Stable isotope ecology of a hyper-diverse community of scincid lizards from arid Australia

Maggie R. Grundler; Eric R. Pianka; Nicolás Pelegrin; Mark A. Cowan; Daniel L. Rabosky

doi:10.1371/journal.pone.0172879

Abstract

We assessed the utility of stable isotope analysis as a tool for understanding community ecological structure in a species-rich clade of scincid lizards from one of the world's most diverse lizard communities. Using a phylogenetic comparative framework, we tested whether δ¹⁵N and δ¹³C isotopic composition from individual lizards was correlated with species-specific estimates of diet and habitat use. We find that species are highly divergent in isotopic composition with significant correlations to habitat use, but this relationship shows no phylogenetic signal. Isotopic composition corresponds to empirical observations of diet for some species but much variation remains unexplained. We demonstrate the importance of using a multianalytical approach to questions of long-term dietary preference, and suggest that the use of stable isotopes in combination with stomach content analysis and empirical data on habitat use can potentially reveal patterns in ecological traits at finer scales with important implications for community structuring.

Citation: Grundler MR, Pianka ER, Pelegrin N, Cowan MA, Rabosky DL (2017) Stable isotope ecology of a hyper-diverse community of scincid lizards from arid Australia. PLoS ONE 12(2): e0172879. https://doi.org/10.1371/journal.pone.0172879

Editor: Elena Gorokhova, Stockholm University, SWEDEN

Received: September 29, 2016; Accepted: February 11, 2017; Published: February 28, 2017

Copyright: © 2017 Grundler 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 work was supported by National Science Foundation (NSF) grants OSIE-0612855 and DEB-0814277, and by the David and Lucile Packard Foundation. Field sampling at Matuwa/Lorna Glen Conservation Park was conducted under permit SF0004654 to D. L. Rabosky (Western Australia Department of Parks and Wildlife). E. R. Pianka thanks the National Geographic Society and the Denton A. Cooley Centennial Professorship in Zoology at UT Austin for financial support. N. Pelegrin thanks CONICET for the external postdoctoral fellowship to carry out research activities at the University of Texas at Austin. The funders had no role in the 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.

Introduction

The trophic ecology of vertebrate communities is contingent on an array of both historical and contemporary interactions. By characterizing ecological traits among species and their relation to phylogenetic and trophic structure, researchers can reveal the processes that influence community composition in diverse species assemblages [1–3]. However, detailed profiles of diet can be difficult to obtain and may only provide a “snapshot” view in time (Dalerum and Angerbjörn 2005; Inger et al. 2006; Lugendo et al. 2006; Cherel et al. 2007). For studying trophic interactions in vertebrate communities, stable isotope analysis can potentially provide integrated measures of resource use through time, without requiring intensive sampling over broad temporal and spatial scales [4–5]. Stable isotope profiles thus afford complementary insights to analysis of stomach contents and resulting estimates of community trophic structure.

These analyses are relatively robust to variability through time because an organism’s tissues carry a unique isotopic signature based on ecophysiological properties and the sources from which nutrients are assimilated, providing a proxy that can be used for the analysis of a species’ trophic niche [4, 6–8]. This is a result of the relatively minimal fractionation of the stable isotope of carbon and the predictable enrichment of the stable isotope of nitrogen during trophic transfer, resulting in the preservation of isotopic signature of the primary sources of dietary carbon and reliable measures of trophic position [5].

A variety of existing studies have compared isotopic signatures to multivariate ecological data and phylogenies to answer questions about community assembly ranging from niche space occupancy to trait differentiation in ecomorphological radiations [5, 9–13]. While many advances in isotope ecology have involved marine food webs, fewer studies have combined isotope ratios with both ecological and phylogenetic data in terrestrial community assemblages; those that do focus heavily on birds and mammals, with little data for reptile or amphibian communities [14–17]. However, isotopic signatures of squamate reptiles are expected to capture temporally-averaged patterns of trophic resource use, due to the slow turnover rates of δ¹⁵N and δ¹³C isotopes in tissues of poikilothermic vertebrates [15–16].

