revises past population estimates of the species which ranged from 100 to 300 individuals (Coimbra-Filho, 1990; Coimbra-Filho and Mittermeier, 1977). The latter were numbers that resulted from broad surveys which attempted to cover large geographic areas in short periods of time. This study relies on long term field study on L. chrysopygus to calculate an estimate of metapopulation size for the black lion tamarins. Short-term estimates should be taken with caution in management plans (Rumiz, 1989). The 1,000 individuals estimated for the total number of black lion tamarins in the wild casts a great deal more optimism on the conservation status of the species. However, these sub-population sizes of L. chrysopygus might still not represent a viable population for the long-term. Numbers might be comparatively large but the

PAGE 149

131 Table 4-2. Estimates of available habitat for each area where black lion tamarins are found, average home range size, group densities and total population sizes. Area Forest Size (Km2) Morro do Diabo State Park Caitetus Biological Reserve Ponte Branca Farm Rosanella Farm Tucano Farm Sub-Total Rio Claro Total Population Size Average Group Size = 4 75 Average Home Range 137 75 ha or 1.375 km2 Group density=O. 73groups/Km2 Individual density=3.45 individuals/Km2 Pop. Size (Average individ. densityxarea size) 238 821 20 69 8 28 8 28 4 1 4 278 960 8 44 1004

PAGE 150

132 most important question is the effective population size (Ne) for the species. Black lion tamarins are monogamous animals with only two indiv i duals reproducing in each group (Kleiman 1981 ; Baker 1991 ). The 0 73 groups per km2 for an inhabited area of 278 km 2 excluding the Fazenda Ri9 Claro gives a total number of 200 groups in the wild If I use the number of breeders to roughly estimate Ne will be equivalent to the number of black lion tamarins individuals that reproduce, or 200 groups x 2 reproductive individuals per group the actual Ne would be 400 individuals As suggested by Franklin ( 1980) a population mutation and genetic drift will only be at equilibrium by when it has an effective population size of at least 500. If this number was accepted as its goal for the black lion tamarin management the metapopulation size must be increased by nearly 100 reproductive individuals to avoid deleterious genetic effects in the metapopulation (Seal, 1984; Seal and Foose 1984) This of course assumes that one is able to successfully manage all sub-populations as a sole metapopulation. As I mentioned in the beginning of this discussion, there is no single magic number that will assure the viability of a population. Populations with number of founders smaller than 500 can indeed survive. There is a series of examples suggesting the existence of populations that started with a small number of founders and survived. This seems to be the case of the Drosophila sp. flies of Hawaii or the North American golden hamster population (Carson 1971; Carson and Yoon 1982; Ralls and Ballou, 1982). Although the.re are several threats to small populations, such as the probability of catastrophe demographic instability, among others, another major problem seems to be

PAGE 151

133 inbreeding depression (Ralls and Ballou, 1983). Domestic animal breeders demonstrated by trial and error that to avoid the rapid fixation of deleterious genes in a population it is necessary to have a rate of inbreeding per generation lower then one percent (Franklin, 1980). This one percent rule was called by Soule (1980) as the survival threshold of a population. Based on the formula that the rate of loss of heterozygosity is equal to 1 /2 Ne, a population of 50 would be the threshold point of loss of heterozygosity per generation ( 1 /2 x 50 = one percent). Based on this rule only the Morro do Diabo sub-population barely has a chance to survive. The other 5 sub-population are far below the minimum Ne of 50 individuals. Therefore, for black lion tamarin, the question is not: what is a minimum viable population, but rather what is the minimum viable metapopulation? A more dynamic and modern approach for calculating a minimum viable metapopulation size is to use a number of computer techniques that can simulate the events that occur during the lives of organisms. There are modeling tools that allow one to correlate the survey results of sub-population sizes and their degree of insularization of the different subpopulations with their probability of extinction Moreover it permits the design and testing of different minimum viable metapopulation management scenarios as will demonstrate for the black lion tamarins in the next chapter.

PAGE 152

MET APOPUL.A TION EXTINCTION MODEL Int reduction The black lion tamarin, whose population once occupied more than half of the State of Sao Paulo, is now reduced to six sub populations isolated from any and all other conspecifics. Such reduction in size and isolation is usually extremely detrimental to the long-term survival of a species Species can go rapidly extinct from random fluctuations in birth and death rates as well as marked changes in sex ratios. Genetic factors can also be important: genetic variation can be lost through genetic drift and catastrophes can wipe out small populations (Soule, 1980; Allendorf and Leary 1986; Ballou, 1990) They are vulnerable to extrinsic factors such as disease epidemics or deforestation (Lacy, 1990). There are, however, many possible solutions to ameliorate this situation in situ. In this chapter I will emphasized the following three: (a) integration of the six sub-populations through management; (b) augmentation of habitat carrying capacity; and (c) reduction of the natural degradation of the forest habitat fragments by controlling fire and edge deterioration. The major goal of this chapter then is to test if a combination of these three solutions in a management scenario will increase the minimum metapopulation population size that will ensure the long-term persistence of black 134

PAGE 153

135 lion tamarins. To accomplish this goal I used simulation modeling as a tool for exploring the viability of the species under different management scenarios. The model I used was the Vortex model written by Lacy (Chicago Zoological Society) and I set the goals of the simulation to be the same established by conservation biologists for similar cases. The metapopulation must be of sufficient size to have lower than 2 percent probability of going extinct and be able to retain 90 percent of its heterozygosity over 100 years (Seal et al, 1990). To test the conservation implications of managing black lion tamarins as a metapopulation, I created seven different scenarios: one for each known sub-population of black lion tamarins independently and one for the metapopulation (all the sub populations combined). Based on my recent work with the conservation of the species, I assumed that the level of protection for the species habitat would be satisfactory to the point that there will be no logging or hunting in the metapopulation habitats. I also assumed an increase in the Morro do Diabo carrying capacity by adding 30 percent of forest habitat to it. This is the area of the park that lost its forest cover in the past but which is now naturally regenerating. I also added 2,419 ha of uninhabited forest to the metapopulation carrying capacity. This represents forest isolates that are left in their original range but that are not now inhabited by tamarins and that have landowners who have agreed to work with conservation efforts. Finally, I assumed that the species has enough behavioral flexibility to be able to tolerate translocations and managed dispersals as a regular management procedure.

PAGE 154

136 Vortex Overview According to its author Lacy ( 1990): "the Vortex computer simulation model is a Monte Carlo simulation of the effects of deterministic forces as well as demographic, environmental and genetic stochastic events on wildlife populations. Vortex models population dynamics as discrete sequential events that occur according to defined probabilities. The probabilities of events are modeled as constants or as random variables that follow specified distributions. Vortex simulates a population by stepping through the series of events that describe the typical life cycle of a sexually reproducing, diploid organism. The program was designed to model long-lived species with low fecundity, such as mammals birds and reptiles. Vortex iterated life events on an annual cycle although a user could adapt the year concept to more or less than twelve months duration. The simulation of the population is itself iterated to reveal the distribution or the fates the population might experience. Methods To model the probability of extinction of the known black lion tamarins sub-populations, I ran simulations for each known sub population Each of these sub-populations was considered an isolated sub-population and was modeled for a period of 100 years; each simulation was repeated 1000 times. In the same fashion I ran a simulation for the metapopulation (all sub-populations combined). Both the probability of extinction and the expected heterozygosity

PAGE 155

137 were tracked over time. When a population size dropped to zero it was counted as an extinction event during that year. The probability of extinction was calculated to be the average extinction rate over all simulations. Expected heterozygosity was calculated by monitoring allele frequencies over time in each sub-population as well as in the metapopulation. I considered the metapopulation carrying capacity to be not only the integrated sub-populations but I added three forest isolates in the same region to their habitat (Fazenda Mosquito Fazenda Nova Pontal and Fazenda Vista Bonita) These fazendas do not presently contain black lion tamarins, but are inside the boundaries of the original distribution of the species. In addition they are well protected by their land owners and their forest cover is similar to the River study site of the Morro do Diabo park. Data Sources The models were based mainly on data I collected on the ecology and conservation biology of the black lion tamarins over the last seven years However, even after so many years of study of one species I do not have enough information on many aspects of its life history. Since the results of probabilistic models like Vortex are highly correlated with the quality of the input data sources wherever necessary and possible my data was complemented by data on the congeneric golden lion tamarins (Tables 5-1 and 5-2).

