The positional behavior of pygmy marmosets (Cebuella pygmaea) in northwestern Bolivia
- First Online:
- Cite this article as:
- Jackson, C.P. Primates (2011) 52: 171. doi:10.1007/s10329-011-0237-7
- 238 Views
Pygmy marmosets are distinctive given their diminutive body size, their year-round reliance upon exudates, and their use of morphologically adapted tegulae to engage in a high degree of claw-clinging behaviors associated with exudate exploitation. This project examined the positional behavior and habitat preferences of one group of pygmy marmosets in a secondary forest within the Department of Pando, northwestern Bolivia. Results from this study indicate that pygmy marmosets primarily use claw-clinging during feeding (89.6%) with preferential use of large vertical trunks. Claw-clinging was also the dominant postural mode during exudate foraging (43.1%) with preferential use of large vertical trunks. Quadrupedalism was the dominant locomotor mode during travel (55.7%) with preferential use of bamboo and medium-sized substrates. These results support previous notions that claw-climbing is a solution to overcome the constraints of small body size while suggesting that quadrupedalism is a habitat-dependent locomotor mode.
KeywordsPygmy marmosetsPositional behaviorHabitat useHabitat preference
Previous studies have shown that pygmy marmosets use clawed positional behaviors while on large vertical trunks (Kinzey et al. 1975; Soini 1988; Youlatos 1999a, 2009) and possess key cranial and postcranial features that enable this species to exploit exudates (Vinyard et al. 2003; Ford and Davis 2009). Given that pygmy marmosets exhibit a suite of morphological characters that enable this species to utilize exudate resources, their high frequencies of claw-climbing and claw-clinging during feeding and traveling would suggest that this positional behavior repertoire and morphology are closely related to gouging and feeding on exudates (Kinzey et al. 1975; Soini 1988; Youlatos 1999a, 2009; Vinyard et al. 2003; Ford and Davis 2009).
Youlatos (2009) presented the most thorough pygmy marmoset positional behavior study to date by incorporating detailed descriptions of habitat availability at his site in Ecuador. He demonstrated how detailed habitat descriptions differentiate the evolutionary significance of preferential substrate use from habitat-dependent locomotor modes; for example, pygmy marmosets strongly preferred vertical substrates while they used horizontal substrates according to availability. Vertical substrate use is evolutionarily significant whereas horizontal substrate use is habitat dependent. A comparative positional behavior and habitat use study would provide the opportunity to examine variation in habitat-dependent locomotor modes and preferential substrate use between study sites.
For these reasons, I studied the positional behavior of wild Bolivian pygmy marmosets and the associated environmental contexts to draw comparisons between habitat availability and behavioral choices between different study sites. Since little is known about pygmy marmoset positional behavior, it is essential to collect more data to address the following questions: Does pygmy marmoset positional behavior vary by site, specifically between this study in Bolivia and Ecuador? Does substrate availability vary between sites, and does this variation influence positional behavior choices? If pygmy marmoset positional behavior is influenced by diet and habitat availability, then animals at sites with different forest structures and vegetation types are expected to vary in their choice of positional behaviors, thus underlining the adaptive flexibility of similar behaviors.
The study site was located within the Department of Pando, Bolivia in a small patch of secondary bamboo forest (11 25.521S, 069 00.286W). This study was conducted between June 1 and August 1, 2007. I observed one group of pygmy marmosets that consisted of five individuals within a 0.015-ha mixed-liana secondary bamboo forest patch. Liana patches, defined as a tight networks of randomly arranged lianas (Youlatos 2009), were present within the study area. Data were collected using an instantaneous scan sampling technique at 1-min intervals in order to maximize the data pool, since the Camp Callimico study group was not as well habituated, and not as consistently visible, as the group in Yasuni National Park.
I collected information on the following activities as described by Bergeson (1996): feed, forage, rest, and travel. I scored feed exudates if the animal was holding or processing exudates (Bergeson 1996), and feed insects if the animal was holding or processing insects (Bergeson 1996). I scored forage exudates if the animal traveled within an exudate site to exploit exudates. I scored forage insects if the animal traveled within a resource site to exploit insects. Travel was scored if the animal moved throughout space between resource sites, the sole function of travel being spatial displacement (Bergeson 1996).
