Nutritional content explains the attractiveness of cacao to crop raiding Tonkean macaques

Abstract

Nutritional ecology has been linked to crop raiding behavior in a number of wildlife taxa. Here our goal is to explore the role nutrition plays in cacao crop raiding by Tonkean macaques Macaca tonkeana in Sulawesi, Indonesia. From June - Sept. 2008 we collected fruit samples from 13 species known to be important Tonkean macaque foods and compared their nutritional value to that of cacao Theobroma cacao, an important cash crop in Sulawesi. Cacao pulp was significantly lower in protein, but lower in dietary fiber, and higher in digestible carbohydrates and energy content compared to forest fruits. These findings, combined with the fact that cacao fruits are spatially concentrated and available throughout the year, likely explain why Tonkean macaques are attracted to this cultivated resource. We use these data along with published feeding ecology data to propose strategies to minimize human-macaque conflict. Namely, we recommend the deliberate protection of Elmerillila tsiampaccca, Ficus spp. and Arenga pinnata, fruit species known to be regularly consumed and of considerable nutritional value. We also identify the A. pinnata palm as a potential buffer resource to curb cacao crop raiding by macaques. Cacao is a hard-to-process food because the pods have a thick outer skin that encases the seeds and pulp. Aren palm fruit, although lower in digestibility, is easier-to-process, higher in protein, and also available year round. In addition, because the palm has considerable cultural and economic significance for local people, the strategy of planting Aren palm in a buffer corridor is likely to garner local community support [Current Zoology 59 (2): 160-169, 2013].

Full text

Current Zoology 59 (2): 160–169, 2013
Received June 15, 2012; accepted Aug. 1, 2012.
∗
Corresponding author. E-mail: epriley@mail.sdsu.edu
© 2013 Current Zoology
Nutritional content explains the attractiveness of cacao to
crop raiding Tonkean macaques
Erin P. RILEY
1*
, Barbara TOLBERT
1
, Wartika R. FARIDA
2
1
Department of Anthropology, San Diego State University, San Diego, CA 92182-6040, USA
2
Research Center for Biology, Indonesian Institute of Sciences, Cibinong – Bogor, Indonesia
Abstract Nutritional ecology has been linked to crop raiding behavior in a number of wildlife taxa. Here our goal is to explore
the role nutrition plays in cacao crop raiding by Tonkean macaques Macaca tonkeana in Sulawesi, Indonesia. From June – Sept.
2008 we collected fruit samples from 13 species known to be important Tonkean macaque foods and compared their nutritional
value to that of cacao Theobroma cacao, an important cash crop in Sulawesi. Cacao pulp was significantly lower in protein, but
lower in dietary fiber, and higher in digestible carbohydrates and energy content compared to forest fruits. These findings, com-
bined with the fact that cacao fruits are spatially concentrated and available throughout the year, likely explain why Tonkean ma-
caques are attracted to this cultivated resource. We use these data along with published feeding ecology data to propose strategies
to minimize human-macaque conflict. Namely, we recommend the deliberate protection of Elmerillila tsiampaccca, Ficus spp.
and Arenga pinnata, fruit species known to be regularly consumed and of considerable nutritional value. We also identify the A.
pinnata palm as a potential buffer resource to curb cacao crop raiding by macaques. Cacao is a hard-to-process food because the
pods have a thick outer skin that encases the seeds and pulp. Aren palm fruit, although lower in digestibility, is easier-to-process,
higher in protein, and also available year round. In addition, because the palm has considerable cultural and economic signifi-
cance for local people, the strategy of planting Aren palm in a buffer corridor is likely to garner local community support [Current
Zoology 59 (2): 160–169, 2013].
Keywords Human-macaque conflict, Foraging efficiency, Buffer crop, Arenga pinnata, Ficus, Fiber content
Crop raiding is undoubtedly one of the most serious
challenges to wildlife conservation. This behavior is
particularly pervasive among the nonhuman primates,
given the high levels of sociality and intelligence and
keen problem-solving skills of the taxa commonly re-
ported as culprits (e.g., capuchins, McKinney, 2011;
macaques, Riley and Priston, 2010; baboons, Hill, 2000;
chimpanzees, Hockings et al., 2009). Because of the
detrimental impact crop raiding can have on human
livelihoods and the likelihood of local support for con-
servation efforts, scholars and conservation managers
worldwide are increasingly focusing attention on the
development and implementation of effective mitigation
programs (Dickman, 2010; Osborn and Hill, 2005;
Strum, 1994).
Predicting the patterns of raiding is an important first
step in developing mitigation strategies. The ability to
develop a successful strategy is largely dependent on
behavioral and ecological data from raiding primates.
For example, ranging patterns, substrate use, habitat
composition, and patterns of forest fruit availability all
can shed light on the causes of crop raiding and its fre-
quency and timing (e.g., Naughton-Treves et al., 1998;
Siex and Struhsaker, 1999; Riley, 2007b; Hockings et al.,
2009). Because crop raiding is a foraging strategy
(Strum, 1994), the factors that influence the choice of
wild foods are also likely at play when primates raid
crops. Foraging theory predicts that animals will forage
in a way that maximizes the rate of energy intake
(Stephen and Krebs, 1986). Cultivated foods tend to be
highly palatable, easier to process, and are typically
distributed in a way that minimizes searching costs.
Therefore, they often represent the “smarter choice”,
particularly in areas where anti-raiding strategies by
farmers are infrequent or ineffective. Nutritional content,
also an important aspect of foraging efficiency, has been
shown to play a role in crop raiding behavior (e.g., ele-
phants, Osborn, 2004; Rode et al., 2006a; white-tailed
deer, Dostaler et al., 2011). For example, Rode et al.
(2006b) suggest that elephants in Kibale National Park,
Uganda may raid crops such as maize and bananas to
compensate for limited sodium availability in wild

RILEY EP et al.: Nutritional content cacao crop raiding 161
foods. Among nonhuman primates, the greater digesti-
bility of human foods, such as maize, has been proposed
to explain crop raiding patterns by olive baboons
(Forthman Quick and Demment, 1988). Linking crop
raiding behavior by nonhuman primates to specific nu-
tritional components of cultivated foods, however, re-
mains largely understudied. In this paper, we explore
this relationship in a nonhuman primate endemic to Su-
lawesi, Indonesia: the Tonkean macaque Macaca
tonkeana.