PAGE 156

138 Table 5-1. Different population parameters used in the Vortex model: (a) Survival percentage of golden lion tamarins (L. rosalia) (Baker and Dietz, 1990); (b) Age distribution in months of the golden lion tamarin (L. rosalia) (Baker and Dietz, 1990): and (c) Reproduction rates in percentage for the four study groups of black lion tamarins (L. chrysopygus) at the Morro do Diabo State Park (Ballou and Valladares Padua, 1990). Survival % Survival (years) 01 1 2 2-3 Adults Age Distribution: Age in Months <9 9-1 8 18-30 >30 Reproduction % Reproduction 25 25 50 Males 13. 14 10. 12 33.21 19 11 Males(%) 16 18 18 48 100% Females 13. 14 10.12 33.21 19. 11 Females(%) 1 9 21 1 9 tl 100% Young/Year 0 1 2

PAGE 157

139 Table 5-2. Black lion tamarin carrying capacity based on the available known habitat in the original home range of the species. Metapop~ation Carrying Capacity: Caetetus Ecological Station Morro do Diabo State Park Fazenda Rosanella Fazenda Tucano Fazenda Ponte Branca Fazenda Rio Claro F azenda Mosquito Fazenda Nova Pontal Fazenda Vista Bonita Metapopulation Individuals 69 1,067 28 1 4 28 44 69 6 9 1 334

PAGE 158

140 All the currently existing sub-populations except that in the Morro do Diabo State Park are assumed to be at carrying capacity In the case of Morro do Diabo, 30 percent of the State Park's area has native forest in its initial re-growth stage. I expect that in the future this will become forest habitat for the black lion tamarins. Consequently, I increased the park's carrying capacity by 30 percent. As previously mentioned, I also added three new forest fragments to the total carrying capacity (Fazenda Mosquito, Fazenda Nova Pontal and Fazenda Vista Bonita). Environmental Variation: Effects of environmental variation on mortality and reproduction in areas containing black lion tamarins are not available. Their effect on reproduction and mortality of the golden lion tamarins has been observed to be low in the Pogo das Antas Reserve in Rio de Janeiro, Brazil (Jim Dietz, person. comm ). As environmental variation plays a critical role, and in the absence of a good estimate of its effects in black lion tamarins reproduction and mortality, I decided to assume that these values would be the same as the previously calculated for the golden lion tamarin population at the Pogo das Antas biological reserve (Baker and Dietz, 1990).

PAGE 159

141 Catastrophes It is not possible to run more than two catastrophes with the vortex model. Hunting and logging would be catastrophes most detrimental to the population. However, their risk of occurrence is small due to a regional environmental education program which resulted in a community involvement in the long-term conservation program for the species (Padua et al., 1990). I thus decided to use fire and diseases as the two possible and independent sources of catastrophes for the simulations. Fire was assigned a 5 percent chance of occurring each year and its effect would be a 50 percent reduction in survival. Disease was assigned a one percent chance of occurring each year, and its effect would be a 50 percent reduction in the survival. In both cases the catastrophe will affect the modeled populations if it is either a sub-population or the metapopulation. Inbreeding Depression Although the reduction in survival and reproduction is considered a common phenomenon in small populations of mammals, I did not incorporate inbreeding depression into the model because it is very difficult to get a good estimate of the effects of inbreeding depression on wild populations (Ralls et al., 1983). An alternative approach to use in the model is the average effect of inbreeding on the mortality and on the reduction on litter size in captive colonies of the same species. For black lion tamarins however, the captive populations are not large enough to estimate levels of genetic load.

PAGE 160

142 Secondly, the results of previous simulations demonstrated that a reduction in the inbreeding depression slightly reduces the extinction probability of the species (Ballou and Valladares-Padua 1990). Finally the genetic survey I conducted with 25 blood enzymes in the Morro do Diabo and Caetetus sub-populations of black lion tamarins showed no heterozygosity and no polymorphism but no apparent decrease in their vigor (see chapter 3). Based on these results I did not incorporate inbreeding depression in the model. Results The model uses simple life-table calculations to estimate the population growth rates and generation lengths that will be used in the simulations. Extinction rates in all modeled sub-populations failed to meet the minimum tolerance criteria of 2 percent probability of extinction in 100 years (Fig. 5-1 ). However Morro do Diabo's sub-population which is by far the largest surviving in the wild, showed results close to the pre-established demographic conservation goal. It presented a 6. 7 percent probability of extinction in 100 years. The model's results concerning the genetics of the sub populations are very similar to the demographic results. The difference lies in the fact that the Morro do Diabo sub-population

PAGE 161

143 I I 0.9 Caete t us 0 9 Morro do D i abo gJ, 0 .8 a, 0.8 co 0 7 g' 0 7 0 6 c 0 .6 0 5 0.5 cf 0.4 Q) 0.4 0 .3 a.. 0.3 0 .2 0 2 0.1 0.1 0 0 C 0 0 0 ;;::; 3c C 0 C C C 0 C'I r'" -c C ri 3c 0 tr. r0 C Y ear Ye a r I 1 0.9 Rosanella 0 9 Tucano a, 0 8 a, 0 8 en 0 7 g' 0 .7 co c 0 .6 c 0 6 a, 0 5 0 5 a, 0.4 ai 0.4 a.. 0 3 a.. 0 3 0 2 0 2 0 1 0 1 0 0 0 C ;:;; C 8 xi C g C 0 C 0 0 ri ;:;; ;;::; -.:; rx =' c Ye ar Year I 1 0 .9 Pon t e Branca 0.9 R i o C l a r o gJ, 0 .8 a, 0 8 co 0 .7 g' 0 .7 0 .6 c 0 .6 0 5 0. 5 cf 0.4 o5 0 4 0 .3 a.. 0.3 0.2 0 2 0 1 0.1 0 0 0 0 C C 0 0 C 0 C C C 0 0 0 0 C 8 01 er tr. SC, rX 0 0 ri r'" ,r sc r00 0 Y e ar Yea r 0.1 Metapopulation 0.08 gJ, 0.06 co c a, Q) 0.04 a.. 0.02 0 0 0 0 0 0 0 0 C 0 0 0 C"I ""' tr. SC, r00 Cl 0 Year Figure 5-1 Ext i nction probability over t i me for the S I X sub populat i ons and the me t apopulation. Notice the c h ange of sca l e i n t h e metapopulation f i gure

PAGE 162

144 meet the retention of 90 percent of its original heterozygosity criteria (Fig. 5-2). Discussion The traditional recommendations for the conservation of endangered species of primates include (a) protecting habitat for particularly endangered species ; ( b) creating and maintaining parks and reserves ; ( c) creating awareness to encourage coexistence between human and non human primates and, (d) establishing conservation-oriented captive breeding programs for endangered species (Mittermeier and Cheney, 1987). I will not argue here for or against these recommendations They are all valid conservation policies However I want to argue that in many cases they are not enough to assure the long-term survival of the species. They do not appropriately take into consideration what Gilpin ( 1987) called the ex t ension of the spatial stage on which the extinction drama is played. Spatial distribution is especially dramatic in metapopulations where discrete patches of the area are habitable and the intervening regions are not. This pattern of habitat patchiness precludes the normal individual dispersal fragments the population into small sub-populations and increase their chance of extinction (Falconer, 1978; Ralls and Ballou, 1983). As, in the case of primates this distribution pattern is usually positively correlated to anthropocentric habitat destruction it is considered to be a possible sign of endangerment of the species that lives i n that habitat. Ex-situ preservation ( captive breeding) is increasingly

PAGE 163

l 0.9 0.8 ~ 0 .7 en 0 .6 0 Ol 0.5 >, N 0.4 e 0.3 Q) w 0 2 I 0 1 Caetetus 0 0 0 C C 0 5 g ri "1" ,r, --c r-Y e ar l 0.9 0.8 ~ 0 .7 en 0.6 0 05 N 0.4 e Q) 0.3 w 0.2 Rosanella I 0. 1 0 0 r, -;;: 0 ;:::: x 5 V"\ Ye ar l 0 .9 0.8 ~ 0 .7 en 06 0 Ol 0 5 >, 0.4 N e 0 3 Q) w 0.2 Ponte Branca I 0.1 0 C s C 0 0 ::c 0 0 0 (" l -;:;<. "1" ,r, r-00 =-Y ear 0.98 Z:0 96 ;;; 0 Ol >, N e 0.94 Q) ai I 0 92 Metapopulation 0 9 C 0 0 0 0 N <'" -.:i145 l 0.9 0.8 ~ 0.7 en 0 .6 0 Ol 0 5 >, N 0.4 e 0.3 Q) w 0.2 I 0 1 0 l 0 9 0 8 ~ 0 7 en 0 .6 0 Ol 0 5 >, N 0 4 0 ai 0.3 w 0 2 I 0 1 0 l 0 .9 0 8 0 .7 ;;; 0 .6 0 Ol 0 5 >, N 0.4 e 0 3 Q) w 0 .2 I 0.1 0 0 ,r, Year Morro do Diabo 0 ;::, Tucano C 0 r, Rio Claro 0 0 0 N 0 ..c; -;:;:. -;;:, -;:;<. 0 r-0 ;;:. "1" Ye ar "'1" ir. Yea r 0 0 -.:iV"\ Y ear C 00 C ;:::: -:; C ;::: -:; C --c r-3c 5 3c 5 g 0 0 0 Figure 5 2. The l oss of heterozygosity over time for the six sub po p ulations and the metapopulation. Notice the change of scale i n the metapopulation figure.