Substrate type, orientation, and size
Substrate types were classified either as trunk, branch, liana, bamboo or terminal. Substrate orientations followed Garber and Leigh’s (2001) descriptions: horizontal (0°–15°), oblique (16°–74°), and vertical (75°–90°). I visually assessed substrate size relative to the body size of the individual. Small substrates were scored if the animal could grasp the substrate with a single hand (<5 cm circumference) (Davis 2002). Medium substrates were scored if the animal could wrap both of their arms around the substrate from front to back (6–15 cm circumference) (Davis 2002). Large substrates were scored if the animal could not wrap both their arms around the substrate from front to back (>16 cm circumference) (Davis 2002).
To collect substrate availability data, I randomly selected thirty sample plots of 1 m3. A 150 m × 50 m × 25 m home range produces 187,500 m3 while the total volume of the sample plots produces 30 m3. The ratio of total volume to sampled volume is 6,250:1. I randomly assigned these thirty locations to different visually estimated heights set at 1–2 m, 6–7 m, and 11–12 m above ground, in order to sample broader height classes of 0–5 m, 6–10 m, and >11 m respectively, producing ten samples for each height class. For substrates at 1–2 m above ground, I counted all substrates within the one meter cube, and I counted additional substrates for each change in size or orientation. For example, if a horizontal branch shifted to a vertical orientation, then this one branch would be counted twice as both a horizontal and a vertical branch. To provide a sample of substrate availability at 5–6 m and 10–11 m heights, heights that did not facilitate counting, I described a cube (1 m3) either as sparsely covered, moderately covered or densely covered. A sparsely covered cube was defined as any cube with fewer than 100 substrates. A moderately covered cube was defined as any cube with any number of substrates between 100 and 250. A densely covered cube was defined as any cube with any number of substrates greater than 250. I then nonrandomly selected 1 m cubes close to the ground that were visually estimated as similar to cubes at heights above 5 m. Within these cubes, I counted each substrate in terms of type, size, and orientation, and then I used these counts as an equivalent measure to the densities at the higher levels, which I could not directly count. For example, I assumed that the number of small substrates within a densely covered cube at a lower level was equal to the number of substrates within a densely covered cube at 10 m height.
“standing on the hindlimbs” (Hunt et al. 1996, p. 371)
“the [hind-foot claws] anchor the lower body to a stable support in order for the individual to reach out and snatch insects” (Hunt et al. 1996, p. 372)
“a flexed-limb posture [in which the claws are anchored] on vertical-subvertical substrates” (Hunt et al. 1996, p. 369)
“torso pronograde posture on a relatively horizontal substrate with the weight borne principally by the torso” (Hunt et al. 1996, p. 373)
“four-limbed standing on horizontal or subhorizontal supports and the elbow and knee are relatively extended with the trunk near horizontal” (Hunt et al. 1996, p. 371)
“a posture in which the ischia bear substantial portion of the body weight and the torso is relatively orthograde” (Hunt et al. 1996, p. 373)
“a gap crossing mode involving active or passive compliance of initial and terminal supports” (Youlatos 1999b, p. 544)
“quadrupedal progression using the tegulae along large, vertical supports” (Youlatos 1999b, p. 544)
“vertically ascended a substrate with grasping hands” (Hunt et al. 1996, p. 378)
“fall after releasing a support” (Hunt et al. 1996, p. 381)
“a gap crossing mode involving an airborne phase” (Youlatos 1999b, p. 544)
“fast locomotion using asymmetrical or regular gaits and with a period of free flight” (Hunt et al. 1996, p. 377)
I ran chi-square tests to test the statistical significance of the results within this study. To conduct these analyses, I eliminated all unknowns from the data pool as well as any data sets that were lacking data points (i.e., quadrupedal walking on vertical substrates, grasping large substrates, etc.). I considered p values less than 0.05 as statistically significant. To test statistically for substrate preferences, I used Jacob’s D index (Jacobs 1974): D = u − a/u + a – 2 × u × a, where u is the percentage of substrates used and a is the percentage of substrates available (Jacobs 1974). A positive result indicates preference while a negative result indicates avoidance, and the distance away from zero indicates the degree of preference or avoidance.
