Received: 4 January 2019 | Revised: 2 August 2019 | Accepted: 13 August 2019 DOI: 10.1002/ajp.23044

RESEARCH ARTICLE

Nutritional composition of the diet of the western gorilla (Gorilla gorilla): Interspecific variation in diet quality

Jessica L. Lodwick1,2 | Roberta Salmi1,3

1Interdepartmental Doctoral Program in Anthropological Sciences, Stony Brook Abstract University, Stony Brook, New York To meet nutritional needs, primates adjust their diets in response to local habitat 2Department of Ecology and Evolutionary differences, though whether these dietary modifications translate to changes in Biology, University of Connecticut, Storrs, Connecticut dietary nutrient intake is unknown. A previous study of two populations of the 3Department of Anthropology, University of mountain gorilla (MG: Gorilla beringei) found no evidence for intraspecific variation in Georgia, Athens, Georgia the nutrient composition of their diets, despite ecological and dietary differences Correspondence between sites. One potential explanation is that nutritional variability in primate diets Jessica L. Lodwick, Department of Ecology and Evolutionary Biology, University of requires greater ecological divergence than what was captured between MG sites, Connecticut, Storrs, Connecticut 06269. underpinning environmental differences in the nutrient quality of plant foods. To test Email: [email protected] whether Gorilla exhibits interspecific variation in dietary composition and nutrient Funding information intake, we studied the composition and macronutrients of the western gorilla (WG: National Science Foundation Directorate for Social, Behavioral and Economic Sciences; The Gorilla gorilla) staple diets and compared them with published data from the two MG Leakey Foundation, Grant/Award Numbers: populations. We recorded feeding time and food intake of four adult female WGs General Research Grant, Great Ape – Fellowship; Conservation International, Grant/ from one habituated group over a period of 11 months (December 2004 October Award Number: Margot Marsh Biodiversity 2005) at the Mondika Research Center, Republic of Congo, allowing for assessment Fund; U.S. Fish and Wildlife Service, Grant/ Award Number: Great Ape Conservation Fund of seasonal patterns of nutrient intake. Staple diets of WGs and MGs diverged in their dietary and macronutrient composition. Compared to MGs, the staple diet of WGs (by intake) contained higher proportions of fruit (43%) and leaf (12%) and a lower proportion of herb (39%), resulting in a higher intake of total nonstructural carbohydrate and fiber and a lower intake of crude protein. Staple gorilla fruits and herbs differed in nutrient quality between sites. Gorillas exhibit nutritional flexibility that reflects ecological variation in the nutrient quality of plant foods. Since dietary quality typically affects rates of growth and reproduction in primates, our results suggest that interspecific differences in nutrient intake and food quality may shape differences in gorilla nutrient balancing and female life history strategies.

KEYWORDS ape, ecology, feeding, nutrition, primate

1 | INTRODUCTION Davies, Oates, & Dasilva, 1999; Potts, Watts, & Wrangham, 2011), socioecology (Boinski, 1999; Borries, 1993; Sterck, Watts, & van For primates with broad geographical distributions, habitats typically Schaik, 1997), and life history traits (Borries, Koenig, & Winkler, vary widely in their climates, plant communities, and resource 2001; Stoinski, Perdue, Breuer, & Hoff, 2013). To meet nutritional characteristics, leading to intraspecific and interspecific differences needs, primates may respond to environmental variability by in primate diets (Chapman & Chapman, 1999; Chapman et al., 2004; adjusting activity budgets, feeding behavior, and dietary composition

Am J Primatol. 2019;e23044. wileyonlinelibrary.com/journal/ajp © 2019 Wiley Periodicals, Inc. | 1of16 https://doi.org/10.1002/ajp.23044 2of16 | LODWICK AND SALMI

(Dasilva, 1992). For example, baboons (Papio spp.) display eclectic 1992). They have also been shown to drive socioecological variation and geographically variable diets (Hill & Dunbar, 2002) reflective of within and among closely related species, including differences in local habitat differences (Swedell, 2011), including variations in dispersal patterns (Barton et al., 1996; Boinski, 1999), feeding altitude, climate, vegetation, food quality, and predation risk and ranging behavior (Chapman et al., 2004; Harris & Chapman, (Altmann, 1998; Barton, Byrne, & Whiten, 1996; Byrne, Whiten, 2007), and female sociality (Barton et al., 1996; Borries, Sommer, & Henzi, & McCulloch, 1993). In some populations, fruit and seeds Srivastava, 1991; Koenig, Beise, Chalise, & Ganzhorn, 1998). Differences accounted for more than half of baboon feeding time (e.g., P. anubis in in food quality among sites may also drive variation in primate group Ghana and Ethiopia; P. cynocephalus in Tanzania), whereas in others size, as suggested in black‐and‐white colobus monkeys (Fimbel, Vedder, underground plant parts composed the bulk of feeding time (e.g., P. Dierenfeld, & Mulindahabi, 2001). To enhance our understanding of cynocephalus in Kenya; P. ursinus in South Africa: Altmann, 1998). how nutritional variation may underpin ecologically‐driven differences While variation in primate diets is well‐documented, the in primate life history and socioecology, we compare the dietary nutritional consequences of such variation may vary greatly, ranging composition and macronutrient intake between two closely related from minimal to pronounced, depending on differences in food species, the western gorilla (WG: G. gorilla) and the mountain gorilla quality. Although blue monkeys (Cercopithecus mitis stuhlmanni) and (MG: G. b. beringei). Gorillas present an ideal taxon for studying how golden monkeys (C. m. kandti) exhibited habitat and dietary ecological variability leads to variation in primate socioecology and life differences in Uganda, the macronutrient content of their 10 most history because they occupy a range of habitats at varying altitudes, commonly eaten foods was remarkably similar (Twinomugisha, resulting in striking environmental differences among sites (Doran & Chapman, Lawes, Worman, & Danish, 2006). Correspondingly, for McNeilage, 2001; Robbins, 2010; Robbins & Robbins, 2018; Rothman, two savannah baboon species in close geographic proximity to one Nkurunungi, Shannon, & Bryer, 2013; Rothman, Pell, Nkurunungi, & another (i.e., P. anubis at Laikipia; P. cynocephalus at Amboseli), as well Dierenfeld, 2006). as groups of black‐and‐white colobus monkeys (Colobus angolensis at The eastern gorilla (Gorilla beringei, incl. MGs) diverged recently Diani Forest) and red colobus monkeys (Procolobus rufomitratus at from the WG (Scally et al., 2012; Thalmann, Fischer, Lankester, Kibale) inhabiting areas with differing degrees of disturbance, dietary Pääbo, & Vigilant, 2006; Xue et al., 2015; Zinner, Groeneveld, Keller, flexibility did not translate to nutritional differences (Altmann, Post, & Roos, 2009). There are two populations of MG situated in close & Klein, 1987; Barton, Whiten, Byrne, & English, 1993; Dunham, proximity (25 km) to one another in the herbaceous highlands of East 2017; Ryan, Chapman, & Rothman, 2012). Nonetheless, when field Africa. The best known population, studied at Karisoke, resides in sites are separated by wide geographic distances and/or diverge in high altitude (2500–3700 m) montane forest of the Virunga their altitude, climate, plant composition, and resource character- Mountains (Harcourt & Stewart, 2007; Robbins, Sicotte, & Stewart, istics, then the macronutrient composition of primate diets may 2001; Watts, 1984). The second population of MGs occupies differ. For example, in a comparison of two Bornean orangutans intermediate‐to‐high altitude (2100–2500 m) areas of the Bwindi (Pongo pygmaeus) populations living in hydrologically distinct peat‐ Impenetrable Forest (Goldsmith, 2002; Robbins, 2003; Rothman, swamp forests (separated by only 63 km), the nutritional content of Plumptre, Dierenfeld, & Pell, 2007). WG, by contrast, inhabit the their diets differed markedly as a result of differences in food quality tropical rainforests of Western Central Africa (Doran & McNeilage, between sites (Vogel et al., 2015). Black‐and‐white colobus monkeys 1998), with populations distributed among the region’s lowland (Colobus angolensis) in a degraded forest also differed in their dietary (<800 m) forests (Goldsmith, 2002). composition and macronutrient intake from those in primary forest, a Compared with the highland habitats of MGs, the lowland finding attributed to differences in food quality and availability rainforests of Central Africa contain lower densities of herbs (Ganas, (Dunham & Rodriguez‐Saona, 2018). More broadly, habitat char- Ortmann, & Robbins, 2009; Goldsmith, 2002; Kuroda, Nishihara, acteristics related to climate and forest composition play an Suzuki, & Oko, 1996; Rogers et al., 2004), higher densities of fruiting important role in determining food quality. For example, the diet of trees, more seasonality in fruit availability (Goldsmith, 2002), and olive baboons (P. anubis) inhabiting a tropical forest site in Uganda greater plant diversity (Lieberman, Lieberman, Peralta, & Hartshorn, (Johnson, Swedell, & Rothman, 2012) contained more fibrous foods 1996; Richards, 1952). Furthermore, researchers have documented compared to that of chacma baboons (P. ursinus) at a temperate site interspecific variability in many aspects of gorilla behavioral ecology, dominated by cultivated land and fynbos vegetation (i.e., a diverse such as differences in diet (Doran‐Sheehy, Mongo, Lodwick, & floral community composed of heaths and shrubs, unique to the Cape Conklin‐Brittain, 2009; Rogers et al., 2004; Watts, 1984; Williamson, region) in South Africa (Johnson, Raubenheimer, Rothman, Clarke, & Tutin, Rogers, & Fernandez, 1990), ranging (Doran‐Sheehy, Greer, Swedell, 2013). Mongo, & Schwindt, 2004; Ganas & Robbins, 2005; Robbins, 2003; For primate species living in diverse habitats, changes in food quality Tutin, 1996; Watts, 1998), rates of reproduction and development and density, among other resource characteristics, have been linked to (Stoinski et al., 2013), and grouping patterns, including differences in intraspecific and interspecific variation in life history traits, such as birth average female group size and the frequency of multimale groups seasonality (Borries et al., 2001; Koenig, Borries, Chalise, & Winkler, (Robbins & Robbins, 2018), female dispersal behavior (Harcourt & 1997), weaning age (Stoinski et al., 2013), and reproductive rates Stewart, 2007; Robbins, Stoinski, Fawcett, & Robbins, 2009; Stokes, (Borries et al., 2001; Newton, 1987; Sommer, Srivastava, & Borries, Parnell, & Olejniczak, 2003), and female social relationships (Lodwick, LODWICK AND SALMI | 3of16