PAGE 164

146 recommended for such situations and is accepted as having an important role to play in the conservation of endangered species Coimbra-Filho, 1976; Seal, 1990). There are however many problems in adopting captive breeding as a conservation management strategy. Endangered species in captivity are more susceptible to the stochastic processes previously mentioned in this chapter (Ralls and Ballou, 1983). Furthermore, zoos have limitations in their space (Conway, 1986) and even if a species is reproducing successfully, they will be separated in different zoos, increasing the patchiness and reducing the size of each population. In other words, keeping small groups in different zoos increases the risk of genetic depauperation and domestication of the captive populations. This can be partly avoided through interactive management programs like the Species Survival Plan (SSP) of the American Association of Zoological Parks and Aquariums (AAZPA) or the equivalent program of the European zoos. There is also an increasing number of zoos cooperating with the in-situ conservation efforts for many species One of the most successful of these cooperative conservation programs is that of the Golden Lion Tamarin Conservation Program. The species' population growth has been so successful that managers are attempting to stabilize its population at 500 animals. This is done by using hormonal devices to preclude the reproduction of some over represented lineage and by a reintroduction program in animals the wild (Ballou, 1990). This is probably the first reintroduction program for the conservation of a primate ever conducted (Kleiman et al., 1991 ) However, so far the most important conservation result of the golden lion tamarin reintroduction program has nothing

PAGE 165

147 to do with reintroduction itself. It was the involvement of many regional land owners in the conservation effort by setting aside fo r est fragments in their properties for the release of the primates (Kleiman et al., 1991 ). This is not to say that the program is perfect; the mortality rate is too high and the program is expensive. The managers of the golden lion tamarin reintroduction program calculated that to reintroduce enough animals to reach the same goals that I established in this chapter for the black lion tamarins would cost more than one million dollars (Ballou, person. comm.) I want to use the golden lion tamarin example to advocate the position that captive breeding should only be used as a conservation strategy after exhausting all the conservation efforts in the wild. At the moment, even if the species is successfully bred in captivity there is no assurance that the reintroduction of the surpluses will be possible, as there might not have any habitat left for the species when the surpluses are available. Even if captive breeding is successful and there is some habitat left for the reintroduction, there is still the risk of introducing diseases or introducing undesired new behaviors into the wild populations (Woodford and Rock, 1991 ). The metapopulation management approach offers many advantages for the conservation of the black lion tamarins. It is possible to minimize the problems and costs of captive breeding and reintroduction of the species in the wild. The conditions to adopt the metapopulation approach for the black lion tamarins are almost ideal. The captive population is still very small and far from the size that would encourage a reintroduction. There are large sub

PAGE 166

148 populations like the Morro do Diabo s and, at least in the Pontal region it would be possible to establish corridors for movement of organisms among the different sub-populations (Harris 1984). The latter, if adopted would avoid one of the most difficult decisions i n managing a metapopulation: the rate of dispersal necessary for the metapopulation long-term survival. With corridors the animals could naturally disperse among the different sub-populations Simberloff et al. ( 1992) cautioned against a wholesale acceptance of the metapopulation paradigm (Levins, 1970). In their view metapopulation is still an untested hypothesis and not broadly representative of nature as suggested by Noss ( 1992) Metapopulation management is of course possible even without co r ridors. As pointed out by Merrian (1991 ), what it requires is movement and not corridors. Movements can, in the absence of co rr idors be achieved by translocations and managed dispersals. This is the case of the black lion tamarins where corridors are no longer completely feasible. The distance among some different sub po p ulations may be of 400 km A metapopulation program based on translocations and managed dispersals seems to be ecologically and bio-politically feasible, as well as being cost effective. Translocations have already been done successfully among many primates. Golden lion tamarins have been successfully translocated in a rescue operation (Pinder, 1986). Another example is the common marmoset that had its geographical range tremendously extended by the release of confiscated animals by the wildlife au t horities or by unhappy pet owners (Coimbra-Filho 1990) The effectiveness of the metapopulation management for the

PAGE 167

149 conservation of the black lion tamarins must be tested in the field. However, I have demonstrated that the black lion tamarin is a highly adaptable, generalist species provided that it has a forest as its habitat. In Morro do Diabo at least, they inhabit four different types of forest. They are capable of major dietary shifts in response to habitat change and therefore may be less sensitive to environmental variation (see chapter 4). What is suggested in such cases is the combination of the available data with mathematical models to estimate the probability of the species going extinct based on different scenarios (Seal et al., 1990). I did that for black lion tamarins and the results of the simulations I processed show that there is a high probability that the available sub-populations and the remaining habitat is sufficient to create and maintain, through translocations and managed dispersal, a metapopulation that will assure the species' long-term survival. Conclusions The results of my data modeled by the Vortex model suggest that if treated individually, black lion tamarin sub-populations, with the exception of the Morro do Diabo's have more than 50% of chance of going extinct in the next 100 years. Even the Morro do Diabo sub population, with its comparatively large size, does not meet the survival criteria established for the species. These results call for immediate action to manage the sub populations in order to obtain a minimum viable size. Their ability to survive in different habitats, their recently discovered sub populations, and their ecological flexibility leads to the conclusion

PAGE 168

150 that black lion tamarins will probably respond well to metapopulation management procedures. The fact that the possibility for such management procedures still exists only gives cause for hope. This is of course a major shift from the total extinction prediction of the seventies (Coimbra-Filho, 1970a) or from the estimated 100 individuals on the verge of extinction in the wild of the eighties (Coimbra-Filho and Mittermeier, 1977). While the modeling results are pretty straight forward, my conclusions are based on an comprehensive interpretation of the results. By that mean that the model is simple, the data are the best available though sometimes scarce, and the involved biological and ecological processes are so complex that there is no way I can rely completely on the extinction probabilities of black lion tamarins. But despite the comprehensive interpretation of the results and the many assumptions in modeling the sub and metapopulation, they certainly show that a managed total metapopulation is more secure than the isolated sub-populations.

PAGE 169

CONCLUSIONS AND RECOMMENDATIONS This dissertation compares the use of time an space and feeding ecology in relation to habitat differences among four groups of black lion tamarins in the Morro do Diabo State Park and demonstrates the variation of the tamarins' behaviors in response to habitat differences The four study groups of Leontopithecus chrysopygus showed no significant difference of their overall time budget although they adjusted their use of space to the different habitats. This flexibility enabled them to live in different habitat conditions and to be able to respond to habitat variations through major dietary shifts. These results are in accordance to studies conducted with laboratory animals in regard to animals use of space (K l opfer and Hailman, 1965). The flexibility of black lion tamarin behavior makes them well suited for almost any type, shape and size of forest within their original range. This flexibility becomes even clearer with the results of the general survey I conducted for the species I obtained a higher number for the total population size of L. chrysopygus than the previous estimates for the species. I also found new sub populations of the tamarins living in a variety of forest fragments. As a result of this survey, I estimate the total number of black lion tamarins in the wild to be slightly higher than 1,000 individuals 151

PAGE 170

152 living in six structurally and dimensionally different islands of forest. In the Morro do Diabo park the species inhabits forests ranging from relatively tall, undisturbed forests such as the River area, to a homogeneous "cerrado" type of forest as found in the Caldeira.a area. In the Caldeira.a area the ten most common species made up 88.1 percent of its total number of individual trees and the Myrtaceae family is so heavily represented that it comprises 61 .1 percent of the individual trees. Outside the park I found the species living in small forest fragments composed of narrow strips of gallery forest interspersed with Eucalyptus plantations as in the Fazenda Rio Claro, or in small patches of degraded secondary forest such as in the Fazenda Ribeirao Bonito. The total population of black lion tamarins living in this network of fragmented sub-populations constitutes a good example of what Hanski (1989) called a finite metapopulation. Hanski's definition contrasted with the original definition of metapopulation presented by Levins (1969). While Levins assumed the existence of an infinite number of habitat patches of the same size and quality, Hanski assumed a more realistic a finite number of sub-populations. Small sub-populations of a finite metapopulation of an exclusively arboreal primate such as the black lion tamarin are very susceptible to extinction for two main reasons: (1) they inhabit discontinuous habitats that may result in the total impossibility of natural migration among sub-populations (Gilpin, 1987); and (2) when the number of sub-populations is small it is possible that the

PAGE 171

153 metapopulation go extinct simply because all sub-populations happen to go extinct at the same time (Nisbet and Gurney 1982). The information I obtained from my study would not be complete without a long-term survival prediction for the black lion tamarins. For this purpose, I used the Vortex model to develop a viability analysis for the species (Lacy, 1990). The results of this analysis showed that if the sub-populations were considered individually, L. chrysopygus would have a high chance of going extinct in the next 100 years. If all the sub-populations were managed as a metapopulation, however, they have a high probability of long-term survival. Past Management Measures Based on the best estimate of the black lion tamarin population size (300 individuals), there seemed to be few conservation actions that could enhance the species survival in the wild (reviewed by Coimbra-Filho in 1976). At the time Coimbra Filho, conducted his study there were major threats to L. chrysopygus existing known habitat. In the beginning of the 1970's, landowners used agent orange to deforest extensive areas in the tamarin's original range (Valladares-Padua, 1987). The major conservation initiatives suggested by Coimbra-Filho and colleagues at the time were to increase the level of protection of the reserves, which is not an easy task in a country like Brazil, due to the limited resources available for this purpose, and to start a captive breeding program for future reintroduction of the species. His