The Tonkean macaque is a medium-sized, primarily
frugivorous primate that exists in multimale-multife-
male social groups (Riley, 2007a). This primate shows
considerable ecological flexibility in its ability to use
multiple forest strata, including a considerable amount
of time on the ground, and to persist in heavily-altered
habitats (Riley, 2008). The latter is made possible in part
by the ecological strategy of crop raiding. While it is
likely that the raiding of subsistence crops has a long
history on Sulawesi, the planting of highly palatable
cash crops in the post-colonial era has undoubtedly ex-
acerbated human-macaque conflict in the area. For ex-
ample, in the 1980s the Sulawesi landscape experienced
an enormous expansion of cacao acreage, mostly by
smallholder production (Akiyama and Nishio, 1996).
Cacao Theobroma cacao, a perennial tree crop that is
frequently planted under forest shade, is one of the pri-
mary cash crops raided by macaques and other wildlife
(Riley, 2007b; Supriatna et al., 1992). Previous research
has found that Tonkean macaques raid cacao crops
year-round regardless of levels of forest fruit availability
and the frequency of cacao crop guarding by farmers
(Riley, 2007b).
Our goal in this article is threefold. First, we assess
the nutritional value of a portion of the Tonkean ma-
caque fruit diet. Second, we compare the nutritional
value of forest fruits and cacao pulp to explore whether
nutritional factors play a role in cacao crop raiding by
Tonkean macaques. Our third goal is to use the nutri-
tional data to inform conservation and human-macaque
conflict management strategies. Namely, we identify
specific fruit tree species of high nutritional value that
should be protected in macaque habitat as part of in situ
conservation efforts. In terms of mitigating human-
wildlife conflict, one existing strategy is to plant buffer
crops at the forest-farm edge that act as barriers to de-
sirable cash crops (Hockings and Humle, 2009;
Dickman, 2010). This strategy has been employed using
unpalatable cash crops (e.g., tea and chili) with some
success (Parker and Osborn, 2006; Southworth et al.,
2010; Hockings and Humle, 2009). An alternative ver-
sion of this strategy is to establish an exploitable buffer
that diverts feeding away from important cash crops.
For example, Yuwono et al. (2007) suggest that a buffer
zone planted with an abundance of fruit trees could dis-
suade orangutans Pongo pygmaeus from entering oil
palm plantations. Here we identify the Arenga pinnata
palm as a potential species that could be used to test the
efficacy of the exploitable buffer strategy.
1 Materials and Methods
1.1 Study area
This research was conducted in the Lake Lindu high-
land plain (1° 19' 07" S, 120° 04' 51" E) in Lore Lindu
National Park (LLNP), Central Sulawesi, Indonesia.
Comprising approximately 2,200 km
2
, LLNP is the
second largest terrestrial protected area in Wallacea. The
park is designated as a UNESCO Man and the Bio-
sphere Reserve due to the critical role it plays in the
conservation of endemic fauna and flora, including one
of the seven endemic macaques, Macaca tonkeana. The
Tonkean macaque is currently listed as “Vulnerable”
(IUCN, 2011) with an estimated remaining population
of 150,000 (Riley, 2010). As noted earlier, Tonkean
macaques are primarily frugivorous with fruit account-
ing for an average of 81% of feeding records (2 groups;
monthly range: 40%−93%), but they also consume in-
sects, young leaves, and other vegetative matter (Riley,
2007a).
The Lindu plain is one of two enclaves allowed to
exist within the National Park because it is a major rice
growing area and has long established human settle-
ments. In its designation as a National Park, the granting
of enclave status meant that the Lindu people could
maintain their existing agricultural fields and continue
to engage in small-scale forest production collection.
Although wet-rice agriculture (sawah) predominates in
Lindu, tree cash crops, such as coffee and cacao, have
grown in importance as part of the Lindu economy since
the 1980s. Cacao is typically grown as a spatially con-
centrated monocrop on plantation plots ranging in size
from 1–5 ha, with the number of trees in the plantation
ranging between 124–1903 (Riley and Fuentes, 2011).
Cacao pods are harvested by farmers when the fruit
reaches a deep yellow or red color, depending on the
varietal. The pods are then cut open, and the wet beans
covered in pulp are removed (Fig. 1). It is the sweet
pulp that macaques and other raiding wildlife (e.g., fore-
st rats and squirrels) consume (Riley, 2007b). Although
the percent contribution of cacao pulp to the diet of ma-

162 Current Zoology Vol. 59 No. 2
Fig. 1 Cacao tree (top) and a farmer removing the beans
and pulp (white part) from a cacao pod (bottom).
caque groups is unknown, previous research found that
macaque crop raiding occurred throughout the course of
a six-month period (with raiding peaks coinciding with
peaks in cacao fruiting), with one farmer experiencing a
loss of 7% of his total cacao harvest due to macaque
raiding (Riley, 2007b).
1.2 Sample collection
We opportunistically collected fruit samples from
tree species that were fruiting during our study period
(June–September 2008). Every attempt was made to
collect samples from specific trees where the macaques
had previously been observed feeding. If this was not
possible, samples were collected from trees within the
home range of known macaque groups. Our sample
comprised 13 fruit species (Table 1). We focused on
fruit because it constitutes the primary component of the
Tonkean macaque diet and due to the ease of its collec-
tion. We acknowledge that our sample represents a por-
tion of the overall fruit diet; however, our samples did
come from species known to be important food re-
sources for Tonkean macaques (Table 1; Riley, 2007b,
unpublished data). For each of the species, the ripe fruit
Table 1 Forest fruit species sampled (n = 13) and known
contribution to Tonkean macaque diet
Contribution to diet
a
(% feeding records)
Species Family
Vernacular
name
Anca
group
CH group
Areca vestiaria Palmae Pinang merah
c
0.26 —
Arenga
pinnata
Palmae Enau
d
52 1.3
Artocarpus
teysmanii
Moraceae Teah
d
4.5 —
Artocarpus
vriescana
Moraceae Baloli
d
3.1 —
Elmerillia
tsiampacca
Magnoliaceae Takasa
d
2.8 20.7
Ficus elmeri Moraceae 0.52 1.6
Ficus virgata Moraceae Ara
c
— 6.3
Ficus sagittata Moraceae 1.1 1.3
Ficus sp. 1
b
Moraceae n n
Ficus sp. 2 Moraceae n n
Pandanus sp. Pandanaceae Pandan
c
0.77 12.5
Pinanga sp. Palmae Pinang hitam
c
1.9 4.1
Timonius
teysmanii
Rubiaceae Kalambio
d
0.77 0.33
a
These data come from previous research conducted on two groups of
Tonkean macaques (Riley 2007b; unpublished data); percent feeding
records derive from scan sampling data (total number of feeding re-
cords: Anca = 479; CH = 359; Riley 2007a).
