Leaping and differential habitat use in sympatric tamarins in Amazonian Peru

Abstract

Differential habitat use in sympatric species can provide insight into how behavior relates to morphological differences and as a general model for the study of biological adaptations to different functional demands. In Amazonia, closely related sympatric tamarins of the genera Saguinus and Leontocebus regularly form stable mixed-species groups, but exhibit differences in foraging height and locomotor activity. To test the hypothesis that two closely related species in a mixed-species group prefer different modes of leaping regardless of the substrates available, we quantified leaping behavior in a mixed-species group of Saguinus mystax and Leontocebus nigrifrons. We studied leaping behavior in relation to support substrate type and foraging height in the field for 5 months in the Amazonian forest of north-eastern Peru. Saguinus mystax spent significantly more time above 15 m (79%) and used predominantly horizontal and narrow supports for leaping. Leontocebus nigrifrons was predominantly active below 10 m (87%) and exhibited relatively more trunk-to-trunk leaping. Both species preferred their predominant leaping modes regardless of support type availability in the different forest layers. This indicates that the supports most commonly available in each forest layer do not determine the tamarins’ leaping behavior. This apparent behavioral adaptation provides a baseline for further investigation into how behavioral differences are reflected in the morphology and species-specific biomechanics of leaping behavior and establishes callitrichid primates as a model well-suited to the general study of biological adaptation.

Full text

1
Journal of Mammalogy, XX(X):1–13, 2021
https://doi.org/10.1093/jmammal/gyab121
© The Author(s) 2021. Published by Oxford University Press on behalf of the American Society of Mammalogists, www.mammalogy.org.
Leaping and differential habitat use in sympatric tamarins in
AmazonianPeru
PB,,* EW.H, FG,  JA.N
1AG Morphologie und Formengeschichte, Institut für Biologie, Humboldt-Universität zu Berlin, Philippstraße 12, D-10115
Berlin, Germany
2Verhaltensökologie & Soziobiologie, Deutsches Primatenzentrum – Leibniz-Institut für Primatenforschung, Kellnerweg 4,
D-37077 Göttingen, Germany
3Institut für Deutsche Sprache und Linguistik, Humboldt-Universität zu Berlin, Unter den Linden 6, D-10117 Berlin, Germany
*To whom correspondence should be addressed: patricia.berles@hu-berlin.de
Differential habitat use in sympatric species can provide insight into how behavior relates to morphological
differences and as a general model for the study of biological adaptations to different functional demands. In
Amazonia, closely related sympatric tamarins of the genera Saguinus and Leontocebus regularly form stable
mixed-species groups, but exhibit differences in foraging height and locomotor activity. To test the hypothesis
that two closely related species in a mixed-species group prefer different modes of leaping regardless of the
substrates available, we quantied leaping behavior in a mixed-species group of Saguinus mystax and Leontocebus
nigrifrons. We studied leaping behavior in relation to support substrate type and foraging height in the eld
for 5 months in the Amazonian forest of north-eastern Peru. Saguinus mystax spent signicantly more time
above 15 m (79%) and used predominantly horizontal and narrow supports for leaping. Leontocebus nigrifrons
was predominantly active below 10 m (87%) and exhibited relatively more trunk-to-trunk leaping. Both species
preferred their predominant leaping modes regardless of support type availability in the different forest layers.
This indicates that the supports most commonly available in each forest layer do not determine the tamarins’
leaping behavior. This apparent behavioral adaptation provides a baseline for further investigation into how
behavioral differences are reected in the morphology and species-specic biomechanics of leaping behavior
and establishes callitrichid primates as a model well-suited to the general study of biological adaptation.
Key words: Callitrichidae, habitat use, leaping behavior, Leontocebus, Saguinus
El uso diferencial del hábitat por parte de especies simpátricas puede constituir un modelo adecuado en estudios
de adaptaciones biológicas a diferentes demandas funcionales que proporcione información sobre cómo el
comportamiento se relaciona con las diferencias morfológicas. En la Amazonía, tamarinos de especies simpátridas
y estrechamente relacionados de los géneros Saguinus y Leontocebus forman regularmente grupos estables y
mixtos en su composición especíca. Los miembros de las distintas especies exhiben diferencias en la altura de
alimentación y la actividad locomotora. Las estructuras vegetales usadas como soporte dieren en sus propiedades
según las capas del bosque. Para poner a prueba la hipótesis de que dos especies estrechamente relacionadas en
un grupo mixto usan diferentes modos de salto, independientemente de los sustratos disponibles, se cuanticó
el comportamiento de salto en un grupo mixto de las especies Saguinus mystax y Leontocebus nigrifrons. Se
estudió el comportamiento de salto en relación con el tipo de sustrato de soporte y la altura de alimentación en
campo durante 5 meses en la selva amazónica del noreste de Perú. Saguinus mystax pasó signicativamente más
tiempo por encima de los 15 m (79%) y utilizó soportes predominantemente horizontales y estrechos para saltar.
Leontocebus nigrifrons fue predominantemente activo por debajo de los 10 m (87%) y exhibió relativamente
más saltos de tronco a tronco. Ambas especies prerieron sus modos de salto predominantes independientemente
de la disponibilidad del tipo de soporte en las diferentes capas del bosque. Esto indica que los soportes más
comúnmente disponibles en cada capa del bosque no determinan el comportamiento de salto de los tamarinos.
Esta aparente adaptación conductual observada en este grupo de estudio proporciona una línea de base para
investigaciones adicionales sobre cómo las diferencias conductuales se reejan en la morfología y la biomecánica
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

2 JOURNAL OF MAMMALOGY
especie especíca del comportamiento de salto y establece a los primates callitrícidos como un modelo adecuado
para el estudio general de la adaptación biológica.
Palabras clave: Callitrichidae, comportamiento de salto, Leontocebus, Saguinus, uso del hábitat
Habitat characteristics such as spatial location and food supply,
predator refugia, and locations for nocturnal shelter all play
important roles in the survival of individuals (Ganzhorn etal.
2011). Behavioral studies of sympatric and syntopic species in
natural habitats may offer insights into functional and adaptive
signicance of behavioral, morphological, and ecological dif-
ferences, while minimizing the effects of confounding vari-
ables such as different habitat structures.
Tropical rainforests typically consist of diverse microhabi-
tats. While some authors argue against the use of the term
‘forest stratum’ (Dial et al. 2004) because it implies planar,
uniform layering, most authors agree on a vertical organization
of rainforests worldwide (Mortensen 1932; Klinge 1973; King
1998; King and Maindonald 1999; Dial etal. 2004; Iida et al.
2011). At least three layers are usually distinguished (Dial etal.
2004). In our study, it is sufcient to summarize the salient dif-
ferences between a general lower layer and the canopy layer.
In the lower layers, vegetation in tropical rainforests generally
is marked by multi-trunked, small (< 1m) bushes, ferns, palms,
and dominant woody, high-stemmed vertical tree trunks whose
branches branch out only near the canopy (Mortensen 1932;
Richards 1952; Leigh 1975; Corlett and Primack 2011). In con-
trast, the canopy layer is characterized by dense foliage and nu-
merous and often horizontal branches. Woody climbing plants
such as lianas also are abundant and often form tangles in the
canopy (Garber and Pruetz 1995; DeWalt and Chave 2004).
Different substrate types (e.g., various branching patterns, bark,
and branch orientation) occur with differing abundances in the
various forestlayers.
The small-bodied tamarins and marmosets (Primates,
Callitrichidae) present an opportunity for the comparative anal-
ysis of niche partitioning, habitat utilization, and the evolution
of morphological correlates for the different functional de-
mands posed by the various microhabitats in tropical rainforests.