PAGE 172

154 recommendation for a captive breeding program was carried out, but regional habitat destruction never completely stopped. A captive breeding program has been established for the species with the aim of contributing to the black lion tamarins conservation. From seven founders captured in Morro do Diabo in 1973 and another 16 wild individuals added in 1976 from the same reserve, the captive population by 1992, grew to around 70 animals distributed in three captive colonies (Valladares-Padua, 1991 ). In twenty years the captive population increased by only 47 individuals. These results seem to indicate that the future of the black lion tamarin can not rely exclusively on captive breeding programs. The Future of the Black Lion Tamarin There are a series of intra-specific comparative studies on the ecology and behavior of primates which show the adaptive importance of their behavior flexibility (Dawson 1976; Richard, 1978 ; Rudran, 1978; Kinnaird, 1990). The results of this study confirms that black lion tamarins showed flexibility in adapting to the significantly different habitats they occupied. This increases their chance of occupying uninhabited habitat and if appropriately managed "in situ" increases their long term probability of survival. Based on the results of the past, it seems clear that the chances of survival of black lion tamarin is dim if reliance is placed only on captive bred reintroduction programs. If captive breeding programs are to contribute to the conservation of L. ch rysopygus. it is important that they incorporate evolutionary aspects of the behavior

PAGE 173

155 and the ecology of the species. Survival under these different contexts demands different adaptations. To breed a species in captivity necessarily imposes some sort of artificial selection on the animals. This artificial selection may occur either consciously or unconsciously on the part of the manager where personal biases and preferences may influence the selection of the breeding stock (Muntzing, 1959). But no matter the reasons for the occurrence of artificial selection, it creates changes in the behavior flexibility of a species (Redshaw and Mallinson 1991 ). One of the most important behavior changes is a reduction in responsiveness to environmental changes. In any domestication process for example food provision and human control over the breeding process have reduced competition for important resources and thus permitted artificial selection (Price, 1984). It seems of critical importance that captive breeding programs for endangered species consider what captivity as a selective force may do to the adaptive individual variation in response to social and physical change (Clark, 1991 ). Another important point in relation of breeding endangered species in captivity for future reintroduction is that reintroduction by itself cannot efficiently compensate for the common declining causes of primate population reduction in the wild. Primate habitats are being destroyed at such accelerated rates that reintroduction measures might not be possible in the future. In the case of L. chrysopygus it is still possible to consider management strategies other than captive breeding and reintroduction. The existence of forest fragments in the original range of the species with or without black

PAGE 174

156 lion tamarins, allows the development of long-term management strategies for the black lion tamarins. Results of this study demonstrate that the wild sub populations of L. chrysopygus I studied are able to maintain the necessary behavioral flexibility to effectively respond to certain environmental variations. In order to combat the loss of genetic variability, behavior flexibility and evolutionary plasticity of the existing sub-populations I recommend three major actions to be taken First, all sub-populations should be managed as a metapopulation which may include the shifting (reintroduction, translocation and managed dispersal) of individuals among its many sub-populations (metapopulation management). I demonstrated by a Vortex simulation of the black lion tamarin metapopulation that if all sub-populations were incorporated as a metapopulation, there would be a high probability of the long term survival for the black lion tamarins (chapter 5). Second, new reserves need to be established and the established ones enlarged. Forest fragments outside the existing protected areas should preferably be chosen for their similarity to the areas inhabited by L. chrysopygus. One can compare the suitability of the potential habitat fragment by statistically comparing it with the multivariate sample space obtained for the four Morro do Diabo areas (Fig 2-5) as I have done. The areas that have average variables not statistically different to those of the four ellipses produced from the Morro do Diabo groups, may have an enhanced chance of becoming appropriate additional habitats for

PAGE 175

157 black lion tamarins (Shugart, 1980) Under these circumstances, there may be at least 2,419 ha of uninhabited forest that could be used for repopulate with the species (Fazenda Mosquito, Fazenda V i sta Bonita, Fazenda Santa Zelia and Fazenda Nova Pontal). The existing available habitat could also be increased through the use of degraded areas in the Morro do Diabo park Only 70 percent of Morro do Diabo is covered by forest, and from the remaining 30 percent 20 percent could be restored. Another way to increase and improve habitat conditions would be to create, whenever possible, forest corridors between neighboring but non-contiguous sub-populations Despite recent controversy addressed by Simberloff et al. ( 1992) about the costs and benefits of forest corridors in the case of the L. chrysopygus metapopulation management corridors can play a major role in reducing human intervention in the managed dispersal among the connected sub-populations Finally, an environmental education and community involvement program is a must for successful conservation of L. chrysopygus. A program established for the Morro do Diabo region showed that the community became aware of the importance of the park as a conservation site and contributed to its protection (Padua, 1991 ). In the black lion tamarin conservation scenario, each and every remaining habitat fragment is an important piece to its metapopulation survival puzzle. The appreciation and involvement of local populations and land owners can enhance the protection of these remaining forest habitats. My study shows that even a highly endangered species such as the black lion tamarin, has survival chances in the wild if the

PAGE 176

158 appropriate management measures are taken. The ten-year study i nvolving genetics demography, ecology behavior, captive breeding and environmental education has led to the conclusion that the future of the species lies not through the captive breeding and surplus reintroduction but rather through the metapopulation management of its wild sub-populations.

PAGE 177

APPENDIX A LIST OF TREE SPECIES FOUND IN THE MORRO DO DIABO PARK Taxon Anacardiaceae Astronium graveolens Tapirira guianensis Annonaceae Duguetia lanceolata Rollinea sp. Xylopia brasiliensis Apocynaceae Aspidosperma cylindrocarpum Aspidosperma polyneuron Aspidosperma sp. Peschiera fuchsiefolia Araliaceae Didymopanax morototoni Bignoniaceae Tabebuia avellanedae Tabebuia sp. Zeyheria tuberculosa Bombacaceae Chorisia speciosa Boraginaceae Cordia ecalyculata Cactaceae Epiphylum sp. Cereus sp. 159 Common name Guarita Peito-de-pomba Pindafba Araticum Ata/Pinha Perobapoca Peroba Pereiro Leiteiro Mandioqueiro lpe Roxo lpe Amarelo lpe Tabaco Paineira Cafe de Bugre Cacto Palma Mandacaru

PAGE 178

Caricaceae Jaracatia spinosa Com bretaceae Terminalia sp. Euphorbiaceae Croton floribundus Croton sonderianus Joannesia princeps Savia dyctiocarpa Sebastiania serrata Flacourtiaceae Casearia sp. Guttiferae Rheedia gardneriana Lauraceae Enidlicheria paniculata Nectandra saligna Ocotea sp Lecythidaceae Cariniana estrellensis 160 Leguminosae (Caesalpinioideae) Cassia ferruginea Copaifera langsdorffii Holocalyx balansae Peltophorum dubium Leguminosae (Faboideae) Inga sp. Lonchocarpus guilleminianus Lonchocarpus leucanthus Macherium brasiliensis Macherium sp. Poecilanthe parviflora Pterogyne nitens Jaracatia Amarelinho do Brejo Capixingui Marmeleiro Boleiro Guaraiuva Desconhecida Espeteiro Lima6zinho Canela.a Canelinha Canela Jequitiba Canaff stula Oleo-de-copaiba Alecrim Guarucaia Inga Feijao-cru Embi ra-de-sapo Sapuvao Sapuva Coragao-de-negro Amendoim

PAGE 179

1 61 Leguminosae (Mimosoideae) Acacia polyphylla Anadenanthera falcata Pithecellobium edwallii Meliaceae Cabralea canjerana Cedrela fissilis Trichilia pallida Moraceae Cecropia cf. cinerea Ficus enormis Myrtaceae Campomanesia guagumaefolia Campomanesia sp. Eugenia spp. Eugenia spp. Eugenia uvalha Myrceugenia ovata Myrcia rostrata Myrcia sp. Myrcia sp Myrciaria sp. Myrciaria sp. Psidium sp. Psidium sp. Psidium sp. Nyctag in aceae Bouganvilea arborea Palmae Syagrus romanzoffiana Phytolaccaceae Gallesia gorazema Polygonaceae Ruprechtia laxiflora Triplaris surinamensis Monjoleiro Angico Farinha seca Canjerana Cedro Marinheiro Embauba Figueira Capota Guabiroba Ara9a-branco Ara9a-verde Ara9a Pitanga Carvaozinho Araca-jambo Uvaia Cambuf Jaboticaba Piuna Goiabinha Ara9a-vermelho Flor-roxa Coqueiro (Tupi ou Jeriva) Pau-d'alho Arco-de-penei ra Pau-de-formiga

