Macronutrient signature of dietary generalism in an ecologically diverse primate in the wild

Macronutrient signature of dietary generalism in an ecologically diverse primate in the wild Abstract A question of considerable importance is why some animals are able to succeed on a wide range of diets whereas others are more tightly constrained. Theory predicts that generalists should show a flexible response for macronutrient regulation in the face of ecologically driven constraint on the nutritional balance of available foods, which in the modeling framework of nutritional geometry has been quantitatively characterized as an “equal distance” regulatory model. This prediction, which has empirical support from several laboratory studies on insects, has not been tested for any generalist animal in the wild, nor for any vertebrate. We performed the first such test, using dawn-to-dusk focal animal observations over 3 years (2013–2015) of rhesus macaque monkeys (Macaca mulatta tcheliensis), a species that among primates is second only to humans in ecological generalism. Results showed, as predicted, that macronutrient regulation conformed closely to the equal distance pattern and differed markedly from the other, ecologically more-specialized primate species that have been studied to date. The same was independently true for lactating and non-lactating macaques, but lactating females had substantially higher intake of macronutrients, as well as the non-nutritional food components, indigestible fiber and tannins. This demonstrates that equal distance regulation by non-lactating monkeys was not an artefact of confounding constraints such as restricted food availability or an upper limit to the ingestion of dietary fiber or plant tannins, but a strategic regulatory response to variation in dietary macronutrient balance. We discuss implications of our results for the most generalist primate of all, humans. INTRODUCTION Nutritional ecology aims to understand the relationships between animal habitat use, food selection, diet composition, and nutrition. These are complex relationships, considering that foods are comprised of many nutrients, each required at particular levels that change with circumstances (e.g. developmental stage, activity levels, health), and in most cases diets are comprised of several to many foods (Raubenheimer and Simpson 2016). They are also important, determining such fundamental issues as the position occupied by a species on the spectrum of ecological generalism-specialism, and hence the range of habitats the animal can occupy, its robustness to ecological change (Futuyma and Moreno 1988), and its potential as an invasive species (Machovsky-Capuska et al. 2016). Nutritional geometry (Raubenheimer and Simpson 1993; Simpson and Raubenheimer 1993) was developed to deal with the complexity of nutrition. This framework has been applied in many controlled laboratory studies to examine nutritional regulation and its consequences across diverse taxa (reviewed in Simpson and Raubenheimer 2012). In general, these studies show that, when suitable food combinations are available, animals select a diet that provides a specifically targeted amount and balance of macronutrients, termed an “an intake target,” and that the selected blend optimizes outcomes associated with fitness, for example growth and fecundity (e.g. Simpson et al. 2004, Jensen et al. 2012). In many ecological circumstances, however, combinations of foods that enable animals to achieve their intake target are not available. In such cases, the animal is forced into a trade-off between over-eating some and/or under-eating other nutrients (Raubenheimer and Simpson 1997), as we describe schematically in Figure 1. The nutritional regulatory strategy that the animal adopts to deal with this trade-off is termed a “rule of compromise.” As with intake targets, rules of compromise have been examined in laboratory studies for a wide range of captive animal species (Behmer 2009; Simpson and Raubenheimer 2012). Figure 1 View largeDownload slide Nutrient balance, nutritional targets and ingestive trade-offs in the nutritional geometry framework. (a) The intake target represents the amounts and ratios of nutrients selected by animals in an unconstrained situation. The light grey lines represent the ratio of protein to non-protein energy (carbohydrate and fat, NPE) of 4 different foods, which increases from top-left to bottom-right. Points are hypothetical intakes by animals that have access only to one of the 4 nutritionally imbalanced foods, with blue, red and green representing different regulatory responses to constrained dietary macronutrient imbalance. If the strategy was to maintain the intake of NPE regardless of the dietary P/NPE ratio, allowing protein (P) intake to vary with macronutrient balance, then intakes would resemble the blue configuration (NPE prioritization, as observed in mountain gorillas—see Introduction). Conversely, P prioritization would resemble the green pattern (observed in spider monkeys, orangutans, and humans). The red pattern would be observed if the animals prioritized neither nutrient over the other, but fed to a position of constant energy intake regardless of the proportional contributions of P and NPE. (b) While the red constant energy configuration superficially resembles indiscriminate energy targeting, in reality it involves regulating to a point where the deficit of one nutrient (represented by superscripted “−”) matches the surplus (superscripted “+”) of the other, that is, P+ = NPE− on surplus-protein diets, and NPE+ = P− on surplus NPE diets. This pattern of response to macronutrient imbalance, called “equal distance” regulation, is expected to be shown by generalist feeders with flexible physiological capabilities, for example the ability to use surplus ingested protein (P+) to offset the deficit in non-protein energy (NPE−). For comparison, the nutrient intakes corresponding with the P prioritization and NPE prioritization responses are also shown. In P prioritization, P+ and P− remain 0 with changes in dietary P/NPE ratio, and NPE intakes are negative or positive depending on the direction of dietary P/NPE imbalance. The converse is true for NPE prioritization. Figure 1 View largeDownload slide Nutrient balance, nutritional targets and ingestive trade-offs in the nutritional geometry framework. (a) The intake target represents the amounts and ratios of nutrients selected by animals in an unconstrained situation. The light grey lines represent the ratio of protein to non-protein energy (carbohydrate and fat, NPE) of 4 different foods, which increases from top-left to bottom-right. Points are hypothetical intakes by animals that have access only to one of the 4 nutritionally imbalanced foods, with blue, red and green representing different regulatory responses to constrained dietary macronutrient imbalance. If the strategy was to maintain the intake of NPE regardless of the dietary P/NPE ratio, allowing protein (P) intake to vary with macronutrient balance, then intakes would resemble the blue configuration (NPE prioritization, as observed in mountain gorillas—see Introduction). Conversely, P prioritization would resemble the green pattern (observed in spider monkeys, orangutans, and humans). The red pattern would be observed if the animals prioritized neither nutrient over the other, but fed to a position of constant energy intake regardless of the proportional contributions of P and NPE. (b) While the red constant energy configuration superficially resembles indiscriminate energy targeting, in reality it involves regulating to a point where the deficit of one nutrient (represented by superscripted “−”) matches the surplus (superscripted “+”) of the other, that is, P+ = NPE− on surplus-protein diets, and NPE+ = P− on surplus NPE diets. This pattern of response to macronutrient imbalance, called “equal distance” regulation, is expected to be shown by generalist feeders with flexible physiological capabilities, for example the ability to use surplus ingested protein (P+) to offset the deficit in non-protein energy (NPE−). For comparison, the nutrient intakes corresponding with the P prioritization and NPE prioritization responses are also shown. In P prioritization, P+ and P− remain 0 with changes in dietary P/NPE ratio, and NPE intakes are negative or positive depending on the direction of dietary P/NPE imbalance. The converse is true for NPE prioritization. From the perspective of behavioral ecology, and in order to integrate the patterns of nutritional regulation with ecological and evolutionary theory, it is fundamentally important to learn not only whether animals select intake targets and how they resolve nutritional conflicts when constrained from doing so, but to understand the functional reasons for their responses. Theoretical and laboratory studies examining this have already begun. Theory predicts that generalists should show a rule of compromise in which macronutrients are regarded as interchangeable resources, such that a deficit in one can be offset by an equivalent surplus in another (Simpson et al. 2002; Raubenheimer and Simpson 2003). Physiologically, this prediction derives from the expectation that generalism is associated with flexible metabolic systems capable of dealing with nutrient surpluses and deficits. Physiological flexibility could, for example, enable protein (P) surpluses ingested on high protein diets to be channeled into energy metabolism to offset carbohydrate and fat (henceforth non-protein energy, NPE) deficits, and surplus NPE ingested on low protein diets to be stored as fat for later use if subsequently NPE limited (Raubenheimer and Simpson 2003, Simpson et al. 2004). Ecologically, it is consistent with the expectation that the costs for generalists of over-eating a nutrient (e.g. NPE when confined to low P/NPE diets) are lower than for specialists, because the broad diet of generalists makes it more likely they will subsequently encounter complementary foods that enable them to redress the imbalance (i.e. with high P/NPE; the “diet heterogeneity hypothesis,” Simpson et al. (2002)). Nutritional geometry enables these predictions to be set up in a quantitative model and empirically tested. Numerically, the generalist pattern of regulation corresponds with the situation where total ingested energy is equal regardless of dietary P/NPE ratio. This is because the theory predicts that across a range of dietary P/NPE ratios, all diets will be eaten in amounts where the surplus of one nutrient matches the deficit of the other and therefore energy intakes do not vary with dietary macronutrient balance (Figure 1). Geometrically, it corresponds with a pattern in which intakes on all diets align along the same iso-energy line, a pattern called “equal distance” regulation (Figure 1; Raubenheimer and Simpson 2003). Several laboratory studies have demonstrated, as predicted, that generalist herbivorous insects do show a strategy consistent with equal distance regulation, whereas matched (e.g. closely related) specialists do not (reviewed in Simpson and Raubenheimer 2012). Figure 2 View largeDownload slide Geographic distribution of Macaca mulatta. The cross shows the location of our study site. Figure 2 View largeDownload slide Geographic distribution of Macaca mulatta. The cross shows the location of our study site. No study, however, has tested the predictions of this theory for any vertebrate, nor any animal in the wild. Previous field studies of rules of compromise, almost all of which involve primates, have shown a variety of regulatory patterns, none of which resembles the equal distance pattern predicted for ecological generalists. Bolivian spider monkeys (Ateles chamek) (Felton et al. 2009) and Borneo orangutans (Vogel et al. 2015) show the protein prioritization rule, in which absolute P intake is maintained tightly constant while NPE varies with variation in dietary macronutrient ratios (as depicted schematically in Figure 1). Mountain gorillas (Gorilla beringei) show the opposite response, over-eating leaf-derived protein to maintain NPE intake constant in periods when fruit-derived NPE is scarce (Rothman et al. 2011). Among primates, rhesus macaques (Macaca mulatta), are matched only by humans in their ecological generalism (Sayers 2013), having an exceptionally broad geographical distribution that spans diverse ecological habitats from cold temperate to sub-tropical latitudes (Fooden 2000; Figure 2). We capitalized on the opportunity offered by M. mulatta to test the predictions of diet breadth theory in a free-ranging primate. We chose to study the Taihangshan sub-species (M. mulatta tcheliensis), which live at the edge of the species range (Figure 2), in harsh high-latitude habitats and are exposed to substantial seasonal (Guo et al. 2011; Tian et al. 2013; Cui et al. 2015) and inter-annual (Lü et al. 2002; Guo et al. 2011; Zhang 2014; Cui et al. 2015) variation in the availability of various dietary foods. We predicted that if diet breadth theory applies beyond herbivorous insects, and if macronutrient flexibility plays a role in the ecological generalism of rhesus macaques, then the regulatory response by these monkeys to constrained variation in dietary macronutrient balance will resemble the equal distance rule (Figure 1). Our study system offered a powerful opportunity to test this. First, superimposed on seasonal variation (Song and Qu 1996; Lü et al. 2002) is appreciable inter-annual variation in the primary energy-providing food source, seeds of the oak (Quercus variabilis) (Lü et al. 2002). This enabled us to use inter-annual variation in diet within the same season as the independent variable, thus ensuring that the observed intakes were due to ecological constraint on diet quality and not seasonal variation in nutrient requirements (e.g., due to seasonal differences in nutritional costs for thermoregulation or reproduction). Second, inter-annual variation in diet quality was most pronounced in spring, the period in which reproducing females have accentuated nutritional requirements for lactation. The increased food intake by lactating females provided evidence that non-lactating monkeys could themselves have eaten more, and their observed macronutrient intakes thus reflected strategic choice rather than constraint. We discuss the implications of our results for wild primate studies and for diet breadth theory, and consider their relevance for understanding diet and macronutrient regulation by the most generalist