Aconsiderable amount of work on the ecology and behavior of
tamarins has been published (Garber 1991; Buchanan-Smith
1999; Buchanan-Smith et al. 2000), but comparison of the
results is hampered by diverse acquired parameters and dif-
ferent research interests in these studies (see Supplementary
Data SD1 for an overview of published studies). In addition
to the frequent occurrence of tamarin species sympatrically,
some tamarins also use the same areas through the formation
of “mixed-species groups” and therefore are syntopic. These
interspecic associations allow us to study closely related spe-
cies under identical ecological conditions (Rylands etal. 2016;
Heymann 2018).
Despite an overlap in resource exploitation, sympatric cal-
litrichids often show different spatial preferences, the use of
different sizes of locomotor substrates (Norconk 1990; Smith
1997; Buchanan-Smith etal. 2000), and morphological differ-
ences (Garber 1991; Bicca-Marques 1999; Dunham etal. 2020).
The differential use of forest layers (indicated by the height
from the ground) by the different species thus appears to be
an important part of niche separation in mixed-species groups
of callitrichids, allowing different species to travel and forage
together (Garber 1988a; Buchanan-Smith 1991; Heymann and
Buchanan-Smith 2000; Porter 2004).
As in tamarins and marmosets, sympatric and syntopic mam-
mals generally display different preferences in microhabitat
use (Mayr 1947; Heymann and Buchanan-Smith 2000; Cunha
and Vieira 2002; Mauffrey and Catzeis 2003; Porter 2004;
Dammhahn, etal. 2013). Habitats used by tamarins exhibit a
wide range of substrate connectivity and spacing (Madden etal.
2010). To cross gaps between supports in the lower rainforest
layer without moving down to the ground, tamarin species use
trunk-to-trunk leaps from a vertical clinging position (Garber
1991; Youlatos 1999; Buchanan-Smith et al. 2000). This be-
havior is comparable with the vertical clinging and leaping of
strepsirrhine primates (Demes et al. 1995; Hunt et al. 1996;
Delciellos and Vieira 2009, Hemingway etal. 2020). In con-
trast, in the upper forest layer where vertical supports are rare,
tamarins usually cross gaps with quadrupedal leaps between
terminal branches (Garber and Leigh 2001; Nyakatura and
Heymann 2010). While trunk-to-trunk leaps (Fig. 1a) start and
end in a completely crouched and static upright vertical posture,
horizontal leaps (Fig. 1b) usually start and end in a pronograde,
quadrupedal posture that continues into a quadrupedal locomo-
tion sequence. Moreover, locomotion on slender and exible
terminal branches in the canopy layer is linked to a notable
requirement for balancing and grasping (Shapiro and Young
2010; Hesse etal. 2015; Youlatos etal. 2015; Nyakatura 2019),
whereas vertical clinging to large diameter tree trunks (often
much larger than the diameter of the callitrichids torso) poses
specic demands for gripping onto bark with claws (Toussaint
et al. 2020). Studying differences in habitat use by syntopic
callitrichids therefore may identify behavioral specializations
that also are reected in the species’ functional morphology
(Garber 1991; Dagosto and Yamashita 1998; Demes etal. 1998;
Biewener 2002; James etal. 2007; Garber and Porter 2009).
We quantify habitat use and assess how it relates to leaping
in two closely related and syntopic species of Callitrichidae:
moustached tamarins (Saguinus mystax) and saddleback tam-
arins (Leontocebus nigrifrons, previously named Saguinus
fuscicollis nigrifrons—Rylands etal. 2016; Fig. 2). Saguinus
mystax feed mainly on fruit and search for these while
moving on horizontal branches in the canopy layer (Garber
1988a, 1988b, 1998). Leontocebus nigrifrons prefers exud-
ates and insects and use trunks as primary feeding platforms
(Garber 1988a, 1988b, 1998). Studies by Norconk (1990),
Smith (1997), and Nyakatura and Heymann (2010), focused
on the effects of support size and orientation in S. mystax
and L. nigrifrons and briey described the leaping behavior
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 3
0 ms 0.12 ms 0.22 ms
a: trunk-to-trunk leaping
0.42 ms 0.52 ms
0.32 ms
0.62 ms
0.72 ms 0.92 ms
0 ms 0.10 ms
b: horizontal leaping on thin and flexible branche
s
0.26 ms
0.50 ms
0.60 ms
0.30 ms
0.20 ms
Fig.1.—Comparative illustration of the two modes of leaping. a) Vertical leap between inexible trunks of L.nigrifrons over a distance of 1.5
m.b) Horizontal leap between terminal exible branches of S.mystax over a distance of 1 m.
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

4 JOURNAL OF MAMMALOGY
of these tamarins. Saguinus mystax was found more often in
upper forest layers and predominantly used horizontal supports
during above-branch locomotion, while L. nigrifrons spent
most of the time in the lower forest layers clinging vertically
to tree trunks and using trunk-to-trunk leaps (Norconk 1990;
Smith 1997). However, it remains unclear whether the leaping
behavior merely was a function of the available support types in
the preferred forest layer or whether the different tamarin spe-
cies display a general preference for one of the leaping modes
independent of the available supporttypes.
If the two sympatric species of tamarins are adapted behav-
iorally to different components of the habitat, we expect each to
species prefer a different leaping mode. Thus, we hypothesize
that S. mystax leaps horizontally more often and prefers hori-
zontal supports for launching and landing, regardless of the rel-
ative availability of other substrates and the height at which the
activity occurs. In contrast, the same forest layer, we hypoth-
esize that L.nigrifrons primarily will use vertical supports for
take-off and landing. To test these hypotheses, we character-
ized habitat conditions within the home range of a free-ranging
mixed-species group containing both species at a eld station
in the Peruvian Amazon. We observed leaping behavior and the
rate at which this behavior occurred relative to forest layer and
support properties (i.e., the properties of the vegetative struc-
tures that were used by the monkeys). Finally, we analyzed
preference for leaping mode in both species when navigating
within the same forest layer. This information provides insight
for studies of behavioral and morphological responses to selec-
tion in sympatric species with different patterns of habitatuse.
M  M
Study site and study group
Field work was carried out in the Amazonian lowland in north-
eastern Peru, at the Estación Biológica Quebrada Blanco
(EBQB). The station is located at 4°21′ S and 73°09′ W, north
of Quebrada Blanco, a small tributary of the Rio Tahuayo,
which empties into the Amazon River (Fig. 3a and b) The study
area consists of approximately 115 ha. The data were collected
during the dry season from June to October 2017. While mul-
tiple seasons of data collection likely would enable a more
complete assessment of habitat use and foraging patterns, our
efforts specically focused on leaping behavior and were con-
ned to a single season during which the animals are observed
and followed most easily. The habitat at EBQB is predomi-
nantly primary forest with a dense canopy at a height of 25–30
m and some giant trees (Encarnación 1985). The predominant
plant families in the study region have been assessed for the
nearby Area de Conservación Regional Comunal Tamshiyacu-
Tahuayo and consist of 11 species of Fabaceae (including
Erythrina fusca and Campsiandra angustifolia), eight species
of Moraceae (including Ficus insipida and Artocarpus altilis),
six species of Solanaceae (including Solanum sessiliorum
and Brugmansia suaveolens), and Palmae (Siegel etal. 1989;
Acostupa etal. 2013).
Observed primates consisted of a mixed-species group con-
taining six adult (three ♂ and three ♀) and two juvenile (one
♂ and one ♀) S.mystax, and three adult (two ♂ and one ♀)
L.nigrifrons. We did not consider sex as a variate during data
collection or analysis. Other mixed-species groups of tam-
arins with bordering home ranges were present occasionally.
Tamarins at EBQB are habituated to the presence of humans.
Our study involved no aspects subject to regulatory or ethical
approval. Field work was permitted by the Servicio Nacional
Forestal y de Fauna Silvestre (SERFOR) of the Peruvian
Ministry of Agriculture (permission no.: AUT-IFS-2017–062).
