Environmental Enrichment in Captive Marmosets and Tamarins

December 29th, 2011 by primatecare Leave a reply »

By Dr. Hannah Buchanan-Smith

The author discusses the three main criteria which constitute a good captive environment.  They are: good physical health, breeding success, and an animal’s ability to acquire and retain behavioral skills needed to cope successfully with his or her natural environment. Through her field studies, the author has learned that a variety in diet, innovative foraging devices, indoor/outdoor enclosures, natural settings, and a comfortable social environment are vital for the well-being of captive primates. She illustrates how altering conditions can promote natural behaviors in captive marmosets and tamarins.

Diet and Foraging
Cages and Furnishings
Social Environment

About the author


Concern for the psychological well-being of captive primates is not new, but a number of recent publications have illustrated that it is perhaps becoming more prominent. In this paper I shall briefly describe what the criteria are for a good captive environment, and I shall illustrate this with some recent studies which have looked at how altering conditions can promote natural behaviors in marmoset and tamarin monkeys. The same principles can be applied to almost any wild animal kept in captivity.

First, what constitutes a good captive environment? There have been three main criteria put forward in a recent article by Snowdon and Savage (1989). The first is a veterinary medical criterion, defined as good physical health and freedom from disease. The second is a biological criterion, which basically relates to the breeding success of the animals. However, this criterion should be divided into two levels: breeding and rearing.   Many captive animals do breed successfully, but the infants need to be hand-reared, because the parents will not do so successfully. Generally the hand-reared individuals are incapable of rearing their own young, as they have lacked a normal social development. This is particularly true of marmosets and tamarins, and I shall be elaborating on this point later. The third criterion is a behavioral ecological criterion, which relates to the animal’s ability to acquire and retain behavioral skills they need to cope successfully with their natural environment.

It should be noted that these three criteria overlap and directly affect each other in that an animals’ physical well-being is directly affected by its psychological well-being. For instance, it is now known that the social environment can affect the immune system. Animals housed in good social environments are better able to defend against disease (Coe, Rosenberg, Fischer & Levine, 1987).

The starting points for developing a good captive environment are to reproduce some of the salient features of the wild environment, and to create opportunities for captive animals to develop skills they might need in the wild. Obviously, to be able to do this we require data from the field. I have done research on the red-bellied  tamarin (Saguinus labiatus) and the saddle backed tamarin (S. fuscicollis) in northern Bolivia, and I shall be taking examples from my own data throughout the paper.   These two species of tamarins formed stable mixed species troops (Buchanan-Smith, 1990). I collected detailed data on the ranging behavior and activity patterns of the tamarins. I trapped individuals of both species and put radio-collars on certain individuals, and then radio-tracked them from dawn to dusk. This made them easy to locate and observe what they were feeding on, and it allowed me to follow them from a distance to find out what sizes their ranges were and how often the two species were together (Buchanan-Smith, 1990, 1991).  However, not all species of marmosets and tamarins have been studied in the wild, and from those that have been studied, we know that different species may behave very differently. Before creating or improving a captive environment for marmosets and tamarins, we must first find out all we can about that specific species. If the species has not been studied in the wild, we should look to the data for the most closely related species which has. The captive environment should then be modeled on what is known about the natural habitat and behavior in the wild.

Many would argue that there is no way to evaluate the effect of environmental manipulations on psychological well-being. Basically, my view is that a change in behavior is shown to be beneficial if it becomes closer to the natural behavior of the monkeys. This includes not only the presence of the behavior, but the frequency of it as well. For instance, not only is it important to give the monkeys opportunities to forage for food in ways that they do in their natural habitat, but they should spend about the same amount of time foraging in captivity as they do naturally.

I shall now go through examples of how researchers have changed aspects of the environment of marmosets and tamarins, especially in three areas of the environment: diet and foraging, cages and furnishings, and the social environment.


We know that marmosets and tamarins have a very varied diet in the wild. My research on the red-bellied tamarins and saddle backed tamarins in Bolivia shows that the tamarins consume fruits from at least 16 different species of trees. They also consume nectar from flowers (family Guttiferae; Symphonia globuliferae), and resin from a bean-like pod (family Mimosaceae; Parkia pendula [Willd.] Benth ex Walp.). In addition to this they spend much of their time foraging for insects (Buchanan-Smith, 1990, 1991). Although I never saw the tamarins eat any other animal material, field data from other species indicates that they eat birds’ eggs, frogs, and lizards (Neyman, 1977; Sussman & Kinsey, 1984).

