Stable isotope analysis is an increasingly used molecular tool to reconstruct the diet and ecology of elusive primates such as unhabituated chimpanzees. The consumption of C4 plant feeding termites by chimpanzees may partly explain the relatively high carbon isotope values reported for some chimpanzee communities. However, the modest availability of termite isotope data as well as the diversity and cryptic ecology of termites potentially consumed by chimpanzees obscures our ability to assess the plausibility of these termites as a C4 resource. Here we report the carbon and nitrogen isotope values from 79 Macrotermes termite samples from six savanna woodland chimpanzee research sites across equatorial Africa. Using mixing models, we estimated the proportion of Macrotermes C4 plant consumption across savanna woodland sites. Additionally, we tested for isotopic differences between termite colonies in different vegetation types and between the social castes within the same colony in a subset of 47 samples from 12 mounds. We found that Macrotermes carbon isotope values were indistinguishable from those of C3 plants. Only 5 to 15% of Macrotermes diets were comprised of C4 plants across sites, suggesting that they cannot be considered a C4 food resource substantially influencing the isotope signatures of consumers. In the Macrotermes subsample, vegetation type and caste were significantly correlated with termite carbon values, but not with nitrogen isotope values. Large Macrotermes soldiers, preferentially consumed by chimpanzees, had comparably low carbon isotope values relative to other termite castes. We conclude that Macrotermes consumption is unlikely to result in high carbon isotope values in either extant chimpanzees or fossil hominins.
Citation: Phillips S, Scheffrahn RH, Piel A, Stewart F, Agbor A, Brazzola G, et al. (2021) Limited evidence of C4 plant consumption in mound building Macrotermes termites from savanna woodland chimpanzee sites. PLoS ONE 16(2): e0244685. https://doi.org/10.1371/journal.pone.0244685
Editor: Matheus C. Carvalho, Southern Cross University, AUSTRALIA
Received: July 10, 2020; Accepted: December 14, 2020; Published: February 10, 2021
Copyright: © 2021 Phillips et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was generously funded by the Max Planck Society, the Max Planck Society Innovation Fund, and the Heinz L. Krekeler Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Our understanding of wild chimpanzee (Pan troglodytes) feeding ecology has been primarily informed by direct observations of feeding behavior within the limited number of chimpanzee communities consistently monitored by long-term research projects (e.g. [1, 2]). This bias towards a small number of chimpanzee communities has been tempered by the increasing use of indirect methods, such as stable isotope analysis, that enable large-scale cross site comparisons of the various feeding behaviors of both habituated and unhabituated chimpanzee communities [3–10]. Insights obtained from such studies are only as good as our understanding of the various stable isotope ratios, such as carbon (δ13C) and nitrogen (δ15N), of consumed organisms by chimpanzees at different locales [3–5, 11–13]. Chimpanzee δ13C values, from both habituated and unhabituated communities, thus far corroborate observational data that indicate chimpanzees primarily feed on vegetation relying on the C3 photosynthetic pathway as well as organisms consuming such vegetation [3–8, 14–17]. Still, small variation in δ13C values within the observed range of chimpanzee values may be representative of differences in feeding behavior across communities or across demographic classifications within communities [3–5, 8, 14]. For example, some savanna woodland chimpanzee communities exhibit δ13C values that are higher than baseline C3 vegetation even if they are still consistent with a predominately C3 based diet [3, 4, 8, 15]. The consumption of C4 plant reliant termites has been posited as a potential contributor to relatively high δ13C values in some savanna woodland chimpanzee communities .
Termites contribute key nutrients to primate diets [18, 19] and even termite soil is particularly rich in nutrients [20–23]. Across Africa many, yet far from all, chimpanzee populations have been observed to forage for termites  with a common preference for the large, mound-building and fungus-growing termites of the genus Macrotermes (Macrotermitinae) (summarized in [25–29]). The ecology of Macrotermes, and termites in general, is often cryptic and difficult to observe under natural conditions . Additionally, many behaviors within the genus Macrotermes [27, 31–33] may vary significantly intraspecifically depending on the ecological context. Macrotermes is a genus of termites found exclusively in the Old World tropics. Some species, such as M. bellicosus and M. subhyalinus, are broadly distributed across the African continent [31, 34]. In general, Macrotermes spp. build large epigeal mounds and spend most of their time in the subterranean chambers and galleries throughout the year. Macrotermes workers forage for cellulosic debris such as leaf litter, dead grass, woody litter, and wood  including live crop plants . Foraged items are returned to the nest as nutritional substrate for the growth of Termitiomyces fungus combs that consists of fungal biomass and partly decayed plant matter [37–40]. Older workers primarily feed on the fungus combs and the soldier castes rely on the workers to feed them directly with pieces of these fungal combs. By contrast, younger workers may subsist primarily on foraged plant matter as well as on the protein-rich Termitiomyces nodules to some extent . Though primatologists have begun to account for the specific isotopic values of chimpanzee plant foods within their environments [3–5, 11–13] in cross site comparisons, there remains a need for complimentary data on insect food sources, such as Macrotermes, given their relevance to the diets of several chimpanzee communities (summarized in ).
The isotopic values of Macrotermes within savanna woodland ecosystems may be of specific utility in elucidating the source of comparably high δ13C values, particularly found in some savanna chimpanzee communities. In the unhabituated eastern chimpanzees (P. t. schweinfurthii) at the field site of Issa, Tanzania, van Casteren and colleagues  reported higher δ13C values than cannot be explained by the consumption of the sampled C3 plant foods alone. Relatively high δ13C fractionation factors (Δ13C) between chimpanzees and a selection of C3 plants suggested that C4 plants or open canopy C3 plant foods could be potential contributors to these chimpanzees’ diet and hence δ13C values. However, given that termite consumption is well documented in this population , the authors also posited that C4 plant harvesting termites may contribute to δ13C value enrichment in the chimpanzees. Wessling and colleagues  reported even higher Δ13C values in five western chimpanzee (P. t. verus) communities at the very edge of the species range in Senegal . Four of the five communities in that study are unhabituated and the source of these high Δ13C values remained largely unclear. In the habituated community of Fongoli in Senegal however, chimpanzees have been observed to occasionally consume C4 plants . Additionally, Fongoli chimpanzees are exceptional with regards to the frequency and intensity that they consume Macrotermes . This consistent consumption of an organism that may rely on C4 vegetation is another possible explanation for the relatively high δ13C values observed in Senegalese chimpanzees [3, 4]. Though parsimony would suggest unhabituated chimpanzees in Senegal consume Macrotermes, the extent to which they do, if at all, is not yet known. Only in the Kayan chimpanzees were termites from the genus Macrotermes identified in feces . Still, the degree to which termite consumption could contribute towards high δ13C values in savanna chimpanzees is not as of yet clear due to the limited dataset available on the stable isotope ratios of this termite genus within sub Saharan Africa.