Habitat characteristics
We assessed habitat characteristics and obtained baseline data
on tamarin home ranges and movements by following the study
group over 42days in June and July, 2017. Waypoints were
recorded with a Garmin eTrex 30x (Garmin International Inc.,
Olathe, Kansas). We established seven modied Whittaker
plots within the study area to provide data on forest struc-
ture and the substrates available to the tamarins (Dagosto
and Yamashita 1998). The Whittaker plots covered an area of
843 m2 and contained 362 trees, or approximately 43 trees
per 100 m2 (Table 1). Each Whittaker plot was subdivided
into 12 smaller rectangular nested subplots of different sizes
(9.6 ± 6.3 m2) This method was used by the Smithsonian’s
Program for Monitoring and Assessment of Biodiversity in
Peru (Stohlgren and Chong 1997; cited in Ganzhorn et al.
2011) and is most efcient for all types of density-related
characteristics. However, different sized plots may generate
different density estimates even when the same area is cov-
ered (Ganzhorn etal. 2011). In each of these 12 subplots, we
determined tree density, tree diameter at breast height (dbh),
and tree height (Chavel etal. 2017). Following Garber (1991),
we dened a tree as a woody plant with a dbh > 5cm. Tree
density was determined by the number of trees relative to the
area of the plot, and the trees were classied into categories
(dbh: 5–10, 10–20, and >20 cm; height: ≤ 5, 5–10, 10–15,
15–20, and >20 m). DBH was taken with a measuring tape,
and tree height was estimated visually.
Visual estimates of support properties are crucial to our
study. Bezanson etal. (2012) noted the importance of observer
training on the accuracy of visual estimates and the importance
Fig.2.—Photographs of a) S.mystax and b) L.nigrifrons at the EBQB.
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 5
of a single observer making the estimates to eliminate inter-
observer variance. Accordingly, prior to our eld work, the rst
author trained in the visual estimation of support inclination
and tree height in Steigerwald, a deciduous forest with full foli-
ation in central Germany, and subsequently undertook all visual
estimations in Amazonia.
Fig.3.—Location of the study site. a) Map of Peru. b) Location of the study site (EBQB in Amazonian Peru). c) Daily routes (represented by each
colored line; N=42 within the home range of the mixed-species group of S.mystax and L.nigrifrons). d) Selected areas within the home range
where habitat properties were documented (yellow squares). The straight lines indicate the trail system and the red triangle depicts the location
of the eld station. Grid size of the trail system is 100× 100 m.The black contours represent the percent set 10%, 20%, 50%, 70%, 90%, 95%,
and 99% of the home range area point estimate calculated with the kernel method. The blue crosses represent the individual observation points
recorded with GPS.e) Three examples of habitat properties.
Table 1.—Characteristics of trees within the seven Whittaker plots with the total number (N) and the percent abundance of trees within each
size category for each vertical stratum during focal animal sampling at the Estación Biológica Quebrada Blanco (EBQB).
Tree height Tree diameter (dbh) Vertical height of monkeys (category)
5 ≤ 10cm > 10≤ 20cm > 20cm
≤ 5 m 0.60 (N=12) 0.3 (N=6) 0.1 (N=2) Very low layer
> 5 ≤ 10 m 0.88 (N=49) 0.13 (N=7) N/A (N=0) Low layer
> 10≤ 15 m 0.59 (N=20) 0.18 (N=6) 0.24 (N=8) Intermediate layer
> 15≤ 20 m 0.68 (N=74) 0.21 (N=23) 0.11 (N=12) Upper layer
> 20 m 0.13 (N=19) 0.29 (N=41) 0.58 (N=83) Top layer
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

6 JOURNAL OF MAMMALOGY
Although we do not quantify all available supports, our
mixed-species group travelled and foraged together and thus
use the same trees. The same supports were available to all
members of the mixed-species group and thus enable a direct
comparison of support typeuse.
The recorded waypoints were used to determine home range
area with a 95% isopleth contour by using the kernel density es-
timation method (KDE) with the default version of the standard
package in R (adehabitatHR, version: 0.4.18; Calenge 2006).
In the kernel method, density lines are drawn around areas
with different intensities of utilization (Seaman et al. 1999;
Nascimento etal. 2011). At times we lost sight of the mixed-
species group or were unable to follow them. Days in which
less than 90% of the daily activity was recorded therefore were
excluded from further analysis of the average daily distance
travelled (Choi etal. 2007).
Quantitative behavioral description
The tamarins were observed from the time they left their
sleeping site (usually between 0545 and 0615 h) until they
returned to a sleeping site in the afternoon (usually between
1600 and 1715h). We obtained a total of 586h of observation
for S.mystax, and 519h for L. nigrifrons (mostly parallel to
S.mystax). To obtain baseline data on the locomotor activity of
the tamarins, we used instantaneous scan sampling. Choi etal.
(2007) suggested sampling intervals spaced less than 10min
apart for accurate behavioral observations, and thus we used
8-min intervals between records of activity. Behaviors included
locomotor (running, leaping) or stationary behaviors (e.g., sit-
ting, clinging) of all group members that were visible within
1min (Table 2).
Locomotor behavior was documented and quantied in
greater detail using additional focal animal sampling. Every
30min, the locomotor activity and postural behavior of the rst
individual that came into sight were recorded continuously for
as long as the animal remained visible. We also estimated the
vertical distance of the animal above the ground; a measure
traditionally used to describe arboreal position in trees (Garber
and Sussman 1984, Porter 2004) for each behavior (Table 2).
We collected data on adult individualsonly.
For each recorded leap, we estimated inclination of the sup-
port. The angles were estimated in 10° steps and later binned
into categories. We dened all leaps from substrates with an
inclination of 80° or more as vertical leaps. We dened all leaps
from substrates with an orientation of 0° to 30° as horizontal
leaps. In order to emphasize the contrast of horizontal and ver-
tical leaping, we excluded leaps from substrates with inclin-
ations between 30° and 80° from the analysis.
Data analysis
Data collected during June were used for training purposes and
were not analyzed statistically. Because our sample included
only one observer-habituated mixed-species group (S.mystax
and L. nigrifrons) at the EBQB eld site, all measurements
were derived from the same habitat with the same available
substrate properties. Also, because we collected our data at
short, regular (8 min) intervals, they may not be statistically
independent (Seitz 2001). The activity undertaken during one
interval may not be independent of the activity noted in the next
interval (Sussman etal. 1979). However, by using daily aver-
ages, we minimize temporal autocorrelations and rely on the
central limit theorem to assure the averaged data are normally
distributed. The data set that excluded the June data showed no
indication of temporal autocorrelation.
The positional height of the primate was documented inde-
pendently of habitat properties to allow a more ne-grained
analysis (see above). To assess daily vertical position of each
species, we determined the proportion of time during which
each species stayed in the upper (> 15 m) or lower (< 10 m)
forest layer. This enabled us to assess (without temporal au-
tocorrelation) any vertical stratication within a day. No auto-
correlation was expected between consecutive days. Because
the height data were not normally distributed, we evaluated
differences between species using a Wilcoxon test with paired
samples.
For each day and both species, we computed the mean of
the natural log of the ratio of observed locomotor behavior (in
seconds) relative to the log observed stationary behavior (in
seconds). Observations without locomotor behavior were not
used in the analysis because they yielded an innite logarithm.
Table 2.—Behavior categories, with denitions per Hunt etal. (1996), and number of observations for each behavior by both species recorded
using instantaneous scan sampling at Estación Biológica Quebrada Blanco (EBQB), and duration in seconds of each behavior for both species
observed using focal animal sampling, excluding leaps because they were too brief to measure.