Because of the diversity in their natural diet, it is important to provide a wide variety of different food types in captivity. However, we  should remember that in the wild, animals rarely encounter food in great abundance, and often they must search for it. In the wild, foraging occurs throughout the day and can occupy up to 50-60% of the time available (Garber, 1984; Yoneda, 1984). Therefore, in order to more closely approximate natural conditions, we should distribute food over the course of the day and make it more difficult to find.

A recent study has demonstrated how introducing unfamiliar foraging devices which are explicitly designed to make it harder to secure food treats can have a significant influence on the activity budgets of golden lion tamarins (Leontopithecus rosalia; Molzen and French,

This species are described as being extractive foragers, and feeding devices were fashioned to stimulate these extractive foraging skills. These were food containers with small holes drilled in them which were filled with broken corn-cob and a few raisins. These devices were suspended approximately 27cm below the top of the cage, so that the only way the tamarins could get access to the raisins was either by hanging upside down by his or her hind limbs, which they are very proficient at doing, or by sitting on the bowl and grasping one of the device’s suspension wires for support.

Once the tamarins discovered that the suspended bowls did contain some raisins, the devices then became objects of focused attention. To introduce uncertainty in food availability, the tamarins were given both a baited and an unbaited device. These devices were used in addition to the normal rations which the tamarins received in ceramic dog bowls. Behavior in normal feeding protocols was compared with normal feeding protocols and the foraging devices. The results show that time spent actively exploring the environment increased, foraging increased, and adults travelled more, while time spent eating and drinking decreased, and adults rested less and engaged in fewer social interactions such as grooming when the device was present than when it was not. The authors made no comparison between the activity budgets of wild golden lion tamarins and their captive tamarins, and so it is not known whether the changes in behavior profiles drew closer to the activity budgets of the  tamarins in their natural habitat. However, because the tamarins were using natural foraging skills which they had not had the opportunity to use before, the change in behavior was positive.

Another simple technique to increase time spent in food acquisition is to spear whole apples or oranges on to the end of a stick, like a bamboo cane, and suspend this from a branch in the cage. Marmosets and tamarins will hang  upside down and spend much time picking off bits of apple or orange (pers. obs.) in a similar fashion to how they would forage naturally.

I prefer these foraging situations to scattering food items in the wood shavings or other floor coverings, which is another method of  increasing foraging time (McKensie, Chamove & Feistner, 1986). Many species of marmosets and tamarins do not go to the ground regularly in the wild, or if they do they are especially cautious as they may be more vulnerable to predation. I do not think that this technique of increasing foraging time should be encouraged in captivity unless it is known that the species does go to the ground regularly, and for long periods, in the wild. The tamarins I observed in Bolivia were extremely reluctant to descend to less than 2-3m in  the forest canopy. In relation to this, it should also be noted that consideration should be taken about the positioning of food dishes in the cage. A recent observation on cotton-top tamarins showed that group members carrying infants were reluctant to approach food dishes placed near floor level. This was remedied by placing food dishes at least a meter from floor level — whereupon tamarins carrying infants readily approached food dishes and fed (Snowdon & Savage, 1989).  Furthermore, by providing a meal in two sets of dishes in larger groups, competition between group members may be reduced, and each individual is more likely to get a more varied diet and an equal share of the preferred food items (Price & McGrew, 1989).

My final point on diet and foraging relates to gum feeding in marmosets. Marmosets have teeth which are specialized for gnawing holes in trees to get the gum to exude, which they then eat. One study on the pygmy marmoset (Cebuella pygmaea) reports that these monkeys spend 32% of their total activity time in the procurement and ingestion of gum (Ramirez, Freese & Revilla, 1977). To mimic gum feeding in the wild, as we see in the marmosets, a simple inexpensive sap feeder has been devised, which consists of sections of dowel drilled length-wise with holes which are filled with gum arabic.  Captive common marmosets (Callithrix jacchus) readily learn to gnaw at this device to extract gum arabic using the full range of gum foraging patterns as in the wild (McGrew, Brennan & Russell, 1986).