Analysis of the isotopic signature of Macrotermes from a range of sub Saharan African sites may also have implications for paleodietary analysis of hominins. The genus Macrotermes diversified 6–23 million years ago as savannas spread across the African continent and remained relatively unchanged today despite climactic shifts [44, 45]. Thus, it is likely that Macrotermes and early hominins coexisted in the African savanna landscapes. We can further hypothesize that these termites would have had similar diets as we see in extant Macrotermes diets from African savanna woodlands today. Hominins in east, south, and central Africa began to consume foods enriched in 13C approximately 3.5 million years ago [46–57]. In relation to the high δ13C signatures observed in Paranthropus robustus and Australopithecus africanus specimen from Sterkfontein, modern δ13C values of termites and sedges from nearby Kruger National Park were analyzed to investigate the hypothesis that either sedges or termites may be account for the observed high δ13C values in hominins . Accordingly, termite taxa across the park had an average δ13C value of -20.1‰, with a mean of -15.3‰ (n = 10) for open environment termites and a mean of -21.7‰ (n = 30) for termites from closed environments. Based on these results, the authors concluded that termites could reveal a C4 plant dependent isotopic signature, yet their consumption could not solely account for the enriched δ13C values detected in the compared hominins. However, this study did not report the taxonomic classification of the termite specimen sampled (excepting a brief reference to the harvester termite genera Trinervitermes and Hodotermes) that would provide insights into whether these species could have been subject to hominin predation, nor was further ecological information on sampling locations provided.
There are a handful of other studies examining the relative contributions of C3 and C4 resources to termite diets in sub-Saharan Africa. The first systematic study of termite foraging ecology utilizing stable isotope analysis investigated the relative dependence of M. michaelseni on herbaceous (C4) vegetation versus woody (C3) vegetation by sampling from the termite head tissue and using a mixed modeling approach . Both woody and herbaceous food sources contributed to the diets of M. michaelseni at two Kenyan savanna grassland sites but varied in their relative contributions. The dietary contribution from herbaceous vegetation utilizing the C4 photosynthetic pathway was estimated to be 70% at one site and 36% at another site. These results indicate that termites of the same species can vary significantly with regard to δ13C values in two, ecologically similar yet geographically distinct, environments. Additionally, this study provides some preliminary support for the hypothesis of Macrotermes as a partial C4 resource for chimpanzees as well as hominins.
A similar termite isotope study at a humid savanna site (defined as grass savanna, shrub savanna and semideciduous plateau forest) in Côte d’Ivoire also found that the relative contributions of C3 versus C4 plants of four sympatric termite species within the termite subfamily Macrotermitinae was considerably varied even among the same species depending on habitat type and seasonality of sampling effort . More recently, Vesala and colleagues  investigated the isotopic values between termite castes within four Macrotermes colonies located in southern Kenya. While three of the four colonies exhibited δ13C values in range with herbaceous vegetation, one mound with more abundant grass surrounding had relatively enriched δ13C values. Additionally, the authors reported significant differences in the δ13C and δ15N values between castes of the same colony suggesting that the nutritional contribution of fungal symbiont (Termitomyces) varies between castes within the same colony . These studies demonstrate the utility of adding to the African termite isotope database while distinguishing between taxa, habitat type, seasonality, as well as caste (see also [60–62]).
Termite-fishing broadly describes a behavior in which a chimpanzee inserts a vegetative tool into a passageway at the surface of a termite mound to consume the soldier termites that bite the tool. In some habitats, chimpanzees preferentially forage on Macrotermes during the onset of the rainy season, which may be due to an increase in accessibility during the colony’s reproductive cycle . By contrast, other communities reliably consume Macrotermes year-round, which may be attributable to more sophisticated tool-sets [18, 64–66] or a dependence on the termites as a source of protein .
In the present study we analyze Macrotermes spp. samples collected at six chimpanzee savanna woodland field sites across equatorial Africa in the interest of further elucidating possible contributions of this genus of termites in chimpanzee isotope signatures. We focus here on Macrotermes spp. termites due to their status as the most commonly consumed genera among chimpanzees that termite-fish [24–27]. However, it is worth noting here that some populations do not termite-fish despite the presence of mound building termites, such as at the site of Gashaka, in Nigeria, that we report on here [67, 68]. With this study we seek to address the following two questions:
- Do we find evidence for substantial C4 plant consumption by Macrotermes across chimpanzee field sites via stable isotope analysis?
- Do we find intra-specific (between colonies) and intra-colony (between castes) isotopic variation in Macrotermes from the same field site?
Material and methods
Sample collection and isotope analysis
In this study we collected Macrotermes from six savanna woodland sites across Africa, that represent relatively dry and open environments inhabited by chimpanzees today. Savanna woodland habitats are more likely to have substantial amounts of C4 vegetation termites may rely on, as compared to forest habitats. We opportunistically collected 39 termite samples from fungus-growing mound builders at five savanna woodland chimpanzee field sites in West Africa (see Table 1 for further details and season of sampling), following a standardized sample and data collection protocol within the framework of the “Pan African Program—The Cultured Chimpanzee” project [69–71]. Permissions to conduct research were issued under the research permits No NPL/GEN/378/V/504 (Ministère de l‘Ecologie et de la Protection de la Nature, Direction des Eaux, Forêts, Chasses et de la Conservation des Sols, Nigeria), No 078/2015/OGIPAR/MEEF/Ck (Ministère de l’Environment, Eaux et Forêts, Office Guinéen des Parks et Reserves, Guinea), No 01316/DEF/DFG (Direction des Eaux, Forêts, Chasses et de la Conservation des Sols, Senegal) and No 219/MESRS/DGRSIT/TM (Ministère de l’Enseignement Superior et de la Recherche Scientifique, Direction Generale de la Recherche Scientifique de de l’Inovation Technologique, Côte d’Ivoire). At all field sites, except for Gashaka in Nigeria, evidence indicates that chimpanzees termite fish [24, 67]. We collected termites, predominantly of the major soldier caste, directly from mounds and recorded data on the habitat type surrounding the mound location following the protocol of the Pan African Programme [72, 73]. Assigned habitat categories broadly describe the immediate surrounding vegetation at termite mounds in terms of the dominant vegetation type and sometimes the density of canopies as well as understories (e.g. “forest-mixed, closed understory” or “savanna-wooded”). Given our interest in termites as a potential food source for chimpanzees, we decided to combine multiple individual termites of the same caste and the same colony in one measurement.
For stable isotope analysis, we initially submerged the termites in ethanol and then stored them dry on silica in 15 or 50ml tubes. From each termite colony we obtained a second sample stored in ethanol for subsequent taxonomic identification that revealed all samples indeed contained Macrotermes termites of undetermined species.