Behavioral category Number of observations
(instantaneous scan sampling)
Observation time in sec
(focal animal scan)
Active/passive
S.mystax L.nigrifrons S.mystax L.nigrifrons
Quadrupedal locomotion 985 (17%) 8 (0.4%) 31726 (9%) 6845 (3%) Active
leap vertical (from and to vertical substrate) 137 (2%) 787 (35%) n.a. n.a. Active
leap horizontal (from and to horizontal substrate) 2067 (35%) 136 (6%) n.a. n.a. Active
leap from horizontal to vertical substrate 291 (5%) 145 (6%) n.a. n.a. Active
leap from vertical to horizontal substrate 356 (6%) 262 (12%) n.a. n.a. Active
Vertical climbing 56 (1%) 81 (4%) 19501 (6%) 62270 (29%) Active
Cling 162 (3%) 358 (16%) 28776 (8%) 28531 (13%) Passive
Sit 750 (13%) 96 (4%) 192883 (57%) 93807 (44%) Passive
Stand 753 (13%) 196 (9%) 44275 (13%) 8337 (4%) Passive
Lie 345 (6%) 179 (8%) 20489 (6%) 13616 (6%) Passive
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 7
The remaining data were approximately normally distrib-
uted and therefore analyzed using a 2-way repeated measures
ANOVA, with factors for species membership and height range
class. Results of a Levene test (P=0.33) showed the data to be
homoscedastic.
We investigated leaping activity using a generalized mixed
effects model tted with a Poisson link and implemented using
the R package ‘lme4’ (Bates etal. 2021). The number of leaps
per observation interval was the response variable and interval
length was entered into the model as the offset. Leaps s−1 was
included in the analysis because it is reasonably assumed that
the observed number of leaps is proportional to the observa-
tion time; leaps s−1 therefore is a valid indicator of leap activity.
Fixed effects for the model included species identity and ver-
tical height. Because we did not expect activity to be equal
across days, we included day of observation as a random effect.
Also, because we did not assign each leap to a specic indi-
vidual, and because individuals could exhibit sequential leaps,
we used a unique identication number for each observation
and included this in the model as a second random effect. We
expected over-dispersion of the data because of the lack of in-
dividual assignation and this was the case, with a dispersion pa-
rameter of 2.43 indicating considerable over-dispersion. Use of
the observation ID as a random effect alleviates this problem.
R
Habitat characteristics
The home range of the mixed-species group covered 36 ha
(Kernel 95%; Fig. 3c and d). Average daily distance travelled
for both species was 1.4 km ± 0.2 km (mean ± SD) and ranged
from approximately 1 to 1.7km.
The tamarins’ home range was heterogeneous (Fig. 3e), con-
taining areas characterized by a dense lower forest layer, with
many small and thin trees and a small number of trees taller
than 5 m while other areas within the home range consisted
almost exclusively of tall trees with large trunks and a canopy
taller than 15 m.Also, there were areas within the home range
with dense understory vegetation, but with large trees present.
These differences in habitat structure are expected to inuence
vertical positions of the tamarins, but because both species
groups travelled and foraged together, we expected both spe-
cies to use the same forest types for equal amounts of time.
Vertical position and type of locomotor activity
Saguinus mystax was found primarily in the upper and
crown forest layers (> 15 m; 79% of all observations), while
L. nigrifrons was found in forest layers below 10 m (87%;
Fig. 4). The difference between species in time spent in dif-
ferent forest layers was signicant (Wilcoxon rank sum test:
W=79, P <0.001).
There was no effect on locomotor activity either of vertical
position (ANOVA: F1,12=0.08, P > 0.05) or species identity
(ANOVA: F1,12 = 2.1, P > 0.05). Thus, overall level of ac-
tivity was independent of forest layer or species. However,
the interaction between species identity and vertical position
was signicant: both species exhibited greater locomotor ac-
tivity when outside of their preferred forest layer (ANOVA:
F1,12=18.0, P<0.001). Saguinus mystax showed greater loco-
motor activity in the lower forest layer, while L.nigrifrons was
more active in the upper forest layer (Fig. 5).
Both species frequently used leaping and with similar fre-
quencies (Table 2). Tamarins performed about 0.013 leaps per
second (46.8 leaps per hour) for all combinations of species and
vertical positions (Fig. 6), although S.mystax leapt consider-
ably less above 15 m.The tted generalized mixed model with
a Poisson link showed a signicant interaction between species
and vertical position (P=0.03). In layers below 10 m, there was
no difference in leaping behavior between species (P=0.495;
Fig. 6). However, leaping activity differed signicantly above
fraction of day spent > 15m
0.00
0.25
0.50
0.75
1.00
species
L. nigrifrons S. mystax
Fig.4.—Point diagram of proportion of time spent above 15 m height
for both species (S. mystax and L. nigrifrons) studied at the EBQB
(each point represents a daily average).
Fig.5.—Boxplots of the logarithmic ratio of locomotor to stationary
time for each species (S.mystax and L. nigrifrons) in lower (<10 m)
and upper (>15 m) forest layers at the EBQB. The outlier for S.mystax
in the lower forest stratum was omitted from analyses following the
convention of removing observation that deviate by more than 2.5
standard deviations from the mean.
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

8 JOURNAL OF MAMMALOGY
15 m, with L.nigrifrons leaping at a higher rate than S.mystax
(P<0.001; Fig. 6).
Analysis of launch and landing positions (Fig. 7) provides
insight into differential habitat use. Specically, S.mystax leapt
more frequently from horizontal-to-horizontal supports than
L.nigrifrons. This pattern was independent of vertical position
and thus independent of the relative abundance of support types
(Fig. 7a). In contrast, L.nigrifrons leapt more frequently be-
tween vertical supports than S.mystax with a minor variation
related to vertical position (Fig. 7d). Leaping from horizontal to
vertical supports was slightly more frequent than from vertical
to horizontal supports (Fig. 7b and c). The generalized mixed
model with a Poisson link for leaps between horizontal sup-
ports contained only species identity as a signicant main effect
(P< 0.001). Neither vertical position nor the interaction term
was signicant. Leaps using purely vertical supports (vertical
clinging and leaping) showed a similar pattern, and both spe-
cies identity and vertical position were signicant, as was the
interaction term (all P<0.001). Signicance of the interaction
term reects the fact that L. nigrifrons leapt more frequently
using vertical supports, and that both species leapt more fre-
quently in the lower forest layer. For mixed leaps (vertical to
horizontal or vice versa) all main effects (species identity and
vertical position) were signicant (P <0.001).
D
Home range and daily distance travelled
Our estimate of the home range area used by this mixed spe-
cies group of tamarins is consistent with the ndings of other
studies. The mixed-species group occupied a shared home
range of 36 ha during our ve month observation period in
the dry season, while home range areas between 31.5 and 45
ha have been reported for mixed-species groups of S.mystax
and L. nigrifrons at EBQB during both dry and rainy seasons
(Ramírez 1989; Smith 1997; Heymann 2000). In two studies
of mixed-species groups of S.mystax and L.nigrifrons, Garber
(1988a) and Norconk (1986) identied home ranges of 40 ha
(rainy season, and overlap with 7 other groups) and 29–32 ha
(rainy and dry season), respectively. We did not assess home
ranges of overlappinggroups.
The average daily distance travelled (1.4± 0.2 km; mean ±
SD) also is consistent with previous ndings of 1.25–1.7 km for
mixed-species groups comprising of S.mystax and L.nigrifrons
in and around the EBQB (Norconk 1986; Smith 1997). The
fact that our home range and distance travelled estimates agree
closely with values from other studies suggests that the loco-
motor behaviors we observed are typical for the species.
Habitat characteristics and use
Most tamarin species are widespread throughout the Amazon
basin and are considered habitat generalists (Rylands and
Mittermeier 2013). Garber (1991) studied S. mystax and
L.nigrifrons at a site ca. 1 km from EBQB and found that tam-
arins used a variety of habitat types within their home range
and that about 20% of the trees were taller than 15 m and 3.4%
of the trees were taller than 35 m.Tree density was 0.38 trees/
m2, similar to the tree density of 0.43 trees/m2 on our site. The
presence of a well-formed canopy in both studies affords the
monkeys the opportunity to travel continuously on horizontal
supports in the upper forest layers (Garber 1991).