Marmosets and tamarins range over large areas in the wild, and there is now a trend to release groups of tamarins into woods, or on to  islands which are obviously more naturalistic settings (Price et al., 1989). Most zoos provide both indoor and outdoor enclosures for their marmosets and tamarins, and these outdoor enclosures are particularly important as they give the monkeys the opportunities to face the elements, take shelter from the rain, and find sunny spots to bask themselves in.  Outdoor enclosures also often attract insects, and the marmosets and tamarins will readily forage for insects, as they do in the wild, if given the opportunity (pers. obs).

If their home cage cannot be reasonably sized in captivity, it is important to provide the monkeys with large exercise areas. These areas can be shared by several different groups, and the monkeys can be given voluntary access to these cages using plastic tunnelling of 15cm diameter (Price & McGrew, 1990). This ducting can also be used to move monkeys to new housing, and obviates the need for catching and handling of monkeys, which is likely to be stressful.

Let us now turn to cage furnishings.  When the golden lion tamarin (L. rosalia) reintroduction program began, the tamarins appeared to be physically unable to cope with natural substrates and were deficient in appropriate locomotory behavior (Kleiman et al., 1986). So, if possible, captive enclosures should be furnished with a high density of natural branches as well as ropes to simulate vines. These provide surfaces on which the claws can get a good grip, and they should be arranged so that they have a variety of textures, diameters and degrees of firmness. Some of the branches should be quite firm so that they do not give when the animal lands on them, while others should be quite loose, so that the animals must learn to accommodate their jumping to these different surfaces.

As most marmosets and tamarins are highly arboreal, the branches should be positioned well above ground level, although many captive monkeys do not go to the ground in captivity to forage for dropped food; therefore a few branches should be positioned near the ground so that if the monkeys are startled when on the ground, they can easily leap to a higher position. The diameter and orientation of the branches within the cage should also be chosen with care. I found that in the wild, saddle backed tamarins (S. fuscicollis) often locomoted by jumping between vertical trunks, while red bellied tamarins (S. labiatus) more often jumped from branch to branch and used thick near-horizontal branches for quadrupedal travel. Certainly from my own field observations, it seemed that most social interactions were performed on large horizontal or gently sloping branches which were relatively out in the open — possibly to give good predator detection viewing (pers. obs.). A recent comparative study on captive callitrichids has revealed that species do have preferences for certain diameters and orientations of branches (Seymour & Kinghorn, pers. comm.), so again the important point here is to provide variability in branch orientation and diameter.

Branches should obviously be cleaned to provide adequate  sanitation. To my knowledge, no research has been done on the effect of cleaning marmoset and tamarin cages. However, as scent marking is an important means of communication for these monkeys, care should be taken not to clean the branches too regularly as this might result in abnormally high frequencies of scent marking. It may be that it is best to clean only half the branches at any one time, which would give the monkeys more stability in their habitat. It may also be important to replace branches in a new configuration, as this would reduce the likelihood of any motor stereotypies and facilitate the maintenance of the essential skill of acquiring spatial knowledge of new environments (Kleiman, 1989; Redshaw & Mallinson, 1991).

Cover should also be provided for animals, so that they can hide if they wish to. At the Jersey Wildlife Preservation Trust, marmosets and tamarins can freely move between their indoor and outdoor enclosure, and the indoor enclosure is out of bounds to the public, which allows the marmosets and tamarins some degree of privacy (Mallinson, 1975). They also have a reasonable amount of dense natural foliage in their outdoor enclosures to provide cover. If natural foliage cannot be provided to give cover, hanging screens, such as strips of curtains, can visually divide a large cage area, and this does have positive behavioral results (Mackensie, Chamove & Feistner, 1986).

My final point to note on cage furnishings is the benefit of providing more than one nest box. If there is any aggression in the group an individual may be excluded from the nest box, and if there is only one, s/he would not have a safe sleeping area. However, I know of no research in this area.