PAGE 180

Proteaceae Roupala brasiliensis Rhamnaceae Colubrina rufla Rubiaceae Coutarea lexandra Rutaceae Balfourodendron riedelianum Citrus sp Esenbekcia sp. Helietta longifol i ata Metrodorea atropurpura Zanthoxylum sp. Sapindaceae Celtis spinosa Diatenopteryx sorbifolia Matayba sp. Sapotaceae Chrysophyllum gonocarpum Chrysophyllum sp Tiliaceae Luehea divaricata Vochysiaceae Vochysia tucanorum 162 Carne-de-vaca Sobrasil Pau-de-Anta Paumarfim Laranjeira Pau-Preto / Penei ra Amarelinho Canela-de-Cotia Mamica-de-porca Grao-de-Galo Correeira Camboata Guatambu Carrapatei ro AQoita-cavalo Caramuru

PAGE 181

APPENDIXB DATA SHEET FOR PROCESSING BLACK LION TAMARINS Record Number Animal number Species Tattoo No Date Capture Location Radio Collar Bead Collar Code Weight Feces collected Pelage color Face color Identifying marks or scars Total length Head width Knee to Heel Dental State: Up canine length Up canine breadth Discoloration Wear Cavities Personnel at site: At processing: 0 0 0 Sex Ectoparasites collected Tail length Hand length Ear height Up perm canines Broken teeth + + + ++ ++ ++ Observations: 163 Head length Foot length Ear width

PAGE 182

LIST OF REFERENCES Allendorf, F. W. 1986. Genetic drift and the loss of alleles versus heterozygosity Zoo Biol.5: 181-190. Arita, H. T., and S. R. Humphrey. 1988. Revision taxon6mica de las murcielagos magueyeros del genera Leptonycteris (Chiroptera: Phyllostomidae). Acta Zoologica Mexicana 29:1-60 Assump9ao C T., H. F Leitao Filho, and 0. Cesar. 1982. Descri9ao das matas da Fazenda Berreiro Rico SP. Revista Brasileira de Botanica 5 ( 12):53-66 August P. V. 1983 The role of habitat complexity and heterogeneity in structuring tropical communities. Ecology 64 (6):14951507. Baitello J. B., J. A. Pastore, 0. T. Aguiar, F C Serio and C E Silva. 1988. A vegeta9ao do Parque Estadual do Morro do Diabo municfpio de Teodoro Sampaia, Estado de Sao Paulo. Acta Botanica Brasileira 1 (2):221-30. Baker, A. 1991. Evolution of the Social System of the Golden Lion Tamarin (Leontopithecus rosalia): Mating System Group Dynamics, and Cooperative Breeding. Ph.D. diss. University of Maryland. Baker, A., and J. Dietz. 1990. Golden Lion Tamarin: Current Status in the Wild. In Leontopithecus: Population Viability Analysis. ed. U. S. Seal, J. D. Ballou, and C. Valladares-Padua. Bloomington, MN: Captive Breeding Specialist Group (IUCN/SSC/CBSG). Ballou, J. D. 1990. Small population overview In Population Viability Workshop. ed. U. S. Seal, J. D Ballou, and C Valladares-Padua. Bloomington, MN: Captive Breeding Specialist Group (IUCN/SSC/CBSG). 164

PAGE 183

165 Ballou J D. and C. Valladares-Padua. 1990. Population extinction model of lion tamarins in current protected areas. In Population Viability Workshop., ed. U. S. Seal, J. D. Ballou and C. Valladares-Padua. Bloomington MN: Captive Breeding Specialist Group (IUCN/SSC/CBSG) Barcikowsky R. S. ed. 1983. Computer Packages and Research Design. 1st ed. Lanham, MD: University Press of America Bearder S. K. and R. D. Martin 1980. Acacia gum and its use by bush babies Galago senegalensis (Primates, Lorisidae). International Journal of Primatology 1 : 103-28. Beardmore J. A. 1983. Extinction survival and genetic variation. I n Genetics and Conservation ., ed. C. M Schonewald-Cox S M. Chambers B. MacBryde and L. Thomas. Menlo Park California: The Benjamin/Cummings Publishing Company. Bernardes, A. T A. B. Machado, and A. B. Rylands. 1 990 Brazilian Fauna Threatened with Extinction. Belo Horizonte, MG: Funda~ao Biodiversitas. Box H. 0. 1991 Responsiveness to environmental change: interrelationships among parameters. In Primate Responses to Environmental Change., ed. H 0. Box. London: Chapman and Hall. Brockelman W. and R. Ali. 1987. Methods for surveying and sampling forest primate populations. In Primate Conservation i n the Tropical Rain Forest ., Vol. 9, ed. C. Marsh and R. Mittermeier. New York, NY: Alan R. Liss. Campos J. and D. Heinsdijk. 1970. A Floresta do Morro do Diabo Silvicultura em Sao Paulo 7: 45-58. Capen, D. E. 1981. The Use of Multivariate Statistics in Studies of Wildlife Habitat. Fort Collins, CO: Forest ServiceUSDA. Carey A. B. 1981. Multivariate analysis of niche, habitat and ecotope. In The Use of Multivariate Statistics in Studies of Wildlife Habitat., ed. D. E. Capen Fort Collins, CO: Forest ServiceUSDA

PAGE 184

166 Carson H. L. 1971. Speciation and the founder principle. Stadler Genetics Symposia 3:51-70. Carson, H. L., and 1982 Yoon. 1982. Genetics and evolution of Hawaiian Drosophila. In The Genetics and Biology of Drosophila., ed. M. Ashbumer, H. L. Carson, and J. N. Thompson Jr. Vol. 3b New York NY: Academic Press. Carvalho C. T ., and C. F. Carvalho. 1989. A organiza9ao social dos sauins-pretos, Leontopithecus chrysopygus (Mikan 1823) na reserva em Teodoro Sampaia, Sao Paulo (Primates, Callithricidae). Revista Brasileira de Zoologia 6 (4) : 707-17 Carvalho C. T ., A. L. Albernaz and C A. de Lucca 1989 Aspectos da bionomia do mico-leao-preto (Leontopithecus chrysopygus, Mikan). (Mammalia Callitrichidae). Revista do lnstituto Florestal 1 ( 1 ): 67-83. Castro, R. and P. Soini. 1977. Field studies on Saguinus mystax and other Callitrichids in Amazonian Peru. In The Biology and Conservation of the Callitrichidae ed, D. G. Kleiman Washington DC: Smithsonian Institution Press. Cesar, 0., and H F. Leitao Filho. 1990. Estudo florfstico quantitativo de mata mes6fila semi-decfdua na Fazenda Barreiro Rico Municfpio de Anhembi, SP. Revista Brasileira de Biologia 50: 133-47 Chivers D. J 1986. Current issues and new approaches in primate ecology and conservation. In Primate Ecology and Conservation., ed. J. G. Else and P. C. L Lee Cambridge : Cambridge University Press. 1991. Species differences in tolerance to environmental change. In Primate Responses to Environmental Change., ed. H. 0. Box. Cambridge: Chapman and Hall. Chivers, D. J., B. A. Wood, and A. Bilsborough, eds. 1984. Food Acquisition and Processing in Primates. 1st ed. New York NY : Plenum Press.

PAGE 185

167 Clark A. B. 1991 Individual variation in the responsiviness to environmenta l change. In Primates Response to Environmental Change., ed. H. 0. Box. New York NY: Chapman and Hall. Glutton Brock T H. 1975 Feeding behaviour of red colobus and black and white colobus in East Africa. Folia Primatologica 23: 165207 1977. Some aspects of intraspecific variation in feeding and ranging behaviour in primates. In Primate Ecology: Studies of feeding and ranging behaviour in lemurs monkeys and apes. ed. T. H Glutton-Brock London: Academic Press. Glutton Brock, T. H. and P. H. Harvey. 1977. Species differences in feeding and ranging behaviour in primates. In Primate Ecology : Studies of feeding and ranging behaviour in lemurs monkeys and apes. ed. T. H. Glutton-Brock London: Academ i c Press Coimbra-Filho A. F. 1970a. Considera~6es gerais e situa~ao atual dos micos-le6es escuros Leontideus chrysomelas (Kuhl, 1820) e Leontideus chrysopygus (Mikan, 1823). Revista Brasileira de Biologia 30: 249-68. 1970b. Acerca da redescoberta de Leontideus chrysopygus (Mikan 1823) e apontamentos sobre sua ecologia. Revista Brasileira de Biologia 30: 609-15. 1971. Mamfferos amea~ados de extin~ao no Brasil. In Especies da Fauna Brasileira Amea9ada de Extin9ao ed. Anonimous Rio de Janeiro RJ: Academia Brasileira de Ciencias. 1976. Leontopithecus rosalia chrysopygus (Mikan 1823) o mico leao do Estado de Sao Paulo. Silvicultura em Sao Paulo 1 0 : 1-36. 1977. Natural shelters of Leontopithecus rosalia (Linnaeus, 1766) and some ecological implications (Callitrichidae Primates). In The Biology and Conservation of the Callitrichidae., ed. D. G. Kleiman. Washington, DC: Smithsonian Institution Press. 1985 Saui-preto ou mico-leao-preto. FBCN lnformativo 9(3) : 3.