2014; Robbins, 2008; Robbins, Robbins, Gerald‐Steklis, & Steklis, balance nutrient contributions by prioritizing nonprotein energy. A 2005; Stokes, 2004; Watts, 2003). similar nutritional composition of the diet in WGs would indicate Fundamental differences in the quality and availability of fruits either a set of phylogenetically conserved nutrient balancing may explain much of the dietary variation seen in gorillas. While requirements for gorillas or minimal nutrient differences between fruits are somewhat scarce in MG habitats, herbs are abundant, high MG and WG foods. Alternately, if WGs consume sugary fruits and quality, and available year‐round as a predictable source of nutrition other high‐quality foods to a greater degree than do MGs as (Ganas et al., 2009; Watts, 1984). The MG diet in both populations suggested by time‐based measures (Doran‐Sheehy et al., 2009; Masi consists primarily of protein‐rich leaves from a few species of et al., 2009, 2015; Remis, 1997b), then we would expect interspecific abundant terrestrial herbs (Rothman, Pell et al., 2006; Rothman et al., differences in the nutritional composition of gorilla diets. 2013; Watts, 1984). Although MGs at Karisoke sometimes consume Here, we provide original data on WG food and nutrient intake, seasonal raspberries as a very small component of their diet (Watts, using data on feeding time, feeding rates, and the nutritional content 1984), only the Bwindi MGs routinely ingest seasonal fruits (Robbins, of commonly fed/staple foods (i.e., food items that account for >1% of 2008; Rothman, Plumptre et al., 2007). Compared to MGs, WGs eat a the diet) to (a) describe the diversity and composition of the annual greater quantity and diversity of seasonal fruits (Masi et al., 2015; diet, (b) assess the nutritional composition of the diet, and (c) confirm Rothman, Chapman, & Pell, 2008), with fruit and seed comprising whether major food parts vary in nutritional content. We also roughly 70% of feeding time during the high‐frugivory season consider seasonal patterns of nutrient intake in WGs to determine (Doran‐Sheehy et al., 2009; Masi, Cipolletta, & Robbins, 2009). whether intake of total nonstructural carbohydrate peaks during the Although gorillas are highly selective feeders wherever studied high‐frugivory season. Finally, we evaluate whether Gorilla exhibits (Ganas et al., 2009; Rogers, Maisels, Williamson, Fernandez, & Tutin, interspecific variation in diet and nutrition by comparing the 1990; Rothman, Pell et al., 2006; Waterman, Choo, Vedder, & Watts, diversity, composition, and nutrient quality of the staple diet of 1983) and will eat sugary fruits when available (Doran‐Sheehy et al., WGs to published reports on MGs. 2009; Rothman, Dierenfeld, Hintz, & Pell, 2007; Tutin & Fernandez, If measures of dietary composition based on food intake are 1993; Tutin et al., 1991), the amount of fruit in the diet should consistent with those based on feeding times, we expect that WGs depend primarily on the spatial and seasonal availability of ripe fruit will consume more fruit and leaf and less herbs than MGs. in the environment. The nutrient quality of fruits may also vary Furthermore, assuming that the nutrient content of primate foods between sites, which may also affect diet quality and nutrient intake. can vary both within and between sites (Chapman, Chapman, Rode, WGs travel further than MGs in search of ripe fruit, resulting in Hauck, & McDowell, 2003; Ganzhorn & Wright, 1994), we predict longer daily path lengths (Cipolletta, 2004; Doran‐Sheehy et al., that WG and MG staple foods will differ in their average nutrient 2004; Ganas & Robbins, 2005; Goldsmith, 1999; Remis, 1997a), content, resulting in interspecific variation in the nutritional larger home ranges (Cailluad, Ndagijimana, Giarrusso, Vecellio, & composition of gorilla diets. We ground our predictions on previous Stoinski, 2014; Doran‐Sheehy et al., 2004; Ganas & Robbins, 2005; nutritional studies that found fruits consumed by WGs contained less Robbins & McNeilage, 2003), and reduced group cohesion (Salmi & protein and more soluble sugar than did leaves eaten by WGs Doran‐Sheehy, 2014). Yet despite access to energy‐rich fruits and (Doran‐Sheehy et al., 2009; Masi et al., 2015; Popovich et al., 1997; seeds consumed primarily during the high‐frugivory season (Doran‐ Remis, Dierenfeld, Mowry, & Carroll, 2001; Rogers et al., 1990), Sheehy et al., 2009; Masi et al., 2015), wild WGs reproduce at a whereas fruits eaten by Bwindi MGs contained the same amount of slower rate and wean their offspring later than do MGs (Stoinski protein (Rothman, Dierenfeld, Molina, Shaw, Hintz et al., 2006) and et al., 2013), consistent with the hypothesis that WGs evolved a potentially similar sugar content as leaves (in nearby forest: Danish, slower pace of development to cope with heightened ecological risk Chapman, Hall, & Worman, 2006). We conclude by examining how (Janson & van Schaik, 1993) arising from seasonal fluctuations in dietary and nutritional flexibility in gorillas may impact patterns of food availability and quality. WGs also differ from MGs in their nutrient balancing and prioritization, shape female life history grouping patterns, such that average female group size is smaller in strategies, and promote variation in socioecology between the two WGs and their groups tend to contain only one adult male, species. in contrast with the 40% and 45% of MG groups that are multimale in the Virunga and Bwindi populations, respectively (Robbins & 2 | MATERIALS AND METHODS Robbins, 2018). No nutritional differences have been reported between gorilla 2.1 | Study site and subjects populations, though prior nutritional analyses were restricted to the two MG populations in close geographic proximity to one another Research was conducted at the Mondika Research Center (02°21′ (Rothman, Plumptre et al., 2007). In a comparative study, both MGs 859′′N, 016°16′465′′E), a tropical lowland forest site (altitude populations consumed a diet rich in crude protein (CP; Rothman, <420 m; area of 50 km2) situated within the unlogged forest block Plumptre et al., 2007), attributed to the high CP content of their of the Djeke triangle that borders two national parks, Nouabale‐ herbaceous foods. Rothman et al. (2011) have hypothesized that Ndoki and Dzanga‐Ndoki, along the boundary of the Republic of while MGs consume large amounts of protein, they ultimately Congo and Central African Republic. Rainfall (annual mean = 1,600 4of16 | LODWICK AND SALMI mm) and fruit production vary seasonally at the site, with the major prior feeding analyses at the site and is sufficient in size to describe fruiting season (or high‐frugivory period) occurring between June gross patterns of intake over the annual period. and September (Doran et al., 2002; Doran‐Sheehy et al., 2009). We recorded feeding data on four adult females in one social group of 2.2.2 | WG food sample collection and calculation habituated WGs comprising one adult male, six adult females, and of ingestion rates offspring between December 2004 and October 2005. Data on adult females were available as part of a broader study on female social Samples from 40 plant species and one termite species eaten by relationships and feeding competition (Lodwick, 2014). Sex gorillas (Table 1; Table S1) were collected, prepared, and preserved differences in the diet should be minimal within the study group for nutritional analysis as described previously (Doran‐Sheehy et al., based on feeding records from the same period showing that adult 2009), following standard protocols (e.g., Conklin‐Brittain, Wrang- females and the adult male spent similar amounts of time feeding on ham, & Hunt, 1998). These 41 food items accounted for 77% of the fruit and herb (Doran‐Sheehy et al., 2009), despite large differences diet by feeding time (n = 106 food items). Many food items were in body size. For comparative purposes here, we focus analyses on seasonally available for a short duration, which precluded repeated food and nutrient intake and exclude energy intake from sampling of food items over time. Analyses thus cannot account for consideration. seasonal variation in food nutrient quality within and among plants. Food sample preservation was achieved by field‐drying leaves 2.2 | Data collection and analyses and herbs at <60°C and by preserving fruits and termites in unboiled ethanol. Most food samples (63%) were collected directly from trees, 2.2.1 | WG behavioral data: Feeding records lianas, shrubs, herbs, and termite mounds fed upon by study subjects J. L. conducted half‐day focal animal follows (n = 213; mean [SD] within 24 hrs of observation, with a smaller, supplementary collection duration = 4.67 [0.67] hr; Altmann, 1974) for a total of 995 focal made by J. L. in the following year (while following the same group in hours. Throughout follows, behavioral data on the focal animal were the same area). We preserved only the eaten portions of food collected during four 10‐min sampling periods each hour. At 1‐min samples by simulating gorilla processing techniques, including intervals, J. L. recorded focal activity (feed, rest, travel, social) and, stripping leaves from branches, peeling stems to access pith, when appropriate, food item (food part‐species combination). separating fruit pulp from skin and seeds, and shaking termites out Feeding included time spent preparing, processing, chewing, and of clay. The juice from fruit pulp was included as completely as ingesting foods, consistent with Rothman, Plumptre et al. (2007). possible to retain sugars present in preserved fruit samples. We Food parts consumed by gorillas included fruit (pulp, seeds, and skin), included three additional fruit species commonly consumed in other herb (piths, shoots, and roots), leaf (dicotyledonous tree, shrub, and years (Doran‐Sheehy, unpublished data) in basic nutrient compar- liana leaves and young bark from Celtis trees eaten in the same isons by food part (Table 2) to increase fruit sample size, however, we mouthful as leaves), insect (termites from Cubitermes colonies and excluded these fruits from all other analyses. weaver ants), flower (buds and inflorescences), and other (e.g., soil An average fresh weight (g) was determined for each food item scratch and gorilla feces). containing sufficient feeding rate data by weighing at least 10 J. L. recorded individual feeding rates opportunistically during specimens per whole item or handful/mouthful from plants focal feeding bouts as the number of whole items (for pith, shoots, (or termite colonies) visited by the gorillas using a calibrated field roots, and fruits) or handfuls/mouthfuls of food (for leaves and scale. Data on feeding rates and fresh weights were used to calculate insects) ingested per 60 s of observation. In the latter case, we a mean ingestion rate (g/min) for each of the 38 food items approximated the average number of leaves or termites ingested per commonly consumed by gorillas as the average feeding rate (items/ observation period by multiplying the number of handfuls/mouthfuls min) multiplied by the average fresh weight (g/item), accounting for by the average number of units per handful/mouthful, determined 89% of female feeding time (Table 1). from representative samples of terminal branches and termite clay. Following the approach utilized by Vogel et al. (2015) to increase the We acknowledge that variability in the size of representative samples total proportion of diet represented, we substituted data on (a) ingestion should present a source of error in ingestion rates. For most rates (g/min) and (b) macronutrient content from congeneric species commonly consumed foods, a feeding rate was based on at least 10 (where available) or the average value per food part (for leaves, herbs, observations per food, following Koenig et al. (1997). flowers) or part subcategory (for fruits: subdivided by size and form). Since the number of focal follows per month varied across the 11‐ Food items with substituted ingestion rates (n = 30 food items) together month period (mean = 19.45 ± 9.17/month; range = 8–37 follows/ accounted for <10% of average monthly feeding time (Table 1). Food month), we minimize biases in feeding behavior by time of day and intake (i.e., grams of fresh weight consumed) data were accordingly uneven sampling per month by randomly selecting three mornings available for a total of 68 food items (Table 1). From the complete set of and three afternoon samples in each of the 11 months using the food items (n = 106) with feeding time data and the subset of food items sample function in R (R Development Core Team, 2011), resulting in with intake data (n = 68), we calculated mean (+ standard deviation [SD]) a total of 66 follows and 201 focal activity hours. This balanced, monthly feeding time and food intake per food item, respectively, from randomized set of observations (n = 66 follows) is consistent with records compiled across the focal females (n = 4). Though we cannot LODWICK AND SALMI | 5of16

TABLE 1 Staple diet of western gorillas, in descending order of mean percent food intake, compared to mean percent feeding time