b
Many species of figs in
Lore Lindu National Park are unidentifiable. Overall, consumption of
fig fruits accounted for 22% and 51% of the feeding records for Anca
and CH groups, respectively (Riley 2007b).
c
Bahasa Indonesia.
d
Bahasa Lindu/Tado. n: Observations of macaques consuming this
species come only from ad libitum data (Riley, unpublished data).
—: Not observed
pulp was collected as this represented the part consumed
by macaques. When trees were deemed impossible to
safely climb, samples were collected from the ground.
Across the study period, we collected between 2−4 sam-
ples from multiple trees of each species for most fruits.
We also collected the pulp from cacao pods from
plantations located at the forest edge. All samples col-
lected were placed on aluminum trays and sun-dried in
the field and in a generator-powered drying box until
completely dry.
The fruits collected were cut into small pieces to ac-
celerate the drying process, and ultimately, for chemical
analyses. Because the seed fraction of fig fruits is not
typically digested by primates, we separated the pulp
fraction (including the skin) from the seed fraction and
only used the pulp in chemical analyses (Urquiza-Haas
et al., 2008). Once dry, samples were stored in labeled
paper bags with silica gel. As soon as the required dry
weight was obtained (i.e., 50– 80 g per species), we sent
samples to the Laboratory of Nutrition Testing Research

RILEY EP et al.: Nutritional content cacao crop raiding 163
Center for Biology in Cibinong, Bogor for analysis. We
attempted to collect multiple samples from multiple
trees of a given species, but sampling was not conducted
simultaneously with behavioral observations and it was
limited to a four-month time period. We therefore ac-
knowledge that our data can say little about the nutri-
tional composition of the foods at the time they were
consumed or the potential variation in fiber and macro-
nutrient content over a more extended period of time.
1.3 Nutritional and statistical analyses
The samples were analyzed for dry matter content,
crude protein, lipids, and ash using standard proximate
analysis procedures (AOAC, 1990). Given the important
role calcium and phosphorus play in fig nutrition
(O’Brien et al., 1998), these minerals were also ana-
lyzed (AOAC, 1990). Crude protein was determined
using the standard formula: N (total nitrogen) × 6.25
(AOAC, 1990). The detergent fiber analysis (van Soest,
1994), which renders the neutral detergent fiber (NDF =
the total insoluble fiber in plant cell wall, primarily cel-
lulose, hemicelluloses, and lignin) and the acid deter-
gent fiber (ADF = primarily cellulose and lignin), was
also conducted. In the results we report NDF (often
considered the best index of total insoluble fiber and of
energy available from fiber [e.g., Conklin-Brittain et al.
2006]) as well as ADF (a better index of indigestible
fiber if macaques are able to digest hemicellulose). The
digestible carbohydrates or the total nonstructural car-
bohydrates (TNC) were calculated as: %TNC = 100 – %
lipid – % crude protein – % total ash – % NDF. TNC
values were calculated using percentages of organic
matter because ash, which is included in dry matter
(DM), does not contribute energy to food (Conklin-
Brittain et al., 2006). We also calculated available or
metabolizable energy (ME) of the foods using standard
physiological fuel values for carbohydrates, crude pro-
tein, and lipids (4, 4, and 9 kcal/g, respectively) and a
fiber digestion coefficient of 0.463 that was determined
for captive Japanese and rhesus macaques (Sakaguchi et
al., 1991). The physiological fuel value of fiber was
calculated as 3 x 0.463 = 1.389 kcal/g. Assuming
maximal NDF fermentation, we calculated energy per
food species as ME
h
kcal/g OM = [(4 × %TNC) + (4 ×
%CP) + (9 × %lipid) + (1.389 × %NDF)]/100 (Conk-
lin-Brittain et al., 2006).
We compared our nutritional data to those published
on other frugivorous catarrhines. T-tests were used to
compare nutritional composition of fig and non-fig
fruits. To test for differences in nutritional composition
between the forest fruits and cacao pulp, we used one
sample t-tests. We performed arcsine transformations on
proportional data to meet assumptions of normality and
equal variances required by parametric techniques
(Sokal and Rohlf, 1981). Results were considered sig-
nificant at P <0.05.
2 Results
2.1 Nutritional composition: Forest fruits
Among the 13 forest fruit species sampled, diffe-
rences in the principal macronutrients were evident (Ta-
ble 2). Crude protein content varied between 3.05% to
9.48% (DM). The mean protein content (5.39%) fell
below values reported (plant portion of diet) for African
cercopithecines, but was within the range reported for
other macaque species (Table 3). Lipid content ranged
between 0.16% to 25.72%. The mean lipid content
(5.39%) was higher than that reported for African cer-
copithecines and chimpanzees (plant portion of diet),
but was within the range reported for semi-free ranging
Japanese macaques (Table 3). NDF values also varied
considerably, from 18.63% to 58.31% DM. The mean
NDF content (34.42%) was similar to the fruit (or plant)
diet of other frugivorous primates (Table 3). ADF values
ranged between 14.84% to 43.67% DM. The mean ADF
value (27.60%) was similar to those reported for redtail
monkeys but higher than those found for blue monkeys,
grey-cheeked mangabeys, and chimpanzees (Table 3).
Mean calcium (0.48) and phosphorus (0.14) values were
lower than those reported for red-tailed guenons in Ki-
bale National Park, Uganda (Table 3).