primate of all, humans. MATERIALS AND METHODS Behavioral observations We collected data across 2013, 2014, and 2015 in the Taihangshan Macaque National Nature Reserve (TMNNR, 34°54’–35°40’N, 112°02’–113°45’E, 566 km2 in area) in Jiyuan, Henan Province, China. The area has a continental monsoon climate with average annual precipitation 634 mm and temperature 14.3 °C, and is dominated by temperate deciduous broadleaf forest. A wild macaque group (WangWu-1, WW-1) was assigned our target group. The group has been habituated to human observers since 2005 (Guo et al. 2011; Cui et al. 2015) and could thus be observed at close range (ca 2 m). In 2013, group size was 77 (including 20 adult females and 6 adult males), in 2014 88 (23 adult females and 8 adult males), and by August 2015 group size had steadily increased to 100 (27 adult females and 8 adult males). We studied the focal group from March 2013 to February 2016, systematically spanning different seasons across replicate years. In our study area, delineation of the seasons is March to May (spring); June to August (summer); September to November (autumn); December to February (winter) (Song and Qu 1996). For reasons explained below (Data selection), in the current paper we report on only the spring data. Our data covered 23 focal animals in each observation year. These included 17 principle animals (11 females and 6 males) selected to span ranks across the group. We prioritized obtaining at least one whole day observation for each of the principal animals, and thereafter switched to the other 6 animals. In cases where we lost sight of the animal (e.g. if it left the group temporarily for mating), we abandoned the data from that days follow and re-followed the same focal animal on a subsequent day. To avoid bias due to sex and reproductive state, the sex ratio and proportion of reproducing females were maintained as similar as possible across replicate years. The number of lactating females observed in 2013, 2014, and 2015 was 7, 10, and 10, respectively. In each of the 3 years, the number of non-lactating monkeys was 19 (13 non-lactating females and 6 adult males). To avoid pseudoreplication, monkeys that were observed on more than one occasion within the same year and same reproductive state (lactating or not lactating) were represented in the analysis by the average of the repeat observations. A focal animal was selected and followed continuously from dawn to dusk. If the focal animal went out of view for more than 5 min, that day’s observation was abandoned and the data excluded from the analysis. The same focal animal was followed for at least one whole day (range: 1–4 days). We recorded the time that each FA spent in feeding, excluding events <5 s. which we considered rejections. Simultaneously, we recorded the category of food item consumed (seeds, leaves etc., see below) and the source species. Unit counts (Altmann 1998; Vogel 2005; Rothman et al. 2008) were used to estimate the mass of each food item consumed. Sampling units used in our study were diet items including a single plant part (e.g. a leaf or seed), the approximate dimensions of a food item (for bark, twig), or, in the case of small leaves or fruits, the average number of items consumed (Rothman et al. 2008). The wet weight of each food unit was measured immediately after collection in the field (n = 50) (Rothman et al. 2008), and the units were calibrated monthly to account for intraspecific variation in plant morphology. Food units were sun dried at temperatures <40 °C, then transported to the laboratory for nutrient analysis. When the absolute number of ingested items could not be recorded for an event, we multiplied the ingestion time by the feeding rate that most closely matched the corresponding event. The feeding rates were derived using the same categories of food items as distinguished in the nutritional analysis (i.e. are specific to species and plant part). Food intakes were calculated by multiplying ingestion time with the feeding rate. A total of 21 different food items belonging to 19 species of food plants were observed to be eaten by focal animals in spring from 2013 to 2015, and grouped into 4 food types: 1) seeds, 2) leaves, 3) herbaceous plants and 4) buds. Among these, 4 items that contributed <1% of the dry weight intake of the focal animals were considered incidental and excluded from further analysis. The 17 most commonly eaten food items were dried, ground and 51 samples (3 replicates for every food item) were analyzed for their nutrient content. We measured body mass non-invasively, using small amounts of corn to coax the monkeys to step onto an electronic scale (Qianxuan TS-2010A4, accuracy 0.1 kg; Zhang et al. 2016) and recording the digital readout following stabilization for 3 s (Figure 3). Subjects were excluded from observations for 2 days after weighing as a precautionary measure to avoid any effect on behavior. As it was difficult to weigh particular animals on demand, we opportunistically weighed sub-sets of monkeys that represented the relevant age–sex classes, and used the means of these in our calculations. These comprised 6 males (mean± SD = 8.8 ± 1.2 kg), 9 non-reproductive females (6.8 ± 0.4 kg), 8 pregnant females (7.0 ± 0.5 kg), and 8 lactating females (6.6 ± 0.4 kg). Figure 3 View largeDownload slide (a) Macaque feeding on leaves. (b) Measuring the body mass of adult female macaques using an electronic scale (Qianxuan TS-2010A4 type). Figure 3 View largeDownload slide (a) Macaque feeding on leaves. (b) Measuring the body mass of adult female macaques using an electronic scale (Qianxuan TS-2010A4 type). Seed density Previous observations have shown that in our study population fallen Quercus seeds are an important dietary component that is heavily targeted when available (Zhang 2014; Cui et al. 2015). We therefore surveyed the density of these seeds monthly during the study. This was done using 225 sampling plots (1 m × 1 m), set within 20 quadrats (20 m × 20 m). The quadrats were laid at 100 m altitude intervals across a transect, which spanned 800–1700 m above sea level and were selected to include the range of topographies comprising the monkey’s home range (e.g. slopes, ditches etc.). Laboratory methods Dried food items were transported to the School of life sciences, Northwest University and College of Animal Science and Veterinary Medicine, Henan Agricultural University for the chemical analysis. The sampled food items were used to calculate daily and total dry mass of each item consumed by the focal animals. The nutritional content of the 44 most commonly eaten food items, as defined above, are reported on a dry matter (DM) basis in Supplementary Table 1. Ash was determined by combustion (Rothman et al. 2012); crude protein was measured by the Kjehdahl assay (BUICHI, KjelFIex K-360; Brooks et al. 1995; Nioi et al. 2012); available protein (AP) was measured through subtraction of acid detergent insoluble nitrogen from crude protein (Licitra et al. 1996). Fat was measured by ether extract with a fat analyzer (FOSS, SCINOTMST310) (Rothman et al. 2012). Samples were analyzed for neutral detergent fiber with residual ash (NDF; with sodium sulfite and a-amylase), then for acid detergent lignin (ADL) and acid detergent fiber (ADF) with an automatic fiber analyzer (ANKOM, A2000i; Goering and Van Soest 1970; Van Soest et al. 1991; Rothman et al. 2008). The percentages of total nonstructural carbohydrates (TNC) were calculated by subtracting the contributions of NDF, crude protein, fat, and ash from the rest of the nutrients (Johnson et al. 2013). The energetic contribution from neutral detergent fiber (NDF) was estimated following equation #2 in Rothman et al. (2008): NDFdigestibilityDM=100−(100%ADL in diet%ADL in feces×%NDF in feces%NDF in diet) The energy value of NDF was calculated by multiplying the conversion values of 16.7 kJ per g TNC by the NDF digestibility coefficients, and then subtracting the 4 kJ per g estimated to be consumed by gut microbes (Conklin-Brittain et al. 2006). To estimate the NDF digestion coefficient, we collected the fecal samples (n = 36) from 6 females and 6 males monthly from their sleep sites. As there were no significant differences in sex for NDF digestibility coefficient (P > 0.05), we averaged the data across sexes. This gave NDF digestibility coefficients of 0.533, which were entered into the standard equation to estimate conversion values: [(16.7-4)* NDF digestibility coefficient kJ/g] NDF. For macronutrient energy contributions, we used the conventional conversion values of 37.7 kJ per g crude fat, 16.7 kJ per g AP, and 16.7 kJ per g TNC (NRC 1989). We summed the energetic contributions from crude fat, non-structural carbohydrates and the available fraction of NDF to yield non-protein energy (NPE). In addition, quantitative determination of tannin in diets also was evaluated on a DM basis by P-B Spectrophotometry (Price and Larry 1977; Supplementary Table 2). Non-digestible fiber was calculated by subtracting the microbial digestible component of NDF from the total NDF intake on a dry matter basis using the following equation: Non−digestible fiber=total NDF×(1−NDF digestibility coefficient) Calculating feeding data Feeding variables were calculated as follows: 1) Observed daily nutrient intake of focal animals on a dry matter basis: Mj=∑i=1kni⋅hi⋅cij⋅ (1) MNDF=∑i=1kni⋅hi⋅ciNDF (2) where, i……..k represent the number of food items; j = nutrients (TNC, AP, and Fat); Mj= daily nutrient intake of j (g); MNDF= daily nutrient intake of NDF (g); cij= the concentration of j in i; ciNDF= the concentration of NDF in i; ni= the number of observed daily food units intake; hi= the weight of i on a dry matter basis (g). 2) Observed daily nutrient intake per focal animal on an energy basis. NVj=Mj⋅Ej+ MNDF⋅(ENDF−4)*BNDF (3) where, j= nutrients (TNC, AP and Fat); NVj (Nutritional Value, NV) = observed daily energy intake of j (kJ). Mjand MNDF are from equations (1) and (2). EFat = 37.7kJ/g, ETNC = 16.7kJ/g, ENDF = 16.7kJ/g, and EAP = 16.7kJ/g (Song et al. 1996; NRC 1989). BNDF=NDF digestibility coefficients in spring. 3) Observed daily nutrient intake per kilogram of body weight of focal animal on an energy basis. I=NVjN·w (4) where I represents the observed daily nutrient intake per focal animal on an energy basis kJ/(day·BWkg). In addition, N represents the number of samples and w is body weight. 4) We also calculated percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring (on a dry matter basis or an energy basis). Pi=ki∑i=1kki×100% (5) where Pi represents seasonal percentage contribution of different foods (plant parts) i and ki= mean intake of i. Data selection The current analysis arises from a larger program involving data over all 4 seasons and 3 years (2013, 2014, and 2015). However, in order to test the hypothesis of the present study, namely, that macaques show a generalist pattern of macronutrient regulation, it is important to establish that observed variance in intakes reflects the response of macronutrient regulatory systems to constrained dietary imbalance (Figure 1) rather than seasonal changes in macronutrient requirements (for example due to seasonal temperature differences, Terrien et al. 2011). We were able to do this by analyzing data within a single season, spring (March–May), in which available foods, and hence dietary macronutrient balance, varied substantially across the 3 study years. Seasonal comparisons are the subject of a separate publication (Cui et al., in preparation). As detailed above, we distinguished lactating females from non-lactating females and males. As nutrient and energy intakes of males and non-lactating females did not differ statistically (see Results), we combined these into a single category for comparison with lactating females. The contrast of lactating and non-lactating monkeys not only provided a separate instance in which to test whether rhesus macaques show the equal distance pattern of macronutrient regulation, but also provided an important control in the test of our hypothesis. Specifically, since lactating females have increased nutrient requirements, we expected that they would have increased food intakes (Thompson 2013). If so, this would show that non-lactating monkeys could likewise have eaten more than they did, providing confidence that their intakes reflected macronutrient regulation rather than restricted availability of food or upper limits on the ingestion of the non-nutrient dietary components undigestible fiber or tannins. Statistical analysis and hypothesis testing Spearman Rank Correlation was used to test the correlation between the proportional contribution of Quercus seeds to the diet of Taihangshan macaques and their availability across the 24 months during the 3-year study period in which these seeds were available. We used Kruskal–Wallis H to test for differences across the three study years in the number of seeds remaining on the ground in spring. Linear mixed models were used to analyze the influence of sex (non-lactating females vs. males) and years on energy intake and dietary macronutrient ratios. Sex was designated a fixed effect, while year was considered a random effect. We also used Kruskal–Wallis H to test for differences in the dietary P/NPE ratio in spring between 2013, 2014, and 2015. We used linear mixed model to compare intakes of energy, tannin, and indigestible fiber between non-lactating monkeys and lactating females across 3 years. Lactating status was designated a fixed effect, and year a random effect. To examine the patterns of NPE and AP intake of adult macaques, we plotted macronutrient and energy intakes within the nutritional geometry framework, a multidimensional approach in which each axis represents a different food component (Raubenheimer et al. 2009, Simpson and Raubenheimer 2012). This enabled us to visually compare observed data to 3 alternative models of regulation, P prioritization (protein intake is maintained constant), NPE prioritization (NPE intake constant) and the equal distance rule predicted for generalists (macronutrient energy intake constant; Figure 1). We statistically tested the fit of the data to these models as follows. First, we calculated the mean intake across the 3 years of P, NPE, and total energy. For each variable, we then expressed the observed intake of individual monkey as a percentage of the grand mean, and compared these percentages across years using linear mixed models with year designated a fixed effect. The equal distance