Species (staple food items) Family Local name Food Part % Feed Time % Food Intake Haumania danckelmanianaa,b Maranthaceae Genye FR 12.0 12.1 Palisota ambiguaa,b Commelinaceae Doto HE 2.6 9.9 Haumania danckelmanianaa,b Maranthaceae Basele HE 6.2 9.6 Klainedoxa gabonensisa,b Irvingiaceae Bokoko FR 3.9 8.5 Palisota brachythyrsaa,b Commelinaceae Mangabo HE 3.6 6.5 Cubitermes sp.a,b Termitidae Kusu IN 5.6 4.7 Aframomum limbatuma,b Zingiberaceae Njombo HE 8.9 4.2 Gambeya lacourtianaa,b Sapotaceae Bambu FR 0.5 3.9 Unknowna Unknown Ekombe LE 1.8 3.2 Celtis mildbraediia,b,c Ulmaceae Ngombe LE 4.9 3.1 Landolphia sp.1–2d Apocynaceae Pembe‐Ndembo FR 0.8 3.0 Duboscia macrocarpaa,b Tiliaceae Nguluma FR 2.3 2.8 Aframomum subsericeuma,b Zingiberaceae Njokoko HE 3.7 2.1 Pterocarpus soyauxiia,b Papilionaceae Embema FR 0.7 1.9 Ficus sp. 1d Moraceae Ngumu LE 1.4 1.6 Gilbertiodendron dewevreiiva,b Caesalpiniaceae Bemba FR 0.7 1.6 Megaphrynium macrostachyuma,b Maranthaceae Ngungu HE 1.3 1.3 Milicia excelsd Moraceae Mobangi FR 1.6 1.2 Whitfieldia elongataa,b Acanthaceae Indolu LE 2.6 1.1 Unknownd Unknown Nganda LE 1.0 1.0 Anonidium manniia,b Annonaceae Mobei FR 0.7 1.0 Tetrapleura tetrapteraa,b Mimosaceae Ekombolo FR 1.1 1.0 N = 21 species N = 14 families (min) N = 22 food items Abbreviations: FR, fruits; HE, herbs; IN, insects; dicotyledonous leaves. aFood items: Analyzed for macronutrient content. bFood items: Recorded feeding rates (items/min). cFood items: Leaf and bark analyzed separately. dFood items: Substituted feeding rates and macronutrient content. account for individual variation in feeding rates (because of substantial exhibited no differences in rates of energy intake, feeding time, or % of dietary diversity, which limited sample size partitioned among each food feeding time on fruit, herb, or leaf (Lodwick, 2014). Additionally, feeding item and individual), we have no reason to expect individual dietary rate variation among food items should eclipse any individual variation, differences among female subjects in the present study given that they according to Schülke, Chalise, and Koenig (2006).

TABLE 2 Macronutrient and fiber content of fruits (FR), leaves (LE), and herbs (HE) in the WG diet at Mondika

Mean % dry matter (SD)K–W testb Post hoc comparisons Fruit Assaya (n = 18) Leaf (n = 15) Herb (n =7) χ2 p FR vs. LE FR vs. HE LE vs. HE CP 7.4 (4.2) 19.7 (6.3) 15.7 (4.2) 24.1 *** *** *** ns FSS 16.5 (8.6) 2.8 (1.8) 3.7 (1.0) 26.0 *** *** *** (*) TNC 45.0 (21.0) 19.9 (11.2) 11.1 (7.0) 19.4 *** *** *** (*) NDF 42.3 (17.5) 50.7 (10.0) 53.8 (5.7) 4.2 ns ns ns ns HC 11.9 (9.5) 14.1 (6.0) 18.1 (5.4) 6.9 * ns (*) ns ADF 30.4 (15.5) 36.6 (10.0) 35.6 (8.6) 1.3 ns ns ns ns Cs 17.6 (8.5) 20.3 (5.6) 24.5 (6.9) 3.4 ns ns ns ns Ls 12.8 (8.8) 16.3 (8.4) 11.1 (4.3) 2.2 ns ns ns ns aAbbreviations: ADF, acid detergent fiber; CP, crude protein; Cs, cellulose; FSS, free simple sugars; HC, hemicellulose; Ls, lignin; NDF, neutral detergent fiber; ns, not significant; SD, standard deviation; TNC, total nonstructural carbohydrates. bKruskal–Wallis rank sum (K–W) tests were conducted with multiple comparisons adjusted by the Holm method to test for differences among fruits, leaves, and herbs. ***, p < .001; **, .001 > p < .01; *, .01 > p < .05; (*), .05 > p < .10; ns, p > 0.05. 6of16 | LODWICK AND SALMI

2.2.3 | Comparison of WG diet composition based (Conklin & Wrangham, 1994; Conklin‐Brittain et al., 1998). TNC on feeding time and food intake content was calculated after the DM correction as 100 − % NDF+%LP+%CP+%ash(Conklin‐Brittain et al., 1998; Rothman, To characterize WG diet composition, we first calculated the Plumptre et al., 2007). Though there is evidence that the energy proportion of feeding time spent on each food part (i.e., fruit, herb, content of important WG gorilla foods differs between the high‐ leaf, termite) within a follow (e.g., number of fruit‐ or leaf‐feeding and low‐frugivory seasons (Masi et al., 2015) and that gorillas have sample points divided by the total number of feeding sample points). capacious hindguts that permit microbial fermentation of NDF for We then averaged across follows to determine the proportion of energy (e.g., see Remis & Dierenfeld, 2004), comparisons of energy feeding time that gorillas spent on each food part on a daily basis that account for fiber fermentation are not available across MG separately for all food items eaten during the study (n = 106 food sites and will not be made here. items) and for the subset of 68 items that had corresponding food intake data. We followed a similar approach to calculate the average proportion of total food intake composed of each food part on a daily basis as the number of fresh grams ingested from each food part 2.2.5 | Staple food identification and nutritional divided by the total fresh grams ingested for the subset of 68 items, composition of the diet averaged across follows. Following Rothman, Plumptre et al. (2007), we defined and identified staple foods in the WG diet as those items accounting for at least 1% of the overall diet (operationally defined as average 2.2.4 | Macronutrient content of WG foods: monthly fresh weight intake of food items). Once staple food items Chemical assays and nutrient quality were identified, we calculated a nutritionally weighted dry matter We determined macronutrient content of food samples in the contribution of each staple food item to the diet, first within each Nutritional Ecology Lab at Harvard University, Massachusetts, using month and then averaged across months, following a three‐step standard chemical analyses described in previous nutrition studies on process described by Rothman, Plumptre et al. (2007)—Step 1: chimpanzees (e.g., Conklin‐Brittain et al., 1998) and WGs (Doran‐ Convert monthly food intake by fresh weight (g) into field dry Sheehy et al., 2009). Before analysis, all samples were dried at 50°C weight (g) by multiplying fresh weight intake (g) by % FDM; Step 2: in a convection oven and hot‐weighed to account for fluctuations in Convert monthly field dry weight (g) into dry matter intake (g) by moisture content within the lab environment, ground in a Wiley Mill multiplying field dry weight (g) by % DM; and Step 3: Adjust using a Number 2 screen, and divided into two replicates to verify the nutrient content by its % DM (e.g., CP % DM) and multiply the reliability of nutritional assays. CP was quantified using the Kjeldahl adjusted percent nutrient content by dry matter intake (g) to method as total nitrogen × 6.25 (Pierce & Haenisch, 1948). Free calculate amount of nutrient (g) ingested on a dry matter basis. simple sugars (FSS) were measured using colorimetric assay (DuBois, We defined and calculated the nutritional composition of the Gilles, Hamilton, Rebers, & Smith, 1956) with sucrose as a standard overall WG diet as the assemblage of macronutrient concentrations (Strickland & Parsons, 1972). Total fiber content (neutral detergent (i.e., CP, TNC, NDF, HC, and Cs) expressed as a percentage of total fiber [NDF]) and fiber fractions of the total cell wall, including dry matter intake. Specifically, nutrient intake (g) was summed across cellulose (Cs), hemicellulose (HC), and lignin (Ls), were measured staple food items within each month and expressed as a percentage using sequential fiber filtration analysis (Conklin‐Brittain et al., 1998; of total dry matter intake, averaged across months. Van Soest, 1994). Lipids (LP) were measured as the total amount extracted in petroleum ether (AOAC, 1984). Dry matter (DM) and ash content were determined from samples heated at 100°C and 2.2.6 | Interspecific variation in nutritional 520°C, respectively, for 8 hr each (Conklin & Wrangham, 1994; composition of the diet and food quality Conklin‐Brittain et al., 1998). Insoluble nitrogen fractions (that bind with fiber: Rothman et al., 2008) were not analyzed in the present We compared dietary composition (i.e., % fruit, % herb, % leaf, % study because of limited sample material. LP and ash concentrations insect), dietary diversity (i.e., n of food items), mean macronu- contributed to the calculation of total nonstructural carbohydrate trient and fiber content, and the nutritional composition of the (TNC) content (see below). However, LP content of staple foods was staple diet of WGs to published reports from a comparative study excluded from intersite comparisons, given that fat concentrations of MGs at two neighboring sites, Bwindi and Virunga (Rothman, are negligible in gorilla foods and usually do not differ among food Plumptre et al., 2007). Although staple food items accounted for parts (Doran‐Sheehy et al., 2009; Masi et al., 2015). 90–96% of MG diets at both sites, they composed only 85% of the We report mean macronutrient and fiber content of WG staple WG diet in the present study because of greater dietary diversity. food items as a percentage of dry matter (% DM) for comparison Therefore,tocomparestapledietsdirectly,weadjustedthe with nutritional data from MGs. To convert to % DM from % field percentage of food intake attributed to each staple food such dry matter (% FDM), we calculated a DM correction factor for each that the total contribution of staple foods at all three sites food item as the weight of each subsample multiplied by its % DM summed to 100%. LODWICK AND SALMI | 7of16