Sympatric and syntopic primate species prefer to occupy dif-
ferent heights in the forest (Sussman and Kinzey 1984; Cant
1992). This vertical stratication may represent niche differen-
tiation (Schreier etal. 2009). Generally, among sympatric pri-
mates the larger species spend more time in upper forest levels
while smaller species spend more time in lower forest levels
(Youlatos 1999). Previous studies have shown S.mystax pre-
ferred upper forest layers, while L.nigrifrons was found signif-
icantly more frequently in lower forest layers (Norconk 1990;
Fig. 7.—Boxplots of leaps per second for each studied species
(S.mystax and L. nigrifrons) at the EBQB separated by support ori-
entation of launch and landing sites and height of stay. Grey circles
show individual data records. Circles on the bottom of each panel are
the data that do not contain leaps. a) Horizontal to horizontal leaps. b)
Vertical to horizontal ‘mixed’ leaps. c) Horizontal to vertical ‘mixed’
leaps. d) Vertical to vertical leaps.
Fig. 6.—Boxplots of leaps per second for each studied species
(S.mystax and L. nigrifrons) at the EBQB in lower and upper forest
layers. Grey circles show daily averages. Circle size represents the
total observation time (in min).
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 9
Nadjafzadeh and Heymann 2008; Nyakatura and Heymann
2010). Norconk (1990) showed that S.mystax generally used
higher vertical layers (> 8 m rather than the > 15 m threshold
used in our study and in Smith 1997) of the forest compared to
L.nigrifrons (< 8 m). Peres (1993a) found an average height
of vertical position for L.nigrifrons to be 11± 9.5 m (mean
± SD) and 14.1± 8.6 m (mean ± SD) for S.mystax. In a study
at EBQB, the mean preferred vertical height of activity for
L. nigrifrons (8.9± 6.8 m) was lower than that of S. mystax
(13.9 ± 5.6 m) during various activities (Smith 1997). These
apparent differences in height preference may represent niche
separation within the mixed-species groups, allowing syntopy
and shared territories, which could provide protection from
predators (Garber 1988a; Buchanan-Smith 1991; Peres 1993b;
Porter 2004) through a high degree of coordination in detecting
and defending against predators at different heights in the forest
(Peres 1993b; Buchanan-Smith 1999). While L.nigrifrons ex-
hibits greater vigilance against potential terrestrial predators,
S. mystax exhibits greater vigilance against aerial predation
threats (Peres 1993b).
Tree squirrels are similar in size to callitrichids but do not
possess some morphological attributes associated with lo-
comotion in primates, including grasping autopodia and at-
tened nails. They have been used as a comparative model for
primates (e.g., Garber and Sussman 1984; Orkin and Pontzer
2011; Nyakatura 2019). Sciurus granatensis is sympatric with
Saguinus geoffroyi in Panama and spends twice as much time in
the lower forest layer and prefers trunk-to-trunk leaps (Garber
and Sussman 1984), similar to L.nigrifrons. Saguinus geoffroyi
in Panama restricted much of its locomotor activity (69%) to
the upper forest layers of the canopy (Garber and Sussman
1984), similar to S. mystax at the EBQB. Garber and Leigh
(2001) observed a similar pattern in the leaping behavior of
a mixed-species group of Leontocebus weddelli and Saguinus
labiatus. Fleagle (1977) found signicant quantitative differ-
ences in locomotor pattern use consistent with different habitat
use strategies within the same forest of a sympatric group of
Presbytis obscura and Presbytis melalophos. Presbytis obscura
usually moves by quadrupedal walking and running along large
branches in the canopy, while P. melalophos more often are
found in the discontinuous understory of the forest and are
more likely to move by leaping. These examples of sympatric
species exhibit notable signicant differences in their overall
pattern of vertical spatial preference, substrate preference, and
locomotor behavior (Garber and Sussman 1984), similar to our
detailed observations of leaping behavior in a mixed-species
group of tamarins. This suggests that our observations may be
reective of a general pattern of behaviors in mixed species
groups.
Leaping behavior
Leaping was the second most frequent locomotor activity (31–
41%) in the tamarin species studied by Garber (1991). Saguinus
mystax and S.geoffroyi leapt mainly from horizontal branches;
however, 20% of all leaps by L.nigrifrons were initiated from
vertical trunks in the lower layer. Leontocebus weddelli were
found to leap from trunk to trunk or from trunk to branch (and
vice versa) 80% of the time in the lower forest layer (< 5 m), in
which 60% of all leaps occurred (Yoneda 1981). Interestingly,
in our study, L.nigrifrons also undertook ‘mixed’ leaps between
vertical and horizontal substrates much more than S.mystax.
Saguinus mystax carried out these ‘mixed’ leaps less often than
leaps between horizontal substrates, suggesting that S.mystax
prefers to leap predominantly between horizontal substrates.
We hypothesized that leaping behavior is not dependent on
prevalent habitat characteristics and found that S.mystax leap
mainly between horizontal branches even when they use their
less-preferred forest layer. They tend not to use trunk-to-trunk
leaps in the lower layer even though trunks predominate below
10 m (Richards 1952; Kinzey et al. 1975; Dial et al. 2004).
Likewise, in higher layers L. nigrifrons has a higher proba-
bility of undertaking leaps between vertical supports than hor-
izontal supports, contrary to the suggestions of Porter (2004)
that leaping behavior is dependent on vertical position and
availability and properties of supports. It appears that support
availability in a given forest layer does not strictly determine
the leaping behavior in either species. Instead, preference
for a leaping mode is maintained even when support availa-
bility differs from that found in their preferred vertical layer.
While both species in our study displayed vertical leaps be-
tween vertical supports with a similar frequency in the lower
forest layer, this indeed may be a consequence of a paucity of
horizontal branches in this layer (Richards 1952; King 1998;
Dial etal. 2004) sensu Porter (2004). Alternatively, horizontal
branches in the lower layer often represent a poor substrate for
launching (e.g., palms and brambles). Nevertheless, we de-
termined that S.mystax favored horizontal leaps regardless of
forest layer, while L.nigrifrons favored leaps between vertical
supports in all forest layers. This is consistent with the idea that
postcranial anatomy inuences locomotor efciency and con-
sequently tamarins exhibit a consistent pattern of support use
and positional behavior (Garber and Pruetz 1995). Forelimb
and hindlimb lengths in free-ranging tamarins suggest that the
relatively longer limbs of L.nigrifrons are related to positional
maneuvering during the ight phase and deceleration during
landing (Garber 1991). Our behavioral data suggest that there
may be additional morphological differences related to leaping
behavior but this requires additional study in a phylogenetic
context.
Morphological outlook
Although, as expected, the support characteristics differed
among forest layers and the monkeys preferred different heights
in the forest, we did not nd locomotor behavior to be deter-
mined by the support properties available in a specic forest
layer. The species showed preferences for a specic leaping be-
havior even when confronted with a lack of preferred supports
outside of their preferred forest layers. Leontocebus nigrifrons
preferred vertical leaps even in forest layers where horizontal
supports were dominant. These observations, together with
the morphological differences identied by Garber (1991),
may represent species-specic adaptations. Clearly, observing
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

10 JOURNAL OF MAMMALOGY
primate behavior in a natural habitat is essential if we wish to
understand the relationship between functional morphology
and behavior (Ward and Sussman 1979; Cant 1992; McNamara
etal. 2019; Hemingway etal. 2020). Demonstrating behavioral
and morphological adaptations to use of forest layers and to
disentangle these effects from phylogenetic inertia will require
the use of the comparative method with morphological and
behavioral data representing the broad phylogenetic context
(Bicca-Marques 1999; Botton-Divet and Nyakatura 2021).