While sample storage of termites between the field and the lab is essential for taxonomic identification, there is the possibility of the introduction of slight isotope ratio bias due to storage method. In tissue samples of fish δ13C values have been reported to become enriched by ~0.5 to 1.5‰, whereas δ15N values increased by 0.5 to ~1‰ when fish samples were stored in 80% ethanol . However, Arrington and Winemiller (2011) found a similar trend in fish samples yet concluded that these shifts are so small that they should not have considerable consequences for the use of preserved specimens in ecological research . For insects, such as ants, crickets and flies, ethanol preservation was not observed to affect δ15N values, but δ13C values shifted by ~0.6 to 1.5‰ [76–78]. However, non-chemically preserved samples (frozen, freeze-dried, fresh) also appeared to differ in their isotope values, suggesting that inter-sample variation may be just as large as rival biases introduced by preservation method . Other isotopic work on ground beetles and aquatic consumers did not indicate that storage in ethanol significantly affects values δ13C [reviewed in 79, 80]. Despite the contradictory evidence in the literature, potential small-scale shifts in δ13C values in our samples ultimately neither obscures nor aggravates the identification of the larger-scale isotopic differences between C3 and C4 food resources in termites we wish to identify here that typically exceeds 10‰ in tropical habitats. We thus follow the recommendation by Arrington and Winemiller (2002), suggesting that the tradeoff between specimen taxonomic preservation and isotopic integrity is sufficient to address the major ecological questions raised in this study.
We rinsed all isotope samples thoroughly with ethanol, dried them down and then homogenized them to a fine powder in a pebble mill (Retsch MM400). Subsequently, we weighed 500μg of homogenized termite sample into tin capsules for stable isotope analysis performed in parallel to IAEA standards and several internal standard materials in a FLASH HT Plus coupled to a MAT 253 Isotope Ratio Mass Spectrometer (both by Thermo Scientific, Waltham, MA, USA) at the commercial laboratory IsoDetect GmbH in Leipzig, Germany. Stable isotope ratios of carbon and nitrogen are here expressed as the ratio of 13C/12C and 15N/14N using the delta (δ) notation in parts per thousand or permil (‰) relative to the international standard materials Vienna PeeDee Belemite (vPDB) and atmospheric N2 (AIR), respectively. Analytical error calculated from repetitive measurements of international and lab-internal standard materials in each run is lower than 0.2‰ (2σ) for δ15N and δ13C.
We collected 47 additional M. subhyalinus samples from 12 different mounds at the Issa Valley chimpanzee field site in western Tanzania between November 2017 and May 2018 under the research permit No 2017-336-NA-2017-341 (Tanzanian Commission for Science and Technology). These samples were identified to species level . At Issa, we primarily collected samples at mounds known to be used by chimpanzees as well as at two active mounds not observed to be used by chimpanzees but located within the chimpanzee home range . We recorded habitat types surrounding each mound sampled following the same protocol as mentioned above . The Issa samples included termites from three separate castes (major soldiers, minor soldiers, and workers—major and minor workers not differentiated here) that we analyzed separately in order to detect potential isotopic differences within the same termite colony. We transported the samples in 85% ethanol, then dried them down and homogenized them into a fine powder using a pebble mill (Retsch MM400). We weighed between 500μg and 800μg of this powder into tin capsules for stable isotope analysis at the University of California, Santa Cruz Stable Isotope Laboratory. Isotopic and elemental composition was determined by Dumas combustion using a Carlo Erba 1108 elemental analyzer coupled to a ThermoFinnigan Delta Plus XP isotope ratio mass spectrometer and corrected towards the same international standard materials as specified above. Analytical precision of internationally calibrated in-house standards was better than 0.2‰ for both δ13C and δ15N.
Mixing models to estimate C4 plant proportions
We performed all statistical data analyses in R, version 3.4.4 (R Core Team, 2018). To determine the proportion of C3 versus C4 plants in the diets of Macrotermes termite colonies, we employed a Mixing Model based on Gaussian likelihood running Markov Chain Monte Carlo (MCMC) analyses in the R-package “siar” version 4.2  on the normally distributed variables δ13C and δ15N values we measured in termites. We determined the mean δ13C and δ15N values and their standard deviations for non-reproductive parts of C3 plant sources of various species of trees and shrubs (leaves, bark) we measured from the three of the six field sites presented here. We limited the plant data to plants found in the same habitat types from which the termite samples were derived (as shown in Tables 1 and 2). We refer here to 13 samples from Issa , 14 from Kayan [3, 4] and five new isotope datapoints from Comoé GEPRENAF (see S3 Table). Given the environmental similarities between sites, these C3 plants resulted in remarkably similar δ13C values for all three sites (mean δ13C -28.6 ±1.3‰, mean δ15N 2.3 ±2.3‰). Although all sites are considered savanna woodland sites and presumably C4 grasses are abundant in the landscape, no systematic taxonomic survey of these plants was conducted. Consequently, C4 plants were comparatively scant within our plant reference sample due to the emphasis on collecting species of known chimpanzee dietary relevance. As a result, we refer to the average δ13C value for African savanna C4 plants of -12.5±1‰ [58, 59, 83, 84]. Given the lack of published δ15N values for savanna woodland C4 plants, we assigned the same mean δ15N value for the C4 plants as we had calculated for C3 plants.
To correct the mixing model for isotopic fractionation (Δ) between diet and body tissue measured, we included a trophic enrichment factor of 2.3‰ for δ13C and 0.3‰ in δ15N following the only published Δ-data available, at the time of analysis, for fungi-cultivating African termites . We did not alter the trophic enrichment factor based on caste nor did we record the age-class of termites, although these parameters affect differential consumption of plant matter, fungus combs, and Termitomyces nodules [37, 39] that ultimately lead to differences in δ13C fractionation [40, 62]. Termites are observed to associate with N-recycling bacteria and contain a rich diversity of microbes in their gut aiding digestion resulting in fractionation factors that can range from –1.6‰ to + 8.8‰ in δ15N and from –2.2‰ to + 3.0‰ in δ13C in different termite species with different dietary specializations within the same forest .
Assessing the effects of habitat and termite caste
We ran two linear mixed models (LMM) with Gaussian error structure using the lmer-function . Our models tested the effect of the predictor’s habitat type and termite caste on the responses δ13C and δ15N values measured in individual termites. By including the random effect of individual termite mounds, we accounted for multiple measurements per termite mound . We obtained p-values by conducting likelihood ratio tests comparing each full model with a null model excluding the fixed effects. We tested the variance inflation factors (vif ) for each model and consistently obtained values around one. Finally, we inspected the normality and homogeneity of the residuals shown in a histogram, a qq-plot, and residuals plotted against fitted values and found no violation of model assumptions.