Substrate profile of the study area
Substrate availability in Camp Callimico, Bolivia
Activity profile of the pygmy marmosets
Activity profile of Cebuella pygmaea in Bolivia
Feeding within exudate sites: claw-clinging
Positional behaviors by activity of Cebuella pygmaea in Bolivia
Foraging (gouging) within exudate sites: claw-clinging
Traveling between exudate sites: quadrupedalism
Traveling between exudate sites: leaping
Traveling between exudate sites: claw-climbing
This study expands the overall data pool regarding pygmy marmoset positional behavior, habitat use, and habitat preferences. Pygmy marmosets are feeding and foraging specialists that claw-climb within the understory in order to exploit exudates on large-diameter substrates of vertical orientation (Kinzey et al. 1975; Soini 1988; Garber 1992; Youlatos 1999a, 2009). The results from this study demonstrate that pygmy marmosets are adapted for feeding and foraging on large trunks, but these behavioral adaptations are not necessarily used for other activities—specifically while traveling between exudate sites.
Youlatos (2009) has shown that claw-clinging and claw-climbing are used most during feeding. I also found that claw-clinging was the principle postural mode of feeding for Camp Callimico pygmy marmosets, but there were differences in claw-climbing frequencies. Youlatos (2009) reported that claw-climbing accounted for 93.6% of all feeding activity. Claw-climbing in this study was not recorded during feeding due to the way in which foraging and feeding were defined. In order to make results more comparable to Youlatos (2009), I examined total counts of claw-climbing locomotion during foraging and then compared them with total counts of foraging locomotion to determine what percentage of claw-climbing locomotion was associated with exudativory and comparable to Youlatos’ (2009) claw-climbing figures while feeding. For Camp Callimico pygmy marmosets, claw-climbing accounted for 54% of exudate-associated locomotion including foraging, although the sample size was extremely small. These results indicate higher frequencies of claw-climbing within the Yasuni National Park group. This difference could be related to the differences in habituation between study groups. Claw-climbing on trunks within the understory is a vulnerable position, and the Camp Callimico study group may have preferred alternative routes to exudates sites that facilitated more concealed movements.
When examining habitat use during feeding between the two study sites, the pygmy marmosets within this study used trunks and branches, substrates of vertical orientation, and substrates of medium and large size. The pygmy marmosets in Yasuni National Park showed preferential use of trunks, substrates of large size, and substrates of vertical orientation. Both results agree on these points of habitat preference.
However, both studies do not agree on the use of lianas while gouging and feeding on exudates. Why did the Camp Callimico pygmy marmosets not feed and forage from lianas? This difference could be related to difference in habitat profiles. Youlatos (2009) reported twice as many lianas (80.6%) within his habitat profile. Pygmy marmosets in Yasuni National Park also spent significantly more time on lianas, as was reflected in their frequent foraging for insects on this substrate type (Youlatos 2009). With fewer lianas and few observations of insect feeding or foraging within this study sample, the differences between studies appear to be a result of habitat differences. Pygmy marmosets in Yasuni National Park preferred to engage in activities on lianas more so than Camp Callimico pygmy marmosets because lianas were more prevalent within their range.
Youlatos (2009) reported equivalent proportions of claw-climbing, leaping, and quadrupedal behaviors during travel between feeding and foraging sites. When subsuming Youlatos’ (2009) quadrupedal bounding and quadrupedal walking under the category quadrupedalism and then comparing traveling positional behaviors, his pygmy marmosets primarily used quadrupedalism (35.2%) during travel more than any other positional behavior. This study reports the same primary locomotor mode during travel but a higher frequency of occurrence (55.7%). Although it has been suggested that claw-climbing represents a highly specialized form of positional behavior that is correlated with vertical ranging within the understory (Garber 1992; Youlatos 2009), the results from this and Youlatos’ (2009) study indicate that pygmy marmosets are not obligated to use claw-climbing during travel, and that they can incorporate more quadrupedal locomotion into their locomotor repertoire.