PAGE 186

168 1990. Sistematica, distribuiQao geografica e situaQao atual dos sImIos brasileiros (Platyrrhini-Primates). Rev. Bras. Biol 50(4): 1063-1079. Coimbra-Filho, A. F ., and R. A. Mittermeier. 1972. Taxonomy of the genus Leontopithecus (Lesson, 1840). In Saving the Lion Marmoset., ed. D. D. Bridgewater. Wheeling, WV: Wild Animal Propagation Trust. 1973. Distribution and ecology of the genus Leontopithecus in Brazil. Primates 14(1): 47-66. 1977 Conservation of the Brasilian lion tamarins Leontopithecus rosalia. In Primate Conservation., ed. P. Rainier and G. H. Bourne. New York, NY: Academic Press. Conway, W. G. 1 986. The pratical difficulties and financial implications of endangered species breeding programmes. International Zoo Yearbook 24/25:210-19. Crook, J. H. and J. S. Gartlan. 1966. Evolution of primate societies Nature 210:1200-1203. Dawson, G. A. 1976. Behavioral Ecology of the Panamanian Tamarin Saguinus oedipus (Callitrichidae: Primates). Ph.D. diss. Michigan State University. de Vries, P. G. 1986 Sampling Theory for Forest Inventory. Berlin: Springer-Verlag. Deshler, W. 0. 1975. RecomendaQ6es para o manejo de Morro do Diabo. Pub/. lnstituto Florestal de Sao Paulo 6: 1-32. Dunbar R. I. M. 1976. Some aspects of research design and their implications in the observational study of behaviour. Behaviour 58: 77-98. Eisenberg, J. F., N A Muckenhirn, and R. Rudran. 1972. The relation between ecology and social structure. Science 176:863-874. Emmons, L. E. 1984. Geographic variation in densities and diversities of non-flying mammals in amazonia. Biotropica 16: 210-22.

PAGE 187

169 Fagen, R. 1982. Evolutionary Issues in the Development of Behavioral Flexibility In Perspectives in Ethology., ed. P P. Bateson and P. H. Klopfer. Vol. 5. New York, NY: Plenum Press. Falconer, D. S. 1978. Introduction to Quantitative Genetics. New York NY: Ronald Press. Fe r rari S. F. 1988. The Behaviour and Ecology of the Buffy-Headed Marmoset, Callithrix flaviceps. Ph.D. diss., University College London. Ferri M. G. 1980. A Vegeta9ao Brasileira 1st ed. Sao Paulo: EDU SP. Fonseca G. A. B. 1989. Small mammal species diversity in Brazilian tropical primary and secondary forests of different sizes Revista Brasileira de Zoologia 6(3): 381-422. Foose, T 1990. Interactive management of small wild and captive populations. In Leontopithecus : Population Viability Analysis ed. U S. Seal J. D. Ballou, and C. Valladares-Padua Apple Valley, MN: Captive Breeding Specialist Group (IUCN/SSC/CBSG). Forman L. D. G. Kleiman, R. M. Bush, L. Phillips J. M. Dietz, A. F Coimbra-Filho, and S. O'Brien. 1986. Genetic variation within and among lion tamarins. American Journal of Physical Anthropology 71: 13-27. Franklin, I. R. 1980 Evolutionary changes in small populations. In Conservation Biology An Evolutionary-Ecological Perspective. 1st ed., eds. M. E Soule and B. A. Wilcox. Sunderland MA: Sinauer Associates Gaines M. S., and L. R. J. Mcclenaghan. 1980. Dispersal in small mammals. Annual Review of Ecology and Systematics 11: 16396. Garber P. A. 1984. The ecology of exudate feeding and the importance of plant exudates in the diet of the Panamanian tamarin. International Journal of Primatology 5: 1-15.

PAGE 188

170 Garber, P. A. 1 988. Foraging decisions during nectar feeding by tamarin monkeys (Saguinus mystax and Saguinus fuscicollis, Callitrichidae, Primates). Biotropica 20(2): 100-106. Gilpin M. E. 1987. Spatial structure and population vulnerability. In Viable Populations for Conservation. 1st ed ed M. E. Soule. Cambridge: Cambridge Univ. Press. Goldizen, A. W. 1987. Tamarins and marmosets: Communal care of offspring. In Primate Society., ed. B. B. Smuts D L. Cheney, R. M. Seyfarth, R W. Wrangham, and T. T. Struhsaker. Chicago : The University of Chicago Press. Gu i l laumon J. R. 0. C. Negreiros, A. J. Faria A. C. Dias D E. Brettas C. T Carvalho E N. Domingues F C. Serio, H. Y Ogawa and R. M Pfeifer 1983. Proposta Para o Manejo da Reserva Estadual do Morro do Diabo Sao Paulo Sao Paulo SP: lnstituto Florestal. Hanski I. 1989. Metapopulation dynamics : does it help to have more of the same. Trends in Ecology and Evolution 4 (4):113-14 Hanski, I., and M. Gilpin. 1991. Metapopulation dynamics: brief history and conceptual domain. Biological Journal of the Linnean Society 42:3-16. Harris, L. D. 1984 The Fragmented Forest: Island Biogeographic Theory and the Preservation of Biotic Diversity. Chicago IL: University of Chicago Press. Harvey, M. J., and R. W Barbour. 1965. Home range of Microtus ochrogaster as determined by a modified minimum area method. Journal Mammalogy 46: 398-402. Hershkowitz, P. 1972. The recent mammals of the Neotropical region : A zoogeographic and ecological review. In Evolution Mammals and Southern Continents. ed. A. Keast, F C. Erk, and B Glass Albany: State University of New York. 1977 Living New World Primates (Platyrrhini). 1st ed. Chicago, IL: Univ Chicago Press. Huck S. W W H Cormier, and W. G. Bounds. 1974 Reading Statistics and Research. New York, NY: Harper & Row Publishers.

PAGE 189

171 Hueck, K. 1978. Vegetation von Sudamerika. Stuttgart: Gustav Fischer Verlag. Hutchinson G E 1965. The Ecological Theater and the Evolutionary Play New Haven, Conn.: Yale University Press. Johns A. D. 1991 Forest disturbance and Amazonian primates. In Primate Responses to Environmental Change., ed H. 0. Box. London: Chapman and Hall. Keuroglian, A 1990. Observations on the Behavioral Ecology of the Black Lion Tamarin (Leontopithecus chrysopygus) at Caetetus Reserve, Sao Paulo, Brazil. Master's thesis, West Virginia University. Kinnaird M. F 1990. Behavioral and Demographic Responses to Habitat Change by the Tana River Crested mangabey (Cercocebus galeritus galeritus). Ph.D. diss. University of Florida. Kinzey W. G ., ed. 1987 The Evolution of Human Behavior: Primate Models 1st ed. Albany, NY : State University of New York Press Kleiman, D. G. 1981. Correlations among life history characteristics of mammalian species exhibiting two extreme forms of monogamy In Natural Selection and Social Behavior eds D. Alexander, and D. W. Tinkle, 332-344. New York NY: Chiron Press. Kleiman, D. G. R. J. Hoage, and K. M. Green. 1988 The lion tamarins genus Leontopithecus. In Ecology and Behavior of Neotropical Primates 1st ed., Vol. 2, eds R. A. Mittermeier, A. B. Rylands A. F. Coimbra-Filho and G. A. B Fonseca. Washington DC: World Wildlife Fund. Kleiman D. G., B. B. Beck, J. M. Dietz, and L. A. Dietz. 1991. Costs of re-introduction and criteria for success: accounting and accountability in the Golden Lion Tamarin Conservation Program. In Beyond Captive Breeding: Reintroducing Endangered Mammals to the Wild ed. J. H. Gipps. Oxford: Claredon Press.

PAGE 190

172 Klopfer, P. H., and J. P Hailman 1965. Habitat selection in birds. Adv. Study Behav.:279-303. Krebs, C. J. 1989. Ecological Methodology. 1st ed. New York, NY: Harper Collins Lacy R. C. 1990. Vortex : A Computer Simulation Model for Population Viability Analysis. Wild. Resear. 34: 1-25. Lee P C. 1991. Adaptations to environmental change : an evolutionary perspective. In Primate Responses to Environmental Change ed. H. 0. Box Cambridge: Chapman and Hall. Leger D W. 1977. An empirical e~aluation of instantaneous and one zero sampling of chimpanzee behavior. Primates 1 8 : 387-93. Leite J. F. 1981. A Ocupagao do Pontal do Paranapanema. Ph.D. diss. Universidade do Estado de Sao Paulo-UNESP. Levins, R. 1969. Some genetic and demographic consequences of environmental heterogeneity for biological control. Bulletin of the Enthomologycal Society of America 15: 237-40. 1970. Extinction. In Some Mathematical Questions i n Biology. 1st ed., Vol. 2 ed. M. Gerstenhaber. Providence RI: American Mathematical Society. Ludlow, M. 1986. Home Range, Activity Patterns and Food Habits of the Oncelot (Felis pardal i s) in Venezuela. Master's thesis, University of Florida. Ludwig, J. A., and J. F. Reynolds. 1988. Statistical Ecology A primer on methods and computing. New York, NY: John Wiley & Sons. Mack, D. S. and D. G. Kleiman 1978. Distributions of scent marks in different contexts in captive lion tamarins, Leotopithecus rosalia (Primates). In Biology and Behavior of Marmosets, ed. H. Rothe H J. Wolters and J. P Hearn. Gotiingen: E H. Rothe. Gottingen: Gottingen University Press. Manly B. F. J. 1986. Multivariate Statistical Methods A Primer. New York NY: Chapman and Hall.