rule predicts that P intake and NPE intake will differ across years, whereas total energy intake will remain more constant. In contrast, the P prioritization and NPE prioritization models predict that P and NPE intake will remain constant, respectively, while the other 2 variables differ with dietary macronutrient balance. We used IBM SPSS version 19.0 (IBM Corp., Armonk, NY) for data analysis. All data were expressed as means ± SE, and the significance level for statistical tests was set at α = 0.05. RESULTS Foods The proportional contribution to the diet of different food categories varied substantially across years (Table 1). The most notable dietary component that varied was seeds of Quercus trees, which when available are strongly targeted by the macaques, most likely for their high TNC content (~57% by weight; Supplementary table 1). There was a significant positive correlation between the availability (density on the ground) of these seeds and their contribution to the diet of Taihangshan macaques on a dry matter basis (r = 0.788, P < 0.001, n = 24). Seed production occurred over a limited time of the year (autumn) in our study area, but in mast years seeds were sufficiently prolific that they were not depleted by the macaques and remained abundantly available until the end of the following spring. Our seed density estimates suggested that the number of seeds remaining on the ground in spring varied significantly across the 3 study years (2013 = 1.4 ± 0.1/m2, 2014 = 0.0 ± 0.0/m2, 2015 = 1.2 ± 0.1/m2; Kruskal–Wallis test: χ2 = 151.98, df = 2, 674, P< 0.001). Table 1 Percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring from 2013 to 2015 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 View Large Table 1 Percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring from 2013 to 2015 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 View Large Energy and macronutrient intake Neither energy intakes (F = 0.13, df = 1, 55, P = 0.721) nor dietary macronutrient ratios (F = 0.04, df = 1, 55, P = 0.838) differed between non-lactating females and males, and we therefore combined these groups to contrast in the analysis with lactating females. Figure 4 shows the mean energy and macronutrient intakes in spring of 2013, 2014, and 2015 in non-lactating monkeys and lactating female monkeys. The dietary P/NPE ratio differed significantly across years (Kruskal–Wallis test: χ2 = 69.65, df = 2, 83; P < 0.001), having been lower in 2013 and 2015, when seed availability was high (Table 1), and high in 2014 when seeds were scarce. As predicted, however, lactating females ingested substantially and significantly more energy than non-lactating monkeys (F = 124.52, df = 1, 82, P < 0.001). For lactating monkeys, the mean energy intake across the 3 years was 413 kJ/day·BWKg (shown by the dashed diagonal line in Figure 4), compared with 318 kJ/day·BWKg for non-lactating monkeys (the solid diagonal line). Figure 4 View largeDownload slide Mean daily energy and macronutrient intakes of macaques in spring of 2013, 2014, and 2015. Data are shown separately for non-lactating adults (solid circles) and lactating females (hollow circles). Gray diagonals are energy isolines representing mean total energy intakes across the 3 years for non-lactating monkeys (solid line) and lactating females (dashed line). Mathematically, these isolines are derived from the equation X + Y = constant, where X and Y represent the 3-year mean of available protein (AP) and non-protein energy intake, respectively. As shown in Figure 1, these lines represent the equal distance model of macronutrient regulation. Figure 4 View largeDownload slide Mean daily energy and macronutrient intakes of macaques in spring of 2013, 2014, and 2015. Data are shown separately for non-lactating adults (solid circles) and lactating females (hollow circles). Gray diagonals are energy isolines representing mean total energy intakes across the 3 years for non-lactating monkeys (solid line) and lactating females (dashed line). Mathematically, these isolines are derived from the equation X + Y = constant, where X and Y represent the 3-year mean of available protein (AP) and non-protein energy intake, respectively. As shown in Figure 1, these lines represent the equal distance model of macronutrient regulation. As our hypothesis of equal distance regulation predicts that total energy intakes will be more constant in the face of variation in dietary macronutrient balance than P or NPE intakes (Figure 1), for each variable we compared across years the deviations from the 3 year mean separately for non-lactating and lactating monkeys (Figure 5). For non-lactating monkeys, there were substantial inter-annual differences both in P (F = 308.92, df = 2, 54, P < 0.001) and NPE intake (F = 17.40, df = 2, 54, P < 0.001), but there was no significant difference in total energy intake (F = 1.90, df = 2, 54, P = 0.159). This demonstrates, as predicted, that the data fitted the equal distance model of regulation, in which total energy intakes are constant, but did not fit the P prioritization or NPE prioritization patterns. For lactating females, there were likewise marked inter-annual differences in P (F = 114.01, df = 2, 24, P < 0.001) and NPE intake (F = 21.70, df = 2, 26, P < 0.001). There was also a significant inter-annual difference in total energy intake (F = 4.33, df = 2, 24, P = 0.025). This difference, which was small compared to the inter-annual differences in P and NPE intake (Figure 5), was due to marginally reduced energy intake in 2014 when dietary P/NPE was highest (Figures 4 and 5), with no difference between 2013 and 2015 (F = 0.516, df = 1, 15, P = 0.484). This shows that, in common with their non-lactating counterparts, lactating monkeys showed neither P prioritization nor NPE prioritization, with their pattern of regulation most closely resembling the equal distance rule. Figure 5 View largeDownload slide Annual deviations from the 3 year mean for intakes of total energy, available protein and non-protein energy. The horizontal lines represent the levels of intake that would be expected under (a) the equal distance rule (constant total energy), (b) protein prioritization, and (c) prioritization of non-protein energy. Figure 5 View largeDownload slide Annual deviations from the 3 year mean for intakes of total energy, available protein and non-protein energy. The horizontal lines represent the levels of intake that would be expected under (a) the equal distance rule (constant total energy), (b) protein prioritization, and (c) prioritization of non-protein energy. Dietary constraint due to non-nutritional factors The fact that lactating females ingested more macronutrients than non-lactating monkeys demonstrates that the equal distance pattern observed for the latter group was not due to restricted food availability. We also examined two further potential confounds, both of which have previously been implicated in restricting food intake by monkeys, non-digestible fiber (Milton 1999) and plant tannins (Felton et al. 2009). Figure 6 shows that across the 3 study years intakes of both tannin (F = 211.97, df = 1, 82, P < 0.001) and indigestible fiber (F = 142.32, df = 1, 82, P < 0.001) were significantly higher for lactating than non-lactating females. This demonstrates that the maximum thresholds for fiber and tannin intakes were not reached by non-lactating monkeys, providing further evidence that their pattern of macronutrient intake reflects a nutritional regulatory strategy. Figure 6 View largeDownload slide Intakes of non-digestible fiber (a) and tannin (b) by non-lactating and lactating monkeys in the spring of 2013, 2014, and 2015. Figure 6 View largeDownload slide Intakes of non-digestible fiber (a) and tannin (b) by non-lactating and lactating monkeys in the spring of 2013, 2014, and 2015. DISCUSSION As briefly reviewed in the Introduction, several studies have examined the responses of free-ranging primates to ecologically driven constraints on the ratios of available macronutrients. In addition to extending the comparative database of nutritional regulatory responses in free-ranging primates, the present study is unique in two important respects: it is a targeted test of a hypothesis generated by theory, and the study system provided an unprecedented opportunity to test the predictions of the theory in free-ranging animals in an ecological context. Previous studies of macronutrient regulation patterns in free-ranging primates have not been predictive, but largely descriptive, for two reasons. Firstly, the macronutrient regulatory patterns have been studied in too few primates to generate worthwhile predictions inductively. Secondly, there exists very little theory linking macronutritional regulatory responses to specific ecological circumstances, and this limits the scope for generating predictions deductively. An exception, however, is the dietary specialist-generalist spectrum, and its links to ecological niche theory (Simpson and Raubenheimer 2012; Machovsky-Capuska et al. 2016). Drawing on this theory, and based on the broad geographical distribution and wide range of habitats occupied by M. mulatta, we anticipated that this species would show the equal distance regulatory response to macronutrient imbalance, which as explained in the Introduction is predicted on theoretical grounds for generalist species. Our analysis provides strong support for this prediction. The only previous empirical evidence for the association of equal distance regulation with dietary generalism comes from laboratory studies of insects (reviewed in Behmer 2009; Simpson and Raubenheimer 2012). In these studies, experimental animals were confined to one of several synthetic diets formulated to vary systematically in macronutrient ratios, and the ad-libitum intakes of the different dietary treatment groups compared. The results showed that generalist insect species conformed more closely to the equal distance pattern of regulation than matched species that are more specialized. Furthermore, hybrids between a generalist and specialist species were found to have an intermediate pattern of macronutrient regulation (Lee et al. 2006), suggesting a heritable component to nutritional regulatory strategies. Finally, studies of insects in which the same genotype generates alternative developmental pathways in different environments provide intriguing evidence for the link between equal distance macronutrient regulation and ecological generalism. Locusts reared at high population densities develop phenotypes that are morphologically, behaviorally, and ecologically very different than those reared at low densities. Among the differences, is that locusts reared in high densities (form gregaria) are nutritionally more generalist than those reared at low densities (form solitaria). Studies have shown, as predicted by the theory, that the generalist morph has a pattern of macronutrient regulation that resembles the equal distance pattern, whereas the solitarious morph does not (Simpson et al. 2002). Similar results were found in caterpillars that have gregarious and solitarious developmental phenotypes (Lee et al. 2004). It is substantially more challenging to perform such tests on animals in the wild, because the ecological requirements for conducting the relevant studies non-invasively are stringent. Our study system provided a powerful opportunity to test the equal distance-generalist hypothesis, in a way that goes some distance towards resolving the trade-off between experimental control afforded by captive animal studies and the ecological realism that justifies field studies. First, circumstances are needed in which animals are constrained to eat diets that differ in their macronutrient ratios. If not, and animals have free choice of a wide range of nutritionally complementary foods from which to compose their diet, then the results are likely to reflect a regulated intake target rather than a strategic response to constrained macronutrient imbalance, as observed in prolonged observations of a chacma baboon (Johnson et al. 2013). Our study system satisfied this requirement, because inter-annual variance in the relative availability of different foods imposed constraints on the macronutrient ratios of the diets that could be assembled from those foods in different years. Second, it is important to establish that the animals are restricted only in terms of dietary macronutrient ratios, and not the overall quantity of food available or non-nutrient factors such as indigestible fiber and tannins, which at high levels can limit food intake by primates (Amato and Garber 2014; Lambert and Rothman 2015). This is because rules of compromise are a measure of how ad-libitum intakes vary across diets that differ in their nutrient balance, and if factors other than macronutrients constrain consumption then recorded intakes do not exclusively reflect homeostatic regulation. We were able to address this by showing that intakes of energy, fiber, and tannins by lactating females were substantially higher than non-lactating monkeys that were observed concurrently. This suggests that the intakes of non-lactating monkeys were not due to food shortage or limits on the ability to ingest fiber or tannin, but reflected macronutritional priorities. Third, the between-group comparisons should not confound the effects of dietary macronutrient balance with differences in macronutrient requirements that arise, for example, from seasonal differences in thermoregulatory requirements. A powerful benefit of our study system is that it enabled us to compare the responses of monkeys to variable macronutrient balance across 3 years within the same season, thereby controlling for such factors as seasonal temperature differences that are known to influence nutrient requirements and food intake (Terrien et al. 2011). While the lactating females played an important role in our study by allowing us to discount quantitative food limitation and non-nutrient factors (tannin and fiber) as the causes of the intake pattern by non-lactating females, we cannot interpret their pattern of intake with equal confidence. This is because without further controls it is difficult to be sure that the equal distance-like intake pattern of lactating females was not influenced by limitations on the amount of food that they could access, or by upper limits to the amounts of fiber or tannins that they could ingest. Indeed, the marginally lower energy consumption that we observed for lactating