2.2.7 | Statistical analyses 38 dicotyledonous leaf, 13 herbs, a minimum of 6 insects, and 5 flower items, encompassing a minimum of 96 species across 26 We conducted nonparametric Spearman’s rank correlations (Siegel & families (Table 1). Fruits constituted the most diverse part of the diet Castellan, 1988) on the monthly intake of TNC and CP to test whether (42% of 106 food items). Food intake data (n = 68 of 106 food items) nonstructural carbohydrate intake was negatively associated with protein accounted for a substantial proportion (89%) of the diet based on intake, both overall and during periods of high frugivory and high folivory. feeding time. Of the nonstaple food items with missing food intake To determine whether the composition of the WG diet differed on the data, only two items (i.e., weaver ants and soil scratch, noted for their basis of feeding time versus food intake, particularly the percent of daily time‐consuming extraction) composed >1.0% (but not more than feeding time spent on each food part, we performed paired t tests on 1.5%) of the diet by feeding time (Table S1). Fruit (mean = 39.6% + logit‐transformed dietary proportions (Warton & Hui, 2011). To evaluate 28.6; n = 66 follows) and herb (35.4% + 22.3; n = 66 follows) whether fruits, leaves, and herbs differed in their mean macronutrient comprised the chief portion of the WG diet by intake, with content, we used Kruskal–Wallis rank sum tests (with multiple dicotyledonous leaf (20.0% + 21.4; n = 66 follows) and termite comparisons: family‐wise error rate of alpha = 0.05; p values adjusted (5.0% + 7.12; n = 66 follows) composing the remainder (Figure 1). for the number of comparisons using the Holm method (Holm, 1979). To The nutritional composition of the diet included CP (mean = examine whether WG staple food items differed nutritionally from those 11.9% + 2.9, n = 11 months), TNC (33.5% + 8.8, n = 11 months), and at MG sites, we tested for differences in the mean macronutrient content NDF (53.9% + 6.7, n = 11 months), with nearly equal contributions of (a) staple fruits between the two frugivorous populations using from HC (17.5% + 8.8, n = 11 months) and Cs (19.4% + 5.6, n =11 Mann–Whitney U tests and (b) staple herbs among the three populations months). Monthly intake of TNC was negatively correlated with that using Kruskal–Wallis rank sum tests (Siegel & Castellan, 1988) for small of CP (n = 11 months; Spearman’s rho = −0.74, S = 382.4, p = .0095) and unbalanced data. All statistical tests were two‐tailed at p <.05and and NDF (n = 11 months; Spearman’s rho = −0.67, S = 368.3, p = .023; runusingR(RDevelopmentCoreTeam,2011). Figure 2). During the high‐frugivory period (i.e., June–September in Masi et al., 2009), the percentage of TNC in the WG diet reached its 2.2.8 | Ethics statement for animal studies maximum (max monthly TNC = 49.0% in August; mean TNC = 38.9% + 11.7, median TNC = 42.3%, n = 4 months) and CP its minimum (min Research protocols complied with ASP Principles for the Ethical monthly CP = 8.4% in August; mean CP = 10.0% + 1.4, median CP = Treatment of Primates and were ethically reviewed and approved by 10.0%, n = 4 months; Figure 2). Intake of CP peaked during the high‐ the Stony Brook University Institutional Animal Care and Use folivory period (i.e., December–April in Doran‐Sheehy et al., 2009; Committee. To minimize the risk of zoonotic disease transmission, max monthly CP = 18.4% in March; mean CP = 13.7% + 3.3, median researchers observed animals at a distance of >7 m, following regulations CP = 13.7%, n = 5 months; Figure 2). established for Gorilla (Cooper & Hull, 2017; Homsy, 1999). Based on average monthly feeding time (for n = 106 food items), the WG diet was composed of 35% fruit (±26 SD, 11 months), 30% herb (±18 SD, 11 months), 24% leaf (±22 SD, 11 months), 6% termite 3 | RESULTS (±9 SD, 11 months), and 5% other food items (±11 SD, 11 months). When comparing the diet based on food intake to that based on the 3.1 | Dietary diversity and nutritional composition feeding time subset (n = 68 food items), the former included of the WG diet significantly more fruit and less leaf, but no difference in herb The WG diet was diverse and composed of a minimum of 106 food (Figure 1, paired t test: fruit: df = 65, t = −2.49, p = .020; leaf: df = 65, items (i.e., food part‐species combination), including 44 fruit, at least t = −1.36, p < .001; herb: df = 65, t = −1.36, p = .180). Though the

FIGURE 1 Composition of the western gorilla diet based on mean percent (a) feeding time of all foods (% Feed Time Overall), (b) feeding time of the subset (% Feed Time Subset), and percent food intake 8of16 | LODWICK AND SALMI

FIGURE 2 Monthly and seasonal patterns of variation in WG nutrient intake. Horizontal gray bars denote the high‐frugivory season (Masi et al., 2015). The high‐folivory season occurs between December and April (Doran‐Sheehy et al. 2009). CP, crude protein; NDF, neutral detergent fiber; TNC, total nonstructural carbohydrate

feeding time subset showed no difference in the proportion of 3.3 | Comparison of WGs and MGs diet dietary fruit compared to the overall data on feeding time, the subset The staple diet of Mondika WGs was more diverse than the staple overrepresented the proportion of herb and leaf in the diet (Figure 1, diets of either MG population. It included 22 staple food items, paired t test of complete vs. subset of feeding time data: herbs: compared to 15 and 9 staple food items reported for Bwindi and df = 65, t = −3.76, p < .001; leaves: df = 65, t = −6.05; p < .001; fruits: Virunga, respectively (Rothman, Plumptre et al., 2007). Compared to df = 65, t = 0.23, p = .820). both MG sites, the WG staple diet included a greater number of fruit and dicotyledonous leaf species and fewer herb species (Figure 3a). Gorilla staple diets varied in composition across sites (Figure 2b). At Mondika, the consumption of fruit was nearly three times greater 3.2 | Nutritional differences among WG food parts than that at Bwindi, composing 43.5% of the staple diet, compared to Major food parts in the diet of WGs (fruits: n = 18 species; herbs: 0% at Virunga. Herb consumption at Mondika (39.5% of staple diet) n = 7 species; leaves: n = 15 species) differed significantly in their was much reduced compared to Virunga (100%) and nearly half that average macronutrient content (see Table 2 for statistics). Compared of Bwindi (73.1%; Figure 3b). Dicotyledonous leaves accounted for a to leaves and herbs, fruits contained a higher mean concentration of greater percentage of the staple diet at Mondika (11.8%) compared TNC and FSS and lower CP content (Table 2). Herbs contained to either Bwindi (7.3%) or Virunga (0%; Figure 2b). Although termites similar CP and fiber content as leaves, though the mean concentra- were a staple item at Mondika, insects were not staples at either tion of FSS trended higher and TNC lower in herbs compared to Bwindi or Virunga (Figure 3b). leaves (Table 2). No differences in fiber content were detected When compared to MG diets, the WG diet included nearly twice among food parts, except for a trend for fruits to contain lower mean as much TNC (TNC in WG: 33.5 vs. MG: 18.8, 18.2), and substantially HC content compared to herbs (Table 2). less CP (CP in WG: 11.9 vs. MG: 18.2, 17.2) as a percentage of dry

FIGURE 3 Western gorilla (Mondika) and mountain gorilla (Bwindi; Virunga) staple dietary (a) diversity (N of species) and (b) composition (percent food intake) classified by food part LODWICK AND SALMI | 9of16 matter intake. NDF consumption, particularly the HC fraction, was neighboring MG sites (Rothman, Plumptre et al., 2007), in which no highest at Mondika (Table 3). nutritional differences were indicated. This interspecific nutritional variation, coupled with differences in the predictability of the seasonal food supply, offers an explanation as to one of the primary 3.4 | Interspecific differences in food quality mechanisms driving life history differences between gorilla species. The nutrient quality of staple food items, matched by food part, The WG diet was diverse, as found at other sites, though slightly varied significantly between WG and MG sites (Figure 3). The mean less so on account of the shorter study duration and different concentration of TNC in staple fruits at Mondika (53.1 + 22.5%) was sampling methods, with 106 different food items in the diet significantly higher than at Bwindi (15.4 + 8.3%; the Mann–Whitney compared with 132–239 food items at other sites (Masi et al., U test: df =1, U = 29, p = .010; Figure 4a). Staple herbs at Mondika 2015; Remis, 1997b; Rogers et al., 2004; Williamson et al., 1990). differed little from those at Virunga, but when compared to Bwindi Fruits accounted for a major portion (42%) of dietary diversity, falling (mean: TNC = 23.5 + 7.9; HC = 10.5 + 3.0; NDF = 37.4 + 12.8), Mondi- in the mid‐range of other sites (Remis, 1997b; Williamson et al., ka’s staple herbs contained significantly less TNC (11.0 + 7.6% 1990). WGs at this site differed in their dietary composition, with Kruskal–Wallis rank sum test for TNC: df =2, χ2 = 7.4, p = .025; more time spent feeding on herb (30%) and less on dicotyledonous Mondika vs. Bwindi post hoc test TNC: p = .009) and more NDF leaf and bark (24%), but similar time on fruit (35%), compared to (54.7 + 5.7; Kruskal–Wallis rank sum test for NDF: df =2, χ2 = 7.8, WGs at Bai Hokou, a neighboring site (herb: 10%, leaf: 37%, fruit: p = .025; Mondika vs. Bwindi post hoc test NDF: p = .033) and HC 36%: Masi et al., 2015) separated by 60 km of lowland rainforest (19.1 + 5.3; Kruskal–Wallis rank sum test for HC: df =2, χ2 = 9.2, (Robbins et al., 2016). Differences in the presence/absence of p = .010; Mondika vs. Bwindi post hoc test HC: p = .012; Figure 4b). swamps and herb densities between Mondika and Bai Hokou The CP content of staple herbs was similar across sites (Kruskal–- (Goldsmith, 2002) could potentially explain why the two sites Wallis rank sum test for CP: df =2, χ2 = 1.8, p = .400). differed in the proportions of herb and leaf in the diet. Interannual variation in fruit availability and sex differences in leaf consumption during the high‐folivory period (Doran‐Sheehy et al., 2009) are other 4 | DISCUSSION factors expected to influence WG dietary composition. The percen- tage of TNC in the diet varied between the high‐ (median: 42%) and 4.1 | Dietary and nutritional variation in gorillas low‐ (median: 31%) frugivory season, whereas CP varied much less This comparison of nutrient intake patterns in gorillas inhabiting (median: 10% vs. 14%), consistent with previous reports from the distinct environments further refines our understanding of how neighboring WG site (median TNC: 58% vs. 38%, median CP: 10% vs. ecological variation in plant resource characteristics influences 10% in Masi et al., 2015). primate nutritional variability. Our study revealed that differences We confirmed that WG fruits contained higher concentrations of in food quality and diversity between WG and MG forests have led to soluble sugars and lower protein than did leaves or herbs, as interspecific variation in the nutrient composition of gorilla diets, predicted and shown in other studies (e.g., Doran‐Sheehy et al., 2009; with WGs consuming much more simple carbohydrate and fiber and Masi et al., 2015; Rogers et al., 1990), signaling that WG fruits are less protein than MGs. Evidence suggests that the nutritional selected primarily for their nonprotein energy (NPE) content. Leaves differences demonstrated between WGs and MGs reflect a greater displayed similar protein content as herbs, suggesting that WGs may degree of ecological variation in the nutrient composition of plant use them interchangeably as key protein (i.e., available protein foods than that described in a previous comparison of two energy) sources, and may explain why WGs can consume varying

TABLE 3 Nutrient contributions (percentage of dry matter intake) to the diet of western gorillas (Mondika) in comparison with previously reported values for apes and monkeys

Site Species CP TNC NDF HC Cs Mondika Western lowland gorilla (Gorilla gorilla)a 11.9 33.5 53.9 17.5 19.4 Bwindi Mountain gorilla (Gorilla beringei)b 18.2 18.8 42.9 10.8 17.5 Virunga Mountain gorilla (Gorilla beringei)b 17.2 18.2 41.2 12.5 19.8 Kibale Chimpanzee (Pan troglodytes schweinfurthii)b 9.5 38.8 33.6 13.7 11.8 Salonga Bonobo (Pan paniscus)b 8.3 – 26.8 10.1 11.5 Kibale Blue monkey (Cercopithecus mitis stuhlmanni)b 17.6 35.3 32.3 11.8 12.3 Kibale Red‐tailed monkey (Cercopithecus ascanius schmidti)b 17.6 36.5 31.3 11.4 11.6 Kibale Mangabey (Lophocebus albigena johnstoni)b 16.3 34.0 32.0 12.0 11.9 Abbreviations: CP, crude protein; Cs, cellulose; HC, hemicellulose; NDF, neutral detergent fiber; TNC, total nonstructural carbohydrate. aData come from this study. bData compiled previously in Rothman, Plumptre et al. (2007). 10 of 16 | LODWICK AND SALMI