PAGE 191

173 Marsh, C. W., A. D. Johns and J. M. Ayres. 1987. Effects of habitat disturbance on rain forest primates. In Primate Conservation in the Tropical Rain Forest. 1st ed., Vol. 9, ed. C. W. Marsh and R. A. Mittermeier. New York, NY: Alan R. Liss. Martin P., and P. F. Bateson. 1986. Measuring Behaviour. Cambridge: Cambridge University Press. M'Closkey, R. I. 1976. Community structure in sympatric rodents. Ecology 57:728-39. Merrian G. 1991. Corridors and connectivity: animal populations in heterogenous environments In The Role of Corridors ., eds. D. A. Saunders and R. J. Hobbs. New South Wales: Surrey Beaty. Mikan J. C. 1823. Delectus Florae et Faunae Brasiliensis. Vienna: Anthony Strauss. Mil t on K 1980. The Foraging Strategy of Howler Monkeys A Study in Primate Economics. New York NY : Columbia University Press. Mittermeier, R. A., and D. L. Cheney 1987. Conservation of primates and their habitats. In Primate Societies., eds. B. B. Smuts D. L. Cheney, R. M. Seyfarth, R. W. Wrangham and T. T Struhsaker Chicago, IL: The University of Chicago Press. Mittermeier R. A., A. B. Rylands, and A. F. Coimbra-Filho. 1988. Systematics:Species and subspecies. In Ecology and Behavior of Neotropical Primates., eds. R. A. Mittermeier, A. B. Rylands A. F. Coimbra-Filho, and G. A. B. Fonseca Vol. 2. Washington, DC: World Wildlife Fund. Mittermeier, R. A., and M. G. M. van Roosmalen. 1981. Preliminary observations on habitat utilization and diet in eight Surinam monkeys. Folia Primatologica 36: 1-39. Moore, P. D. and S. B. Chapman, eds. 1986. Methods in Plant Ecology, 2nd ed., Vol. 1. Oxford: Blackwell. Moynihan, M. 1970. Some behavioral patterns of Platyrrhini monkeys, 2: Saguinus geoffroyi and some other tamarins. Smithsonian

PAGE 192

174 Contributions to Zoology Washington, DC: Smithsonian Institution Press. Mukenhirn N. A. 1967. The behavior and vocal repertoire of Saguinus oedipus (Hershkovitz, 1965) (Callitrichidae, Primates). Master's thesis, University of Maryland. Murray, 0., and M. Oldfield. 1984. The top ten endangered species of the world. IUCN Bulletin 15: 77. National Research Council. 1981. Techniques for the Study of Primate Population Ecology. 1st ed. Washington, DC: National Academy Press. Natori M., and T. Hanihara. 1989. Morphological variations of the cranium within the genus Leontopithecus. Primates 30(2): 265268. Neter J., W. Wasserman, and M. H. Kutner. 1985. Applied Linear Statistical Models. Homewood: Richard Irwin Inc. Neyman, P. R. 1978. Aspects of the ecology and social organization of free ranging cotton-top tamarins Saguinus oedipus and the conservation status of the species. In Biology and Conservation of the Callitrichidae., ed. D. G. Kleiman. Washington: Smithsonian Institution Press. Nisbet, R. M., and W. S. Gurney. 1982. Modelling Fluctuating Populations. New York, NY: John Wiley and Sons. Noss, R. F. 1992. Wildlife corridors. In Ecology of Greenways., ed. D. Smith and P. Hellmund. Minneapolis: University of Minnesota Press. O'Brien, S. J., D. E. Wildt, and M. Bush. 1986. The Cheetah in genetic peril. Scientific American 5: 84-92. Otte, D., and J. A. Endler. 1989. Speciation and its Consequences. 1st ed. Sunderland, MA: Sinauer Associates Owen, J. G., and M. A. Chmielewsky. 1985. On cannonical variates analysis and the construction of confidence ellipses in systematic studies. Systematic Zoology:336-7 4.

PAGE 193

175 Padua, S ., L.A. Dietz, E. Nagagata, and C. Alves. 1990. Environmental education and the lion tamarins. In Population Viability Workshop., ed. U. S. Seal J. D. Ballou, and C. Valladares-Padua. Bloomington, MN: Captive Breeding Specialist Group (IUCN/SSC/CBSG). Padua, S. 199t. Conservation awareness through an environmental education school program at the Morro do Diabo State Park, Sao Paulo State, Brazil. Master's thesis., University of Florida. Passos, F. C. 1992. Habito alimentar do mico-leao-preto Leontopithecus chrysopygus (Mikan, 1823) (Callitrichidae, Primates)na Estagao Ecol6gica dos Caetetus, Municfpio de Gali a, SP. Master's thesis, Universidade Estadual de Campinas. Peres, C. A. 1986. Costs and benefits of territorial defense in golden lion tamarins, Leontopithecus rosalia. Master 's thesis University of Florida. Pereira M. E., and J. Altmann. 1985. Development and social behaviour in free-living non-human primates. In Nonhuman Primate Models for Human Growth and Development., ed. E. S. Watts. New York, NY: Alan R. Liss. Pielou, E. C. 1979. Biogeography. New York, NY: John Wiley and Sons ---.1984. The Interpretation of Ecological Data New York, NY: John Wiley and Sons. Pinder, L. 1986 Translocagao como tecnica de conservagao em Leontopithecus rosalia. Master's thesis, Universidade do Estado do Rio de Janeiro. Pissinatti, A. 1992 Cromogenia anomala. In A Primatologia no Brasil. 1st ed., Vol. 3, ed A. B. Rylands. Belo Horizonte MG: Sociedade Brasileira de Primatologia. Pook, A. G., and G. A. Pook. 1981. Field study of the socio-ecology of the Goeldi's monkey (Callimico goeldi) in northern Bolivia. Folia Primatologica 35: 288-312.

PAGE 194

176 Powell, J. R. 1983. Molecular approaches to studying founder effects. In Genetics and Conservation, A Reference for Managing Wild Animal and Plant Populations., ed. C. M. Schonewald-Cox, S. M. Chambers B. MacBryde, and L. Thomas. London: The Benjamin/Cummings Publishing Company, Inc Price E 0 1984. Behavioral aspects of animal domestication The Quarterly Review of Biology 59(1): 1-32. Ralls K., and J. Ballou. 1982. Effects of inbreeding on infant mortality in captive primates. International Journal of Primatology 3(4): 491-505. Ralls K. and J. Ballou 1983. Extinctions: Lessons from Zoos. I n Genetics and Conservation. ed. C. M. Schonewald-Cox S. M. Chambers B. MacBryde and L. Thomas. Menlo Park California: The Benjamin / Cummings Publish i ng Company. Ral l s, K., J Ballou, and R. L. Brownell. 1983. Genetic diversity in california sea otters: Theoretical considerations and management implications. Biological Conservation 25 : 209232. Ramirez, M. 1985. Exudate feeding in upper Amazonian Saguinus and Cebuella. American Journal of Physical Anthropology 66:216. Redford, K H. 1983. Mammalian myrmecophagy: feeding foraging and food preference Ph D. diss., Harvard University. Redshaw, M and J. J. Mallinson. 1991. Stimulation of natural patterns of behaviour, studies with golden lion tamarins and gorillas. In Primate Responses to Environmental Change., ed. H. 0 Box. London : Chapman and Hall. Renkonen, 0. 1938. Statisch-okologische Untersuchugen uber die terrestiche kaferwelt der finnischen bruchmoore Ann. Zoo/. Soc. Bot. Fenn. Vanano 6: 1-231. Rhine, R. J., and M Flanigon. 1978. An empirical comparison of one zero, focal animal, and instantaneous methods of sampling spontaneous primate social behavior. Primates 19:353-61