females in 2014 compared with 2013 and 2015 is quite possibly due to limitations on the capacity of lactating monkeys to meet their high energy demands when acorns are scarce. Such uncertainty over the causes of the pattern observed in our study for lactating females underscores the importance of their role in helping to interpret the regulatory pattern of non-lactating monkeys with a high level of certainty. We are thus confident that our data provide strong evidence for the first case of equal distance macronutrient regulation in a primate, and for any animal in the wild. That this was demonstrated in the wild is particularly gratifying, partly because of the challenges of achieving this is in free-ranging animals, but principally because the data reflect regulatory responses to natural, ecologically-generated, variation in available diets. This is, nonetheless, a single-species study, and on its own contributes only one data point towards comparative analysis of the relationship between diet breadth and macronutrient regulatory strategy. On the other hand, the fact that our expectation was developed on theoretical grounds, and has previously been observed in controlled laboratory studies of insect species, provides confidence that the data do reflect broader generality. In so doing, they also substantiate the underlying theory. It is interesting to speculate on the adaptive benefits and ecological consequences for M. mulatta of the flexible macronutrient intake that characterizes the equal distance regulatory strategy. One possibility is that these monkeys tolerate sub-optimal macronutrient ratios, but are otherwise not adapted to them. For example, Nie et al. (2015) found that giant pandas in the wild subsist for several months on nutritionally imbalanced diets, and to reproduce need to migrate between habitats that offer complementary nutrition (Nie et al. 2015). The fact that lactation was maintained across the 3 study years regardless of the dietary macronutrient ratio (Figure 4) suggests that this was not the case for the monkeys in our study. The suggestion, therefore, is that M. mulatta are truly robust to variation in nutrient balance, with a flexible ability to utilize macronutrients inter-changeably. This is predicted by the theory to be a phenotype that is characteristic of a species capable of occupying diverse and variable habitats. Finally, if equal distance regulation is more generally associated with ecological generalism, this gives rise to the fascinating question of why the most geographically widespread and ecologically and dietarily diverse primate of all—humans—does not show this pattern of regulation. Rather, in common with spider monkeys and orangutans, humans show the protein prioritization pattern (Figure 1; Gosby et al. 2014), in which protein intake is maintained relatively constant while non-protein energy is regulated less tightly. One possibility is that humans are dietary generalists only in the sense of being capable of subsisting on a wide range of foods (“food generalists”, sensuMachovsky-Capuska et al. 2016), but at the level of macronutrients are more specialized (“macronutritional specialists”). Consistent with this is the observation that the P:NPE ratio of human diets is remarkably invariant across populations, with a few known exceptions such as the traditional Inuit diet (Simpson and Raubenheimer 2005). Further studies are needed to expand the database of primate macronutrient regulatory strategies to ultimately enable phylogenetically controlled analysis of how these relate to variance in evolutionary ecology, and better understand the implications for foraging theory, niche theory, and human biology. SUPPLEMENTARY MATERIAL Supplementary data are available at Behavioral Ecology online. We thank Jiyuan Administration of Taihangshan Macaque National Nature Reserve for permission to conduct fieldwork and logistic support throughout this study. We are grateful to Mr. Hou Jiafu, Mr. Zhao Guoliang, and Mr. Kong Maojuan for their assistance with field work. We thank Dr. Zhu Shixin (School of Life Science, Zhengzhou University) for plant species identification and Dr. Paul A. Garber (Department of Anthropology, University of Illinois, USA) for his helpful suggestions on the study. We also thank Prof. Li Baoguo, Prof. Guo Songtao, and Mr. Hou Rong for their assistance with the nutritional analyses. This work was supported by the National Natural Science Foundation of China (30970378, 31170503). Data accessibility: Analyses reported in this article can be reproduced using the data provided by Cui et al. (2017). FUNDING This study was supported by the National Natural Science Foundation of China (30970378, 31170503). Conflict of Interest: The authors declare no conflict of interest. REFERENCES Altmann SA . 1998 . Foraging for survival: yearling baboons in Africa . Chicago (IL) : University of Chicago Press . Amato KR , Garber PA . 2014 . Nutrition and foraging strategies of the black howler monkey (Alouatta pigra) in Palenque National Park, Mexico . Am J Primatol . 76 : 774 – 787 . Google Scholar CrossRef Search ADS PubMed Behmer ST . 2009 . Insect herbivore nutrient regulation . Entomology . 54 : 165 – 187 . Google Scholar CrossRef Search ADS Brooks S , Lampi BJ , Sarwar G , Botting HG . 1995 . A comparison of methods for determining total body protein . Anal Biochem . 226 : 26 – 30 . Google Scholar CrossRef Search ADS PubMed Conklin-Brittain NL , Knott CD , Wrangham RW . 2006 . Energy intake by wild chimpanzees and orangutans: methodological considerations and a preliminary comparison . Cam S Bio Evol Anthr . 1 : 445 – 472 . Cui Z , Wang Z , Zhao G , Lu J . 2015 . Seasonal variations in activity budget of adult female rhesus macaques (Macaca Mulatta) at Mt Taihangshan area, Jiyuan, China: effects of diet and temperature . Acta Theriologica Sinica . 35 : 138 – 146 . Cui Z , Wang Z , Shao Q , Raubenheimer D , Lu J . 2017 . Data from: macronutrient signature of dietary generalism in an ecologically diverse primate in the wild . Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.mb342.2. Felton AM , Felton A , Raubenheimer D , Simpson SJ , Foley WJ , Wood JT , Wallis IR , Lindenmayer DB . 2009 . Protein content of diets dictates the daily energy intake of a free-ranging primate . Behav Ecol . 20 : 685 – 690 . Google Scholar CrossRef Search ADS Fooden J . 2000 . Systematic review of the rhesus macaque, Macaca mulatta (Zimmermann, 1780) . Fieldiana Zool . 96 : 1 – 180 . Futuyma DJ , Moreno G . 1988 . The evolution of ecological specialization . Annu Rev Ecol Syst . 19 : 207 – 233 . Google Scholar CrossRef Search ADS Goering HK , Van Soest PJ . 1970 . Forage fiber analyses: apparatus, reagents, procedures, and some applications. Agriculture Handbook No. 379 . Washington : USDA . p. 20 . Gosby AK , Conigrave AD , Raubenheimer D , Simpson SJ . 2014 . Protein leverage and energy intake . Obes Rev . 15 : 183 – 191 . Google Scholar CrossRef Search ADS PubMed Guo X , Wang Z , Chen J , Tian J , Wang B , Lu J . 2011 . Winter and spring food habits of Macaca mulatta tcheliensis in Taihangshan National Nature Reserve in Henan Province of China . Chin J Ecol . 30 : 483 – 488 . Jensen K , Mayntz D , Toft S , Clissold FJ , Hunt J , Raubenheimer D , Simpson SJ . 2012 . Optimal foraging for specific nutrients in predatory beetles. P Roy Soc B-Biol Sci . 279 (1736) : 2212 – 2218 . Google Scholar CrossRef Search ADS Johnson CA , Raubenheimer D , Rothman JM , Clarke D , Swedell L , 2013 . 30 days in the life: daily nutrient balancing in a wild chacma baboon . PLoS One . 8 : e70383 . Google Scholar CrossRef Search ADS PubMed Lambert JE , Rothman JM . 2015 . Fallback foods, optimal diets, and nutritional targets: primate responses to varying food availability and quality . Annu Rev Anthropol . 44 : 493 – 512 . Google Scholar CrossRef Search ADS Lee KP , Behmer ST , Simpson SJ . 2006 . Nutrient regulation in relation to diet breadth: a comparison of heliothis sister species and a hybrid . J Exp Biol . 209 : 2076 . Google Scholar CrossRef Search ADS PubMed Lee KP , Simpson SJ , Raubenheimer D . 2004 . A comparison of nutrient regulation between solitarious and gregarious phases of the specialist caterpillar, Spodoptera exempta (Walker) . J Insect Physiol . 50 : 1171 – 1180 . Google Scholar CrossRef Search ADS PubMed Licitra G , Hernandez TM , Van Soest PJ . 1996 . Standardization of procedures for nitrogen fractionation of ruminant feeds . Anim Feed Sci Tech . 57 : 347 – 358 . Google Scholar CrossRef Search ADS Lü J , Lu J , Li J , Qu W . 2002 . Feeding habits of Macaca mulatta in Taihang Mountains . Chin J Ecol . 21 : 29 – 31 . Machovsky-Capuska GE , Senior AM , Simpson SJ , Raubenheimer D . 2016 . The multidimensional nutritional niche . Trends Ecol Evol . 31 : 355 – 365 . Google Scholar CrossRef Search ADS PubMed Milton K . 1999 . Nutritional characteristics of wild primate foods: do the diets of our closest living relatives have lessons for us ? Nutrition . 15 : 488 – 498 . Google Scholar CrossRef Search ADS PubMed Nie Y , Zhang Z , Raubenheimer D , Elser JJ , Wei W , Wei F . 2015 . Obligate herbivory in an ancestrally carnivorous lineage: the giant panda and bamboo from the perspective of nutritional geometry . Funct Ecol . 29 : 26 – 34 . Google Scholar CrossRef Search ADS Nioi C , Kapel R , Rondags E , Marc I . 2012 . Selective extraction, structural characterisation and antifungal activity assessment of napins from an industrial rapeseed meal . Food Chem . 134 : 2149 – 2155 . Google Scholar CrossRef Search ADS PubMed NRC . 1989 . Recommended dietary allowances . Washington (DC) : National Academy Press . Price Martin L , Butler Larry G . 1977 . Rapid visual estimation and spectrophotometric determination of tannin content of sorghum grain . J Agric Food Chem . 25 : 1268 – 1273 . Google Scholar CrossRef Search ADS Raubenheimer D , Simpson SJ . 1993 . The geometry of compensatory feeding in the locust . Anim Behav . 45 : 953 – 964 . Google Scholar CrossRef Search ADS Raubenheimer D , Simpson SJ . 1997 . Integrative models of nutrient balancing: application to insects and vertebrates . Nutr Res Rev . 10 : 151 – 179 . Google Scholar CrossRef Search ADS PubMed Raubenheimer D , Simpson SJ . 2003 . Nutrient balancing in grasshoppers: behavioural and physiological correlates of dietary breadth . J Exp Biol . 206 : 1669 – 1681 . Google Scholar CrossRef Search ADS PubMed Raubenheimer D , Simpson SJ , 2016 . Nutritional ecology and human health . Annu Rev Nutr . 36 : 603 – 626 . Google Scholar CrossRef Search ADS PubMed Rothman JM , Chapman CA , Van Soest PJ . 2012 . Methods in primate nutritional ecology: a user’s guide . Int J Primatol . 33 : 542 – 566 . Google Scholar CrossRef Search ADS Rothman JM , Dierenfeld ES , Hintz HF , Pell AN . 2008 . Nutritional quality of gorilla diets: consequences of age, sex, and season . Oecologia . 155 : 111 – 122 . Google Scholar CrossRef Search ADS PubMed Rothman JM , Raubenheimer D , Chapman CA . 2011 . Nutritional geometry: gorillas prioritize non-protein energy while consuming surplus protein . Biol Lett . 7 : 847 – 849 . Google Scholar CrossRef Search ADS PubMed Sayers K , 2013 . On folivory, competition, and intelligence: generalisms, overgeneralizations, and models of primate evolution . Primates . 54 : 111 – 124 . Google Scholar CrossRef Search ADS PubMed Simpson SJ , Raubenheimer D . 1993 . A multi-level analysis of feeding behaviour: the geometry of nutritional decisions . Philos T R Soc B . 342 : 381 – 402 . Google Scholar CrossRef Search ADS Simpson SJ , Sibly RM , Lee KP , Behmer ST , Raubenheimer D . 2004 . Optimal foraging when regulating intake of multiple nutrients . Anim Behav . 68 : 1299 – 1311 . Google Scholar CrossRef Search ADS Simpson SJ , Raubenheimer D . 2005 . Obesity: the protein leverage hypothesis . Obes Rev . 6 : 133 – 142 . Google Scholar CrossRef Search ADS PubMed Simpson SJ , Raubenheimer D . 2012 . The nature of nutrition . Princeton (NJ) : Princeton University Press . Google Scholar CrossRef Search ADS Simpson SJ , Raubenheimer D , Behmer ST , Whitworth A , Wright GA . 2002 . A comparison of nutritional regulation in solitarious- and gregarious-phase nymphs of the desert locust Schistocerca gregaria . J Exp Biol . 205 : 121 – 129 . Google Scholar PubMed Song C , Qu W . 1996 . Scientific survey of Taihangshan Macaque nature reserve . Beijing : China Forestry Publishing House . Terrien J , Blanc S , Zizzari P , Epelbaum J , Aujard F . 2011 . Physiological responses to chronic heat exposure in an aging non-human primate species, the gray mouse lemur (Microcebus murinus) . Exp Gerontol . 46 : 747 – 754 . Google Scholar CrossRef Search ADS PubMed Thompson ME . 2013 . Comparative reproductive energetics of human and nonhuman primates . Anthropology . 42 : 287 – 304 . Google Scholar CrossRef Search ADS Tian J , Wang Z , Lu J , Wang B , Chen J . 2013 . Reproductive parameters of female Macaca mulatta tcheliensis in the temperate forest of Mount Taihangshan, Jiyuan, China . Am J Primatol . 75 : 605 – 612 . Google Scholar CrossRef Search ADS PubMed Van Soest PV , Robertson JB , Lewis BA . 1991 . Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition . J Dairy Sci . 74 : 3583 – 3597 . Google Scholar CrossRef Search ADS PubMed Vogel ER ., 2005 . Rank differences in energy intake rates in white-faced capuchin monkeys, Cebus capucinus: the effects of contest competition . Behav Ecol Sociobiol . 58 : 333 – 344 . Google Scholar CrossRef Search ADS Vogel ER , Rothman JM , Moldawer AM , Bransford TD , Thompson ME , Noordwijk MAV , Atmoko SSU , Crowley B , Knott CD , Erb WM , et al. 2015 . Coping with a challenging environment: nutritional balancing, health, and energetics in wild Bornean orangutans . Am J Phys Anthropol . 156 : 314 – 315 . Zhang P , Lyu MY , Wu CF , Chu YMR , Han N , Yang D , Hu K . 2016 . Variation in body mass and morphological characters in Macaca mulatta brevicaudus from Hainan, China . Am J Primatol . 78 : 679 – 698 . Google Scholar CrossRef Search ADS PubMed Zhang Y . 2014 . The interactions among plant seeds, rodent, and insects in Mt. Taihangshan area . Zhengzhou : Zhengzhou University . © The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral Ecology Oxford University Press

Macronutrient signature of dietary generalism in an ecologically diverse primate in the wild

Loading next page...
 
/lp/ou_press/macronutrient-signature-of-dietary-generalism-in-an-ecologically-NtCvzc8wii
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1045-2249
eISSN
1465-7279
D.O.I.