FIGURE 4 Nutrient quality of WG (Mondika) and MG (Bwindi; Virunga) staple (a) fruits and (b) herbs. Error bars are standard deviations (1 SD) from the mean. CP, crude protein; Cs, cellulose; HC, hemicellulose; NDF, neutral detergent fiber; TNC, total nonstructural carbohydrate

proportions of leaf and herb at different sites without it translating to content on average compared to those eaten by MGs (Rothman, differences in CP intake. Plumptre et al., 2007), signifying that WGs can gain more NPE per Our examination of overall WG dietary composition based on unit of feeding time on fruit than can MGs. The pattern was more food intake indicated that the fresh weight contribution of fruit to nuanced for staple herbs, with clear differences in herb nutrient the diet (40% of intake) was indeed substantial, as previously quality indicated between WGs and MGs at Bwindi, while the suggested by feeding time data (Doran et al., 2002; Masi et al., 2009; comparison of WG and MG herbs at Virunga yielded no statistical Remis, 1997b; Tutin & Fernandez, 1993; Williamson et al., 1990), and difference. In the former comparison, WG herbs were more fibrous supports previous reporting of fruit consumption from a neighboring and contained less TNC than those eaten by MGs, supporting site (36% of diet by feeding time: Masi et al., 2009). While intake and previous work showing that plant parts in tropical lowland forests feeding time data indicated a similar diet composition for WGs, we typically are more fibrous compared to their counterparts in highland also found that feeding time slightly overestimated leaf and under- environments (Goldsmith, 2002; Popovich et al., 1997). However, the estimated fruit consumption. These method‐related discrepancies latter comparison poses a paradox since Bwindi and Virunga did not paralleled patterns reported for howler monkeys and langurs, differ in herb quality, although small sample sizes in pair‐wise although were not as severe in magnitude (Chivers, 1998; Gaulin & comparisons may have limited statistical power. Gaulin, 1982; Hladik, 1977) and could be explained by differences in The exceptionally protein‐rich leaves of staple herbs in the MG processing time between food parts (Hladik, 1977; Kurland & Gaulin, diet can explain the high protein diets recorded for MGs in both 1987; Schülke et al., 2006). populations. Though the mean protein content of the WG diet (% CP: Results further indicated that fruit intake was much higher and 11.9) was lower than would be expected based on MG data herb intake much lower in WGs (Mondika) than in both populations describing CP intake (% CP: 17.2 at Bwindi, 18.2 at Virunga: of MGs (Bwindi, Virunga) and that the staple foods of the two gorilla Rothman, Plumptre et al., 2007) and available protein (AP) intake (AP species differed significantly in nutrient quality, contributing to the contribution to total energy: 16–31%: Rothman, Raubenheimer, & interspecific variation seen in the nutritional composition of their Chapman, 2011), we expect that WGs can easily meet minimum diets. Specifically, staple fruits consumed by WGs had greater TNC protein requirements by feeding steadily on herbs throughout the LODWICK AND SALMI | 11 of 16 year (see Doran‐Sheehy et al., 2009) and prioritizing NPE whenever Wrangham, 1996). Yet despite their greater reliance on herbs, preferred foods are available, such as sugary fruits and preferred bonobos did not consume high fiber diets like gorillas (at least at one herbs. Analysis of fiber‐bound nitrogen (Rothman et al., 2008) in WG site, Salonga; Table 3), though herb nutrient quality and the amount staple foods should be made in the future to improve estimates of of herb in the diet may vary across sites. The potential roles of protein intake and assess whether WGs fall short of meeting protein phylogenetic constraints, as well as site‐specific ecological influences recommendations in humans (i.e., AP contributes to 15% of energy: on the nutritional composition of diets, are mirrored in cercopithe- Saint Joer et al., 2001). cine taxa (Conklin‐Brittain et al., 1998) living in a middle‐elevation forest near MGs. Multiple species consistently exhibited high CP intake similar to MGs, however their high TNC intake reflected a 4.2 | Adaptive strategies of nutrient prioritization greater reliance on sugary foods similar to WGs and chimpanzees and nutrient balancing in gorillas and other apes (Table 3). Our results do not support the hypothesis that Gorilla has a phylogenetically fixed strategy of nutrient optimization, although 4.3 | Impacts of variable nutrient intake on female a critical comparison remains with Grauer’s gorilla (G. b. graueri), as they life history strategies and grouping patterns are a subspecies of eastern gorilla with dietary habits similar to WGs (Yamagiwa, Basabose, Kaleme, & Yumoto, 2005). If we assume that In gorillas, ecological differences related to differences in the gorillas are capable of adjusting the nutritional composition of their predictability of the seasonal food supply (Wright et al., 2015) are diets to shifting ecological parameters, we expect that nutritional thought to explain the slower life histories of WGs (Stoinski et al., differences in gorillas may trigger modifications in nutrient prioritization 2013) and Bwindi MGs (Robbins, Gray, Kagoda, & Robbins, 2009) within the geometric framework (Raubenheimer & Simpson, 1997). compared to Virunga MGs. This explanation borrows from the Future studies should examine whether WGs follow a similar pattern of ecological risk hypothesis (Janson & van Schaik, 1993), in which seasonal NPE prioritization as that reported for MGs (Rothman et al., species subject to fluctuating food availability (e.g., orangutans, WGs) 2011). Masi et al. (2015) recently demonstrated that WG foods eaten are expected to face greater ecological risk than those with a steady during the high‐frugivory period contained significantly more energy food supply (e.g., Virunga gorillas) because they experience periods of than foods eaten during the low‐frugivory period. Thus, compared with unpredictable resource availability and fruit scarcity (Stoinski et al., MGs, we predict a steeper slope between NPE and AP in WG diets 2013). To manage ecological risk, WGs are expected to mature more during the high‐frugivory period, a time of increased energy intake. In slowly and reproduce at lower rates than MGs to minimize the orangutans, high‐frugivory periods have been linked to seasonal protein effects of food shortages on their reproductive physiology. This deficits (Vogel et al., 2012), and subjects select non‐fruit items when hypothesis has received empirical support in gorillas (Robbins, Gray fruits are available (DiGiorgio & Knott, 2017). Though WGs are unlikely et al., 2009; Stoinski et al., 2013), baboons (Barrett, Henzi, & Lycett, to experience seasonal protein deficits due to their year‐round herb 2006), and possibly orangutans, since longer inter‐birth intervals are consumption (Doran‐Sheehy et al., 2009), we urge future researchers to reported among the more frugivorous populations in Sumatra test this and other nutrient balancing hypotheses that require knowl- compared with Borneo (Wich et al., 2004, 2010). Baboons, like edge of AP, which we could not measure due to an insufficiency in the gorillas, show environmental variation between sites in the quality dry mass of our food samples. and predictability of their seasonal food supply, which has similarly Considering the nutritional niche of apes inhabiting the same translated into variation in their life history patterns, including environment, our results challenge the long‐held notion that the diets differences in rates of baboon development and survival (Barrett of gorillas and chimpanzees (Pan troglodytes) diverge substantially et al., 2006). Further investigation into the links between seasonal and instead bolster the proposal that sympatric chimpanzees and patterns of nutrient balancing and life history traits are needed to gorillas share a high degree of niche overlap (Tutin & Fernandez, shed light on the mechanisms by which nutrient intake influences 1993). Here, we presented evidence that the nutritional composition female reproduction in gorillas and other primates. of the WG diet more closely resembles that of chimpanzees (Conklin‐ Our study of gorillas confirmed that the percentage of TNC in the Brittain et al., 1998) than that of MGs (Rothman, Plumptre et al., WG diet fluctuated seasonally, reaching its height during the high‐ 2007) concerning the high TNC:CP ratio in their diets. Similar to frugivory season. Although nutritional differences between WGs and chimpanzees, WGs emphasize selection of sugary fruits when MGs detected in the present study leave unanswered questions about available (Doran‐Sheehy et al., 2009; Masi et al., 2015; Rothman what impact such differences have on gorilla life history parameters, we et al., 2008) and likewise exhibit peaks in NPE (in the form of TNC) propose that pronounced seasonal and interannual variability in fruit during periods of high frugivory (Conklin‐Brittain et al., 1998; Masi availability will influence TNC intake in WGs and may explain why WGs et al., 2015). One key difference is that WGs consume considerably hedge against low TNC periods by slowing down growth and more fiber than do chimpanzees (especially HC), largely due to a reproduction. Dramatic differences in the unpredictability of resource greater reliance on fibrous herbs and leaves. Bonobos (Pan paniscus) availability in the environment, from highly unpredictable in WGs to also consume more herbaceous vegetation compared to chimpanzees highly predictable in Virunga MGs, can explain differences in their (Malenky & Wrangham, 1994; Malenky, Kuroda, Vineberg, & developmental life histories. Additionally, we hypothesize that on a 12 of 16 | LODWICK AND SALMI continuum of unpredictable resource availability, Bwindi MGs face provided by M. Gately and E. Stokes at Wildlife Conservation resource unpredictability that is intermediate between Virunga MGs at Society and D. Greer at World Wildlife Fund and to our talented one end of the continuum and WGs at the other. If so, this modest level field assistants, F. Dodonou, J. Dzoni‐Epeni, A. Mayoke, and W. of seasonal resource unpredictability at Bwindi may be sufficient to Meno, and experienced trackers, Bakombo, Dona, Kete Mokonjo, explain the intermediate rates of development and reproduction seen in Mamandele, Ndima, and Samedi. We finally thank C. Janson, A. the Bwindi population (Robbins & Robbins, 2018) compared with those Koenig,D.Watts,J.Rothman,W.Erb,K.Ossi‐Lupo, and four of WGs and Virunga MGs, despite similar patterns of nutrient intake to anonymous reviewers for their insightful comments, and the latter. M. Szczupider for editorial suggestions. Diet quality may play a significant role in determining grouping patterns in gorillas. Variation in leaf quality among sites was thought to explain intraspecific variation in group size of black‐ CONFLICT OF INTEREST ‐ and white colobus monkeys (Fimbel et al., 2001). MGs groups The authors declare that there is no conflict of interest. contain more adult females on average compared with WGs groups, and up to 45% of MGs groups contain more than one adult ORCID male, compared to 5% in WG groups (Robbins & Robbins, 2018). The protein‐rich herbs available year‐round and in high abundance Jessica L. Lodwick http://orcid.org/0000-0003-1524-6065 in MG habitats may explain some of the interspecific differences Roberta Salmi http://orcid.org/0000-0001-5745-2014 we see in gorilla grouping patterns compared with WGs. Continued study of the relationships between ecological condi- tions, nutrient quality of staple foods, and feeding competition REFERENCES may offer further insight into why gorillas at different sites vary in Altmann, J. (1974). Observational study of behavior: Sampling methods. their grouping patterns. Behaviour, 49, 227–267. https://doi.org/10.1163/156853974X00534 Altmann, S. A. (1998). Foraging for survival: Yearling baboons in Africa. Chicago, IL: University of Chicago Press. 4.4 | Primate foods, diets, and nutrition in the Altmann, S. A., Post, D. G., & Klein, D. F. (1987). Nutrients and toxins of anthropocene plants in Amboseli, Kenya. African Journal of Ecology, 25, 279–293. https://doi.org/10.1111/j.1365‐2028.1987.tb01119.x In the age of global climate change and human‐altered landscapes AOAC (1984). Official methods of analysis. Association of official analytical (Hockings et al., 2015), long‐term data on the quality of wild primate chemists. Arlington, VA: Association of Official Analytical Chemists. foods and the nutritional composition of their diets will be needed to Barrett, L., Henzi, P. S., & Lycett, J. E. (2006). Whose life is it anyway? Maternal investment, developmental trajectories, and life history assess how species respond flexibly to changing nutritional land- strategies in baboons. In L. Swedell, & S. R. Leigh (Eds.), Reproduction scapes. In light of recent findings that the protein quality of tropical and fitness in baboons: Behavioral, ecological, and life history perspectives. leaves has decreased over the past 30 years (Rothman et al., 2015), Boston, MA: Springer. another important objective will be to utilize information on primate Barton, R. A., Byrne, R. W., & Whiten, A. (1996). Ecology, feeding competition and social structure in baboons. Behavioral Ecology and nutrient prioritization and nutritional niches (Machovsky‐Capuska, Sociobiology, 38, 321–329. https://doi.org/10.1007/s002650050248 Senior, Simpson, & Raubenheimer, 2016) to direct conservation Barton, R. A., Whiten, A., Byrne, R. W., & English, M. (1993). Chemical practices aimed at protecting critically important food staples for composition of baboon plant foods: Implications for the interpretation primate survival and reproduction. of intra‐ and interspecific differences in diet. Folia Primatologica, 61(1), 1–20. https://doi.org/10.1159/000156722 Boinski, S. (1999). The social organizations of squirrel monkeys: ACKNOWLEDGMENTS Implications for ecological models of social evolution. Evolutionary Anthropology: Issues, News, and Reviews, 8, 101–112. https://doi.org/10. Funding for the study was provided, as part of a long‐term project 1002/(sici)1520‐6505(1999)8:3<101::aid‐evan5>3.0.co;2‐o Borries, C. (1993). Ecology of female social relationships: Hanuman at the site, by U.S. Fish and Wildlife Services (Great Ape langurs (Presbytis entellus) and the van Schaik model. Folia Conservation Fund), L.S.B. Leakey Foundation (Great Ape Fellow- Primatologica, 61,21–30. https://doi.org/10.1159/000156723 ship; General Research Grant), Conservation International (Mar- Borries, C., Koenig, A., & Winkler, P. (2001). Variation of life history traits got Marsh Biodiversity Fund), and National Science Foundation. and mating patterns in female langur monkeys (Semnopithecus – Research permits were granted by the Ministry of l’Enseignement entellus). Behavioral Ecology and Sociobiology, 50, 391 402. https:// doi.org/10.1007/s002650100391 Primaire, Secondaire, et Superieur Charge de la Recherche Borries, C., Sommer, V., & Srivastava, A. (1991). Dominance, age, and Scientifique (Republic of Congo) and the Ministries of Eaux et reproductive success in free‐ranging female hanuman langurs Forets and Recherches Scientifique (Central African Republic). We (Presbytis entellus). International Journal of Primatology, 12(3), – are indebted to our collaborators, including D. Doran‐Sheehy for 231 257. https://doi.org/10.1007/BF02547586 Byrne, R. W., Whiten, A., Henzi, S. P., & McCulloch, F. M. (1993). her contributions to study design and analysis and N. Conklin‐ Nutritional constraints on mountain baboons (Papio ursinus): Implica- Brittain and P. Mongo for conducting nutritional assays at Harvard tions for baboon socioecology. Behavioral Ecology and Sociobiology, 33, University. We are also grateful for logistical project support 233–246. https://doi.org/10.1007/BF02027120 LODWICK AND SALMI | 13 of 16