PAGE 195

177 Richard, A. F 1978. Behavioral Variation Case Study of a Malagasy Lemur. London: Bucknell University Press. Rizzini C. T 1963 Nota previa sabre a divisao fitogeografica do Brasil. Revista Brasileira de Geografia 25(1 ): 1-64 1967 Delimitacao, caracterizavao e relav6es da flora silvestre hileiana Atas do Simposio sabre a Biota Amaz6nica 4 : 13-36 Robinson J. G. 1986. Seasonal variation in the use of time and space by the wedge-capped capuchin monkey Cebus olivaceus: Implications for foraging theory. Smithsonian Contributions to Zoology (431 ): 1-60 Washington, DC: Smithsonian Institution Press. Rosenberger A. L. and A. F. Coimbra-Filho 1984. Morphology taxonomic status and affinities of the lion tamarins Leontopithecus (Callitrichinae Cebidae) Folia Primatologica 42 : 149-79. Rudran R. 1978. Socioecology of the Blue Monkeys (Cercopithecus m.ill.s. stuhlmanni) of the Kibale Forest Uganda. Smithsonian Contributions to Zoology 249: 1-88. Rumiz, D. I. 1989. Aplicacion de metodos de censado para Alouatta caraya em habitats fraccionados: I. Comparacion de metodos de transecta. In La Primatologia em Latinoamerica. 1st ed. eds. C. J. Saavedra, R. A. Mittermeier, and I. B. Santos. Washington : World Wildlife Fund-US. Rylands A. B. 1982. The Behavioural Ecology of the Golden-Handed Lion Tamarin Leontopithecus chrysomelas and the Marmoset Callitrix kuhlii (Callitrichidae, Primates). Ph.D. diss ., University of Cambridge. Scheffe H 1959 The Analysis of Variance. New York NY: Wiley Publications in Statistics. Sch i mper, A. F. W. 1903. Plant-Geography upon a Physiological Basis. Oxford: Groom and Balfour. Seal U. S. 1984 An analysis of the demography and genetics of captive Amur tigers and a preliminary captive propagation plan

PAGE 196

178 (Panthera tigris altaica). lnternationales Tigerzuchtbuch 1984:5-18. Seal U. S., and T J. Foose. 1984. Siberian tiger species survival plan : a strategy for survival. J. Minn. Acad. Sci.:3-9. Seal U., J. Ballou and C Valladares-Padua, eds. 1990. Leontopithecus Population Viability Analysis. Apple Valley MN : IUCN/SSC/CBSG Serra-Filho, R. A. C. Cavalli J. R. Guillaumon, J V. Chiarini, F. P. Nogueira, I. C. M. de Montfort J L. Barbier P L. S. C. Donzeli and I. Bittencourt. 1975. Levantamneto da cobertura natural e de reflorestamento no Estado de Sao Paulo. Boletim Tecnico do l nstituto Florestal de Sao Paulo 11: 1-53 Setzer J 1949. Os Solos do Estado de Sao Paulo 1 s t ed. Rio de Janeiro: Conselho Nacional de Geografia Shugart H. H. 1980. A Computer Simulation Model of Ecological Succession in Australia Subtropical Rainforest Oak Ridge: Environmental Science Public Division. Simberloff, 0 J. Farr J Cox and D. W. Mehlman. 1992. Movement corridors : Conservation bargains or poor investments? Conservation Biology 6: 493-504. Snowdon C. T. and P. Soini. 1988 The Tamarins, Genus Saguinus In Ecology and Behavior of Neotropical Primates 1st ed. eds. R A. M i ttermeier A. B. Rylands, A. F. Coimbra-Filho, and G A B. Fonseca. Vol. 2. Washington: World Wildlife Fund. Snowdon, C T., A. Hodum, A. F. Rosenberger, and A. F Coimbra-filho. 1986. Long call structure and the taxonomy of the lion tamarins. American Journal of Primatology 11 :253-62 Sokal, R. R. and F. J. Rohlf. 1981. Biometry San Francisco: W.H Freeman and Company. Soule, M. E. 1980. Thresholds for survival: maintaining fitness and evolutionary potential. In Conservation Biology an Evolutionary-Ecological Perspective eds. M. E. Soule and B. A. Wilcox. Sunderland, Massachusetts: Sinauer Associates, Inc

PAGE 197

179 1987. Viable Populations for Conservation. Cambridge: Cambridge University Press Stevenson M. F and A. B. Rylands 1988. The Marmosets Genus Callithrix. In Ecology and Behavior of Neotropical Primates. 1st ed., eds. R. A. Mittermeier, A. B. Rylands, A. F. Coimbra-Filho and G. A. B. Fonseca. Vol. 2. Washington, DC: World Wildlife Fund. Struhsaker T. T. 1975. The Red Colobus Monkey. Chicago, IL : The University of Chicago Press. Sussman R. W. and W. G Kinzey. 1984. The ecological role of the Callitrichidae:A review. American Journal of Physical Anthropology 64 : 419-49. Tabachnick, B. G. and L. S. F i dell. 1983. Using Multivariate Statistics. 1st ed. New York, NY: Harper & Row. Terborgh J. 1983. Five New World Primates A Study in Comparative Ecology. Princeton: Princeton University Press. Thornback J., and M. Jenkins 1982. The IUCN Mammal Red Data Book Gland, Switzerland: IUCN. Valladares-Padua C 1987. Black Lion Tamarin (Leontopithecus chrysopygus): Status and Conservation. Master's thesis University of F l orida. Valladares-Padua, C. 1991. The 1990 International Black Lion Tamarin (Leontooithecus chrysopvgus) Studbook. Sao Paulo SP: Sao Paulo Zoo. Valladares-Padua C. 1992. Social Structure and Mating System of Black Lion Tamarins: Preliminary Results from an Intra Specific Field Study Annals of the XIVth Congress of The International Primatological Society. Strasbourg : France. Valladares-Padua, C., and L. Cullen Jr. 1992. Umpublished report to the Primate Project Coordination: Duraflora; Lencois Paulista SP. (Available upon request).

PAGE 198

180 Ventura, A., G. Berengut, and M. A. M. Victor. 1965. Caracterfsticas edafocliamaticas das dependencias do Servivo Florestal do Estado. Silvicultura em Sao Paulo 4 (5):57-104. Victor, M. A. M. 1975. A Devastacao Florestal em Sao Paulo. Sao Paulo, SP: Soc. Bras. Silv. Waser, P. M. 1977. Feeding ranging and group size in the Mangabey Cercocebus albigena. In Primate Ecology: Studies of feeding and ranging behaviour in lemurs, monkeys and apes., ed. T. H. Glutton-Brock. London: Academic Press. Waser P. M. and R. H. Wiley. 1979. Mechanisms and evolution of spacing in animals. In Handbook of Behavioral Neurobiology. eds. P. Marler and J. G. Vanderbergh. New York NY: Plenum Press Wight, H. M. 1938. Field and Laboratory Technic in Wildlife Management. Univ. Migican Press: Ann Arbour. Wilks, S. S.1962. Mathematical Statistics. New York, NY: Wiley and Sons. Winnet-Murray, K. 1986. Behavioral Plasticity of Four Tropical Cloud Forest Wrens. Ph.D. diss., University of Florida, Gainesville. Whitesides, G. H., J. F. Oates, S. M. Green, and R. P. Kluberdanz. 1988. Estimating primate densities from transects in a west african rain forest: A comparison of techniques. Journal of Animal Ecology 57: 345-67. Wolda, H. 1981. Similarity indices, sample size and diversity. Oecologia 50:296-302. Woodford, M. H., and R. A. Kock. 1991. Veterinary considerations in re-introduction and translocation projects. In Beyond Captive Breeding: Reintroducing Endangered Mammals in the Wild., ed. J. H. Gipps. Oxford: Clarendon Press. Wright, S 1977. Evolution and the Genetics of Populations. Chicago: University of Chicago Press.

PAGE 199

181 Zack, R. and Falls, J B. 1976. Ovenbird (Aves: Parulidae) hunting behavior in a patchy environment: An experimental study. Canadian Journal of Zoology 54:1894-1903.

PAGE 200

BIOGRAPHICAL SKETCH Claudio Valladares Padua was born on April 14, 1948, in Rio de Janeiro, Brazil. In December 1974, he received a degree in Business Administration from the University of Economy and Finances of Rio de Janeiro. He worked in the business field until 1980. In 1978 he entered University Gama Filho Rio de Janeiro, seeking a degree in biology. In 1980 he left the business world to start working on the conservation of the black lion tamarin and other endangered primate species at the Rio de Janeiro Primate Center/CPRJ-FEEMA. In 1981 he received a Bachelor of Science degree in biology In 1982, he participated in an intensive training program on the breeding of endangered species in captivity at the Jersey Wildlife Preservation Trust Jersey, Channel Islands. In 1983, he joined University Gama Filho again, where he taught Introduction to Ecology. He started h i s graduate studies at the University of Florida in August 1984 and completed requirements for the Master of Arts degree in January 1987, majoring in Latin American studies, with specialization i n wildlife conservation. In 1986, he received the conservation award of the American Society of Primatology for his work toward the conservation of the black lion tamarin Besides his native language, Portuguese, Claudio is fluent in English, Spanish and French. 182

PAGE 201

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I ources and Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Forest Resources and Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality as a dissertation for the degree of Doctor of Philosophy F. Eisenberg Katharine Ordwa"'y~'-,cAjfo s r of Ecosystem Conservation

PAGE 202

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Sunquist Associate Scientist of Forest Resources and Conservation I certify that I have read this study and that i n my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. H. Jane Brockmann Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ~!f:!fAR Professor of Botany

PAGE 203

This dissertation was submitted to the Graduate Faculty of the School of Forest Resources and Conservation in the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May, 1993 Aw r 1 kGJ.~)K-v ~illA J Director, Forest Resources and Conservation Dean, Graduate School