10.1093/beheco/ary003
Publisher site
See Article on Publisher Site

Abstract

Abstract A question of considerable importance is why some animals are able to succeed on a wide range of diets whereas others are more tightly constrained. Theory predicts that generalists should show a flexible response for macronutrient regulation in the face of ecologically driven constraint on the nutritional balance of available foods, which in the modeling framework of nutritional geometry has been quantitatively characterized as an “equal distance” regulatory model. This prediction, which has empirical support from several laboratory studies on insects, has not been tested for any generalist animal in the wild, nor for any vertebrate. We performed the first such test, using dawn-to-dusk focal animal observations over 3 years (2013–2015) of rhesus macaque monkeys (Macaca mulatta tcheliensis), a species that among primates is second only to humans in ecological generalism. Results showed, as predicted, that macronutrient regulation conformed closely to the equal distance pattern and differed markedly from the other, ecologically more-specialized primate species that have been studied to date. The same was independently true for lactating and non-lactating macaques, but lactating females had substantially higher intake of macronutrients, as well as the non-nutritional food components, indigestible fiber and tannins. This demonstrates that equal distance regulation by non-lactating monkeys was not an artefact of confounding constraints such as restricted food availability or an upper limit to the ingestion of dietary fiber or plant tannins, but a strategic regulatory response to variation in dietary macronutrient balance. We discuss implications of our results for the most generalist primate of all, humans. INTRODUCTION Nutritional ecology aims to understand the relationships between animal habitat use, food selection, diet composition, and nutrition. These are complex relationships, considering that foods are comprised of many nutrients, each required at particular levels that change with circumstances (e.g. developmental stage, activity levels, health), and in most cases diets are comprised of several to many foods (Raubenheimer and Simpson 2016). They are also important, determining such fundamental issues as the position occupied by a species on the spectrum of ecological generalism-specialism, and hence the range of habitats the animal can occupy, its robustness to ecological change (Futuyma and Moreno 1988), and its potential as an invasive species (Machovsky-Capuska et al. 2016). Nutritional geometry (Raubenheimer and Simpson 1993; Simpson and Raubenheimer 1993) was developed to deal with the complexity of nutrition. This framework has been applied in many controlled laboratory studies to examine nutritional regulation and its consequences across diverse taxa (reviewed in Simpson and Raubenheimer 2012). In general, these studies show that, when suitable food combinations are available, animals select a diet that provides a specifically targeted amount and balance of macronutrients, termed an “an intake target,” and that the selected blend optimizes outcomes associated with fitness, for example growth and fecundity (e.g. Simpson et al. 2004, Jensen et al. 2012). In many ecological circumstances, however, combinations of foods that enable animals to achieve their intake target are not available. In such cases, the animal is forced into a trade-off between over-eating some and/or under-eating other nutrients (Raubenheimer and Simpson 1997), as we describe schematically in Figure 1. The nutritional regulatory strategy that the animal adopts to deal with this trade-off is termed a “rule of compromise.” As with intake targets, rules of compromise have been examined in laboratory studies for a wide range of captive animal species (Behmer 2009; Simpson and Raubenheimer 2012). Figure 1 View largeDownload slide Nutrient balance, nutritional targets and ingestive trade-offs in the nutritional geometry framework. (a) The intake target represents the amounts and ratios of nutrients selected by animals in an unconstrained situation. The light grey lines represent the ratio of protein to non-protein energy (carbohydrate and fat, NPE) of 4 different foods, which increases from top-left to bottom-right. Points are hypothetical intakes by animals that have access only to one of the 4 nutritionally imbalanced foods, with blue, red and green representing different regulatory responses to constrained dietary macronutrient imbalance. If the strategy was to maintain the intake of NPE regardless of the dietary P/NPE ratio, allowing protein (P) intake to vary with macronutrient balance, then intakes would resemble the blue configuration (NPE prioritization, as observed in mountain gorillas—see Introduction). Conversely, P prioritization would resemble the green pattern (observed in spider monkeys, orangutans, and humans). The red pattern would be observed if the animals prioritized neither nutrient over the other, but fed to a position of constant energy intake regardless of the proportional contributions of P and NPE. (b) While the red constant energy configuration superficially resembles indiscriminate energy targeting, in reality it involves regulating to a point where the deficit of one nutrient (represented by superscripted “−”) matches the surplus (superscripted “+”) of the other, that is, P+ = NPE− on surplus-protein diets, and NPE+ = P− on surplus NPE diets. This pattern of response to macronutrient imbalance, called “equal distance” regulation, is expected to be shown by generalist feeders with flexible physiological capabilities, for example the ability to use surplus ingested protein (P+) to offset the deficit in non-protein energy (NPE−). For comparison, the nutrient intakes corresponding with the P prioritization and NPE prioritization responses are also shown. In P prioritization, P+ and P− remain 0 with changes in dietary P/NPE ratio, and NPE intakes are negative or positive depending on the direction of dietary P/NPE imbalance. The converse is true for NPE prioritization. Figure 1 View largeDownload slide Nutrient balance, nutritional targets and ingestive trade-offs in the nutritional geometry framework. (a) The intake target represents the amounts and ratios of nutrients selected by animals in an unconstrained situation. The light grey lines represent the ratio of protein to non-protein energy (carbohydrate and fat, NPE) of 4 different foods, which increases from top-left to bottom-right. Points are hypothetical intakes by animals that have access only to one of the 4 nutritionally imbalanced foods, with blue, red and green representing different regulatory responses to constrained dietary macronutrient imbalance. If the strategy was to maintain the intake of NPE regardless of the dietary P/NPE ratio, allowing protein (P) intake to vary with macronutrient balance, then intakes would resemble the blue configuration (NPE prioritization, as observed in mountain gorillas—see Introduction). Conversely, P prioritization would resemble the green pattern (observed in spider monkeys, orangutans, and humans). The red pattern would be observed if the animals prioritized neither nutrient over the other, but fed to a position of constant energy intake regardless of the proportional contributions of P and NPE. (b) While the red constant energy configuration superficially resembles indiscriminate energy targeting, in reality it involves regulating to a point where the deficit of one nutrient (represented by superscripted “−”) matches the surplus (superscripted “+”) of the other, that is, P+ = NPE− on surplus-protein diets, and NPE+ = P− on surplus NPE diets. This pattern of response to macronutrient imbalance, called “equal distance” regulation, is expected to be shown by generalist feeders with flexible physiological capabilities, for example the ability to use surplus ingested protein (P+) to offset the deficit in non-protein energy (NPE−). For comparison, the nutrient intakes corresponding with the P prioritization and NPE prioritization responses are also shown. In P prioritization, P+ and P− remain 0 with changes in dietary P/NPE ratio, and NPE intakes are negative or positive depending on the direction of dietary P/NPE imbalance. The converse is true for NPE prioritization. From the perspective of behavioral ecology, and in order to integrate the patterns of nutritional regulation with ecological and evolutionary theory, it is fundamentally important to learn not only whether animals select intake targets and how they resolve nutritional conflicts when constrained from doing so, but to understand the functional reasons for their responses. Theoretical and laboratory studies examining this have already begun. Theory predicts that generalists should show a rule of compromise in which macronutrients are regarded as interchangeable resources, such that a deficit in one can be offset by an equivalent surplus in another (Simpson et al. 2002; Raubenheimer and Simpson 2003). Physiologically, this prediction derives from the expectation that generalism is associated with flexible metabolic systems capable of dealing with nutrient surpluses and deficits. Physiological flexibility could, for example, enable protein (P) surpluses ingested on high protein diets to be channeled into energy metabolism to offset carbohydrate and fat (henceforth non-protein energy, NPE) deficits, and surplus NPE ingested on low protein diets to be stored as fat for later use if subsequently NPE limited (Raubenheimer and Simpson 2003, Simpson et al. 2004). Ecologically, it is consistent with the expectation that the costs for generalists of over-eating a nutrient (e.g. NPE when confined to low P/NPE diets) are lower than for specialists, because the broad diet of generalists makes it more likely they will subsequently encounter complementary foods that enable them to redress the imbalance (i.e. with high P/NPE; the “diet heterogeneity hypothesis,” Simpson et al. (2002)). Nutritional geometry enables these predictions to be set up in a quantitative model and empirically tested. Numerically, the generalist pattern of regulation corresponds with the situation where total ingested energy is equal regardless of dietary P/NPE ratio. This is because the theory predicts that across a range of dietary P/NPE ratios, all diets will be eaten in amounts where the surplus of one nutrient matches the deficit of the other and therefore energy intakes do not vary with dietary macronutrient balance (Figure 1). Geometrically, it corresponds with a pattern in which intakes on all diets align along the same iso-energy line, a pattern called “equal distance” regulation (Figure 1; Raubenheimer and Simpson 2003). Several laboratory studies have demonstrated, as predicted, that generalist herbivorous insects do show a strategy consistent with equal distance regulation, whereas matched (e.g. closely related) specialists do not (reviewed in Simpson and Raubenheimer 2012). Figure 2 View largeDownload slide Geographic distribution of Macaca mulatta. The cross shows the location of our study site. Figure 2 View largeDownload slide Geographic distribution of Macaca mulatta. The cross shows the location of our study site. No study, however, has tested the predictions of this theory for any vertebrate, nor any animal in the wild. Previous field studies of rules of compromise, almost all of which involve primates, have shown a variety of regulatory patterns, none of which resembles the equal distance pattern predicted for ecological generalists. Bolivian spider monkeys (Ateles chamek) (Felton et al. 2009) and Borneo orangutans (Vogel et al. 2015) show the protein prioritization rule, in which absolute P intake is maintained tightly constant while NPE varies with variation in dietary macronutrient ratios (as depicted schematically in Figure 1). Mountain gorillas (Gorilla beringei) show the opposite response, over-eating leaf-derived protein to maintain NPE intake constant in periods when fruit-derived NPE is scarce (Rothman et al. 2011). Among primates, rhesus macaques (Macaca mulatta), are matched only by humans in their ecological generalism (Sayers 2013), having an exceptionally broad geographical distribution that spans diverse ecological habitats from cold temperate to sub-tropical latitudes (Fooden 2000; Figure 2). We capitalized on the opportunity offered by M. mulatta to test the predictions of diet breadth theory in a free-ranging primate. We chose to study the Taihangshan sub-species (M. mulatta tcheliensis), which live at the edge of the species range (Figure 2), in harsh high-latitude habitats and are exposed to substantial seasonal (Guo et al. 2011; Tian et al. 2013; Cui et al. 2015) and inter-annual (Lü et al. 2002; Guo et al. 2011; Zhang 2014; Cui et al. 2015) variation in the availability of various dietary foods. We predicted that if diet breadth theory applies beyond herbivorous insects, and if macronutrient flexibility plays a role in the ecological generalism of rhesus macaques, then the regulatory response by these monkeys to constrained variation in dietary macronutrient balance will resemble the equal distance rule (Figure 1). Our study system offered a powerful opportunity to test this. First, superimposed on seasonal variation (Song and Qu 1996; Lü et al. 2002) is appreciable inter-annual variation in the primary energy-providing food source, seeds of the oak (Quercus variabilis) (Lü et al. 2002). This enabled us to use inter-annual variation in diet within the same season as the independent variable, thus ensuring that the observed intakes were due to ecological constraint on diet quality and not seasonal variation in nutrient requirements (e.g., due to seasonal differences in nutritional costs for thermoregulation or reproduction). Second, inter-annual variation in diet quality was most pronounced in spring, the period in which reproducing females have accentuated nutritional requirements for lactation. The increased food intake by lactating females provided evidence that non-lactating monkeys could themselves have eaten more, and their observed macronutrient intakes thus reflected strategic choice rather than constraint. We discuss the implications of our results for wild primate studies and for diet breadth theory, and consider their relevance for understanding diet and macronutrient regulation by the most generalist primate of all, humans. MATERIALS AND METHODS Behavioral observations We collected data across 2013, 2014, and 2015 in the