In this study, we combine stable isotope analysis with long-term ecological monitoring data and a community phylogeny to examine whether isotope ratios provide information about the trophic ecology of a diverse clade of scincid lizards from the spinifex deserts of the western Australian arid zone. This clade—the sphenomorphine skinks [18]—includes the genera Lerista and Ctenotus and is well-known as Australia's most diverse terrestrial vertebrate radiation [19]. Closely-related species within these and other genera are divergent in both habitat use and dietary resources [1,20]. We first asked whether species are distinguishable in isotopic space, and determined if there is phylogenetic signal for isotopic composition. We then investigated whether differences in isotopic composition were predicted by dietary differences as revealed by detailed stomach content analyses. We also tested whether species whose diets are dominated by prey from higher trophic levels (e.g., secondary consumers, such as spiders) would demonstrate higher δ¹⁵N values. Finally, we examined whether the variance in isotopic composition among species was related to patterns of habitat use. We predicted that species with similar isotopic signatures would show divergent habitat use patterns, because competition for dietary resources may have mediated shifts in habitat use in arid Australian lizards [1, 21–25].

Materials and methods

Isotopic analysis

We quantified isotopic composition from ethanol-preserved liver tissue from adult sphenomorphine skinks of three genera (Ctenotus, Lerista, and Eremiascincus) that regionally co-occur in the western Australian arid zone. Whole-animal voucher specimens and tissue samples were collected from populations at Lorna Glen (Matuwa) and Yamarna Station in Western Australia’s Great Victoria Desert in 2004–2006 under permit SF0004654 to DLR (Western Australia Department of Parks and Wildlife). These locations are approximately 321.5 km apart but are broadly similar in physiography and have similar squamate reptile communities. For continuity with previous literature [1,24,26], we note that one of the focal species found in both regions, Ctenotus helenae, was recently synonymized with Ctenotus inornatus [27] and the current name is used here. A total of 141 individuals from 14 species were analyzed for isotopic content, with 2–20 individuals sampled per species. Liver tissue preserved in EtOH was dried and ground using a 1/8” metal bead lysing matrix from MP Biomedicals. Pulverized tissue was weighed in tin capsules and stored in a desiccator until used for analysis. Carbon and nitrogen stable isotope ratios (δ¹³C and δ¹⁵N respectively) were measured on a Finnigan MAT Delta Plus IRMS coupled to an elemental analyzer (Carlo Erba NC2500) at the Cornell University Stable Isotope Laboratory. Stable isotope ratios were expressed as a delta (δ) in ‰ (per mil or parts per thousand) according to: where R_sample is the ratio of the heavy isotope to light isotope (¹³C/¹²C or ¹⁵N/¹⁴N) in the sample, R_standard is the ratio of the heavy isotope to light isotope in the working reference gas, which is calibrated against an internationally known IAEA standard (V-PDB for δ¹³C and atmospheric nitrogen for δ¹⁵N), and d_{sample-standard} is the difference in isotopic composition of the sample relative to that of the reference, expressed in units per mil (‰). Errors associated with linearity were corrected at the Cornell Laboratory using a two-point normalization (linear regression) of all δ¹³C and δ¹⁵N data using two additional in-house standards that loosely resemble the samples being analyzed (HCRN, a corn standard, and CBT, a trout standard; these standards are run once every ten samples to identify variability in measurement or long-term drift). Note that preservation in EtOH has not been found to have a significant effect on isotopic composition of tissue from a range of consumer taxa [28–30], although one study has found that for liver tissue specifically, preservation in 95% EtOH leads to an increase in δ¹³C of around 1.5 ‰ [31].

Ecological data

Data on habitat use was collected by DLR and collaborators [1] as a total of 14 habitat variables relating to vegetation structure and substrate type, collected at Lorna Glen for each of 928 individual pitfall traps as part of a long term study on community structure for arid-zone vertebrates. Examples of measurements include a visual estimate, within a 3-m radius of each trap, of the percentage of ground covered by vegetation such as hummock grasses (spinifex) or chenopod shrubs; the percentage of exposed ground composed of substrates such as sand or gravel; and additional measurements of habitat qualities such as soil compaction and volume of woody debris. The mean value was computed for each habitat variable for each species over all pitfall traps where the species was collected, and means were transformed according to the method (log, logit, or arcsine square root) leading to the best approximation of normality. A complete description of habitat variables and methodology used to derive species-specific habitat scores can be found in [1]. Habitat data are available for 9 species in the present study, all of which are represented in the Lorna Glen isotope dataset and 7 of which are represented in the Yamarna isotope dataset.