Figs did not significantly vary in ME, mean protein
and lipid content, or mean ADF (indigestible fiber) from
non-fig fruit (ME: t = 1.164, df = 10, P = 0.272; Protein:
t = 2.013, df = 10, P = 0.072; Lipids: t = -0.487, df = 10,
P =0.637; ADF: t = 0.574, df = 10, P = 0.579). Mean
fiber content (NDF), however, was significantly lower
in figs compared to non-figs (Fig mean: 34.1% OM,
Non-fig mean: 52.8% OM; t = -2.602, df = 10, P =
0.026). Likewise, mean TNC was significantly higher in
figs than in non-fig fruit (Fig mean: 51% OM, Non-fig
mean: 32.3%OM; t = 2.243, df = 10, P = 0.049). In
terms of mineral content, figs and non-fig fruit did not
significantly vary in their CA: P ratio. However, when
Arenga pinnata, a clear outlier was excluded from the
forest fruit sample, the difference in mean CA: P ratio
(Fig: 6.08, Non-fig: 1.42) approached significance (t =
2.174, df = 10, P = 0.055). Among the fig species, there
were differences in all components of the fruit pulp (Ta-
ble 2). Lipid content varied between 2.18%–6.63% OM,
with a mean of 3.87%±1.72%. Fiber content (NDF)

164 Current Zoology Vol. 59 No. 2
Table 2 Results of nutritional analyses: Crude protein, lipids, NDF, ADF, calcium, phosphorus (% DM and % OM) and
metabolizable energy (ME)
% Dry Matter (DM) % Organic Matter (OM) ME
a
DM Protein Lipid NDF ADF Ca P
Ca: P
ratio
OM Protein NDF ADF TNC
Kcal/g
(incl. NDF)
Areca vestiaria 90.03 4.76 9.47 48.66 24.57 0.17 0.10 1.7 85.33 6.20 63.34 32.18 18.13 2.96
Arenga pinnata 86.26 4.31 0.16 32.19 14.84 1.77 0.05 35.4 78.34 6.38 47.64 22.32 45.75 2.77
Artocarpus teysmanii 90.3 9.48 5.95 22.16 20.18 0.42 0.15 2.8 85.61 6.47 58.58 26.26 31.56 2.64
Artocarpus vriescana 91.10 5.28 4.51 24.07 17.56 0.05 0.23 0.21 87.30 6.64 30.26 22.18 57.42 3.49
Elmerillia tsiampacca 92.19 7.98 25.72 39.34 31.63 0.10 0.15 0.66 90.33 9.59 47.24 38.05 12.29 4.31
Ficus elmeri 90.31 5.41 6.63 18.63 16.93 0.38 0.06 6.33 82.96 7.22 24.87 22.82 59.07 3.79
Ficus rirgata 82.65 5.83 3.55 20.62 26.24 1.39 0.09 15.44 76.19 9.25 32.75 42.43 52.35 3.43
Ficus sagittata 85.43 4.40 4.22 23.36 22.51 0.60 0.11 5.45 80.75 6.38 33.86 32.96 53.65 3.42
Ficus sp.1 92.27 6.34 2.18 38.87 38.17 0.43 0.29 1.48 83.46 8.23 50.48 50.01 38.46 2.82
Ficus sp.2 93.44 5.30 2.78 47.94 41.61 0.50 0.29 1.72 87.59 12.26 28.66 51.08 51.37 3.64
Pandanus sp. 90.94 3.86 3.84 58.31 43.67 0.27 0.11 2.45 87.98 4.82 72.88 54.76 17.50 2.34
Pinanga sp. 84.90 4.03 0.59 34.49 28.30 0.04 0.07 0.57 81.41 5.83 49.89 41.25 43.42 2.74
Timonius teysmanii 86.76 3.05 0.54 38.76 32.60 0.11 0.07 1.57 83.28 4.21 53.64 45.41 41.40 2.64
* Theobroma cacao 85.15 3.20 0.04 5.75 4.68 0.04 0.14 0.28 83.38 4.51 8.10 6.62 87.34 3.79
Forest Fruit Mean
(SD)
88.97
(3.36)
5.39
(1.75)
5.39
(6.65)
34.42
(12.4)
27.60
(9.48)
0.48
(0.53)
0.14
(0.08)
5.83
(9.77)
83.89
(4.04)
7.19
(2.17)
45.70
(14.71)
37.50
(11.52)
40.18
(15.83)
3.15
(0.57)
*cultivated resource.
a
Metabolizable energy (ME
h
kcal/g OM) was calculated as = [(4 x %TNC) + (4 x %CP) + (9 x %lipid) + (1.389 x
%NDF)]/100 (Conklin-Brittain et al. 2006).
Table 3 Nutritional composition of Tonkean macaque fruit diet compared to other frugivorous catarrhines
Protein (%) Lipids (%) NDF (%) ADF (%) Ca (%) P (%)
Results from this study
Forest fruit mean 5.39 5.39 34.42 27.60 0.48 0.14
(SD) (1.75) (6.65) (12.4) (9.48) (0.53) (0.08)
Other studies
Cercopithecus ascanius
a
19 – 19.8 7.9 – 9.2 — 16.3 – 27.5 0.58 – 0.86 0.25 – 0.31
Cercopithecus ascanius
b
16.6 – 17.6 3.4 – 3.6 31.3 – 31.7 19.0 – 19.7 — —
Cercopithecus mitis
b
16.2 – 17.6 2.6 – 3.5 32.3 – 33.2 19.9 – 20.2 — —
Lophocebus albigena
b
15.7 – 16.3 3.4 – 3.9 32 – 33.9 19.8 – 21.0 — —
Macaca fuscata
c
3.21 – 15.74 — 5.7 – 95.28 — — —
Macaca fuscata
d
3.89 - 15.55 1.46 - 9.56 39.35 - 64.78 — — —
Macaca silenus
e
4.9 - 6.9 — 44.8 - 54.4 — 0.31 0.20
Pan troglydytes
f
9.5 2.5 33.6 19.6 — —
NRC suggested requirements
g
Macaca sp. 8 — 10 5 0.55 0.33
All values are percent dry matter.