Analyses of morphological variation across species with
different behavioral and locomotor repertoires are essential
to inform broader phylogenetic analyses of form and function
(Ward and Sussman 1979). The differences in locomotor be-
havior documented in our study suggest the presence of mor-
phological differences related to leaping in these species. An
understanding of muscle architecture (Rosin and Nyakatura
2017; Hemingway et al. 2020), bone shape (Botton-Divet
etal. 2016), and bone micro-structure (Amson etal. 2017) is
essential if we wish to understand the selective pressures, and
possible adaptive responses within these closely related spe-
cies. Our behavioral observations in a habitat specic context
suggest that relative muscle masses may differ between the
two species studied herein. Mass of the distal extensors and
proximal exors of the hind limbs may be hypothesized to be
relatively larger in L.nigrifrons than in S.mystax because these
muscles are involved in powerful extension during launch, and
holding onto vertical supports, respectively. In contrast, we
expect the muscle mass of the proximal extensors to be rela-
tively larger in S.mystax than in L.nigrifrons because these
muscles are involved in horizontal leaping and striding (Rosin
and Nyakatura 2017). Osteological differences between tam-
arin species are apparent. For example, L. nigrifrons ex-
hibit greatly elongated forelimbs due to a longer hand and
forearm (Garber and Leigh 2001). This increased braking
distance could enhance the shock-absorbing role of soft tis-
sues (here, muscle and tendon stretch) when leaping between
vertical supports, providing an advantage during this type of
behavior (Garber 1991; Garber and Leigh 2001). In addition,
L.nigrifrons has the highest intermembral index of all tam-
arin species (Jungers 1985; Fleagle 1999). Additional oste-
ological differences of L.nigrifrons relative to other tamarin
species include a longer patellar groove, a shorter articular
facet of the patella and longer femoral condyles (Garber and
Davis 1995; cited in Garber and Leigh 2001; Botton-Divet
and Nyakatura 2021). These features in vertical clinging and
leaping strepsirrhine primates may be related to powerful knee
extension during leaps between vertical substrates (Napier
and Walker 1967; Oxnard 1983; Garber and Davis 1995;
cited in Garber and Leigh 2001). In many leaping species, the
hindlimbs play a dominant role (Ashton and Oxnard 1964).
Even in squirrels, the main force in leaping is generated by
the hindlimbs (Thorington and Thorington 1989). Small body
size in mammals that specialize in leaping (Howell 1944 cited
in Thorington and Thorington 1989) is associated with rel-
atively longer leaps with large gaps between branches, and
concomitantly elongated hindlimbs, with particular emphasis
on distal elements. Small squirrels such as Glaucomys appear
to have a greater mechanical advantage of the biceps femoris
for leaping due to a specialized pelvic and femoral architec-
ture than do large squirrels such as Petaurista (Scheibe etal.
2007; Wölfer etal. 2019). It would be interesting to use a
broader phylogenetic context to link behavioral observations
in the eld with morphological analyses of several species
of callitrichid primates. Our study provides detailed informa-
tion on one mixed-species group and contributes to a better
understanding of behavior in sympatric species. It also is a
motivation for exploration of the relationship between form,
function, and behavior.
A
We thank eld assistant Migdonio Huanuiri Arirama. Funding
is acknowledged from the DFG NY 63/2-1 to J.A.N. We ac-
knowledge critical comments from anonymous reviewers on a
previous version of this manuscript. We thank J.Scheibe for his
numerous and helpful comments, as well as Elizabeth Kerr for
revising the English language.
C  I
The authors declare that they have no conict of interest.
F
This study was funded by the DFG (NY 63/2-1).
D A
The datasets generated during study are available from the cor-
responding author on reasonable request.
S D
Supplementary data are available at Journal of Mammalogy
online.
Supplementary Data SD1.—Summary table of relevant
studies on tamarin ecology and behavior.
L C
Acostupa R.J.H., Bardales J.J.A., Teco R.M.V. 2013. Uso de
las plantas medicinales en la comunidad El Chino del área de
conservación regional comunal Tamshiyacu-Tahuayo, Loreto Perú.
Conocimiento Amazónico 4:77–86.
AmsonE., ArnoldP., vanHeterenA.H., CanovilleA., NyakaturaJ.A.
2017. Trabecular architecture in the forelimb epiphyses of extant
xenarthrans (Mammalia). Frontiers in Zoology 14:52.
Ashton E.H., Oxnard C.E. 1964. Locomotor patterns in primates.
Proceedings of the Zoological Society of London 142:1–28.
BatesD.M., MaechlerM., BolkerB., WalkerS., ChristensenR.H.B.,
Singmann H., Dai B., Scheipl F., Grothendieck G., Green P.,
., 2021. Package ‘lme4’. Linear Mixed-Effects Models using
‘Eigen’ and S4. R package version, 1.1–27.1. http://dk.archive.
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 11
ubuntu.com/pub/pub/cran/web/packages/lme4/lme4.pdf. Accessed
2021 Jan 6.
Bezanson M., Watts S.M., Jobin M.J. 2012. Technical note:
tree truthing: how accurate are substrate estimates in primate
eld studies? American Journal of Physical Anthropology
147:671–677.
Bicca-MarquesJ.C. 1999. Hand specialization, sympatry, and mixed-
species associations in callitrichines. Journal of Human Evolution
36:349–378.
BiewenerA.A. 2002. Future directions for the analysis of musculo-
skeletal design and locomotor performance. Journal of Morphology
252:38–51.
Botton-DivetL., CornetteR., FabreA.C., Herrel A., Houssaye A.
2016. Morphological analysis of long bones in semi-aquatic mus-
telids and their terrestrial relatives. Integrative and Comparative
Biology 56:1298–1309.
Botton-DivetL., Nyakatura J.A. 2021. Vertical clinging and leaping
induced evolutionary rate shifts in postcranial evolution of tam-
arins and marmosets (Primates, Callitrichidae). BMC Ecology and
Evolution 21:132.
Buchanan-Smith H. 1991. Field observations of Goeldi’s monkey,
Callimico goeldii, in northern Bolivia. Folia Primatologica
57:102–105.
Buchanan-Smith H.M. 1999. Tamarin polyspecic associations:
forest utilization and stability of mixed-species groups. Primates
40:233–247.
Buchanan-SmithH.M., HardieS.M., CaceresC., PrescottM.J. 2000.
Distribution and forest utilization of Saguinus and other primates
of the Pando Department, northern Bolivia. International Journal of
Primatology 21:353–379.
CalengeC. 2006. The package “adehabitat” for the R software: a tool
for the analysis of space and habitat use by animals. Ecological
Modelling 197:516–519.
Cant J.G. 1992. Positional behavior and body size of arboreal pri-
mates: a theoretical framework for eld studies and an illustration
of its application. American Journal of Physical Anthropology
88:273–283.
Chavel E.E., Mazerolle M.J., Imbeau L., Drapeau P. 2017.
Comparative evaluation of three sampling methods to estimate
detection probability of American red squirrels (Tamiasciurus
hudsonicus). Mammalian Biology 83:1–9.
Choi C.Y., NamH.Y., Lee W.S. 2007. Measuring the behaviors of
wintering Black-faced Spoonbills (Platalea minor): comparison of
behavioral sampling techniques. Waterbirds 30:310–316.
Corlett R.T., PrimackR.B. 2011. Tropical rain forests: an ecolog-
ical and biogeographical comparison. Chichester, UK: Wiley-
Blackwell; John Wiley & Sons.
CunhaA.A., VieiraM.V. 2002. Support diameter, incline, and vertical
movements of four didelphid marsupials in the Atlantic forest of
Brazil. Journal of Zoology 258:419–426.