Taihangshan Macaque National Nature Reserve (TMNNR, 34°54’–35°40’N, 112°02’–113°45’E, 566 km2 in area) in Jiyuan, Henan Province, China. The area has a continental monsoon climate with average annual precipitation 634 mm and temperature 14.3 °C, and is dominated by temperate deciduous broadleaf forest. A wild macaque group (WangWu-1, WW-1) was assigned our target group. The group has been habituated to human observers since 2005 (Guo et al. 2011; Cui et al. 2015) and could thus be observed at close range (ca 2 m). In 2013, group size was 77 (including 20 adult females and 6 adult males), in 2014 88 (23 adult females and 8 adult males), and by August 2015 group size had steadily increased to 100 (27 adult females and 8 adult males). We studied the focal group from March 2013 to February 2016, systematically spanning different seasons across replicate years. In our study area, delineation of the seasons is March to May (spring); June to August (summer); September to November (autumn); December to February (winter) (Song and Qu 1996). For reasons explained below (Data selection), in the current paper we report on only the spring data. Our data covered 23 focal animals in each observation year. These included 17 principle animals (11 females and 6 males) selected to span ranks across the group. We prioritized obtaining at least one whole day observation for each of the principal animals, and thereafter switched to the other 6 animals. In cases where we lost sight of the animal (e.g. if it left the group temporarily for mating), we abandoned the data from that days follow and re-followed the same focal animal on a subsequent day. To avoid bias due to sex and reproductive state, the sex ratio and proportion of reproducing females were maintained as similar as possible across replicate years. The number of lactating females observed in 2013, 2014, and 2015 was 7, 10, and 10, respectively. In each of the 3 years, the number of non-lactating monkeys was 19 (13 non-lactating females and 6 adult males). To avoid pseudoreplication, monkeys that were observed on more than one occasion within the same year and same reproductive state (lactating or not lactating) were represented in the analysis by the average of the repeat observations. A focal animal was selected and followed continuously from dawn to dusk. If the focal animal went out of view for more than 5 min, that day’s observation was abandoned and the data excluded from the analysis. The same focal animal was followed for at least one whole day (range: 1–4 days). We recorded the time that each FA spent in feeding, excluding events <5 s. which we considered rejections. Simultaneously, we recorded the category of food item consumed (seeds, leaves etc., see below) and the source species. Unit counts (Altmann 1998; Vogel 2005; Rothman et al. 2008) were used to estimate the mass of each food item consumed. Sampling units used in our study were diet items including a single plant part (e.g. a leaf or seed), the approximate dimensions of a food item (for bark, twig), or, in the case of small leaves or fruits, the average number of items consumed (Rothman et al. 2008). The wet weight of each food unit was measured immediately after collection in the field (n = 50) (Rothman et al. 2008), and the units were calibrated monthly to account for intraspecific variation in plant morphology. Food units were sun dried at temperatures <40 °C, then transported to the laboratory for nutrient analysis. When the absolute number of ingested items could not be recorded for an event, we multiplied the ingestion time by the feeding rate that most closely matched the corresponding event. The feeding rates were derived using the same categories of food items as distinguished in the nutritional analysis (i.e. are specific to species and plant part). Food intakes were calculated by multiplying ingestion time with the feeding rate. A total of 21 different food items belonging to 19 species of food plants were observed to be eaten by focal animals in spring from 2013 to 2015, and grouped into 4 food types: 1) seeds, 2) leaves, 3) herbaceous plants and 4) buds. Among these, 4 items that contributed <1% of the dry weight intake of the focal animals were considered incidental and excluded from further analysis. The 17 most commonly eaten food items were dried, ground and 51 samples (3 replicates for every food item) were analyzed for their nutrient content. We measured body mass non-invasively, using small amounts of corn to coax the monkeys to step onto an electronic scale (Qianxuan TS-2010A4, accuracy 0.1 kg; Zhang et al. 2016) and recording the digital readout following stabilization for 3 s (Figure 3). Subjects were excluded from observations for 2 days after weighing as a precautionary measure to avoid any effect on behavior. As it was difficult to weigh particular animals on demand, we opportunistically weighed sub-sets of monkeys that represented the relevant age–sex classes, and used the means of these in our calculations. These comprised 6 males (mean± SD = 8.8 ± 1.2 kg), 9 non-reproductive females (6.8 ± 0.4 kg), 8 pregnant females (7.0 ± 0.5 kg), and 8 lactating females (6.6 ± 0.4 kg). Figure 3 View largeDownload slide (a) Macaque feeding on leaves. (b) Measuring the body mass of adult female macaques using an electronic scale (Qianxuan TS-2010A4 type). Figure 3 View largeDownload slide (a) Macaque feeding on leaves. (b) Measuring the body mass of adult female macaques using an electronic scale (Qianxuan TS-2010A4 type). Seed density Previous observations have shown that in our study population fallen Quercus seeds are an important dietary component that is heavily targeted when available (Zhang 2014; Cui et al. 2015). We therefore surveyed the density of these seeds monthly during the study. This was done using 225 sampling plots (1 m × 1 m), set within 20 quadrats (20 m × 20 m). The quadrats were laid at 100 m altitude intervals across a transect, which spanned 800–1700 m above sea level and were selected to include the range of topographies comprising the monkey’s home range (e.g. slopes, ditches etc.). Laboratory methods Dried food items were transported to the School of life sciences, Northwest University and College of Animal Science and Veterinary Medicine, Henan Agricultural University for the chemical analysis. The sampled food items were used to calculate daily and total dry mass of each item consumed by the focal animals. The nutritional content of the 44 most commonly eaten food items, as defined above, are reported on a dry matter (DM) basis in Supplementary Table 1. Ash was determined by combustion (Rothman et al. 2012); crude protein was measured by the Kjehdahl assay (BUICHI, KjelFIex K-360; Brooks et al. 1995; Nioi et al. 2012); available protein (AP) was measured through subtraction of acid detergent insoluble nitrogen from crude protein (Licitra et al. 1996). Fat was measured by ether extract with a fat analyzer (FOSS, SCINOTMST310) (Rothman et al. 2012). Samples were analyzed for neutral detergent fiber with residual ash (NDF; with sodium sulfite and a-amylase), then for acid detergent lignin (ADL) and acid detergent fiber (ADF) with an automatic fiber analyzer (ANKOM, A2000i; Goering and Van Soest 1970; Van Soest et al. 1991; Rothman et al. 2008). The percentages of total nonstructural carbohydrates (TNC) were calculated by subtracting the contributions of NDF, crude protein, fat, and ash from the rest of the nutrients (Johnson et al. 2013). The energetic contribution from neutral detergent fiber (NDF) was estimated following equation #2 in Rothman et al. (2008): NDFdigestibilityDM=100−(100%ADL in diet%ADL in feces×%NDF in feces%NDF in diet) The energy value of NDF was calculated by multiplying the conversion values of 16.7 kJ per g TNC by the NDF digestibility coefficients, and then subtracting the 4 kJ per g estimated to be consumed by gut microbes (Conklin-Brittain et al. 2006). To estimate the NDF digestion coefficient, we collected the fecal samples (n = 36) from 6 females and 6 males monthly from their sleep sites. As there were no significant differences in sex for NDF digestibility coefficient (P > 0.05), we averaged the data across sexes. This gave NDF digestibility coefficients of 0.533, which were entered into the standard equation to estimate conversion values: [(16.7-4)* NDF digestibility coefficient kJ/g] NDF. For macronutrient energy contributions, we used the conventional conversion values of 37.7 kJ per g crude fat, 16.7 kJ per g AP, and 16.7 kJ per g TNC (NRC 1989). We summed the energetic contributions from crude fat, non-structural carbohydrates and the available fraction of NDF to yield non-protein energy (NPE). In addition, quantitative determination of tannin in diets also was evaluated on a DM basis by P-B Spectrophotometry (Price and Larry 1977; Supplementary Table 2). Non-digestible fiber was calculated by subtracting the microbial digestible component of NDF from the total NDF intake on a dry matter basis using the following equation: Non−digestible fiber=total NDF×(1−NDF digestibility coefficient) Calculating feeding data Feeding variables were calculated as follows: 1) Observed daily nutrient intake of focal animals on a dry matter basis: Mj=∑i=1kni⋅hi⋅cij⋅ (1) MNDF=∑i=1kni⋅hi⋅ciNDF (2) where, i……..k represent the number of food items; j = nutrients (TNC, AP, and Fat); Mj= daily nutrient intake of j (g); MNDF= daily nutrient intake of NDF (g); cij= the concentration of j in i; ciNDF= the concentration of NDF in i; ni= the number of observed daily food units intake; hi= the weight of i on a dry matter basis (g). 2) Observed daily nutrient intake per focal animal on an energy basis. NVj=Mj⋅Ej+ MNDF⋅(ENDF−4)*BNDF (3) where, j= nutrients (TNC, AP and Fat); NVj (Nutritional Value, NV) = observed daily energy intake of j (kJ). Mjand MNDF are from equations (1) and (2). EFat = 37.7kJ/g, ETNC = 16.7kJ/g, ENDF = 16.7kJ/g, and EAP = 16.7kJ/g (Song et al. 1996; NRC 1989). BNDF=NDF digestibility coefficients in spring. 3) Observed daily nutrient intake per kilogram of body weight of focal animal on an energy basis. I=NVjN·w (4) where I represents the observed daily nutrient intake per focal animal on an energy basis kJ/(day·BWkg). In addition, N represents the number of samples and w is body weight. 4) We also calculated percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring (on a dry matter basis or an energy basis). Pi=ki∑i=1kki×100% (5) where Pi represents seasonal percentage contribution of different foods (plant parts) i and ki= mean intake of i. Data selection The current analysis arises from a larger program involving data over all 4 seasons and 3 years (2013, 2014, and 2015). However, in order to test the hypothesis of the present study, namely, that macaques show a generalist pattern of macronutrient regulation, it is important to establish that observed variance in intakes reflects the response of macronutrient regulatory systems to constrained dietary imbalance (Figure 1) rather than seasonal changes in macronutrient requirements (for example due to seasonal temperature differences, Terrien et al. 2011). We were able to do this by analyzing data within a single season, spring (March–May), in which available foods, and hence dietary macronutrient balance, varied substantially across the 3 study years. Seasonal comparisons are the subject of a separate publication (Cui et al., in preparation). As detailed above, we distinguished lactating females from non-lactating females and males. As nutrient and energy intakes of males and non-lactating females did not differ statistically (see Results), we combined these into a single category for comparison with lactating females. The contrast of lactating and non-lactating monkeys not only provided a separate instance in which to test whether rhesus macaques show the equal distance pattern of macronutrient regulation, but also provided an important control in the test of our hypothesis. Specifically, since lactating females have increased nutrient requirements, we expected that they would have increased food intakes (Thompson 2013). If so, this would show that non-lactating monkeys could likewise have eaten more than they did, providing confidence that their intakes reflected macronutrient regulation rather than restricted availability of food or upper limits on the ingestion of the non-nutrient dietary components undigestible fiber or tannins. Statistical analysis and hypothesis testing Spearman Rank Correlation was used to test the correlation between the proportional contribution of Quercus seeds to the diet of Taihangshan macaques and their availability across the 24 months during the 3-year study period in which these seeds were available. We used Kruskal–Wallis H to test for differences across the three study years in the number of seeds remaining on the ground in spring. Linear mixed models were used to analyze the influence of sex (non-lactating females vs. males) and years on energy intake and dietary macronutrient ratios. Sex was designated a fixed effect, while year was considered a random effect. We also used Kruskal–Wallis H to test for differences in the dietary P/NPE ratio in spring between 2013, 2014, and 2015. We used linear mixed model to compare intakes of energy, tannin, and indigestible fiber between non-lactating monkeys and lactating females across 3 years. Lactating status was designated a fixed effect, and year a random effect. To examine the patterns of NPE and AP intake of adult macaques, we plotted macronutrient and energy intakes within the nutritional geometry framework, a multidimensional approach in which each axis represents a different food component (Raubenheimer et al. 2009, Simpson and Raubenheimer 2012). This enabled us to visually compare observed data to 3 alternative models of regulation, P prioritization (protein intake is maintained constant), NPE prioritization (NPE intake constant) and the equal distance rule predicted for generalists (macronutrient energy intake constant; Figure 1). We statistically tested the fit of the data to these models as follows. First, we calculated the mean intake across the 3 years of P, NPE, and total energy. For each variable, we then expressed the observed intake of individual monkey as a percentage of the grand mean, and compared these percentages across years using linear mixed models with year designated a fixed effect. The equal distance rule predicts that P intake and NPE intake will differ across years, whereas total energy intake will remain more constant. In contrast, the