a
Rode et al. (2006b)—Range for three study groups at Kibale National Park, Uganda; Values represent plant and
insect portion of the diet; Ripe fruit comprised 44.6% – 59.7% of the diet (Chapman and Chapman, 2000).
b
Conklin et al. (1998) –Kibale National
Park, Uganda; Range of means for two study groups per primate species; Values represent the plant portion of the diet; Fruit comprised 20 – 25% of
their diets.
c
Iwamoto (1982)—Koshima Islet, Japan; Values represent fruit and seed portion of the diet (between 19.7% – 63.2% of the diet); NDF
estimated by multiplying crude fiber by 2 (cf. NRC 2003).
d
Jaman et al. (2010)—Semi-naturally forested enclosure, Primate Research Institute,
Inuyama, Japan; Values represent fruit portion of the diet; Fruit/nuts comprised 39.5% of feeding records.
e
Dierenfeld and McCann
(1999)—Semi-free ranging on St. Catherine’s Island, Georgia, USA; Range represents temporal variation; Values represent fruit portion of the diet;
Fruit comprised 40.2% of feeding time; Median values reported for mineral content.
f
Conklin et al. (1998)—Kibale National Park, Uganda; Mean of
one group; Values represent the plant portion of the diet; Fruit comprised approximately 75% of the diet.
g
Recommended values for overall diet
(NRC 2003).

RILEY EP et al.: Nutritional content cacao crop raiding 165
varied widely between 24.87%-50.48% OM, with a mean
of 34.12%±9.81%. Indigestible fiber (ADF) also ranged
widely between 22.82%–51.08% OM, with a mean of
39.86% ±11.97%. Protein content also varied widely
between 6.38%–12.26% OM (mean: 8.66% ± 2.28%).
Metabolizable Energy showed the least variation across
fig species (range: 2.82–3.42, mean: 3.42% ± 0.37%).
2.2 Nutritional composition: Cacao pulp compared
to forest fruits
Mean protein content was significantly higher in forest
fruits compared to cacao pulp (t = 4.899, df = 12, P <
0.001; Fig. 2). Lipid content was also significantly
higher in forest fruits (6.93%) than in cacao pulp
(0.06%) (t= 5.472, df = 12, P <0.001). Forest fruits were
significantly higher in NDF (t= 10.751, df = 12, P <.001)
and ADF (t = 11.6, df = 12, P <0.001) and significantly
lower in TNC (t = -11.030, df = 12, P <0.001). There
was no significant difference in phosphorus content (t =
-0.492, df = 12, P = 0.632) but on average calcium was
significantly higher in forest fruits (0.48% DM) com-
pared to cacao (0.04%; t = 4.235, df = 12, P = 0.001).
Mean energy content of forest fruits, as measured by
ME, was significantly lower than cacao pulp ME (Table
3; t = -3.593, df = 11, P = 0.004).
Fig. 2 Comparison of mean values of protein, NDF, ADF
and TNC for forest fruits with cacao pulp values
Error bars represent standard error of the mean. ** P ≤0.001
2.3 Nutritional composition: Arenga pinnata fruit
compared to other forest fruits and cacao pulp
In comparison to other forest fruits, Aren palm fruit
was lower in protein and lipid content, ME, and indi-
gestible fiber (ADF) and higher in fiber content (NDF)
and digestible carbohydrates (TNC) (Table 2). Calcium
bioavailability was substantially higher in Aren palm
fruit (CA: P ratio = 35.4) compared to the other forest
fruit (range: 0.21–15.44). Compared to cacao pulp, Aren
palm fruit was higher in protein and fiber content (NDF)
and lower in ME and digestible carbohydrates (TNC).
The indigestible fiber (ADF) of Aren palm fruit, al-
though higher than that of cacao pulp, was the lowest
among the forest fruits (14.84% DM; Table 2) and lower
than values reported for other frugivorous catarrhines
(Table 3).
3 Discussion
As part of our assessment of the nutritional value of
Tonkean macaque foods, we found that the sampled
portion of the Tonkean macaque fruit diet was similar in
overall nutritional composition to that reported for other
macaque taxa (Table 3). Although fiber is often consid-
ered to be a marker of a low quality diet, there is evi-
dence that it serves a beneficial role in the diets of a
number of herbivores, including nonhuman primates.
For example, Sakaguchi et al. (1991) demonstrated that
captive rhesus and Japanese macaques are able to digest
46.3% of a high-fiber diet, indicating that as hindgut
fermenters they show considerable capacity to gain en-
ergy from fiber (cf. Conklin-Brittain et al., 1998). The
fact that the mean fiber content (NDF) was similar to
the fruit (or plant) diet of Japanese macaques, as well as
other frugivorous primates, suggests that Tonkean ma-
caques may also be able to benefit from a high-fiber diet.
At the same time, primates undoubtedly benefit from
the inclusion of easily digested foods in their diets. Fig
fruits, which are a consistent and staple food for
Tonkean macaques (Riley, 2007a), were lower in fiber
content and higher in digestible carbohydrates compared
to other forest fruit species. These factors combined
with the greater availability of calcium in figs may ex-
plain why Tonkean macaques are attracted to them (cf.
Conklin and Wrangham, 1994; O’Brien et al., 1998).
A key goal of this study was to determine whether
nutritional factors play a role in cacao crop raiding. Be-
cause cacao pulp had significantly lower levels of pro-
tein and lipids compared to forest fruits, Tonkean ma-
caques do not appear to be attracted to cacao fruit for
these nutritional components. In other wildlife taxa,
studies show that fiber concentrations of crops are
linked to food choice and crop raiding behavior. For
example, Rode et al. (2006b) found that crops raided by
elephants in Kibale National Park, Uganda have lower
fiber concentrations than do wild foods. The findings in
our study show that cacao pulp has significantly lower
fiber levels and significantly higher levels of digestible
carbohydrates, indicating greater digestibility compared
to forest fruits. Cacao pulp also had, on average,

166 Current Zoology Vol. 59 No. 2
significantly greater energy content than did the forest
fruits sampled. The fact that cacao fruits are spatially
concentrated and available throughout the year, along
with the findings on its digestibility and energy content,
may explain why Tonkean macaques are attracted to this
cultivated resource.