DagostoM., YamashitaN. 1998. Effect of habitat structure on posi-
tional behaviour and support use in three species of lemur. Primates
39:459–472.
DammhahnM., SoarimalalaV., GoodmanS.M. 2013. Trophic niche
differentiation and microhabitat utilization in a species-rich mon-
tane forest small mammal community of eastern Madagascar.
Biotropica 45:111–118.
Delciellos A.C., Vieira M.V. 2009. Jumping ability in the arboreal
locomotion of didelphid marsupials. Mastozoologia Neotropical
16:299–307.
DemesB., JungersW.L., Gross T.S., Fleagle J.G. 1995. Kinetics of
leaping primates: inuence of substrate orientation and compli-
ance. American Journal of Physical Anthropology 96:419–429.
DemesB., Fleagle J.G., Lemelin P. 1998. Myological correlates of
prosimian leaping. Journal of Human Evolution 34:385–399.
DeWaltS.J., ChaveJ. 2004. Structure and biomass of four lowland
Neotropical forests. Biotropica 36:7–19.
DialR., BloodworthB., LeeA., BoyneP., HeysJ. 2004. The distri-
bution of free space and its relation to canopy composition at six
forest sites. Forest Science 50:312–325.
Dunham N.T., McNamara A., Shapiro L.J., PhelpsT., YoungJ.W.
2020. Asymmetrical gait kinematics of free-ranging callitrichines
in response to changes in substrate diameter and orientation.
Journal of Experimental Biology 223:jeb217562.
Encarnación F. 1985. Introducción a la ora y vegetación de la
amazonía peruana: terminología y descripción de las formaciones
vegetales de la llanura amazónica. Candollea 10:237–252.
Fleagle J.G. 1977. Locomotor behavior and muscular anatomy
of sympatric Malaysian leaf-monkeys (Presbytis obscura and
Presbytis melalophos). American Journal of Physical Anthropology
46:297–307.
Fleagle J.G. 1999. Primate adaptation and evolution. 2nd ed. San
Diego, California: Academic Press.
Ganzhorn J.U., Rakotondranary S.J., Ratovonamana Y.R. 2011.
Habitat description and phenology. In: CurtisD.J., SetchellJ.M.,
editors. Field and laboratory methods in primatology; a practical
guide. 2nd ed. Cambridge, United Kingdom: Cambridge University
Press; p.51–68.
Garber P.A., Sussman R.W. 1984. Ecological distinctions between
sympatric species of Saguinus and Sciurus. American Journal of
Physical Anthropology 65:135–146.
Garber P.A. 1988a. Foraging decisions during nectar feeding by
tamarin monkeys (Saguinus mystax and Saguinus fuscicollis,
Callitrichidae, Primates) in Amazonian Peru. Biotropica
20:100–106.
GarberP.A. 1988b. Diet, foraging patterns, and resource defense in a
mixed species troop of Saguinus mystax and Saguinus fuscicollis in
Amazonian Peru. Behaviour 105:18–34.
Garber P.A. 1998. Within-and between-site variability in mous-
tached tamarin (Saguinus mystax) positional behavior during food
procurement. In: Strasser E., Fleagle J.G., Rosenberger A.L.,
McHenryH., editors. Primate locomotion: recent advances. New
York: Plenum Press; p.61–78.
GarberP.A. 1991. A comparative study of positional behavior in three
species of tamarin monkeys. Primates 32:219–230.
GarberP.A., DavisL.C. 1995. Intraspecic variability in anatomy and
positional behavior in two tamarin species. American Journal of
Physical Anthropology 22:110–111.
Garber P.A., Pruetz J.D. 1995. Positional behavior in moustached
tamarin monkeys: effects of habitat on locomotor variability and
locomotor stability. Journal of Human Evolution 28:411–426.
Garber P.A., Leigh S.R. 2001. Patterns of positional behavior in
mixed-species troops of Callimico goeldii, Saguinus labiatus, and
Saguinus fuscicollis in northwestern Brazil. American Journal of
Primatology 54:17–31.
Garber P.A., Porter L.M. 2009. Trunk-to-trunk leaping in wild
Callimico goeldii in northern Bolivia. Neotropical Primates
16:9–15.
Hemingway H.W., Burrows A.M., Omstead K.M., Zohdy S.,
PastorJ.F., MuchlinskiM.N. 2020. Vertical clinging and leaping
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

12 JOURNAL OF MAMMALOGY
ahead: how bamboo has shaped the anatomy and physiology of
Hapalemur. Anatomical Record 303:295–307.
HesseB., NyakaturaJ.A., FischerM.S., SchmidtM. 2015. Adjustments
of limb mechanics in cotton-top tamarins to moderate and steep
support orientations: signicance for the understanding of early pri-
mate evolution. Journal of Mammalian Evolution 22:435–450.
HeymannE.W. 2000. Spatial patterns of scent marking in wild mous-
tached tamarins, Saguinus mystax: no evidence for a territorial
function. Animal Behaviour 60:723–730.
Heymann E.W. 2018. Tamarine–die etwas anderen Primaten:
Freilandforschung an Affen im Amazonasregenwald. Biologie in
unserer Zeit 48:114–119.
HeymannE.W., Buchanan-SmithH.M. 2000. The behavioural ecology
of mixed-species troops of callitrichine primates. Biological
Reviews of the Cambridge Philosophical Society 75:169–190.
HowellA.B. 1944. Speed in animals. Chicago, Illinois: University of
Chicago Press.
HuntK.D., CantJ.G., GeboD.L., RoseM.D., WalkerS.E., YoulatosD.
1996. Standardized descriptions of primate locomotor and postural
modes. Primates 37:363–387.
Iida Y., Kohyama T.S., Kubo T., Kassim A.R., Poorter L., Sterck
F., Potts M.D. 2011. Tree architecture and life‐history strategies
across 200 co‐occurring tropical tree species. Functional Ecology
25:1260–1268.
JamesR.S., NavasC.A., HerrelA. 2007. How important are skeletal
muscle mechanics in setting limits on jumping performance? The
Journal of Experimental Biology 210(Pt 6):923–933.
Jungers W.L. 1985. Body size and scaling of limb proportions in
primates. In: JungersW.L., editor. Size and scaling in primate bi-
ology. New York: Plenum Press; p.345–381.
King D.A. 1998. Inuence of leaf size on tree architecture: rst
branch height and crown dimensions in tropical rain forest trees.
Trees 12:438–445.
KingD.A., MaindonaldJ.H. 1999. Tree architecture in relation to leaf
dimensions and tree stature in temperate and tropical rain forests.
Journal of Ecology 87:1012–1024.
KinzeyW.G., RosenbergerA.L., RamirezM. 1975. Vertical clinging
and leaping in a neotropical anthropoid. Nature 255:327–328.
KlingeH. 1973. Struktur und Artenreichtum des zentralamazonischen
Regenwaldes. Amazoniana: Limnologia et Oecologia Regionalis
Systematis Fluminis Amazonas 4:283–292.
LeighJr. E.G. 1975. Structure and climate in tropical rain forest.
Annual Review of Ecology and Systematics 6:67–86.
Madden D., Garber P.A., Madden S.L., Snyder C.A. 2010. Rain-
forest canopy-connectivity and habitat selection by a small neo-
tropical primate, Geoffroy’s tamarin (Saguinus geoffroyi). Journal
of Tropical Ecology 26:637–644.
Mauffrey J.F., Catzeis F. 2003. Ecological and isotopic discrimi-
nation of syntopic rodents in a Neotropical rain forest of French
Guiana. Journal of Tropical Ecology 19:209–214.
MayrE. 1947. Ecological factors in speciation. Evolution 1:263–288.