Caillaud, D., Ndagijimana, F., Giarrusso, A. J., Vecellio, V., & Stoinski, T. S. Doran‐Sheehy, D. M., Greer, D., Mongo, P., & Schwindt, D. (2004). Impact (2014). Mountain gorilla ranging patterns: Influence of group size and of ecological and social factors on ranging in western gorillas. group dynamics. American Journal of Primatology, 76, 730–746. https:// American Journal of Primatology, 64, 207–222. https://doi.org/10. doi.org/10.1002/ajp.22265 1002/ajp.20075 Chapman, C. A., & Chapman, L. J. (1999). Implications of small scale Doran‐Sheehy, D., Mongo, P., Lodwick, J., & Conklin‐Brittain, N. L. (2009). variation in ecological conditions for the diet and density of red Male and female western gorilla diet: Preferred foods, use of fallback colobus monkeys. Primates, 40, 215–231. https://doi.org/10.1007/ resources, and implications for ape versus old world monkey foraging BF02557712 strategies. American Journal of Physical Anthropology, 140, 727–738. Chapman, C. A., Chapman, L. J., Cords, M., Gathua, J. M., Gautier‐Hion, A., https://doi.org/10.1002/ajpa.21118 & Lambert, J. E., et al. (2004). Variation in the diets of Cercopithecus DuBois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). species: Differences within forests, among forests, and across species. Colorimetric method for determination of sugars and related In Glenn, M. E., & Cords, M. (Eds.), The guenons: Diversity and substances. Analytical Chemistry, 28, 350–356. https://doi.org/10. adaptation in African monkeys. Developments in primatology: Progress 1021/ac60111a017 and prospects (pp. 325–350). New York, NY: Kluwer Academic Dunham, N. T. (2017). Feeding ecology and dietary flexibility of Colobus Publishers. https://doi.org/10.1007/b100500 angolensis palliatus in relation to habitat disturbance. International Chapman, C. A., Chapman, L. J., Rode, K. D., Hauck, E. M., & McDowell, L. Journal of Primatology, 38, 553–571. https://doi.org/10.1007/s10764‐ R. (2003). Variation in the nutritional value of primate foods: Among 017‐9965‐x trees, time periods, and areas. International Journal of Primatology, 24, Dunham, N. T., & Rodriguez‐Saona, L. E. (2018). Nutrient intake and 317–333. https://doi.org/10.1023/A:1023049200150 balancing among female Colobus angolensis palliatus inhabiting Chivers, D. J. (1998). Measuring food intake in wild animals: Primates. structurally distinct forest areas: Effects of group, season, and Proceedings of the Nutrition Society, 57, 321–332. https://doi.org/10. reproductive state. American Journal of Primatology, 80, e22878. 1079/PNS19980047 https://doi.org/10.1002/ajp.22878 Cipolletta, C. (2004). Effects of group dynamics and diet on the ranging Fimbel, C., Vedder, A., Dierenfeld, E., & Mulindahabi, F. (2001). An patterns of a western gorilla group (Gorilla gorilla gorilla) at Bai Hokou, ecological basis for large group size in Colobus angolensis in the Central African Republic. American Journal of Primatology, 64, Nyungwe Forest, Rwanda. African Journal of Ecology, 39,83–92. 193–205. https://doi.org/10.1002/ajp.20072 https://doi.org/10.1111/j.1365‐2028.2001.00276.x Conklin, N. L., & Wrangham, R. W. (1994). The value of figs to a hind‐gut Ganas, J., & Robbins, M. M. (2005). Ranging behavior of the mountain fermenting frugivore: A nutritional analysis. Biochemical Systematics gorillas (Gorilla beringei beringei) in Bwindi Impenetrable National and Ecology, 22, 137–151. https://doi.org/10.1016/0305‐1978(94) Park, Uganda: A test of the ecological constraints model. Behavioral 90004‐3 Ecology and Sociobiology, 58, 277–288. https://doi.org/10.1007/ Conklin‐Brittain, N. L., Wrangham, R. W., & Hunt, K. D. (1998). Dietary s00265‐005‐0920‐z response of chimpanzees and cercopithecines to seasonal variation in Ganas, J., Ortmann, S., & Robbins, M. M. (2009). Food choices of the fruit abundance. II. macronutrients. International Journal of Primatol- mountain gorilla in Bwindi Impenetrable National Park, Uganda: The ogy, 19, 971–998. https://doi.org/10.1023/A:1020370119096 influence of nutrients, phenolics and availability. Journal of Tropical Cooper, J., & Hull, G. (2017). Gorilla pathology and health. London, UK: Ecology, 25, 123–134. https://doi.org/10.1017/S0266467408005701 Academic Press. Ganzhorn, J. U., & Wright, P. C. (1994). Temporal patterns in primate leaf Danish, L., Chapman, C. A., Hall, M. B., & Worman, C. O. (2006). The role of eating: The possible role of leaf chemistry. Folia Primatologica, 63, sugar in diet selection in redtail and red colobus monkeys. In G. 203–208. https://doi.org/10.1159/000156820 Hohmann, M. M. Robbins, & C. Boesch (Eds.), Feeding ecology in apes Gaulin, S. J. C., & Gaulin, C. K. (1982). Behavioral ecology of Alouatta and other primates: Ecological, physiological and behavioural aspects (pp. seniculus in Andean cloud forest. International Journal of Primatology, 473–488). Cambridge, UK: Cambridge University Press. 3(1), 1–32. https://doi.org/10.1007/BF02693488 Dasilva, G. L. (1992). The western black‐and‐white colobus as a low‐ Goldsmith, M. L. (1999). Ecological constraints on the foraging effort of energy strategist: Activity budgets, energy expenditure and energy western gorillas (Gorilla gorilla gorilla) at Bai Hoköu, Central African intake. The Journal of Animal Ecology, 61,79–91. https://doi.org/10. Republic. International Journal of Primatology, 20,1–23. 2307/5511 Goldsmith, M. L. (2002). Comparative behavioral ecology of a lowland and Davies, A. G., Oates, J. F., & Dasilva, G. L. (1999). Patterns of frugivory in highland gorilla population: Where do Bwindi gorillas fit? In A. B. three west African colobine monkeys. International Journal of Primatol- Taylor, & M. L. Goldsmith (Eds.), Gorilla biology: A multidisciplinary ogy, 20,327–357. https://doi.org/10.1023/A:1020596503533 perspective (pp. 358–384). New York, NY: Cambridge University Press. DiGiorgio, A., & Knott, C. D. (2017). Orangutans, fruit, and the geometric Harcourt, A. H., & Stewart, K. J. (2007). Gorilla society: Conflict, compromise, and framework: Fruit and non‐fruit choice in wild Pongo pygmaeus cooperation between the sexes. Chicago, IL: University of Chicago Press. wurmbii. American Journal of Physical Anthropology, 162, 163. https:// Harris, T. R., & Chapman, C. A. (2007). Variation in diet and ranging of doi.org/10.1002/ajpa.23210 black and white colobus monkeys in Kibale National Park, Uganda. Doran, D. M., & McNeilage, A. (1998). Gorilla ecology and behavior. Primates, 48, 208–221. https://doi.org/10.1007/s10329‐006‐0036‐8 Evolutionary Anthropology: Issues, News, and Reviews, 6, 120–131. Hill, R. A., & Dunbar, R. I. M. (2002). Climatic determinants of diet and Doran, D. M., & McNeilage, A. (2001). Subspecific variation in gorilla foraging behaviour in baboons. Evolutionary Ecology, 16, 579–593. behavior: The influence of ecological and social factors. In M. M. https://doi.org/10.1023/A:1021625003597 Robbins, P. Sicotte, & K. J. Stewart (Eds.), Mountain gorillas: Three Hladik, C. M. (1977). A comparative study of two sympatric species of leaf decades of research at Karisoke (pp. 123–149). Cambridge, UK: monkeys: Presbytis entellus and Presbytis senex. In T. Clutton‐Brock Cambridge University Press. (Ed.), Primate ecology: Studies of feeding and ranging behaviour in lemurs, Doran, D. M., McNeilage, A., Greer, D., Bocian, C., Mehlman, P., & Shah, N. monkeys, and apes (pp. 323–353). London, UK: Academic Press. (2002). Western lowland gorilla diet and resource availability: New Hockings, K. J., McLennan, M. R., Carvalho, S., Ancrenaz, M., Bobe, R., evidence, cross‐site comparisons, and reflections on indirect sampling Byrne, R. W., … Hill, C. M. (2015). Apes in the Anthropocene: methods. American Journal of Primatology, 58,91–116. https://doi.org/ Flexibility and survival. Trends in Ecology & Evolution, 30, 215–222. 10.1002/ajp.10053 https://doi.org/10.1016/j.tree.2015.02.002 14 of 16 | LODWICK AND SALMI