Since fruit is typically a poor source of mineral nutri-
tion (NRC 2003), the mineral concentrations of culti-
vated resources may also explain crop raiding by wild-
life. For example, in addition to having lower fiber
concentrations, crops raided by Kibale elephants had
higher sodium concentrations than wild foods (Rode et
al., 2006a). In our study, we measured the levels of two
minerals: Calcium and Phosphorus. Mineral nutrition,
specifically of Ca and P, is an unlikely contributing fac-
tor to macaque crop raiding for three reasons. First, forest
fruits do not appear to be deficient in calcium; in fact,
the mean value was comparable to that recommended
by the NRC (2003) for the overall diet. Second, cacao
pulp has very low concentrations of calcium compared
to forest fruits (CA: P = 0.28). Third, cacao pulp and
forest fruits show comparable phosphorus levels.
In-situ conservation efforts for nonhuman primates
may be aided by integrating information on feeding
ecology with data on nutritional composition of known
food resources. Although nonhuman primates often raid
crops regardless of pattern of forest fruit availability
(Naughton-Treves et al., 1998; Riley, 2007b; but see Siex
and Struhsaker, 1999), the initial development of crop
raiding is linked to a decrease in natural forage available
as a result of agricultural expansion (Strum, 1994). In
any given primate species, not all groups whose ranges
include matrix (agriculture-forest) habitat will raid crops
(e.g., Riley, 2007b). Therefore, the deliberate protection
of key tree species known to be regularly consumed and
of considerable nutritional value to primates may pre-
vent against the future development of crop raiding and
contribute to broader conservation efforts. A number of
forest fruit species stand out from our sample. For ex-
ample, figs have considerable nutritional value and
show high densities in many areas of Sulawesi [e.g.,
23.3 trees/ha, Karaenta Nature Reserve, South Sulawesi
(Zahrah 1988 as cited in Supriatna et al., 1992); 34.8
trees/ha, Lore Lindu National Park (Riley, 2010a)],
making them an important staple food for the Sulawesi
macaques, and hence worthy of deliberate protection.
Elmerillia tsiampacca, a preferred Tonkean macaque
food, also stands out in our dataset for its high nutri-
tional quality (i.e., high protein, lipid, and energy con-
tent with medium levels of fiber; Table 2). Villagers in
Lindu frequently seek out this species for the construc-
tion of dug-out canoes (Riley, 2007b), and thus it may
be overselected in forest areas close to villages. Given
their generally positive perceptions toward macaques
(Riley and Priston, 2010), villagers could be willing to
switch to other sources of timber if they knew how im-
portant E. tsiampacca is to the diet of Tonkean ma-
caques. The third forest fruit species that should be ac-
tively protected in Sulawesi is Arenga pinnata, the sugar
palm. Previous research has found that aren palm fruit is
an important food for Tonkean macaques; the fruit is
available year round and 52% of the feeding records of
one study group was devoted to this fruit (Fig. 3; Riley,
2007a). The importance of this species for the Tonkean
Fig. 3 Arenga pinnata palms (right) and Availability of ripe Aren palm fruit and percent feeding records devoted to Aren
fruit over an 11-month period (Riley, unpublished data)
Percent feeding records derive from scan sampling data. The ripe fruit index was calculated by multiplying the mean abundance score by the density
of Aren palms in the habitat (Riley, 2007a).

RILEY EP et al.: Nutritional content cacao crop raiding 167
macaque diet and aspects of its nutritional composition
led us to consider how the Aren palm might be incorpo-
rated into a crop raiding mitigation strategy.
The mitigation strategy of buffer cropping has fo-
cused almost exclusively on crops that are not attractive
to wildlife (Parker and Osborn, 2006; Southworth et al.,
2010; Hockings and Humle, 2009). This focus is under-
standable; it is assumed that buffer crops that are highly
palatable will attract wildlife to the edge and increase
the likelihood of crop raiding once the buffer crop has
been depleted (e.g., Goldsmith, 2005). Indeed, if crops
have not already been planted in an area, recommended
crop raiding prevention strategies include planting less
palatable crops near the forest edge (Naughton-Treves et
al., 1998) and setting up an appropriately wide enough
buffer zone to discourage crossing into croplands (e.g.,
more than 20m from the edge for Buton macaques;
Priston et al., 2012). However, in situations where raid-
ing is already occurring, we argue that another strategy
may prove effective: establishing exploitable buffer
zones. There is some evidence that this strategy can be
successful: in Costa Rica, fruiting live fence trees and
plots planted with plantains were successful in drawing
white faced capuchins Cebus capucinus away from im-
portant cash crops (Baker and Shutt, 2005). The deci-
sion regarding what diversionary crop to plant may be
aided by foraging theory. As a foraging strategy, crop
raiding comes with costs and benefits (Strum, 1994).
The consumption of cacao pulp makes sense from a
foraging efficiency perspective; it has a high caloric
content, is easily digested, and exists in a dense spatial
distribution. However, because cacao pods have a thick
outer skin that encases the seeds and pulp, it is a
hard-to-process food. Therefore, in Sulawesi a mitiga-
tion strategy could involve planting palatable resources
that mimic the caloric density and spatial distribution of
cacao but that are easier to process.
Re-enter, Arenga pinnata palm fruit. The sugar palm
grows naturally throughout the Asian humid tropics, but
it can also be easily cultivated (Mogea et al., 1991). In
areas where macaques are already attracted to cacao
gardens, the goal of planting either a corridor or plot of
Aren palms would be to satiate the macaques so that
their foraging impact on cacao is reduced. As noted ear-
lier Aren fruit is an important food for Tonkean ma-
caques and because the fruit is available throughout the
year, it could mimic the perennial nature of cacao fruit-
ing. Although Aren palm had a lower percentage of di-
gestible carbohydrates compared to cacao pulp (Table 2),
it had the lowest level of indigestible fiber (ADF %DM)
of all the forest fruits sampled. In fact, its ADF level
(14.84% DM) is more similar to cacao (4.68%) than to
the forest fruit mean (28.67%), making it an attractive
choice among forest fruit species for a buffer crop. The
overall difference in digestibility between Aren fruit and
cacao pulp may be offset by Aren fruit’s higher protein
content and the fact that it is substantially easier to
process than cacao. Aren fruit is also an important
source of calcium as illustrated by its very high Ca: P
ratio, a measure of calcium availability (Table 2). A high
Ca: P ratio (that is, due to low P intake) has been linked
with positive health benefits. For example, it has been
shown to be favorable for bone mineralization in adult
rats (Koshihara et al., 2005). This attribute might also
contribute to the attractiveness of Aren palm fruit as a
buffer crop.