McNamara A., Dunham N.T., Shapiro L.J., Young J.W. 2019. The
effects of natural substrate discontinuities on the quadrupedal gait
kinematics of free-ranging Saimiri sciureus. American Journal of
Primatology 81:e23055.
Mortensen H. 1932. Das Licht im tropischen (heißfeuchten)
Regenwalde. Naturwissenschaften 20:312–315.
Nadjafzadeh M.N., Heymann E.W. 2008. Prey foraging of red titi
monkeys, Callicebus cupreus, in comparison to sympatric tam-
arins, Saguinus mystax and Saguinus fuscicollis. American Journal
of Physical Anthropology 135:56–63.
Napier J.R., Walker A.C. 1967. Vertical clinging and leaping—a
newly recognized category of locomotor behaviour of primates.
Folia Primatologica 6:204–219.
Nascimento A.T., Schmidlin L.A., Valladares-Padua C.B.,
MatushimaE.R., Verdade L.M. 2011. A comparison of the home
range sizes of mainland and island populations of black-faced lion
tamarins (Leontopithecus caissara) using different spatial analysis.
American Journal of Primatology 73:1114–1126.
Norconk M.A. 1986. Interactions between primate species in a
Neotropical forest: mixed-species troops of Saguinus mystax
and S. fuscicollis (Callitrichidae) [Ph.D. dissertation]. Davis,
California: University of California.
Norconk M.A. 1990. Mechanisms promoting stability in mixed
S.mystax and S. fuscicollis troops. American Journal of Physical
Anthropology 135:56–63.
NyakaturaJ.A. 2019. Early primate evolution: insights into the func-
tional signicance of grasping from motion analyses of extant
mammals. Biological Journal of the Linnean Society 127:611–631.
NyakaturaJ.A., HeymannE.W. 2010. Effects of support size and ori-
entation on symmetric gaits in free-ranging tamarins of Amazonian
Peru: implications for the functional signicance of primate gait
sequence patterns. Journal of Human Evolution 58:242–251.
Orkin J.D., Pontzer H. 2011. The Narrow Niche hypothesis: gray
squirrels shed new light on primate origins. American Journal of
Physical Anthropology 144:617–624.
OxnardC.E. 1983. The order of man: a biomathematical anatomy of
the primates. Hong Kong University Press, HKU.
PeresC.A. 1993a. Structure and spatial organization of an Amazonian
terra rme forest primate community. Journal of Tropical Ecology
9:259–276.
PeresC.A. 1993b. Anti-predation benets in a mixed-species group
of Amazonian tamarins. Folia Primatologica 61:61–76.
PorterL.M. 2004. Forest use and activity patterns of Callimico goeldii
in comparison to two sympatric tamarins, Saguinus fuscicollis and
Saguinus labiatus. American Journal of Physical Anthropology
124:139–153.
RamírezM.M. 1989. Feeding ecology and demography of the mous-
tached tamarin Saguinus mystax in northeastern Peru [PhD disser-
tation]. New York: City University of New York.
RichardsP.W. 1952. The tropical rain forest. Cambridge: Cambridge
University Press.
RosinS., NyakaturaJ.A. 2017. Hind limb extensor muscle architec-
ture reects locomotor specialisations of a jumping and a striding
quadrupedal caviomorph rodent. Zoomorphology 136:267–277.
Rylands A.B., Mittermeier R.A. 2013. Family Callitrichidae (mar-
mosets and tamarins). In: Mittermeier R.A., Rylands A.B.,
Wilson D.E., editors. Handbook of the mammals of the world.
Barcelona, Spain: Lynx Edicions; p.262–346.
Rylands A.B., Heymann E.W., Alfaro J.L., Buckner J.C., Roos
C., Matauschek C., Boubli J.P., Sampaio R., Mittermeier R.A.
2016. Taxonomic review of the New World tamarins (Primates:
Callitrichidae). Zoological Journal of the Linnean Society
177:1003–1028.
ScheibeJ.S., PaskinsK.E., FerdousS., BirdsillD. 2007. Kinematics
and functional morphology of leaping, landing, and branch use in
Glaucomys sabrinus. Journal of Mammalogy 88:850–861.
Schreier B.M., Harcourt A.H., Coppeto S.A., Somi M.F. 2009.
Interspecic competition and niche separation in primates: a global
analysis. Biotropica 41:283–291.
Seaman D.E., Millspaugh J.J., Kernohan B.J., Brundige G.C.,
Raedeke K.J., Gitzen R.A. 1999. Effects of sample size on
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021

BERLES ET AL.—LEAPING IN SYMPATRIC TAMARINS 13
kernel home range estimates. Journal of Wildlife Management
63:739–747.
Seitz S. 2001. Vergleichende Untersuchungen zu Verhalten und
Schauwert von Tapiren (Familie Tapiridae) in Zoologischen
Gärten [PhD dissertation]. Heidelberg, Germany: Ruprecht-Karls-
University of Heidelberg.
ShapiroL.J., YoungJ.W. 2010. Is primate-like quadrupedalism nec-
essary for ne-branch locomotion? A test using sugar gliders
(Petaurus breviceps). Journal of Human Evolution 58:309–319.
SiegelC.E., Hamilton J.M., CastroN.R. 1989. Observations of the
Red-billed Ground-Cuckoo (Neomorphus pucheranii) in associ-
ation with tamarins (Saguinas) in northeastern Amazonian Peru.
Condor 91:720–722.
SmithA.C. 1997. Comparative ecology of saddleback (Saguinus
fuscicollis) and moustached (Saguinus mystax) tamarins
[PhD dissertation]. Reading, United Kingdom: University of
Reading.
Stohlgren T.J., Chong G.W. 1997. Assessment of biological diver-
sity and long-term monitoring plan for the lower Urubamba region.
Biodiversity assessment and monitoring of the lower Urubamba
region, Peru. SI/MAB Series 1:41–44.
SussmanR.W., O’FallonW.M., Sussman L.K., Buettner-JanuschJ.
1979. Statistical methods for analyzing data on daily activity
cycles of primates. American Journal of Physical Anthropology
51:1–14.
Sussman R.W., Kinzey W.G. 1984. The ecological role of the
callitrichidae: a review. American Journal of Physical Anthropology
64:419–449.
ThoringtonJr R.W., ThoringtonE.M. 1989. Postcranial proportions
of Microsciurus and Sciurillus, the American pygmy tree squirrels.
Advances in Neotropical Mammalogy 1989:125–136.
ToussaintS., Llamosi A., Morino L., Youlatos D. 2020. The central
role of small vertical substrates for the origin of grasping in early
primates. Current Biology 30:1600–1613.e3.
Ward S.C., Sussman R.W. 1979. Correlates between locomotor
anatomy and behavior in two sympatric of Lemur. American
Journal of Physical Anthropology 50:575–590.
Wölfer J., Amson E., Arnold P., Botton-Divet L., Fabre A.C.,
vanHeteren A.H., NyakaturaJ.A. 2019. Femoral morphology of
sciuromorph rodents in light of scaling and locomotor ecology.
Journal of Anatomy 234:731–747.
Yoneda M. 1981. Ecological studies of Saguinus fuscicollis und
Saguinus labiatus with reference to habitat segregation and height
preference. Kyoto University Overseas Research Reports of New
World Monkeys 2:43–50.
YoulatosD. 1999. Comparative locomotion of six sympatric primates
in Ecuador. Annales des Sciences Naturelles-Zoologie et Biologie
Animale 20:161–168.
Youlatos D., Karantanis N.E., Byron C.D., Panyutina A. 2015.
Pedal grasping in an arboreal rodent relates to above‐branch
behavior on slender substrates. Journal of Zoology 296:
239–248.
Submitted 17 February 2021. Accepted 26 September 2021.
Associate Editor was John Scheibe.
Downloaded from https://academic.oup.com/jmammal/advance-article/doi/10.1093/jmammal/gyab121/6430835 by guest on 19 November 2021