Holm, S. (1979). A simple sequentially rejective multiple test procedure. intake in wild western gorillas. PLoS One, 10, https://doi.org/10.1371/ Scandinavian Journal of Statistics, 6,65–70. https://doi.org/10.2307/ journal.pone.0129254. e0129254–22. 4615733 Newton, P. N. (1987). The social organization of forest hanuman langurs Homsy, J. (1999). Ape tourism and human diseases: How close should we (Presbytis entellus). International Journal of Primatology, 8, 199–232. get?: A critical review of the rules and regulations governing park https://doi.org/10.1007/BF02735173 management and tourism for the wild mountain gorilla, Gorilla Pierce, W. C., & Haenisch, E. L. (1948). Quantitative analysis (3rd ed.). New gorilla beringei. Kigali, Rwanda: International Gorilla Conservation York, NY: John Wiley & Sons, Inc. Programme. Popovich, D. G., Jenkins, D. J. A., Kendall, C. W. C., Dierenfeld, E. S., Janson, C. H., & van Schaik, C. P. (1993). Ecological risk aversion in Carroll, R. W., Tariq, N., & Vidgen, E. (1997). The western lowland juvenile primates: Slow and steady wins the race. In Pereira, M. E., & gorilla diet has implications for the health of humans and other Fairbanks, L. A. (Eds.), Juvenile primates life history, development, and hominoids. The Journal of Nutrition, 127, 2000–2005. https://doi.org/ behaviour (pp. 57–74). Chicago, IL: University of Chicago Press. 10.1093/jn/127.10.2000 Johnson, C. A., Raubenheimer, D., Rothman, J. M., Clarke, D., & Swedell, L. Potts, K. B., Watts, D. P., & Wrangham, R. W. (2011). Comparative feeding (2013). 30 days in the life: Daily nutrient balancing in a wild chacma ecology of two communities of chimpanzees (Pan troglodytes) in Kibale baboon. PLoS One, 8, e70383–e70387. https://doi.org/10.1371/ National Park, Uganda. International Journal of Primatology, 32, journal.pone.0070383 669–690. https://doi.org/10.1007/s10764‐011‐9494‐y Johnson, C. A., Swedell, L., & Rothman, J. M. (2012). Feeding ecology of R Development Core Team (2011). R: A language and environment for olive baboons (Papio anubis) in Kibale National Park, Uganda: statistical computing. Vienna, Austria: R Foundation for Statistical Preliminary results on diet and food selection. African Journal of Computing. http://www.R‐project.org Ecology, 50, 367–370. https://doi.org/10.1111/j.1365‐2028.2011. Raubenheimer, D., & Simpson, S. J. (1997). Integrative models of nutrient 01316.x balancing: Application to insects and vertebrates. Nutrition Research Koenig, A., Beise, J., Chalise, M. K., & Ganzhorn, U. (1998). When females Reviews, 10, 151–179. https://doi.org/10.1079/NRR19970009 should contest for food – testing hypotheses about resource density, Remis, M. J. (1997a). Ranging and grouping patterns of a western lowland distribution, size, and quality with Hanuman langurs (Presbytis gorilla group at Bai Hokou, Central African Republic. American Journal entellus). Behavioral Ecology and Sociobiology, 42, 225–237. https:// of Primatology, 43, 111–133. doi.org/10.1007/s002650050 Remis, M. J. (1997b). Western lowland gorillas (Gorilla gorilla gorilla) as Koenig, A., Borries, C., Chalise, M. K., & Winkler, P. (1997). Ecology, seasonal frugivores: Use of variable resources. American Journal of nutrition, and timing of reproductive events in an Asian primate, the Primatology, 43,87–109. Hanuman langur (Presbytis entellus). Journal of Zoology, 243, Remis, M. J., & Dierenfeld, E. S. (2004). Digesta passage, digestibility and 215–235. https://doi.org/10.1111/j.1469‐7998.1997.tb02778.x behavior in captive gorillas under two dietary regimens. International Kurland, J. A., & Gaulin, S. J. C. (1987). Comparability among measures of Journal of Primatology, 25, 825–845. https://doi.org/10.1023/B:IJOP. primate diets. Primates, 28,71–77. https://doi.org/10.1007/ 0000029124.04610.c7 BF02382184 Remis, M. J., Dierenfeld, E. S., Mowry, C. B., & Carroll, R. W. (2001). Kuroda, S., Nishihara, T., Suzuki, S., & Oko, R. A. (1996). Sympatric Nutritional aspects of western lowland gorilla (Gorilla gorilla gorilla) chimpanzees and gorillas in the Ndoki Forest, Congo. In W. C. diet during seasons of fruit scarcity at Bai Hokou, Central African McGrew, L. F. Marchant, & T. Nishida (Eds.), Great ape societies (pp. Republic. International Journal of Primatology, 22, 807–836. https://doi. 71–81). Cambridge, UK: Cambridge University Press. https://doi.org/ org/10.1023/A:1012021617737 10.1017/CBO9780511752414.008 Richards, P. W. (1952). The tropical rain forest: An ecological study. London, Lieberman, D., Lieberman, M., Peralta, R., & Hartshorn, G. S. (1996). UK: Cambridge University Press. Tropical forest structure and composition on a large‐scale altitudinal Robbins, M. M. (2003). Home range and frugivory patterns of mountain gradient in Costa Rica. The Journal of Ecology, 84, 137–152. https://doi. gorillas in Bwindi Impenetrable National Park, Uganda. International org/10.2307/2261350 Journal of Primatology, 24, 467–491. https://doi.org/10.1023/ Lodwick, J. L. (2014). Links between foraging strategies, feeding A:1023741328338 competition, and female agonistic relationships in wild western Robbins, M. M. (2008). Feeding competition and agonistic relationships gorillas (Gorilla gorilla) (unpublished doctoral dissertation). Stony among Bwindi Gorilla beringei. International Journal of Primatology, 29, Brook, NY: Stony Brook University. 999–1018. https://doi.org/10.1007/s10764‐008‐9275‐4 Machovsky‐Capuska, G. E., Senior, A. M., Simpson, S. J., & Raubenheimer, Robbins, M. (2010). Gorillas: Diversity in ecology and behavior. In D. (2016). The multidimensional nutritional niche. Trends in Ecology & Campbell, C. J., Fuentes, A., MacKinnon, K. C., Bearder, S. K., & Evolution, 31, 355–365. https://doi.org/10.1016/j.tree.2016.02.009 Stumpf, R. M. (Eds.), Primates in perspective (pp. 326–339). New York, Malenky, R. K., Kuroda, S., Vineberg, E. O., & Wrangham, R. W. (1996). The NY: Oxford University Press. significance of terrestrial herbaceous foods for bonobos, chimpan- Robbins,A.M.,Andrew,M.,Stoinski,T.S.,Fawcett,K.A.,&Robbins,M.M. zees, and gorillas. In R. W. Wrangham, W. C. McGrew, F. B. M. de (2009). Socioecological influences on the dispersal of female mountain Waal, & P. Heltne (Eds.), Chimpanzee cultures (pp. 59–76). Cambridge, gorillas—evidence of a second folivore paradox. Behavioral Ecology and MA: Harvard University Press. Sociobiology, 63,477–489. https://doi.org/10.1007/s00265‐008‐0679‐0 Malenky, R. K., & Wrangham, R. W. (1994). A quantitative comparison of Robbins, M. M., Ando, C., Fawcett, K. A., Grueter, C. C., Hedwig, D., Iwata, terrestrial herbaceous food consumption by Pan paniscus in the Y., … Yamagiwa, J. (2016). Behavioral variation in gorillas: Evidence of Lomako Forest, Zaire, and Pan troglodytes in the Kibale Forest, potential cultural traits. PLoS One, 11, e0160483. https://doi.org/10. Uganda. American Journal of Primatology, 32,1–12. https://doi.org/10. 1371/journal.pone.0160483 1002/ajp.1350320102 Robbins, M. M., Gray, M., Kagoda, E., & Robbins, A. M. (2009). Population Masi, S., Cipolletta, C., & Robbins, M. M. (2009). Western lowland gorillas dynamics of the Bwindi mountain gorillas. Biological Conservation, 142, (Gorilla gorilla gorilla) change their activity patterns in response to 2886–2895. https://doi.org/10.1016/j.biocon.2009.07.010 frugivory. American Journal of Primatology, 71,91–100. https://doi.org/ Robbins, M. M., & McNeilage, A (2003). Home range and frugivory 10.1002/ajp.20629 patterns of mountain gorillas in Bwindi Impenetrable National Park, Masi, S., Mundry, R., Ortmann, S., Cipolletta, C., Boitani, L., & Robbins, M. Uganda. International Journal of Primatology, 24, 467–491. https://doi. M. (2015). The influence of seasonal frugivory on nutrient and energy org/10.1023/A:1023741328338 LODWICK AND SALMI | 15 of 16