Camp Callimico pygmy marmosets primarily used branches and substrates of small and medium size during travel. Pygmy marmosets in Yasuni National Park used lianas, substrates of large size, and substrates of vertical or oblique orientation (Youlatos 2009). Differences in the frequency with which branches were used during travel may result from differences in the abundance of branches at these two sites. This study reports that branches accounted for 28.8% of the habitat profile, whereas Youlatos (2009) reported that branches accounted for 12.2% of his habitat profile.
When comparing substrate availability counts between studies, small substrates accounted for 92.8%, whereas Youlatos (2009) reported that substrates smaller than 10 cm accounted for 89.3% of his total sample. This is well associated with its high availability within both environments, suggesting that quadrupedalism for both groups is a habitat-dependent locomotor mode.
The results from this study diverge from Youlatos (2009) in that bamboo was highly preferred. Can this difference be reconciled with each site’s habitat profile? This study reports a higher frequency of bamboo at the study site than is documented at the Youlatos study site, and the D value for traveling while on bamboo substrates indicates a preference. The bamboo substrates observed within this study offered a long direct passage, ideal for the displacement of horizontal space. Bamboo may be preferred over lianas, which accounted for 80.6% of all substrate types within Youlatos’ (2009) substrate availability profile, for its ease in horizontal navigation. Considering this, it is very likely that this difference between study sites results from habitat differences and potential advantages in travel.
When examining leaping behavior between studies, Youlatos (2009) reported that leaping accounted for 24.5% of all recorded behaviors. This study reports that leaping only occurred while foraging and traveling and accounted for 12.4% of this study sample. This difference may be related to the fact that the Camp Callimico group was not as well habituated as the group in Yasuni National Park. The Camp Callimico sample reflects the unhabituated state of the animals rather than actual frequencies of leaping. These animals might prefer concealment as leaping would draw attention towards the animals.
When comparing claw-climbing behaviors between studies, Youlatos (2009) reported that claw-climbing locomotion accounted for 23.7% of travel locomotion. This study reports that claw-climbing occurred for 14.8% of all travel locomotion. While the overall frequency of claw-climbing locomotion is lower than in Youlatos’ (2009) sample, this can be related to the extremely small sample of claw-climbing locomotion within this study and the unhabituated state of the study group. This study only reports twenty-two total counts of claw-climbing travel locomotion. While the sample size was small, pygmy marmosets within this study did show preferential substrate use while claw-climbing.
Camp Callimico pygmy marmosets primarily used trunks, substrates of vertical orientation, and substrates from medium to large size during claw-climbing travel. Pygmy marmosets in Yasuni National Park primarily used lianas, substrates of large size, and substrates of vertical or oblique orientation during travel (Youlatos 2009). When comparing the results between studies, there are great similarities between substrate size and orientation use. The literature on callitrichid positional behavior shows that clawed positional behaviors are highly associated with body size reduction and exploitation of the large, vertical branch niche within the understory (Ford 1986; Garber 1992; Garber and Leigh 2001; Porter and Garber 2004; Youlatos 2009). In addition, a review of evidence collected from Yasuni National Park on the positional behavior of Sciurus igniventris and Microsciurus flaviventer further substantiates the relationship between clawed locomotion, understory exploitation, and body size reduction (Youlatos 1999b). The evidence from this study also shows that clawed locomotion on vertical substrates in pygmy marmosets is highly associated with an evolutionary reduction in body size, as pygmy marmosets engaged in claw-climbing in order to navigate vertical distances required to reach exudate sites.
Pygmy marmoset use branches while traveling and feeding and bamboo while traveling more often than was previously reported.
The feeding and foraging behavior of the pygmy marmosets within this study supports proposals that they are morphologically and ecologically specialized for exudate exploitation.
Quadrupedalism is a habitat-dependent locomotor mode, whereas claw-climbing during travel is a solution to overcome the constraints of small body size during vertical travel between exudate sites.
This project was financially supported by Dr. Leila Porter’s National Geographic Research and Exploration Grant. I would like to express the deepest gratitude to Dr. Leila Porter, as this project would not have been possible without her devoted maintenance of the Camp Callimico field site. I would also like to thank all the reviewers for their work on all previous and current versions of the paper.