We measured the δ13C and δ15N values in 79 specimens of Macrotermes termites from six savanna woodland sites (Table 1, Fig 1) and present the raw data as well as site averages in Table 2. Our mixing model estimates the relative amount of C4 plant resources in termite diets are consistently below 15%. Average C4 plant proportions in Macrotermes diet range from 5 to 15%, with the lowest C4 proportions in Kayan, Senegal, the highest proportions at Bakoun, Guinea (Fig 2). The δ13C values of all termites measured in this study are indistinguishable from C3 plants and can thus not be considered a C4 food resource. Termite δ13C values across sites averaged at -24.3±1.3‰ (1σ), with the highest average δ13C values found at the site of Bakoun in Guniea with a mean of -23.8±1.2‰ (1σ) (Fig 1). Termite δ15N values varied between sites, ranging from -2.3‰ to 8.0‰, demonstrating considerable differences in plant baselines between sites . Comoé GEPRENAF revealed the highest mean δ15N termite values (4.0±2.2‰ 1σ), whereas Gashaka in Nigeria showed comparatively low values (-1.5 ±0.8‰ 1σ).
The raw data from the Issa termites are presented separately in Table 3. The results of our LMMs suggest a strong influence of both fixed effects habitat and caste on the δ13C values of M. subhyalinus specimens from 12 different mounds at Issa (χ2 = 10.4, df = 4, p < 0.001), but no effect on the δ15N values (χ2 = 2.4, df = 4, p = 0.649). The effect of caste on the δ13C values was highly significant (χ2 = 31.0, df = 2, p < 0.001). Estimates indicate that major soldiers are on average 0.6‰ lower in δ13C than workers and on average 0.7‰ lower than minor soldiers (Fig 3). In the δ13C model, the effect of habitat was significant (χ2 = 9.1, df = 2, p = 0.010) with estimates suggesting 1.7‰ lower δ13C values in savanna woodland (miombo) areas and 0.4‰ lower δ13C values in gallery forest compared to newly colonizing forest areas (Fig 4). We present the estimates of the δ13C and δ15N models in S1 and S2 Tables.
We report here the δ13C and δ15N values of termites belonging to the genus Macrotermes from six savanna woodland chimpanzee field sites in equatorial Africa. All sites except for Gashaka in Nigeria bear evidence that chimpanzees utilize these termites as a feeding resource . We interpret the considerable range δ15N values reported here to be primarily a result of differences in the plant baseline values between the six sites . Although these δ15N values add to the published database on Macrotermes isotope values, our research interests are primarily concerned with δ13C values of which will be the focus of our discussion. Across sites we quantified the potential C4 plant consumption in these termites and found that C4 plants are a marginal and insignificant part of Macrotermes diets (5–15%, Fig 2). The range of Macrotermes δ13C values (mean -24.3±1.3‰) are indistinguishable from C3 plants and thus these termites cannot be considered a C4 food resource (C4 mean = -12.5 ± 1.0‰). The Macrotermes spp. isotope values that we report here are important in the context of subtle 13C-enrichments observed in isotopic values from savanna dwelling chimpanzee populations. Although a preservation effect of up to 1.5‰ may have an influence the δ13C values of the termite samples analyzed here, this effect would not substantively alter the conclusions of this study. If anything, a preservation effect would be likely to enrich δ13C values  and could thus partially explain the slight C4 signal seen in our mixing models across all six sites. In the case of termite sampled from Issa, in which castes and habitat type were analyzed, all samples were preserved in the same manner. Thus, significant differences in observed δ13C values between samples from different castes and habitat types that are less than 1.5‰ are still substantive findings and relevant for discussion.
Relatively high Δ13C values were detected in chimpanzees from Senegal that suggests potential input of C4 resources in chimpanzee diets . Here we provide mixing models results for Macrotermes sp. samples collected at Kayan that suggest minimal (5%) input from C4 vegetation (Fig 2), therefore positioning Macrotermes as an unlikely contributor to the comparatively high δ13C values of -23.0‰  measured in the Kayan chimpanzees. Wessling and colleagues  report even higher δ13C values of -21.7‰ within a population of chimpanzees further to the north of Senegal, Hérémakhono, but Macrotermes samples from that site were not analyzed in that study.
A north to south decline in tree density was observed across Senegalese sites in Wessling et al.  with the lowest tree coverage observed at the site of Hérémakhono. Tree density is consistently lower in both Kayan and Hérémakhono relative to the site of Dindefelo. The latter is a site much more similar in structure to Bakoun and Sobeya . It is possible that Hérémakhono contains fewer trees and a greater proportion of grasses than Kayan and that Macrotermes at Hérémakhono may therefore be further δ13C enriched relative to Kayan samples. However, that we do not by extension see the converse pattern of lower δ13C values from Sobeya and Bakoun, which are two sites that are presumably more heavily forested than Kayan, relative to Kayan Macrotermes samples suggests that variation in grass coverage as a determinant of C4 consumption by Macrotermes is unlikely to be a considerable contribution to chimpanzee isotopic variation.
Further, while it is parsimonious to assume that other chimpanzee communities in Senegal rely on Macrotermes consumption to similar degrees as the nearby Fongoli chimpanzees , our termite isotope data do not suggest that this feeding behavior will considerably affect patterns of δ13C values variation without the unlikely scenario that Kayan Macrotermes samples differ considerably from their Senegalese counterparts. Instead, our results support the hypothesis that these Senegalese chimpanzees may engage in wild C4 plant consumption or even crop-raiding on domestic C4 plants . We cannot exclude the possibility that termites at other chimpanzee sites studied by Wessling and colleagues  rely more heavily on C4 resources as they were not sampled in this study.
Our study also aimed to test the effect of caste on Macrotermes δ13C and δ15N values. Macrotermes worker castes feed other colony members. Workers forage for food that they may eat themselves or store in reserves. Soldiers by contrast, depend on the workers to directly feed them pieces fungal comb material . Additionally, the worker castes support the growth of the Termitomyces fungal comb with their feces that helps to promote the production of fungal nodules that are additionally consumed by some termites within the colony [37, 40]. Carbon fractionation occurs in the plant matter, fungal combs, and Termitomyces nodules within a Macrotermes colony [40, 62]. The consumption of these food sources varies between colony members based on caste and age-class [37, 39], but the exact contributions of Termitomyces nodules in food processing within a mound remains uncertain . We incorporated a trophic enrichment factor into our mixing model according to comparable Δ-data available at the time of analysis (but see ). However, our trophic enrichment factor was not altered based on caste and we did not record the age-class of termites. These uncertainties, in addition to probable differences in carbon fractionation between Macrotermes species, impose possible limitations that may be considered in the context of the results from our mixing model. The data reported here from the Issa Valley, in which we collected individuals from each caste from 12 mounds, suggests that minor soldiers and workers are significantly higher in δ13C than major soldiers (Fig 3) and that there is no detectable effect of caste on the δ15N values. Young workers may subsist partly on 13C-enriched Termitomyces nodules  that may account for the difference in δ13C values observed between workers and major soldiers here and in previous research . Differences in δ13C values between major and minor soldiers has also been reported for Macrotermes in Kenya  but the underlying mechanism remains unclear as both similarly depend on being fed fungal comb material. Depletion of δ13C values in major soldiers compared to other castes adds important context to the data reported for the Issa Valley.