P prioritization and NPE prioritization models predict that P and NPE intake will remain constant, respectively, while the other 2 variables differ with dietary macronutrient balance. We used IBM SPSS version 19.0 (IBM Corp., Armonk, NY) for data analysis. All data were expressed as means ± SE, and the significance level for statistical tests was set at α = 0.05. RESULTS Foods The proportional contribution to the diet of different food categories varied substantially across years (Table 1). The most notable dietary component that varied was seeds of Quercus trees, which when available are strongly targeted by the macaques, most likely for their high TNC content (~57% by weight; Supplementary table 1). There was a significant positive correlation between the availability (density on the ground) of these seeds and their contribution to the diet of Taihangshan macaques on a dry matter basis (r = 0.788, P < 0.001, n = 24). Seed production occurred over a limited time of the year (autumn) in our study area, but in mast years seeds were sufficiently prolific that they were not depleted by the macaques and remained abundantly available until the end of the following spring. Our seed density estimates suggested that the number of seeds remaining on the ground in spring varied significantly across the 3 study years (2013 = 1.4 ± 0.1/m2, 2014 = 0.0 ± 0.0/m2, 2015 = 1.2 ± 0.1/m2; Kruskal–Wallis test: χ2 = 151.98, df = 2, 674, P< 0.001). Table 1 Percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring from 2013 to 2015 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 View Large Table 1 Percentage contribution of different foods (plant parts) to the diet of Taihanshan macaques in spring from 2013 to 2015 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 Year Part On a dry matter basis On an energy basis % Contribution % Contribution 2013 Seeds (Quercus) 59.4 68.4 Leaves 21.3 17.5 Herbaceous leaves 17.1 12.8 Buds 2.2 1.4 2014 Seeds (Quercus) 0 0 Leaves 82.6 84.5 Herbaceous leaves 15.3 13.7 Buds 2.1 1.8 2015 Seeds (Quercus) 45.7 49.9 Leaves 41.4 39.1 Herbaceous leaves 12.0 10.2 Buds 0.9 0.8 View Large Energy and macronutrient intake Neither energy intakes (F = 0.13, df = 1, 55, P = 0.721) nor dietary macronutrient ratios (F = 0.04, df = 1, 55, P = 0.838) differed between non-lactating females and males, and we therefore combined these groups to contrast in the analysis with lactating females. Figure 4 shows the mean energy and macronutrient intakes in spring of 2013, 2014, and 2015 in non-lactating monkeys and lactating female monkeys. The dietary P/NPE ratio differed significantly across years (Kruskal–Wallis test: χ2 = 69.65, df = 2, 83; P < 0.001), having been lower in 2013 and 2015, when seed availability was high (Table 1), and high in 2014 when seeds were scarce. As predicted, however, lactating females ingested substantially and significantly more energy than non-lactating monkeys (F = 124.52, df = 1, 82, P < 0.001). For lactating monkeys, the mean energy intake across the 3 years was 413 kJ/day·BWKg (shown by the dashed diagonal line in Figure 4), compared with 318 kJ/day·BWKg for non-lactating monkeys (the solid diagonal line). Figure 4 View largeDownload slide Mean daily energy and macronutrient intakes of macaques in spring of 2013, 2014, and 2015. Data are shown separately for non-lactating adults (solid circles) and lactating females (hollow circles). Gray diagonals are energy isolines representing mean total energy intakes across the 3 years for non-lactating monkeys (solid line) and lactating females (dashed line). Mathematically, these isolines are derived from the equation X + Y = constant, where X and Y represent the 3-year mean of available protein (AP) and non-protein energy intake, respectively. As shown in Figure 1, these lines represent the equal distance model of macronutrient regulation. Figure 4 View largeDownload slide Mean daily energy and macronutrient intakes of macaques in spring of 2013, 2014, and 2015. Data are shown separately for non-lactating adults (solid circles) and lactating females (hollow circles). Gray diagonals are energy isolines representing mean total energy intakes across the 3 years for non-lactating monkeys (solid line) and lactating females (dashed line). Mathematically, these isolines are derived from the equation X + Y = constant, where X and Y represent the 3-year mean of available protein (AP) and non-protein energy intake, respectively. As shown in Figure 1, these lines represent the equal distance model of macronutrient regulation. As our hypothesis of equal distance regulation predicts that total energy intakes will be more constant in the face of variation in dietary macronutrient balance than P or NPE intakes (Figure 1), for each variable we compared across years the deviations from the 3 year mean separately for non-lactating and lactating monkeys (Figure 5). For non-lactating monkeys, there were substantial inter-annual differences both in P (F = 308.92, df = 2, 54, P < 0.001) and NPE intake (F = 17.40, df = 2, 54, P < 0.001), but there was no significant difference in total energy intake (F = 1.90, df = 2, 54, P = 0.159). This demonstrates, as predicted, that the data fitted the equal distance model of regulation, in which total energy intakes are constant, but did not fit the P prioritization or NPE prioritization patterns. For lactating females, there were likewise marked inter-annual differences in P (F = 114.01, df = 2, 24, P < 0.001) and NPE intake (F = 21.70, df = 2, 26, P < 0.001). There was also a significant inter-annual difference in total energy intake (F = 4.33, df = 2, 24, P = 0.025). This difference, which was small compared to the inter-annual differences in P and NPE intake (Figure 5), was due to marginally reduced energy intake in 2014 when dietary P/NPE was highest (Figures 4 and 5), with no difference between 2013 and 2015 (F = 0.516, df = 1, 15, P = 0.484). This shows that, in common with their non-lactating counterparts, lactating monkeys showed neither P prioritization nor NPE prioritization, with their pattern of regulation most closely resembling the equal distance rule. Figure 5 View largeDownload slide Annual deviations from the 3 year mean for intakes of total energy, available protein and non-protein energy. The horizontal lines represent the levels of intake that would be expected under (a) the equal distance rule (constant total energy), (b) protein prioritization, and (c) prioritization of non-protein energy. Figure 5 View largeDownload slide Annual deviations from the 3 year mean for intakes of total energy, available protein and non-protein energy. The horizontal lines represent the levels of intake that would be expected under (a) the equal distance rule (constant total energy), (b) protein prioritization, and (c) prioritization of non-protein energy. Dietary constraint due to non-nutritional factors The fact that lactating females ingested more macronutrients than non-lactating monkeys demonstrates that the equal distance pattern observed for the latter group was not due to restricted food availability. We also examined two further potential confounds, both of which have previously been implicated in restricting food intake by monkeys, non-digestible fiber (Milton 1999) and plant tannins (Felton et al. 2009). Figure 6 shows that across the 3 study years intakes of both tannin (F = 211.97, df = 1, 82, P < 0.001) and indigestible fiber (F = 142.32, df = 1, 82, P < 0.001) were significantly higher for lactating than non-lactating females. This demonstrates that the maximum thresholds for fiber and tannin intakes were not reached by non-lactating monkeys, providing further evidence that their pattern of macronutrient intake reflects a nutritional regulatory strategy. Figure 6 View largeDownload slide Intakes of non-digestible fiber (a) and tannin (b) by non-lactating and lactating monkeys in the spring of 2013, 2014, and 2015. Figure 6 View largeDownload slide Intakes of non-digestible fiber (a) and tannin (b) by non-lactating and lactating monkeys in the spring of 2013, 2014, and 2015. DISCUSSION As briefly reviewed in the Introduction, several studies have examined the responses of free-ranging primates to ecologically driven constraints on the ratios of available macronutrients. In addition to extending the comparative database of nutritional regulatory responses in free-ranging primates, the present study is unique in two important respects: it is a targeted test of a hypothesis generated by theory, and the study system provided an unprecedented opportunity to test the predictions of the theory in free-ranging animals in an ecological context. Previous studies of macronutrient regulation patterns in free-ranging primates have not been predictive, but largely descriptive, for two reasons. Firstly, the macronutrient regulatory patterns have been studied in too few primates to generate worthwhile predictions inductively. Secondly, there exists very little theory linking macronutritional regulatory responses to specific ecological circumstances, and this limits the scope for generating predictions deductively. An exception, however, is the dietary specialist-generalist spectrum, and its links to ecological niche theory (Simpson and Raubenheimer 2012; Machovsky-Capuska et al. 2016). Drawing on this theory, and based on the broad geographical distribution and wide range of habitats occupied by M. mulatta, we anticipated that this species would show the equal distance regulatory response to macronutrient imbalance, which as explained in the Introduction is predicted on theoretical grounds for generalist species. Our analysis provides strong support for this prediction. The only previous empirical evidence for the association of equal distance regulation with dietary generalism comes from laboratory studies of insects (reviewed in Behmer 2009; Simpson and Raubenheimer 2012). In these studies, experimental animals were confined to one of several synthetic diets formulated to vary systematically in macronutrient ratios, and the ad-libitum intakes of the different dietary treatment groups compared. The results showed that generalist insect species conformed more closely to the equal distance pattern of regulation than matched species that are more specialized. Furthermore, hybrids between a generalist and specialist species were found to have an intermediate pattern of macronutrient regulation (Lee et al. 2006), suggesting a heritable component to nutritional regulatory strategies. Finally, studies of insects in which the same genotype generates alternative developmental pathways in different environments provide intriguing evidence for the link between equal distance macronutrient regulation and ecological generalism. Locusts reared at high population densities develop phenotypes that are morphologically, behaviorally, and ecologically very different than those reared at low densities. Among the differences, is that locusts reared in high densities (form gregaria) are nutritionally more generalist than those reared at low densities (form solitaria). Studies have shown, as predicted by the theory, that the generalist morph has a pattern of macronutrient regulation that resembles the equal distance pattern, whereas the solitarious morph does not (Simpson et al. 2002). Similar results were found in caterpillars that have gregarious and solitarious developmental phenotypes (Lee et al. 2004). It is substantially more challenging to perform such tests on animals in the wild, because the ecological requirements for conducting the relevant studies non-invasively are stringent. Our study system provided a powerful opportunity to test the equal distance-generalist hypothesis, in a way that goes some distance towards resolving the trade-off between experimental control afforded by captive animal studies and the ecological realism that justifies field studies. First, circumstances are needed in which animals are constrained to eat diets that differ in their macronutrient ratios. If not, and animals have free choice of a wide range of nutritionally complementary foods from which to compose their diet, then the results are likely to reflect a regulated intake target rather than a strategic response to constrained macronutrient imbalance, as observed in prolonged observations of a chacma baboon (Johnson et al. 2013). Our study system satisfied this requirement, because inter-annual variance in the relative availability of different foods imposed constraints on the macronutrient ratios of the diets that could be assembled from those foods in different years. Second, it is important to establish that the animals are restricted only in terms of dietary macronutrient ratios, and not the overall quantity of food available or non-nutrient factors such as indigestible fiber and tannins, which at high levels can limit food intake by primates (Amato and Garber 2014; Lambert and Rothman 2015). This is because rules of compromise are a measure of how ad-libitum intakes vary across diets that differ in their nutrient balance, and if factors other than macronutrients constrain consumption then recorded intakes do not exclusively reflect homeostatic regulation. We were able to address this by showing that intakes of energy, fiber, and tannins by lactating females were substantially higher than non-lactating monkeys that were observed concurrently. This suggests that the intakes of non-lactating monkeys were not due to food shortage or limits on the ability to ingest fiber or tannin, but reflected macronutritional priorities. Third, the between-group comparisons should not confound the effects of dietary macronutrient balance with differences in macronutrient requirements that arise, for example, from seasonal differences in thermoregulatory requirements. A powerful benefit of our study system is that it enabled us to compare the responses of monkeys to variable macronutrient balance across 3 years within the same season, thereby controlling for such factors as seasonal temperature differences that are known to influence nutrient requirements and food intake (Terrien et al. 2011). While the lactating females played an important role in our study by allowing us to discount quantitative food limitation and non-nutrient factors (tannin and fiber) as the causes of the intake pattern by non-lactating females, we cannot interpret their pattern of intake with equal confidence. This is because without further controls it is difficult to be sure that the equal distance-like intake pattern of lactating females was not