Robbins, M. M., & Robbins, A. M. (2018). Variation in the social Scally, A., Dutheil, J. Y., Hillier, L. W., Jordan, G. E., Goodhead, I., Herrero, organization of gorillas: Life history and socioecological perspectives. J., … Durbin, R. (2012). Insights into hominid evolution from the gorilla Evolutionary Anthropology: Issues, News, and Reviews, 27, 218–233. genome sequence. Nature, 483, 169–175. https://doi.org/10.1038/ https://doi.org/10.1002/evan.21721 nature10842 Robbins,M.M.,Robbins,A.M.,Gerald‐Steklis, N., & Steklis, H. D. (2005). Long‐ Schülke, O., Chalise, M. K., & Koenig, A. (2006). The importance of term dominance relationships in female mountain gorillas: Strength, ingestion rates for estimating food quality and energy intake. stability and determinants of rank. Behaviour, 142,779–809. https://doi. American Journal of Primatology, 68, 951–965. https://doi.org/10. org/10.1163/1568539054729123 1002/ajp.20300 Robbins, M. M., Sicotte, P., & Stewart, K. J. (Eds.), 2001). Mountain gorillas: Siegel, S., & Castellan, N. J., Jr. (1988). Nonparametric statistics for the Three decades of research at Karisoke. Cambridge, UK: Cambridge behavioral sciences (2nd ed.). New York, NY: McGraw‐Hill. University Press. Sommer, V., Srivastava, A., & Borries, C. (1992). Cycles, sexuality, and Rogers, M. E., Abernethy, K., Bermejo, M., Cipolletta, C., Doran, D., conception in free‐ranging langurs (Presbytis entellus). American Mcfarland, K., … Tutin, C. E. G. (2004). Western gorilla diet: A Journal of Primatology, 28,1–27. synthesis from six sites. American Journal of Primatology, 64, 173–192. Sterck, E. H. M., Watts, D. P., & van Schaik, C. P. (1997). The evolution of https://doi.org/10.1002/ajp.20071 female social relationships in nonhuman primates. Behavioral Ecology Rogers, M. E., Maisels, F., Williamson, E. A., Fernandez, M., & Tutin, C. E. G. and Sociobiology, 41, 291–309. https://doi.org/10.1007/ (1990). Gorilla diet in the Lopé Reserve, Gabon:: A nutritional analysis. s002650050390 Oecologia, 84, 326–339. https://doi.org/10.1007/BF00329756 Stoinski, T. S., Perdue, B., Breuer, T., & Hoff, M. P. (2013). Variability in the Rothman, J. M., Chapman, C. A., & Pell, A. N. (2008). Fiber‐bound nitrogen developmental life history of the genus Gorilla. American Journal of in gorilla diets: Implications for estimating dietary protein intake of Physical Anthropology, 152, 165–172. https://doi.org/10.1002/ajpa. primates. American Journal of Primatology, 70, 690–694. https://doi. 22301 org/10.1002/ajp.20540 Stokes, E. J. (2004). Within‐group social relationships among females and Rothman, J. M., Chapman, C. A., Struhsaker, T. T., Raubenheimer, D., adult males in wild western lowland gorillas (Gorilla gorilla gorilla). Twinomugisha, D., & Waterman, P. G. (2015). Long‐term declines in American Journal of Primatology, 64, 233–246. https://doi.org/10.1002/ nutritional quality of tropical leaves. Ecology, 96, 873–878. https://doi. ajp.20074 org/10.1890/14‐0391.1 Stokes, E. J., Parnell, R. J., & Olejniczak, C. (2003). Female dispersal and Rothman, J. M., Dierenfeld, E. S., Hintz, H. F., & Pell, A. N. (2007). reproductive success in wild western lowland gorillas (Gorilla gorilla Nutritional quality of gorilla diets: Consequences of age, sex, and gorilla). Behavioral Ecology and Sociobiology, 54, 329–339. https://doi. season. Oecologia, 155, 111–122. https://doi.org/10.1007/s00442‐ org/10.1007/s00265‐003‐0630‐3 007‐0901‐1 Strickland, J., & Parsons, T. R. (1972). A practical handbook of seawater Rothman, J. M., Dierenfeld, E. S., Molina, D. O., Shaw, A. V., Hintz, H. F., & analysis (2nd ed.). Ottawa, Ontario: Fisheries Research Board of Pell, A. N. (2006). Nutritional chemistry of foods eaten by gorillas in Canada Bulletin. Bwindi Impenetrable National Park, Uganda. American Journal of Swedell, L. (2011). African papionins: Diversity of social organization and Primatology, 68, 675–691. https://doi.org/10.1002/ajp.20243 ecological flexibility. In C. J. Campbell, A. Fuentes, K. C. MacKinnon, S. Rothman, J. M., Nkurunungi, J. B., Shannon, B. F., & Bryer, M. A. H. (2013). K. Bearder, & R. M. Stumpf (Eds.), Primates in perspective (pp. High altitude diets: Implications for the feeding and nutritional 241–277). New York, NY: Oxford University Press. ecology of mountain gorillas. In N. B. Grow, S. Gursky‐Doyen, & A. Thalmann, O., Fischer, A., Lankester, F., Paabo, S., & Vigilant, L. (2006). The Krzton (Eds.), High altitude primates (pp. 247–264). New York, NY: complex evolutionary history of gorillas: Insights from genomic data. Springer New York. https://doi.org/10.1007/978‐1‐4614‐8175‐1_14 Molecular Biology and Evolution, 24, 146–158. https://doi.org/10.1093/ Rothman, J. M., Pell, A. N., Nkurunungi, J. B., & Dierenfeld, E. S. (2006). molbev/msl160 Nutritional aspects of the diet of wild gorillas: How do Bwindi gorillas Tutin, C. E. G. (1996). Ranging and social structure of lowland gorillas in compare? In Newton‐Fisher, N. E., Paterson, J. D., & Reynolds, V. the Lopé Reserve, Gabon. In W. C. McGrew, L. F. Marchant, & T. (Eds.), Primates of Western Uganda (pp. 153–169). New York, NY: Nishida (Eds.), Great ape societies (pp. 58–70). Cambridge, UK: Springer. Cambridge University Press. Rothman, J. M., Plumptre, A. J., Dierenfeld, E. S., & Pell, A. N. (2007). Tutin, C. E. G., & Fernandez, M. (1993). Composition of the diet of Nutritional composition of the diet of the gorilla (Gorilla beringei): A chimpanzees and comparisons with that of sympatric lowland gorillas comparison between two montane habitats. Journal of Tropical Ecology, in the Lopé Reserve, Gabon. American Journal of Primatology, 30(3), 23, 673–682. https://doi.org/10.1017/S0266467407004555 195–211. Rothman, J. M., Raubenheimer, D., & Chapman, C. A. (2011). Nutritional Tutin, C. E. G., Fernandez, M., Rogers, M. E., Williamson, E. A., McGrew, W. geometry: Gorillas prioritize non‐protein energy while consuming C., Altmann, S. A., et al. (1991). Foraging profiles of sympatric lowland surplus protein. Biology Letters, 7, 847–849. https://doi.org/10.1098/ gorillas and chimpanzees in the Lopé Reserve, Gabon. Philosophical rsbl.2011.0321 Transactions of the Royal Society of London. Series B: Biological Sciences, Ryan, A. M., Chapman, C. A., & Rothman, J. M. (2012). How do differences 334, 179–186. https://doi.org/10.1098/rstb.1991.0107 in species and part consumption affect diet nutrient concentrations? A Twinomugisha, D., Chapman, C. A., Lawes, M. J., Worman, C. O., & Danish, test with red colobus monkeys in Kibale National Park, Uganda. L. M. (2006). How does the golden monkey of the Virungas cope in a African Journal of Ecology, 51,1–10. https://doi.org/10.1111/j.1365‐ fruit‐scarce environment? In N. E. Newton‐Fisher, H. Notman, V. 2028.2012.01346.x Reynolds, & J. D. Paterson (Eds.), Primates of Western Uganda (pp. St. Jeor, S. T., Howard, B. V., Prewitt, T. E., Bovee, V., Bazzarre, T., & Eckel, 45–60). New York, NY: Springer. https://doi.org/10.1007/978‐0‐387‐ R. H. (2001). Dietary protein and weight reduction. Circulation, 104, 33505‐6_3 1869–1874. Van Soest, P. J. (1994). Nutritional ecology of the ruminant. Livestock Salmi, R., & Doran‐Sheehy, D. M. (2014). The function of loud calls (Hoot Production Science (2nd ed.). Ithaca, NY: Comstock Publishing Series) in wild western gorillas (Gorilla gorilla): Gorilla Long Calls and Associates. https://doi.org/10.1016/0301‐6226(95)90011‐X Group Cohesion. American Journal of Physical Anthropology, 155, Vogel, E. R., Harrison, M. E., Zulfa, A., Bransford, T. D., Alavi, S. E., Husson, 379–391. https://doi.org/10.1002/ajpa.22575 S., … Farida, W. R. (2015). Nutritional differences between two 16 of 16 | LODWICK AND SALMI

orangutan habitats: Implications for population density. PLoS One, 10, Journal of Primatology, 21, 265–277. https://doi.org/10.1002/ajp. e0138612–e0138618. https://doi.org/10.1371/journal.pone.0138612 1350210403 Vogel, E. R., Knott, C. D., Crowley, B. E., Blakely, M. D., Larsen, M. D., & Wright, E., Grueter, C. C., Seiler, N., Abavandimwe, D., Stoinski, T. S., Dominy, N. J. (2012). Bornean orangutans on the brink of protein Ortmann, S., & Robbins, M. M. (2015). Energetic responses to bankruptcy. Biology Letters, 8, 333–336. https://doi.org/10.1098/rsbl. variation in food availability in the two mountain gorilla populations 2011.1040 (Gorilla beringei beringei): Energetics of Mountain Gorillas. American Warton, D. I., & Hui, F. K. C. (2011). The arcsine is asinine: The analysis of Journal of Physical Anthropology, 158, 487–500. https://doi.org/10. proportions in ecology. Ecology, 92,3–10. https://doi.org/10.1890/10‐ 1002/ajpa.22808 0340.1 Xue, Y., Prado‐Martinez, J., Sudmant, P. H., Narasimhan, V., Ayub, Q., Waterman, P. G., Choo, G. M., Vedder, A. L., & Watts, D. (1983). Szpak, M., … Scally, A. (2015). Mountain gorilla genomes reveal the Digestibility, digestion‐inhibitors and nutrients of herbaceous foliage impact of long‐term population decline and inbreeding. Science, 348, and green stems from an African montane flora and comparison with 242–245. https://doi.org/10.1126/science.aaa3952 other tropical flora. Oecologia, 60, 244–249. https://doi.org/10.1007/ Yamagiwa, J., Basabose, A. K., Kaleme, K., & Yumoto, T. (2005). Diet of BF00379527 Grauer's gorillas in the montane forest of Kahuzi, Democratic Watts, D. P. (1984). Composition and variability of mountain gorilla diets Republic of Congo. International Journal of Primatology, 26, in the central Virungas. American Journal of Primatology, 7, 323–356. 1345–1373. https://doi.org/10.1007/s10764‐005‐8856‐8 https://doi.org/10.1002/ajp.1350070403 Zinner, D., Groeneveld, L. F., Keller, C., & Roos, C. (2009). Mitochondrial Watts, D. P. (1998). Long‐term habitat use by mountain gorillas (Gorilla phylogeography of baboons (Papio spp.) – Indication for introgressive gorilla beringei). 1. Consistency, variation, and home range size and hybridization? BMC Evolutionary Biology, 9,83–97. https://doi.org/10. stability. International Journal of Primatology, 19, 651–680. https://doi. 1186/1471‐2148‐9‐83 org/10.1023/A:1020324909101 Watts, D. P. (2003). Gorilla social relationships: A comparative overview. In A. B. Taylor, & M. L. Goldsmith (Eds.), Gorilla biology: A multidisciplinary SUPPORTING INFORMATION perspective (pp. 302–327). Cambridge, UK: Cambridge University Press. Wich, S. A., de Vries, H., Ancrenaz, M., Perkins, L., Shumaker, R. W., Suzuki, Additional supporting information may be found online in the A., & van Schaik, C. P. (2010). Orangutan life history variation. In S. A. Supporting Information section. Wich, S. S. Utami‐Atmoko, T. Mitra Setia, & C. P. van Schaik (Eds.), Orangutans: Geographic variation in behavioral ecology and conservation (p. 440). Oxford, UK: Oxford University Press. Wich, S. A., Utami‐Atmoko, S. S., Setia, T. M., Rijksen, H. D., Schürmann, C., How to cite this article: Lodwick JL, Salmi R. Nutritional van Hooff, J. A. R. A. M., & van Schaik, C. P. (2004). Life history of wild composition of the diet of the western gorilla (Gorilla gorilla): Sumatran orangutans (Pongo abelii). Journal of Human Evolution, 47, Interspecific variation in diet quality. Am J Primatol. 385–398. https://doi.org/10.1016/j.jhevol.2004.08.006 Williamson, E. A., Tutin, C. E. G., Rogers, M. E., & Fernandez, M. (1990). 2019;e23044. https://doi.org/10.1002/ajp.23044 Composition of the diet of lowland gorillas at Lopé in Gabon. American