We found that in the Issa Valley, habitat type also had a significant effect on δ13C, but not on δ15N values. According to the habitat description protocol [72, 73], “forest-colonizing” describes habitats in which a mature forest expands into a non-forest area (i.e. savanna woodland in this case), whereas “gallery forest” describes forests in direct proximity to a river. And “savanna-wooded” describes areas that are dominated by grasses or ferns but also contain significant interspersed wooded vegetation. Termites foraging within savanna wooded sites may reasonably be expected to have the highest δ13C values due to the relative abundance of C4 grasses. However, our mixing model demonstrates that Issa Valley termites within forest-colonizing habitats were more enriched in δ13C than either gallery forests or savanna-wooded habitats (Fig 4). These results may be attributable to the “canopy effect” in which dense forest canopies produce depleted δ13C values in understory vegetation less exposed to sunlight and atmospheric carbon [3, 91, 92]. The young and small trees from a colonizing forest segment may not cause a canopy effect as much as the mature trees in the gallery forest or even the sparse, yet larger, trees within a savanna-wooded habitat. Another possibility is that the forest-colonizing habitat provides less suitable food and the termites compensated by foraging on comparatively more C4 sources. However, we did not conduct vegetation plots at the termite-mounds for this study and thus any consequent interpretations are limited.
We collected termites at various wet and dry seasons at the six sites in our study. While Issa and Gashaka samples were collected in both wet and dry seasons, the samples from the other four sites were collected during either one or the other (Table 1). Some chimpanzee communities preferentially feed on Macrotermes during the rainy season while other communities termite-fish throughout the year . It is worth noting that the samples from Kayan were collected during the rainy season and that the nearby chimpanzee community at Fongoli are known to feed on Macrotermes throughout the year . Though it is conceivable that termites could have higher δ13C values in the dry season if they are more dependent on C4 grasses at that time, if anything one would expect a bias towards higher δ13C values during the wet season when grasses are generally more abundant. Our findings suggest that Macrotermes at savanna woodland chimpanzee sites do not depend heavily on C4 resources in either rainy or dry seasons.
Several scholars have proposed that termites could have been exploited by hominins with the use of tools [93, 94] comparable or even more derived than what we see in chimpanzees across Africa today (summarized in ). Bone tool replicas used to dig into Trinervitermes mounds developed striation marks significantly similar to bone tool fossils discovered in Swartkrans . Lesnik  replicated Backwell and d’Errico’s 2001 method of experimental bone tool use on both Trinervitermes and Macrotermes mounds for comparison. Although not able to fully distinguish between Trinervitermes and Macrotermes wear patterns, Lesnik’s analysis introduced Macrotermes as an appealing alternative hypothesis to Trinervitermes as a genus targeted by hominins. Although we cannot refute this hypothesis, our data suggests that Macrotermes at the six savanna woodland sites in our study are isotopically distinct from C4 resources and thus unlikely to have contributed to the enriched δ13C values found in some chimpanzees and early hominins.
Further, one should note that previous studies [58, 60] that found substantially higher δ13C values in Macrotermes, were conducted in environments outside of the range of extant chimpanzees, which suggest that these locations lack the climate and vegetation structure chimpanzees need to survive. Direct comparisons between the Macrotermes isotope values presented here and in previous studies are further complicated due to varying sampling methodologies. Schyra and team  report δ13C values approximately 4–5‰ higher relative to the mean values reported here. However, only the workers were sampled in the former study, which hinders comparisons given the inconsistencies in δ13C between castes of the same colony reported here and elsewhere [40, 62]. Still, it is unlikely that worker termites would be that dissimilar from soldiers and thus, these results are worth careful consideration to the interpretations made here. More notably divergent, however, are the values reported in Boutton, Arshad, and Tieszen’s flagship study on Macrotermes isotope values within two Kenyan grassland habitats . Among the nonreproductive castes, the δ13C values reported were roughly -15‰ at Kaijado and -19‰ at Ruiru. The researchers exclusively sampled termite head tissues in that study so as to minimize isotopic variation due to sampling various body parts . Again, incomparable methodologies obfuscate direct comparison to the present study in which we sampled the complete termite as various body parts differentially affect isotopic signatures [95, 96]. Additionally, as the objective of this study is to assess the isotopic value of Macrotermes as a food source, the samples here also include termite gut content that may have an effect on the resulting isotope values. Nevertheless, these dissimilar results from the Kenyan grassland sites add important context to the present study and further highlight the influence of habitat on Macrotermes diets. Still, the data presented here on whole termite bodies from extant chimpanzee habitats are likely to be more relevant to isotopic ecology of chimpanzees as well as hominins that are hypothesized to live in similar savanna woodlands environments [97, 98]. Our overall results indicate that Macrotermes inhabiting savanna woodland habitats in Africa can reveal C3 plant-based diets and do not seem to uphold as a reliable source of high δ13C values in chimpanzees. Our data further illustrate the value of cross-site comparisons and the importance of corresponding habitat data when considering the isotopic signatures of potential food resources in primate isotope ecology and paleodietary analyses.
S1 Table. Model estimates, standard error (SE), t-values and p-values for each fixed effect in the model testing for habitat and caste differences in Issa termite δ13C values.
S2 Table. Model estimates, standard error (SE), t-values and p-values for each fixed effect in the model testing for habitat and caste differences in Issa termite δ15N values.
We thank our local field assistants for their support in sample collection. We thank the UCSC stable isotope lab staff Colin Carney as well as the staff at IsoDetect for their help with isotope analysis. We also thank Henk Eshuis and Mizuki Murai for their contribution to the PanAf sample and data collection. UCSD/Salk Center for Academic Research and Training in Anthropogeny (CARTA) provided ongoing support to GMERC and the Issa Valley research station. The following institutions supported sample collection in the field: Ministere des Eaux et Forets, Cote d’Ivoire, Ministere de l'Agriculture de l'Elevage et des Eaux et Forets Guinea, National Park Service Nigeria, Direction des Eaux, Forêts, Chasses et de la Conservation des Sols, Senegal, and the Tanzania Wildlife Research Institute. We were kindly supported by our partners from the Wild Chimpanzee Foundation, Emmanuelle Normand, Vincent Lapeyre, and Virginie Vergnes.
- 1. Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V, Sugiyama Y, et al. Cultures in chimpanzees. Nature. 1999;399(6737):682–685. pmid:10385119
- 2. Wrangham RW, McGrew WC, De Waal F. Chimpanzee cultures. Harvard University Press; 1996.