Marmosets and tamarins have a cooperative rearing system where the mother generally gives birth to twins, and the father and other group members care for the young by carrying, possibly food sharing, and perhaps by looking out for predators (e.g. Buchanan-Smith, 1984; Price, 1992; Rylands, 1985; Cleveland & Snowdon, 1984).  There are many published reports that if offspring are removed from their group before they have had experience with rearing infants, they have a much lower likelihood of raising offspring themselves (S. fuscicollis, Epple, 1978; L. rosalia, Hoage, 1977; S. oedipus, Cleveland & Snowdon, 1984, Snowdon, Savage & McConnell, 1985, Tardif, Richter & Carson, 1984a, 1984b). It is recommended that marmosets and tamarins should have experience with at least two sets of rearing episodes; otherwise they will not make good parents themselves (Snowdon & Savage, 1989). This applies to sons as well as daughters, because fathers as well as mothers care for the young. Most marmoset and tamarin keepers do allow this, but I understand that there are still some
laboratories which remove individuals from the group before they have had this experience, either for testing,  or because the cage is too small to house a greater number of monkeys. The problem is that once this habit has started it is difficult to reverse.

If it is really not possible to keep marmosets and tamarins in large family groups, giving them opportunity for a good social development and social activities such as allogrooming, infant care, and in some species food sharing, then there are ways around this. Giving monkeys the opportunity to watch others may increase their quality of life, although being in continual close visual contact and proximity may be stressful for them. A neat way around this is to allow monkeys an opportunity to peep through a small hole at a neighboring group. A recent study on cotton-top tamarins (S. oedipus) found that when a small peephole was made (initially by accident) group members were keen to use it to observe their neighbors (Moore, Cleland & McGrew, 1991).  Their neighbors did not know they were being watched and so it is unlikely that it had any detrimental effects on them.

Another way to provide a varied social environment is to keep marmosets and tamarins in mixed exhibits (Xanten,  1990). My personal feelings are that if mixed exhibits are to be set up, they should be between species which are sympatric in the wild, which not only is likely to be more successful, but which also provides a more realistic zoographic enclosure with greater educational value.  Tamarins often form stable mixed associations with other tamarins — such as the saddle backed and red-bellied tamarins I studied in Bolivia. These two species spent 85% of their time within 50m of each other and often much closer, and the associations were long lasting and stable in that each mixed species group shared a common territory which they jointly defended against neighboring mixed species troops (Buchanan-Smith, 1990). Saddle backed tamarins also form stable associations with emperor tamarins (S. imperator, Terborgh, 1983) and with moustached tamarins (e.g., S. mystax, Garber, 1988), and black mantle tamarins (S. nigricollis, Hernandez-Camacho & Cooper, 1976). We have recently set up mixed groups of red-bellied tamarins and saddle backed tamarins successfully at Belfast Zoological Gardens, Northern Ireland. Further details are provided in: Hardie, S.M., Day, R.T. & Buchanan-Smith, H.M. (1993) Mixed species Saguinus groups at Belfast Zoological Gardens.


There is a clear relationship between the environment and behavior. An impoverished environment leads to a decrease in social behaviors such as grooming, playing and other behaviors such as scent marking and locomotion (C. jacchus, Schoenfeld, 1989). Furthermore, there is clear evidence that environmental stress can also affect the reproductive success of marmosets. A recent study has demonstrated that parity was suppressed and the number of spontaneous abortions increased when a common marmoset (C. jacchus) colony was disrupted by nearby construction work (Johnson et al., 1991).

This has been a rather a brief outline of the many different aspects of how to improve conditions in captivity and promote naturalistic behaviors.  Already many zoos and laboratories are implementing environmental enrichment techniques, not only to improve the quality of life of the monkeys but to enhance the education value of zoo exhibits.

I should like to end by emphasizing the importance of collecting behavioral data to determine if the change you have made in the environment has actually had the desired effect. This data collection will give you some objective measure relating to how the behavior has changed.  The different data recording techniques and information on how to set up a scientific study to measure behavior are provided in Martin and Bateson (1986) and Shepherdson (1989).


Buchanan-Smith, H.M. (1984). Preliminary report on infant development of the black tailed marmoset, Callithrix argentata melanura, at the Jersey Wildlife Preservation Trust. Dodo, Journal of the Jersey Wildlife

Preservation Trust, 21, 57-67.

Buchanan-Smith, H.M. (1990). Polyspecific association of  two tamarin species Saguinus labiatus and Saguinus fuscicollis in Bolivia. American Journal of Primatology, 22, 205-214.