We also consider the Aren palm a good candidate for
buffer cropping because it has considerable cultural and
economic significance for villagers throughout much of
Sulawesi. The Aren palm is considered such an impor-
tant resource because of the many benefits and products
it provides (Mogea et al., 1991). For example, hair fi-
bers from the stalks are used to construct household
items and the sap from the tree is tapped to produce
palm sugar or various forms of alcoholic beverages.
Many of these products are easily obtained from the tree
without its felling, thus human use does not necessarily
inhibit macaque use. If people were afforded access to
the Aren palm as a buffer crop, it is highly likely that
this planting strategy would garner local support; an
important factor to consider when designing efforts to
mitigate human-wildlife conflict (Dickman, 2010). Be-
cause the Aren palm is easily propagated, it has already
attracted attention for use in agroforestry systems and in
efforts to rehabilitate unproductive and erosion-prone
sites (Mogea et al., 1991). The main limitations to using
the Aren palm as a buffer resource are the number of
years before flowering (10–12 years) and its relatively
short life span (12 –20 years) (Mogea et al., 1991).
In this study we examined the nutritional value of the
fruit portion of the Tonkean macaque diet and demon-
strated the significance of nutritional and feeding ecolo-
gy for understanding and mitigating crop raiding be-
havior. There is still much to learn regarding the nutri-
tional ecology of Tonkean macaques. Future research
could 1) expand the nutritional analysis to include other
important food items such as insects, leaves, and other
vegetative matter; 2) measure potential variation in nu-
tritional composition of the diet across space and time
(Chapman et al., 2003); and, 3) examine concentrations

168 Current Zoology Vol. 59 No. 2
of other important minerals such as sodium, potassium,
and magnesium to fully understand the role mineral
nutrition plays in the feeding ecology of macaques, in-
cluding the consumption of cultivated resources. The
effectiveness of crop raiding mitigation strategies, such
as buffer cropping, also needs to be tested in the field.
For example, because the establishment of an exploit-
able buffer could potentially result in negative effects
(e.g., attracting more macaques), we recommend that
the use of the Aren palm as a buffer resource first be
tested for effectiveness in a limited area before being
implemented more broadly.
Acknowledgements We thank the Kementerian Negara
Riset dan Teknologi Republik Indonesia (RISTEK) for per-
mission to conduct research in Indonesia, and the Departmen
Kehutanan and Balai Taman Nasional Lore Lindu for permis-
sion to work in the park. Riley is indebted to Dr. Bambang
Suryobroto for serving as her Indonesian research counterpart.
We are grateful for the financial support provided by San
Diego State University’s University Grant Program. Our
thanks go to Ellen Dierenfeld and Erin Vogel for their help
with analysis and interpretation of the nutritional data and
Karyn Rode for constructive comments on a previous draft of
this manuscript. We also thank the reviewers (Vincent Nijman
and 1 anonymous reviewer) for their insightful critiques. Ri-
ley’s field assistants, Manto, James, Papa Denis, Pias, Pak
Asdi, and Jaima Smith deserve special thanks and recognition.
References
Akiyama T, Nishio A, 1996. Indonesia's Cocoa Boom: Hands-off
Policy Encourages Smallholder Dynamism. World Bank
Policy Research Working Paper No. 1580. Washington DC:
World Bank.
AOAC, 1990. Official Methods of Analysis. Arlington, VA:
Association of Official Analytical Chemists.
Baker M, Schutt A, 2005. Managing monkeys and mangos. In:
Paterson JD, Wallis J ed. Commensalism and Conflict: The
Human-Primate Interface. Norman, OK: American Society of
Primatologists, 445–463.
Chapman CA, Chapman LJ, Rode KD, Hauck EM, McDowell LR,
2003. Variation in the nutritional value of primate foods:
Among trees, time periods, and areas. International Journal of
Primatology 24(2): 317–333.
Conklin-Brittain N, Wrangham R, Hunt KD, 1998. Dietary
responses of chimpanzees and cercopithecines to seasonal
variation in fruit abundance. II. Macronutrients. International
Journal of Primatology 19(6): 971–998.
Conklin-Brittain NL, Knott CD, Wrangham R, 2006. Energy
intake by wild chimpanzees and orangutans: Methodological
considerations and a preliminary comparison. In: Hohmann
GM, Robbins MM, Boesch C ed. Feeding Ecology in Apes and
Other Primates. Cambridge: Cambridge University Press,
445–471.
Conklin NL, Wrangham RW, 1994. The value of figs to a hind-gut
fermenting frugivore: A nutritional analysis. Biochemical
Systematics and Ecology 22(2): 137–151.
Dickman AJ, 2010. Complexities of conflict: The Importance of
considering social factors for effectively resolving human-
wildlife conflict. Animal Conservation 13: 458–466.
Dostaler S, Oueleet JP, Therrien JF, Cote SD, 2011. Are feeding
preferences of white-taile deer related to plant constituents?
Journal of Wildlife Management 75(4): 913–918.
Forthman Quick DL, Demment MW, 1988. Dynamics of
exploitation: Differential energetic adaptations of two troops of
baboons to recent human contact. In: Fa JE, Southwick CH ed.
Ecology and Behavior of Food-Enhanced Primate Groups.
New York: Alan R. Liss, 25–51.
Goldsmith M, 2005. Impacts of habituation for ecotourism on the
gorillas of Nkuringo. Gorilla Journal 30: 11–13.
Hill CM, 2000. Conflict of interest between people and baboons:
Crop raiding in uganda. International Journal of Primatology
21(2): 299–315.
Hockings KJ, Anderson JR, Matsuzawa T, 2009. Use of wild and
cultivated foods by chimpanzees at Bossou, Republic of
Guinea: Feeding dynamics in a human-influenced environment.