- 3. Oelze VM, Fahy G, Hohmann G, Robbins MM, Leinert V, Lee K, et al. Comparative isotope ecology of African great apes. J Hum Evol. 2016;101:1–16. pmid:27886808
- 4. Wessling EG, Oelze VM, Eshuis H, Pruetz JD, Kühl HS. Stable isotope variation in savanna chimpanzees (Pan troglodytes verus) indicate avoidance of energetic challenges through dietary compensation at the limits of the range. Am J Phys Anthropol. 2019;168(4):665–675. pmid:30693959
- 5. Loudon JE, Sandberg PA, Wrangham RW, Fahey B, Sponheimer M. The stable isotope ecology of Pan in Uganda and beyond. Am J Primatol. 2016;78(10):1070–1085. pmid:27188271
- 6. Schoeninger MJ, Moore J, Sept JM. Subsistence strategies of two “savanna” chimpanzee populations: The stable isotope evidence. Am J Primatol. 1999;49(4):297–314. pmid:10553959
- 7. Schoeninger MJ, Most CA, Moore JJ, Somerville AD. Environmental variables across Pan troglodytes study sites correspond with the carbon, but not the nitrogen, stable isotope ratios of chimpanzee hair. Am J Primatol. 2016;78(10):1055–1069. pmid:26513527
- 8. van Casteren A, Oelze VM, Angedakin S, Kalan AK, Kambi M, Boesch C, et al. Food mechanical properties and isotopic signatures in forest versus savannah dwelling eastern chimpanzees. Commun Biol. 2018;1(1):1–10. pmid:30271989
- 9. Crowley BE. Stable isotope techniques and applications for primatologists. Int J Primatol. 2012;33(3):673–701.
- 10. Sandberg PA, Loudon JE, Sponheimer M. Stable isotope analysis in primatology: a critical review. Am J Primatol. 2012;74(11):969–989. pmid:23015270
- 11. Carlson BA, Crowley BE. Variation in carbon isotope values among chimpanzee foods at Ngogo, Kibale National Park and Bwindi Impenetrable N tional Park, Uganda. Am J Primatol. 2016;78(10):1031–1040. pmid:26918258
- 12. Carlson BA, Kingston JD. Chimpanzee isotopic ecology: A closed canopy C3 template for hominin dietary reconstruction. J Hum Evol. 2014;76:107–115. pmid:24993419
- 13. Blumenthal SA, Rothman JM, Chritz KL, Cerling TE. Stable isotopic variation in tropical forest plants for applications in primatology. Am J Primatol. 2016;78(10):1041–1054. pmid:26444915
- 14. Smith CC, Morgan ME, Pilbeam D. Isotopic ecology and dietary profiles of Liberian chimpanzees. J Hum Evol. 2010;58(1):43–55. pmid:19796791
- 15. Sponheimer M, Loudon JE, Codron D, Howells M, Pruetz JD, Codron J, et al. Do “savanna” chimpanzees consume C4 resources? J Hum Evol. 2006;51(2):128–133. pmid:16630647
- 16. Sponheimer M, Lee-Thorp JA. Differential resource utilization by extant great apes and australopithecines: towards solving the C4 conundrum. Comp Biochem Physiol A Mol Integr Physiol. 2003;136(1):27–34. pmid:14527627
- 17. Oelze VM, Head JS, Robbins MM, Richards M, Boesch C. Niche differentiation and dietary seasonality among sympatric gorillas and chimpanzees in Loango National Park (Gabon) revealed by stable isotope analysis. J Hum Evol. 2014;66:95–106. pmid:24373257
- 18. Deblauwe I, Janssens GP. New insights in insect prey choice by chimpanzees and gorillas in southeast Cameroon: the role of nutritional value. Am J Phys Anthropol. 2008;135(1):42–55. pmid:17902166
- 19. O’Malley RC, Power ML. The energetic and nutritional yields from insectivory for Kasekela chimpanzees. J Hum Evol. 2014;71:46–58. pmid:24698197
- 20. Reynolds V, Pascual-Garrido A, Lloyd AW, Lyons P, Hobaiter C. Possible mineral contributions to the diet and health of wild chimpanzees in three East African forests. Am J Primatol. 2019;e978. pmid:31090097
- 21. Mills A, Milewski A, Fey M, Groengroeft A, Petersen A. Fungus culturing, nutrient mining and geophagy: a geochemical investigation of Macrotermes and Trinervitermes mounds in southern Africa. J Zool. 2009;278(1):24–35.
- 22. Seymour C, Milewski A, Mills A, Joseph G, Cumming G, Cumming D, et al. Do the large termite mounds of Macrotermes concentrate micronutrients in addition to macronutrients in nutrient-poor African savannas? Soil Biol Biochem. 2014;68:95–105.
- 23. Kalumanga E, Mpanduji DG, Cousins SA. Geophagic termite mounds as one of the resources for African elephants in Ugalla Game Reserve, Western Tanzania. Afr J Ecol. 2017;55(1):91–100.
- 24. Boesch C, Kalan AK, Mundry R, Arandjelovic M, Pika S, Dieguez P, et al. Chimpanzee ethnography reveals unexpected cultural diversity. Nat Hum Behav. 2020;1–7. pmid:31965067
- 25. Lesnik JJ. Termites in the hominin diet: A meta-analysis of termite genera, species and castes as a dietary supplement for South African robust Australopithecines. J Hum Evol. 2014;71:94–104. pmid:24613098
- 26. McGrew WC, McGrew WC. Chimpanzee material culture: implications for human evolution. Cambridge University Press; 1992.
- 27. Sanz CM, Deblauwe I, Tagg N, Morgan DB. Insect prey characteristics affecting regional variation in chimpanzee tool use. J Hum Evol. 2014;71:28–37. pmid:24602365
- 28. van Lawick-Goodall J. The behaviour of free-living chimpanzees in the Gombe Stream Reserve. Anim Behav Monogr. 1968;1:161–IN12.
- 29. Webster TH, McGrew WC, Marchant LF, Payne CL, Hunt KD. Selective insectivory at Toro-Semliki, Uganda: Comparative analyses suggest no ‘savanna’chimpanzee pattern. J Hum Evol. 2014;71:20–27. pmid:24792877
- 30. Traniello JFA, Leuthold RH. Behavior and Ecology of Foraging in Termites. In: Abe T, Bignell DE, Higashi M, editors. Termites: Evolution, Sociality, Symbioses, Ecology [Internet]. Dordrecht: Springer Netherlands; 2000. p. 141–168. Available from: https://doi.org/10.1007/978-94-017-3223-9_7
- 31. Korb J. Termite mound architecture, from function to construction. In: Biology of termites: a modern synthesis. Springer; 2010. p. 349–373.
- 32. Korb J, Linsenmair K. The effects of temperature on the architecture and distribution of Macrotermes bellicosus (Isoptera, Macrotermitinae) mounds in different habitats of a West African Guinea savanna. Insectes Sociaux. 1998;45(1):51–65.
- 33. Pomeroy D. The abundance of large termite mounds in Uganda in relation to their environment. J Appl Ecol. 1978;51–63.
- 34. Ruelle JE. Revision of the termites of the genus Macrotermes from the Ethiopian region (Isoptera: Termitidae). Brit Mus Nat Hist Bull Entomol. 1970
- 35. Darlington JP. Nutrition and evolution in fungus-growing termites. Nourishment Evol Insect Soc. 1994;105–130.