influenced by limitations on the amount of food that they could access, or by upper limits to the amounts of fiber or tannins that they could ingest. Indeed, the marginally lower energy consumption that we observed for lactating females in 2014 compared with 2013 and 2015 is quite possibly due to limitations on the capacity of lactating monkeys to meet their high energy demands when acorns are scarce. Such uncertainty over the causes of the pattern observed in our study for lactating females underscores the importance of their role in helping to interpret the regulatory pattern of non-lactating monkeys with a high level of certainty. We are thus confident that our data provide strong evidence for the first case of equal distance macronutrient regulation in a primate, and for any animal in the wild. That this was demonstrated in the wild is particularly gratifying, partly because of the challenges of achieving this is in free-ranging animals, but principally because the data reflect regulatory responses to natural, ecologically-generated, variation in available diets. This is, nonetheless, a single-species study, and on its own contributes only one data point towards comparative analysis of the relationship between diet breadth and macronutrient regulatory strategy. On the other hand, the fact that our expectation was developed on theoretical grounds, and has previously been observed in controlled laboratory studies of insect species, provides confidence that the data do reflect broader generality. In so doing, they also substantiate the underlying theory. It is interesting to speculate on the adaptive benefits and ecological consequences for M. mulatta of the flexible macronutrient intake that characterizes the equal distance regulatory strategy. One possibility is that these monkeys tolerate sub-optimal macronutrient ratios, but are otherwise not adapted to them. For example, Nie et al. (2015) found that giant pandas in the wild subsist for several months on nutritionally imbalanced diets, and to reproduce need to migrate between habitats that offer complementary nutrition (Nie et al. 2015). The fact that lactation was maintained across the 3 study years regardless of the dietary macronutrient ratio (Figure 4) suggests that this was not the case for the monkeys in our study. The suggestion, therefore, is that M. mulatta are truly robust to variation in nutrient balance, with a flexible ability to utilize macronutrients inter-changeably. This is predicted by the theory to be a phenotype that is characteristic of a species capable of occupying diverse and variable habitats. Finally, if equal distance regulation is more generally associated with ecological generalism, this gives rise to the fascinating question of why the most geographically widespread and ecologically and dietarily diverse primate of all—humans—does not show this pattern of regulation. Rather, in common with spider monkeys and orangutans, humans show the protein prioritization pattern (Figure 1; Gosby et al. 2014), in which protein intake is maintained relatively constant while non-protein energy is regulated less tightly. One possibility is that humans are dietary generalists only in the sense of being capable of subsisting on a wide range of foods (“food generalists”, sensuMachovsky-Capuska et al. 2016), but at the level of macronutrients are more specialized (“macronutritional specialists”). Consistent with this is the observation that the P:NPE ratio of human diets is remarkably invariant across populations, with a few known exceptions such as the traditional Inuit diet (Simpson and Raubenheimer 2005). Further studies are needed to expand the database of primate macronutrient regulatory strategies to ultimately enable phylogenetically controlled analysis of how these relate to variance in evolutionary ecology, and better understand the implications for foraging theory, niche theory, and human biology. SUPPLEMENTARY MATERIAL Supplementary data are available at Behavioral Ecology online. We thank Jiyuan Administration of Taihangshan Macaque National Nature Reserve for permission to conduct fieldwork and logistic support throughout this study. We are grateful to Mr. Hou Jiafu, Mr. Zhao Guoliang, and Mr. Kong Maojuan for their assistance with field work. We thank Dr. Zhu Shixin (School of Life Science, Zhengzhou University) for plant species identification and Dr. Paul A. Garber (Department of Anthropology, University of Illinois, USA) for his helpful suggestions on the study. We also thank Prof. Li Baoguo, Prof. Guo Songtao, and Mr. Hou Rong for their assistance with the nutritional analyses. This work was supported by the National Natural Science Foundation of China (30970378, 31170503). Data accessibility: Analyses reported in this article can be reproduced using the data provided by Cui et al. (2017). FUNDING This study was supported by the National Natural Science Foundation of China (30970378, 31170503). Conflict of Interest: The authors declare no conflict of interest. REFERENCES Altmann SA . 1998 . Foraging for survival: yearling baboons in Africa . Chicago (IL) : University of Chicago Press . Amato KR , Garber PA . 2014 . Nutrition and foraging strategies of the black howler monkey (Alouatta pigra) in Palenque National Park, Mexico . Am J Primatol . 76 : 774 – 787 . Google Scholar CrossRef Search ADS PubMed Behmer ST . 2009 . Insect herbivore nutrient regulation . Entomology . 54 : 165 – 187 . Google Scholar CrossRef Search ADS Brooks S , Lampi BJ , Sarwar G , Botting HG . 1995 . A comparison of methods for determining total body protein . Anal Biochem . 226 : 26 – 30 . Google Scholar CrossRef Search ADS PubMed Conklin-Brittain NL , Knott CD , Wrangham RW . 2006 . Energy intake by wild chimpanzees and orangutans: methodological considerations and a preliminary comparison . Cam S Bio Evol Anthr . 1 : 445 – 472 . Cui Z , Wang Z , Zhao G , Lu J . 2015 . Seasonal variations in activity budget of adult female rhesus macaques (Macaca Mulatta) at Mt Taihangshan area, Jiyuan, China: effects of diet and temperature . Acta Theriologica Sinica . 35 : 138 – 146 . Cui Z , Wang Z , Shao Q , Raubenheimer D , Lu J . 2017 . Data from: macronutrient signature of dietary generalism in an ecologically diverse primate in the wild . Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.mb342.2. Felton AM , Felton A , Raubenheimer D , Simpson SJ , Foley WJ , Wood JT , Wallis IR , Lindenmayer DB . 2009 . Protein content of diets dictates the daily energy intake of a free-ranging primate . Behav Ecol . 20 : 685 – 690 . Google Scholar CrossRef Search ADS Fooden J . 2000 . Systematic review of the rhesus macaque, Macaca mulatta (Zimmermann, 1780) . Fieldiana Zool . 96 : 1 – 180 . Futuyma DJ , Moreno G . 1988 . The evolution of ecological specialization . Annu Rev Ecol Syst . 19 : 207 – 233 . Google Scholar CrossRef Search ADS Goering HK , Van Soest PJ . 1970 . Forage fiber analyses: apparatus, reagents, procedures, and some applications. Agriculture Handbook No. 379 . Washington : USDA . p. 20 . Gosby AK , Conigrave AD , Raubenheimer D , Simpson SJ . 2014 . Protein leverage and energy intake . Obes Rev . 15 : 183 – 191 . Google Scholar CrossRef Search ADS PubMed Guo X , Wang Z , Chen J , Tian J , Wang B , Lu J . 2011 . Winter and spring food habits of Macaca mulatta tcheliensis in Taihangshan National Nature Reserve in Henan Province of China . Chin J Ecol . 30 : 483 – 488 . Jensen K , Mayntz D , Toft S , Clissold FJ , Hunt J , Raubenheimer D , Simpson SJ . 2012 . Optimal foraging for specific nutrients in predatory beetles. P Roy Soc B-Biol Sci . 279 (1736) : 2212 – 2218 . Google Scholar CrossRef Search ADS Johnson CA , Raubenheimer D , Rothman JM , Clarke D , Swedell L , 2013 . 30 days in the life: daily nutrient balancing in a wild chacma baboon . PLoS One . 8 : e70383 . Google Scholar CrossRef Search ADS PubMed Lambert JE , Rothman JM . 2015 . Fallback foods, optimal diets, and nutritional targets: primate responses to varying food availability and quality . Annu Rev Anthropol . 44 : 493 – 512 . Google Scholar CrossRef Search ADS Lee KP , Behmer ST , Simpson SJ . 2006 . Nutrient regulation in relation to diet breadth: a comparison of heliothis sister species and a hybrid . J Exp Biol . 209 : 2076 . Google Scholar CrossRef Search ADS PubMed Lee KP , Simpson SJ , Raubenheimer D . 2004 . A comparison of nutrient regulation between solitarious and gregarious phases of the specialist caterpillar, Spodoptera exempta (Walker) . J Insect Physiol . 50 : 1171 – 1180 . Google Scholar CrossRef Search ADS PubMed Licitra G , Hernandez TM , Van Soest PJ . 1996 . Standardization of procedures for nitrogen fractionation of ruminant feeds . Anim Feed Sci Tech . 57 : 347 – 358 . Google Scholar CrossRef Search ADS Lü J , Lu J , Li J , Qu W . 2002 . Feeding habits of Macaca mulatta in Taihang Mountains . Chin J Ecol . 21 : 29 – 31 . Machovsky-Capuska GE , Senior AM , Simpson SJ , Raubenheimer D . 2016 . The multidimensional nutritional niche . Trends Ecol Evol . 31 : 355 – 365 . Google Scholar CrossRef Search ADS PubMed Milton K . 1999 . Nutritional characteristics of wild primate foods: do the diets of our closest living relatives have lessons for us ? Nutrition . 15 : 488 – 498 . Google Scholar CrossRef Search ADS PubMed Nie Y , Zhang Z , Raubenheimer D , Elser JJ , Wei W , Wei F . 2015 . Obligate herbivory in an ancestrally carnivorous lineage: the giant panda and bamboo from the perspective of nutritional geometry . Funct Ecol . 29 : 26 – 34 . Google Scholar CrossRef Search ADS Nioi C , Kapel R , Rondags E , Marc I . 2012 . Selective extraction, structural characterisation and antifungal activity assessment of napins from an industrial rapeseed meal . Food Chem . 134 : 2149 – 2155 . Google Scholar CrossRef Search ADS PubMed NRC . 1989 . Recommended dietary allowances . Washington (DC) : National Academy Press . Price Martin L , Butler Larry G . 1977 . Rapid visual estimation and spectrophotometric determination of tannin content of sorghum grain . J Agric Food Chem . 25 : 1268 – 1273 . Google Scholar CrossRef Search ADS Raubenheimer D , Simpson SJ . 1993 . The geometry of compensatory feeding in the locust . Anim Behav . 45 : 953 – 964 . Google Scholar CrossRef Search ADS Raubenheimer D , Simpson SJ . 1997 . Integrative models of nutrient balancing: application to insects and vertebrates . Nutr Res Rev . 10 : 151 – 179 . Google Scholar CrossRef Search ADS PubMed Raubenheimer D , Simpson SJ . 2003 . Nutrient balancing in grasshoppers: behavioural and physiological correlates of dietary breadth . J Exp Biol . 206 : 1669 – 1681 . Google Scholar CrossRef Search ADS PubMed Raubenheimer D , Simpson SJ , 2016 . Nutritional ecology and human health . Annu Rev Nutr . 36 : 603 – 626 . Google Scholar CrossRef Search ADS PubMed Rothman JM , Chapman CA , Van Soest PJ . 2012 . Methods in primate nutritional ecology: a user’s guide . Int J Primatol . 33 : 542 – 566 . Google Scholar CrossRef Search ADS Rothman JM , Dierenfeld ES , Hintz HF , Pell AN . 2008 . Nutritional quality of gorilla diets: consequences of age, sex, and season . Oecologia . 155 : 111 – 122 . Google Scholar CrossRef Search ADS PubMed Rothman JM , Raubenheimer D , Chapman CA . 2011 . Nutritional geometry: gorillas prioritize non-protein energy while consuming surplus protein . Biol Lett . 7 : 847 – 849 . Google Scholar CrossRef Search ADS PubMed Sayers K , 2013 . On folivory, competition, and intelligence: generalisms, overgeneralizations, and models of primate evolution . Primates . 54 : 111 – 124 . Google Scholar CrossRef Search ADS PubMed Simpson SJ , Raubenheimer D . 1993 . A multi-level analysis of feeding behaviour: the geometry of nutritional decisions . Philos T R Soc B . 342 : 381 – 402 . Google Scholar CrossRef Search ADS Simpson SJ , Sibly RM , Lee KP , Behmer ST , Raubenheimer D . 2004 . Optimal foraging when regulating intake of multiple nutrients . Anim Behav . 68 : 1299 – 1311 . Google Scholar CrossRef Search ADS Simpson SJ , Raubenheimer D . 2005 . Obesity: the protein leverage hypothesis . Obes Rev . 6 : 133 – 142 . Google Scholar CrossRef Search ADS PubMed Simpson SJ , Raubenheimer D . 2012 . The nature of nutrition . Princeton (NJ) : Princeton University Press . Google Scholar CrossRef Search ADS Simpson SJ , Raubenheimer D , Behmer ST , Whitworth A , Wright GA . 2002 . A comparison of nutritional regulation in solitarious- and gregarious-phase nymphs of the desert locust Schistocerca gregaria . J Exp Biol . 205 : 121 – 129 . Google Scholar PubMed Song C , Qu W . 1996 . Scientific survey of Taihangshan Macaque nature reserve . Beijing : China Forestry Publishing House . Terrien J , Blanc S , Zizzari P , Epelbaum J , Aujard F . 2011 . Physiological responses to chronic heat exposure in an aging non-human primate species, the gray mouse lemur (Microcebus murinus) . Exp Gerontol . 46 : 747 – 754 . Google Scholar CrossRef Search ADS PubMed Thompson ME . 2013 . Comparative reproductive energetics of human and nonhuman primates . Anthropology . 42 : 287 – 304 . Google Scholar CrossRef Search ADS Tian J , Wang Z , Lu J , Wang B , Chen J . 2013 . Reproductive parameters of female Macaca mulatta tcheliensis in the temperate forest of Mount Taihangshan, Jiyuan, China . Am J Primatol . 75 : 605 – 612 . Google Scholar CrossRef Search ADS PubMed Van Soest PV , Robertson JB , Lewis BA . 1991 . Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition . J Dairy Sci . 74 : 3583 – 3597 . Google Scholar CrossRef Search ADS PubMed Vogel ER ., 2005 . Rank differences in energy intake rates in white-faced capuchin monkeys, Cebus capucinus: the effects of contest competition . Behav Ecol Sociobiol . 58 : 333 – 344 . Google Scholar CrossRef Search ADS Vogel ER , Rothman JM , Moldawer AM , Bransford TD , Thompson ME , Noordwijk MAV , Atmoko SSU , Crowley B , Knott CD , Erb WM , et al. 2015 . Coping with a challenging environment: nutritional balancing, health, and energetics in wild Bornean orangutans . Am J Phys Anthropol . 156 : 314 – 315 . Zhang P , Lyu MY , Wu CF , Chu YMR , Han N , Yang D , Hu K . 2016 . Variation in body mass and morphological characters in Macaca mulatta brevicaudus from Hainan, China . Am J Primatol . 78 : 679 – 698 . Google Scholar CrossRef Search ADS PubMed Zhang Y . 2014 . The interactions among plant seeds, rodent, and insects in Mt. Taihangshan area . Zhengzhou : Zhengzhou University . © The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Behavioral EcologyOxford University Press

Published: Apr 30, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off