American Journal of Primatology 71: 636–646.
Hockings KJ, Humle T, 2009. Best Practice Guidelines for the
Prevention and Mitigation of Conflict between Human and
Great Apes. Gland, Switzerland: IUCN/SSC Primate Specialist
Group (PSG).
IUCN, 2011. IUCN Red List of Threatened Species. Version
2011.1. Downloaded on 24 October 2011.
Koshihara M, Katsumata S, Uehara M, Suzuki K, 2005. Effects of
dietary phosphorus intake on bone mineralization and calcium
absorption in adult female rats. Bioscience, Biotechnology, and
Biochemistry 69(5): 1025–1028.
McKinney T, 2011. The effects of provisioning and crop-raiding
on the diet and foraging activities of human-commensal white-
faced capuchins Cebus capucinus. American Journal of Prima-
tology 73: 439–448.
Mogea J, Seibert B, Smits W, 1991. Multipurpose palms: The
sugar palm Arenga pinnata (Wurmb) Merr. Agroforestry
Systems 13: 111–129.
Naughton-Treves L, Treves A, Chapman C, Wrangham R, 1998.
Temporal patterns of crop-raiding by primates: Linking food
availablility in croplands and adjacent forest. Journal of Applied
Ecology 35: 596–606.
NRC, 2003. Nutrient Requirements of Nonhuman Primates.
Washington DC: National Academies Press.
O'Brien TG, Kinnaird MF, Dierenfled ES, Conklin-Brittain N,
Wrangham R, et al., 1998. What's so special about figs? Nature
392: 668.
Osborn FV, 2004. Seasonal variation of feeding patterns and food
selection by crop-raiding elephants in Zimbabwe. African
Journal of Ecology 42: 322–327.
Osborn FV, Hill CM, 2005. Techniques to reduce crop loss:
Human and technical dimensions in Africa. In: Woodroffe R,
Thirgood S, Rabinowitz A ed. People and Wildlife: Conflict or
Coexistence? Cambridge: Cambridge University Press, 72–85.
Parker GE, Osborn FV, 2006. Investigating the potential for chili
Capsicum annuum to reduce human-wildlife conflict in Zim-
babwe. Oryx 40(3): 1–4.
Priston NEC, Wyper RM, Lee PC, 2012. Buton macaques Macaca

RILEY EP et al.: Nutritional content cacao crop raiding 169
ochreata brunnescens: Crops, conflict, and behavior on farms.
American Journal of Primatology 74: 29–36.
Riley EP, 2007a. Flexibility in diet and activity patterns of
Macaca tonkeana in response to anthropogenic habitat
alteration. International Journal of Primatology 28(1): 107–133.
Riley EP, 2007b. The human-macaque interface: Conservation
implications of current and future overlap and conflict in Lore
Lindu National park, Sulawesi, Indonesia. American Anthro-
pologist 109(3): 473–484.
Riley EP, 2008. Ranging patterns and habitat use of Sulawesi
tonkean macaques Macaca tonkeana in a human-modified
habitat. American Journal of Primatology 70(7): 670–679.
Riley EP, 2010. The endemic seven: Four decades of research on
the Sulawesi macaques. Evolutionary Anthropology 19: 22–36.
Riley EP, Fuentes A, 2011. Conserving social-ecological systems
in Indonesia: Human-nonhuman primate interconnections in
Bali and Sulawesi. American Journal of Primatology 73:
62–74.
Riley EP, Priston NEC, 2010. Macaques in farms and folklore:
Exploring the human-nonhuman primate interface in Sulawesi,
Indonesia. American Journal of Primatology 72: 848–854.
Rode KD, Chapman CA, McDowell LR, Stickler C, 2006a.
Nutritional correlates of population density across habitats and
logging intensities in redtail monkeys Cercopithecus ascanius.
Biotropica 38(5): 625–634.
Rode KD, Chiyo PI, Chapman CA, McDowell LR, 2006b.
Nutritional ecology of elephants in Kibale National Park,
Uganda, and its relationship with crop-raiding behavior.
Journal of Tropical Ecology 22: 441–449.
Sakaguchi E, Suzuki K, Kotera S, Ehara A, 1991. Fibre digestion
and digesta retention time in macaque and colobus monkeys.
In: Kimura T, Takanaka O, Iwamoto M ed. Primatology Today.
New York: Elsevier Science, 671–674.
Siex KS, Struhsaker TT, 1999. Colobus monkeys and coconuts: A
study of perceived human-wildlife conflicts. Journal of
Applied Ecology 36: 1009–1020.
Sokal RR, Rohlf FJ, 1981. Biometry. 2
nd
edn. San Francisco: W.H.
Freeman & Co.
Southworth J, Hartter J, Binford MW, Goldman A, Chapman CA
et al., 2010. Parks, people and pixels: Evaluating landscape
effects of an East African National Park on its surroundings.
Tropical Conservation Science 3(2): 122–142.
Stephen DW, Krebs JR, 1986. Foraging Theory. Princeton, New
Jersey: Princeton University Press.
Strum SC, 1994. Prospects for management of primates pests.
Revue D'Ecologie (La Terre et La Vie) 49: 295–306.
Supriatna J, Froehlich JW, Erwin JM, Southwick CH, 1992.
Population, habitat and conservation status of Macaca maurus,
Macaca tonkeana and their putative hybrids. Tropical
Biodiversity 1(1): 31–48.
Urquiza-Haas T, Serio-Silva JC, Hernandez-Salazar LT, 2008.
Traditional nutritional analyses of figs overestimates intake of
most nutrient fractions: A study of Ficus perforata consumed
by howler monkeys Alouatta palliata mexicana. American
Journal of Primatology 70: 432–438.
van Soest PJ, 1994. Nutritional Ecology of the Ruminant. Ithaca,
NY: Cornell University Press.
Yuwono EH, Susanto P, Saleh C, Andayani N, Prasetyo D et al.,
2007. Guidelines for the Better Management Practices on
Avoidance, Mitigation, and Management of Human-Orangutan
Conflict in and around Oil Palm Plantations. Jayapura:
WWF-Indonesia.