Buchanan-Smith, H.M. (1991). A field study on the red-bellied tamarin, Saguinus l. labiatus, in Bolivia. International Journal of Primatology, 12, 259-276.

Cleveland J. & Snowdon, C.T. (1984). Social development during the first 20 weeks in the cotton-top tamarin (Saguinus o. oedipus). Animal Behavior, 32, 432-444.

Coe, C.L., Rosenberg, L.T., Fischer, M. & Levine, S. (1987). Psychological factors capable of preventing the inhibition of antibody responses in separated infant monkeys. Child Development, 58, 1420-1431.

Epple, G. (1978). Reproductive and social behavior of marmosets with special reference to captive breeding.  In N. Gengozian & F. Deinhardt (Eds.), Primates in medicine: Vol. 10, (pp. 50-62). Basel: Karger.

Garber, P.A. (1984). Use of habitat and positional  behavior in a neo-tropical primate, Saguinus oedipus.  In P.S. Rodman & J.G.H. Cant (Eds.), Adaptations for foraging in nonhuman primates (pp. 112-133). New York: Columbia University Press.

Garber, P.A. (1988). Diet, foraging patterns and resource defense in a mixed species troop of Saguinus mystax and Saguinus fuscicollis in Amazonian Peru. Behavior, 105, 18-34.

Hardie, S.M., Day, R.T., & Buchanan-Smith, H.M. (1993). Mixed species saguinus groves at Belfast Zoological Gardens. Neotropical Primates, 1 (4), 19-21.

Hernandez-Camacho, J. & Cooper, R.W. (1976). The non-human primates of Columbia. In R.W. Thorington & P.G. Heltne (Eds.), Neotropical primates: Field studies and conservation (pp. 35-69). Washington, DC: National Academy of Sciences.

Hoage, R.J. (1977). Parental care in Leontopithecus rosalia rosalia, sex and age differences in carrying behavior and the role of prior experience. In D.G. Kleiman (Ed.), The biology and conservation of the callitrichidae (pp. 293-30). Washington, DC: Smithsonian
Institution Press.

Johnson, E.O., Kamilaris, T.C., Carter, S., Gold, P.W. & Chrousos, G.P. (1991). “Environmental stress” and reproductive success in the
common-marmoset (Callithrix jacchus jacchus). American Journal of Primatology, 25, 191-201.

Kleiman, D.G. (1989). Reintroduction of captive mammals for conservation. Guidelines for reintroducing endangered species into the wild. Bioscience, 39, 152-161.

Kleiman, D.G., Beck, B.B., Dietz, L.A., Ballou, J.D. & Coimbra-Filho, A.F. (1986). Conservation program for the golden lion tamarin: Captive research and management, ecological studies, education strategies, and reintroduction. In K. Benirschke (Ed.), Primates: The road to self-sustaining populations (pp. 959-979). New York: Springer-Verlag.

Mallinson, J.J.C. (1975). The design of two marmoset complexes at the Jersey Zoological Park. Jersey Wildlife Preservation Trust Annual Report, 12, 21-26.

Martin, P. & Bateson, P.(1986). Measuring behavior, an introductory guide. Cambridge: Cambridge University Press.

McGrew, W.C., Brennan J.A. & Russell, J. (1986). An artificial “gum-tree” for marmosets (Callithrix j. jacchus). Zoo Biology, 5, 45-50.

McKensie, S.M., Chamove, A.S. & Feistner, A.T.C. (1986). Floor coverings and hanging screens alter arboreal monkey behavior. Zoo Biology, 5, 339-348.

Molzen, E.M. & French, J.A. (1989). The problem of foraging in captive callitrichid primates; Behavioral time budgets and foraging skills. In E. Segal (Ed.), Housing, care and psychological well-being of captive and laboratory primates (pp. 89-101). Park Ridge, NJ: Noyes Publications.

Moore, K., Cleland, J. & McGrew, W.C. (1991). Visual encounters between families of cotton-top tamarins, Saguinus oedipus. Primates, 32, 23-33.

Neyman, P.F. (1977). Aspects of the ecology and social organization of free-ranging cotton-top tamarins (Saguinus oedipus) and the  conservation status of the species. In D.G. Kleiman (Ed.), The biology and conservation of the Callitrchidae (pp. 39-71). Washington, DC: Smithsonian Institution Press.