- 36. Sands W. Termites as pests of tropical food crops. PANS Pest Artic News Summ. 1973;19(2):167–177.
- 37. Badertscher S, Gerber C, Leuthold R. Polyethism in food supply and processing in termite colonies of Macrotermes subhyalinus (Isoptera). Behav Ecol Sociobiol. 1983;12(2):115–119.
- 38. Leuthold R, Badertscher S, Imboden H. The inoculation of newly formed fungus comb with Termitomyces in Macrotermes colonies (Isoptera, Macrotermitinae). Insectes Sociaux. 1989;36(4):328–338.
- 39. Sieber R, Leuthold R. Behavioural elements and their meaning in incipient laboratory colonies of the fungus-growing Termite Macrotermes michaelseni (Isoptera: Macrotermitinae). Insectes Sociaux. 1981;28(4):371–382.
- 40. Vesala R, Arppe L, Rikkinen J. Caste-specific nutritional differences define carbon and nitrogen fluxes within symbiotic food webs in African termite mounds. Sci Rep. 2019;9(1):1–11. pmid:30626917
- 41. Stewart FA, Piel AK. Termite fishing by wild chimpanzees: new data from Ugalla, western Tanzania. Primates. 2014;55(1):35–40. pmid:23720026
- 42. Pruetz JD. Feeding ecology of savanna chimpanzees (Pan troglodytes verus) at Fongoli, Senegal. Feed Ecol Apes Primates. 2006;326–364.
- 43. Bogart SL, Pruetz JD. Insectivory of savanna chimpanzees (Pan troglodytes verus) at Fongoli, Senegal. Am J Phys Anthropol. 2011;145(1):11–20. pmid:21484757
- 44. Brandl R, Hyodo F, von Korff-Schmising M, Maekawa K, Miura T, Takematsu Y, et al. Divergence times in the termite genus Macrotermes (Isoptera: Termitidae). Mol Phylogenet Evol. 2007;45(1):239–250. pmid:17714956
- 45. Lesnik JJ. Edible Insects and Human Evolution. University Press of Florida; 2018.
- 46. Cerling TE, Mbua E, Kirera FM, Manthi FK, Grine FE, Leakey MG, et al. Diet of Paranthropus boisei in the early Pleistocene of East Africa. Proc Natl Acad Sci. 2011;108(23):9337–9341. pmid:21536914
- 47. Cerling TE, Manthi FK, Mbua EN, Leakey LN, Leakey MG, Leakey RE, et al. Stable isotope-based diet reconstructions of Turkana Basin hominins. Proc Natl Acad Sci. 2013;110(26):10501–10506. pmid:23733966
- 48. Lee-Thorp J, Likius A, Mackaye HT, Vignaud P, Sponheimer M, Brunet M. Isotopic evidence for an early shift to C4 resources by Pliocene hominins in Chad. Proc Natl Acad Sci. 2012;109(50):20369–20372. pmid:23150583
- 49. Lee-Thorp JA, Sponheimer M, Passey BH, de Ruiter DJ, Cerling TE. Stable isotopes in fossil hominin tooth enamel suggest a fundamental dietary shift in the Pliocene. Philos Trans R Soc B Biol Sci. 2010;365(1556):3389–3396. pmid:20855312
- 50. Lee-Thorp JA, van der Merwe NJ, Brain C. Diet of Australopithecus robustus at Swartkrans from stable carbon isotopic analysis. J Hum Evol. 1994;27(4):361–372.
- 51. Sponheimer M, Lee-Thorp JA. Isotopic evidence for the diet of an early hominid, Australopithecus africanus. Science. 1999;283(5400):368–370. pmid:9888848
- 52. Sponheimer M, Lee-Thorp J, de Ruiter D, Codron D, Codron J, Baugh AT, et al. Hominins, sedges, and termites: new carbon isotope data from the Sterkfontein valley and Kruger National Park. J Hum Evol. 2005;48(3):301–312. pmid:15737395
- 53. Sponheimer M, Passey BH, De Ruiter DJ, Guatelli-Steinberg D, Cerling TE, Lee-Thorp JA. Isotopic evidence for dietary variability in the early hominin Paranthropus robustus. Science. 2006;314(5801):980–982. pmid:17095699
- 54. Sponheimer M, Alemseged Z, Cerling TE, Grine FE, Kimbel WH, Leakey MG, et al. Isotopic evidence of early hominin diets. Proc Natl Acad Sci. 2013;110(26):10513–10518.
- 55. Van der Merwe NJ, Masao FT, Bamford MK. Isotopic evidence for contrasting diets of early hominins Homo habilis and Australopithecus boisei of Tanzania. South Afr J Sci. 2008;104(3–4):153–155.
- 56. Van Der Merwe NJ, Thackeray JF, Lee-Thorp JA, Luyt J. The carbon isotope ecology and diet of Australopithecus africanus at Sterkfontein, South Africa. J Hum Evol. 2003;44(5):581–597. pmid:12765619
- 57. Wynn JG, Sponheimer M, Kimbel WH, Alemseged Z, Reed K, Bedaso ZK, et al. Diet of Australopithecus afarensis from the Pliocene Hadar formation, Ethiopia. Proc Natl Acad Sci. 2013;110(26):10495–10500. pmid:23733965
- 58. Boutton T, Arshad M, Tieszen L. Stable isotope analysis of termite food habits in East African grasslands. Oecologia. 1983;59(1):1–6. pmid:25024140
- 59. Lepage M, Abbadie L, Mariotti A. Food habits of sympatric termite species (Isoptera, Macrotermitinae) as determined by stable carbon isotope analysis in a Guinean savanna (Lamto, Côte d’Ivoire). J Trop Ecol. 1993;9(3):303–311.
- 60. Schyra J, Scheu S, Korb J. Cryptic niche differentiation in West African savannah termites as indicated by stable isotopes. Ecol Entomol. 2019;44(2):190–196.
- 61. Tayasu I, Abe T, Eggleton P, Bignell D. Nitrogen and carbon isotope ratios in termites: an indicator of trophic habit along the gradient from wood-feeding to soil-feeding. Ecol Entomol. 1997;22(3):343–351.
- 62. Tayasu I, Hyodo F, Abe T. Caste-specific N and C isotope ratios in fungus-growing termites with special reference to uric acid preservation and their nutritional interpretation. Ecol Entomol. 2002;27(3):355–361.
- 63. McGrew WC, Tutin CE, Baldwin P. Chimpanzees, tools, and termites: cross-cultural comparisons of Senegal, Tanzania, and Rio Muni. Man. 1979;185–214.
- 64. Nishihara T, Suzuki S, Kuroda S. Tool-set for termite-fishing by chimpanzees in the Ndoki Forest, Congo. Behaviour. 1995;132(3–4):219–235.
- 65. Sanz CM, Morgan D, Gulick S. New insights into chimpanzees, tools, and termites from the Congo Basin. Am Nat. 2004;164(5):567–581. pmid:15540148
- 66. Sanz CM, Morgan DB. Ecological and social correlates of chimpanzee tool use. Philos Trans R Soc B Biol Sci. 2013;368(1630):20120416. pmid:24101626
- 67. Sommer V, Buba U, Jesus G, Pascual-Garrido A. Sustained myrmecophagy in Nigerian chimpanzees: Preferred or fallback food? Am J Phys Anthropol. 2017;162(2):328–336. pmid:27779749
- 68. Hicks TC, Kühl HS, Boesch C, Menken SB, Hart J, Roessingh P, et al. The relationship between tool use and prey availability in chimpanzees (Pan troglodytes schweinfurthii) of Northern Democratic Republic of Congo (advance online). Int J Primatol. 2020
- 69. Kühl HS, Kalan AK, Arandjelovic M, Aubert F, D’Auvergne L, Goedmakers A, et al. Chimpanzee accumulative stone throwing. Sci Rep. 2016;6:22219. pmid:26923684
- 70. Tagg N, McCarthy M, Dieguez P, Bocksberger G, Willie J, Mundry R, et al. Nocturnal activity in wild chimpanzees (Pan troglodytes): Evidence for flexible sleeping patterns and insights into human evolution. Am J Phys Anthropol. 2018;166(3):510–529. pmid:29989158
- 71. Vaidyanathan G. Apes in Africa: The cultured chimpanzees. Nat News. 2011;476(7360):266–269. pmid:21850081
- 72. Arandjelovic M, Boesch C, Campbell G, Hohmann G, Junker J, Célestin KY, et al. Pan African Programme—The cultured chimpanzee. Guidelines for research and data collection. Leipzig, 2014; http://panafrican.eva.mpg.de/english/approaches_and_methods.php
- 73. White L, Edwards A. Conservation research in the African. 2000
- 74. Sweeting CJ, Polunin NV, Jennings S. Tissue and fixative dependent shifts of δ13C and δ15N in preserved ecological material. Rapid Commun Mass Spectrom. 2004;18(21):2587–2592. pmid:15468144
- 75. Arrington DA, Winemiller KO. Preservation effects on stable isotope analysis of fish muscle. Trans Am Fish Soc. 2002;131(2):337–342.
- 76. Jesus FM, Pereira MR, Rosa CS, Moreira MZ, Sperber CF. Preservation methods alter carbon and nitrogen stable isotope values in crickets (Orthoptera: Grylloidea). PloS One. 2015;10(9):e0137650. pmid:26390400
- 77. Tillberg C, McCarthy D, Dolezal AG, Suarez A. Measuring the trophic ecology of ants using stable isotopes. Insectes Sociaux. 2006;53(1):65–69.
- 78. Ponsard S, Amlou M. Effects of several preservation methods on the isotopic content of Drosophila samples. Comptes Rendus Académie Sci-Ser III-Sci Vie. 1999;322(1):35–41. pmid:10047952
- 79. Zalewski M, Dudek D, Godeau J-F, Maruszkiewicz M, others. Stable isotopic research on ground beetles. Review of methods. Balt J Coleopt. 2012;12:91–98.
- 80. Sarakinos HC, Johnson ML, Zanden MJV. A synthesis of tissue-preservation effects on carbon and nitrogen stable isotope signatures. Can J Zool. 2002;80(2):381–387.
- 81. Piel AK, Strampelli P, Greathead E, Hernandez-Aguilar RA, Moore J, Stewart FA. The diet of open-habitat chimpanzees (Pan troglodytes schweinfurthii) in the Issa valley, western Tanzania. J Hum Evol. 2017;112:57–69. pmid:29037416
- 82. Parnell A, Jackson A. siar: Stable Isotope Analysis in R. R package version 4.2. Available Available HttpCRAN R-Proj Orgpackage Siar23 March 2014. 2013
- 83. Smith BN, Epstein S. Two categories of 13C/12C ratios for higher plants. Plant Physiol. 1971;47(3):380–384. pmid:16657626
- 84. Tieszen LL. Natural variations in the carbon isotope values of plants: implications for archaeology, ecology, and paleoecology. J Archaeol Sci. 1991;18(3):227–248.
- 85. Baayen RH, Davidson DJ, Bates DM. Mixed-effects modeling with crossed random effects for subjects and items. J Mem Lang. 2008;59(4):390–412.
- 86. Mundry R, Oelze VM. Who is who matters—The effects of pseudoreplication in stable isotope analyses. Am J Primatol. 2016;78(10):1017–1030. pmid:26581644
- 87. Fox J, Weisberg S, Adler D, Bates D, Baud-Bovy G, Ellison S, et al. Package ‘car.’ Vienna R Found Stat Comput. 2012
- 88. Boesch C, Kalan AK, Agbor A, Arandjelovic M, Dieguez P, Lapeyre V, et al. Chimpanzees routinely fish for algae with tools during the dry season in Bakoun, Guinea. Am J Primatol. 2017;79(3):e22613. pmid:27813136
- 89. Korb J, Linsenmair KE. Resource availability and distribution patterns, indicators of competition between Macrotermes bellicosus and other macro-detritivores in the Comoé National Park, Côte d’Ivoire. Afr J Ecol. 2001;39(3):257–265.
- 90. Wessling EG, Dieguez P, Llana M, Pacheco L, Pruetz JD, Kühl HS. Chimpanzee (Pan troglodytes verus) density and environmental gradients at their biogeographical range edge. bioRxiv. 2020
- 91. Graham HV, Patzkowsky ME, Wing SL, Parker GG, Fogel ML, Freeman KH. Isotopic characteristics of canopies in simulated leaf assemblages. Geochim Cosmochim Acta. 2014;144:82–95.
- 92. Van der Merwe NJ, Medina E. The canopy effect, carbon isotope ratios and foodwebs in Amazonia. J Archaeol Sci. 1991;18(3):249–259.
- 93. Backwell LR, d’Errico F. Evidence of termite foraging by Swartkrans early hominids. Proc Natl Acad Sci. 2001;98(4):1358–1363. pmid:11171955
- 94. Lesnik JJ. Bone tool texture analysis and the role of termites in the diet of South African hominids. PaleoAnthropology. 2011;268:281.
- 95. Hyodo F. Use of stable carbon and nitrogen isotopes in insect trophic ecology. Entomol Sci. 2015;18(3):295–312.
- 96. Webb SC, Hedges RE, Simpson SJ. Diet quality influences the δ13C and δ15N of locusts and their biochemical components. J Exp Biol. 1998;201(20):2903–2911.
- 97. White TD, Ambrose SH, Suwa G, Su DF, DeGusta D, Bernor RL, et al. Macrovertebrate paleontology and the Pliocene habitat of Ardipithecus ramidus. Science. 2009;326(5949):67–93. pmid:19810193
- 98. Cerling TE, Wynn JG, Andanje SA, Bird MI, Korir DK, Levin NE, et al. Woody cover and hominin environments in the past 6 million years. Nature. 2011;476(7358):51. pmid:21814275