Price, E.C. (1992). Contributions to infant care in  captive cotton-top tamarins (Saguinus oedipus): The influence of age, sex, and  reproductive status.  International Journal of Primatology, 13, 125-141.

Price, E.C., Feistner, A.T.C., Carroll, J.B. & Young, J.A. (1989). Establishment of a free-ranging group of cotton-top tamarins Saguinus oedipus at the Jersey Wildlife Preservation Trust. Dodo, Journal of the Jersey Wildlife Preservation Trust, 26, 60-69.

Price, E.C. & McGrew, W.C. (1990). Cotton-top tamarins (Saguinus o. oedipus) in a semi-naturalistic captive colony. American Journal of Primatology, 20, 1-12.

Ramirez, M.F. Freese C.H. & Revilla, J.(1977).  Feeding ecology of the pygmy marmoset, Cebuella pygmaea,  in northeastern Peru. In D.G. Kleiman (Ed.), The biology and conservation of the Callitrichidae (pp. 39-71).  Washington, DC: Smithsonian Institution Press.

Redshaw, M.E. & Mallinson, J.J.C. (1991). Stimulating natural patterns of behavior. Studies with golden lion tamarins and gorillas. In H.O. Box (Ed.), Primate responses to environmental change (pp. 217-237). London: Chapman & Hall.

Rylands, A.B. (1985). Infant carrying in a wild marmoset group, Callithrix humeralifer; evidence for a polyandrous mating system. In M.T. de Mello (Ed.), A primatologia no Brazil, II (pp. 131-144). Brasilia: University of Brasilia.

Schoenfeld, D. (1989). Effects of environmental impoverishment on the social behavior of marmosets (Callithrix jacchus). American Journal of Primatology Supplement, 1, 45-51.

Sheperdson, D. (1989). Environmental enrichment — measuring the behavior of animals. Ratel, 16, 134-139.

Snowdon, C.T. & Savage, A. (1989). Psychological well-being of captive primates: General considerations and examples from callitrichids. In E. Segal (Ed.), Housing, care and psychological well-being of captive and laboratory primates (pp. 75-88). Park Ridge, NJ: Noyes Publications.

Snowdon, C.T., Savage, A. & McConnell, P.B. (1985). A breeding colony of cotton-top tamarins (Saguinus oedipus). Laboratory Animal Science, 35, 477-481.

Sussman R.W. & Kinsey, W.G. (1984). The ecological role of the Callitrichidae: A review. American Journal of Physical Anthropology, 64, 419-449.

Tardif, S.D., Richter, C.B. & Carson, R.L. (1984a). Effects of sibling-rearing experience on future reproductive success in two species of Callitrichidae. American Journal of Primatology, 6, 377-380.

Tardif, S.D., Richter, C.B. & Carson, R.L. (1984b). Reproductive performance of three species of Callitrichidae. Laboratory Animal Science, 34, 272-275.

Terborgh, J. (1983). Five new world primates: A study of comparative ecology. Princeton, NJ: Princeton University Press.

Xanten, W.A. (1990). Marmoset behavior in mixed species exhibits at the National Zoological Park, Washington. International Zoo Year Book, 29, 143-148.

Yoneda, M. (1984). Comparative studies on vertical separation, foraging behavior and travelling mode of saddle-backed tamarins (Saguinus fuscicollis) and red-chested moustached tamarins (Saguinus labiatus) in northern Bolivia. Primates, 25, 414-422.

About the author

Hannah Buchanan-Smith
University of Stirling
Stirling, Scotland 

Dr. Hannah Buchanan-Smith graduated from St. Andrews University in Scotland with a B.Sc. in Psychology in 1985. As part of her degree, she studied marmosets at the Jersey Wildlife Preservation Trust. She went on to Reading University, England to conduct her doctoral
research, which combined both behavioral observations of captive tamarins and field research on tamarins and other primates in Bolivia.

Since completing her Ph.D., Hannah has conducted research and taught at St. Andrews and Stirling Universities in Scotland.

For more information, the author can be reached at Department of Psychology, University of Stirling, Stirling, FK9 4LA, Scotland.


Leave a Reply

You must be logged in to post a comment.