Skinks were collected at Yamarna by ERP and collaborators from 1978–1979 and deposited in collections at the Western Australian Museum and the University of Texas. Stomach contents were removed from preserved animals through dissection, and prey items were sorted into 63 categories, including but not limited to centipedes, spiders, termites, ants, and beetles. Percent stomach volume occupied by each prey class was calculated for individual lizards, and means were calculated for each species. Ontogenetic shifts in diet are generally considered rare in small lizards [32–34] and fewer than five percent of lizards in our dataset had body sizes consistent with juvenile status. Hence, we believe ontogenetic dietary change is only a marginal contributor to the overall variation in isotopic composition within the focal taxa. Empirical dietary data were available for only 8 species after selection for sample size (N ≥ 5), and subsequent statistical analyses were conducted separately for Lorna Glen (6 of 8 species represented) and Yamarna (8 of 8 species represented) communities.

Data analysis

All statistical and phylogenetic analyses were performed in the R programming/statistical environment [35]. We first assessed variation in isotopic composition among species using a oneway ANOVA with unequal variances for δ¹³C and δ¹⁵N values separately (δ¹³C: Bartlett’s K-squared = 33.39, d.f. = 16, P = 0.007; δ¹⁵N: Bartlett’s K-squared = 32.68, d.f. = 16, P = 0.008). A oneway ANOVA was also performed for δ¹³C and δ¹⁵N values to determine if there was any effect of geographic region (Yamarna vs. Lorna Glen) on isotopic composition.

We tested for phylogenetic signal for δ¹³C and δ¹⁵N values pooled from both communities using a branch-length transformation test (Pagel's λ; [36]) as implemented in the function phylosig from the package phytools [37]. Briefly, the method compares the likelihood of a phenotypic dataset under a model where patterns of trait covariation are predicted by phylogeny to a null model where no phylogenetic correlation in species trait values exists. Analyses were conducted on the maximum clade credibility tree from a previously-published distribution of time-calibrated species-level phylogenies for Australian sphenomorphine skinks that included all species from the Lorna Glen and Yamarna communities [19]. We also tested for phylogenetic signal for δ¹³C and δ¹⁵N values in each regional community separately, to minimize the potential effects of regional variation on species mean values.

We tested for an association between isotopic composition and habitat use using a Mantel test to compare pairwise Euclidean distance matrices of species’ mean values for δ¹³C and δ¹⁵N to a distance matrix computed from fourteen habitat variables from [1]. Carbon and nitrogen values were analyzed together and separately, factored by location. To assess the relationship between dietary similarity and stable isotope composition, we used a Mantel test to compare species' isotopic means to dietary similarity as inferred from stomach content analyses. Because the majority of species from these datasets belong to the genus Ctenotus, we also conducted these analyses with the non-Ctenotus species removed. Non-Ctenotus taxa included several species from the genus Lerista, a fossorial and highly limb reduced clade that is ecologically distinct from sphenomorphine lineages. We also tested for an association between stomach contents and habitat use to identify correlations between trophic ecology and microhabitat that may have gone undetected by isotopic analysis. We used Mantel tests implemented in the ade4 package [38].

We also conducted phylogenetically informed three-way Mantel tests for habitat and diet associations with isotopic composition separately for the two regional communities using a permutation algorithm, demonstrated to be less prone to type-I error than the traditional Mantel test [39–40].

To further compare results of stable isotope analysis with stomach content analyses, we tested whether isotopic measures of trophic position were correlated with the estimated trophic rank of each species as inferred from stomach content analysis. We used species' positions along the δ¹⁵N axis as a proxy for trophic ordering in isotopic space; these values were compared to species’ ordering predicted by a weighted trophic rank calculated based on the mean proportion of prey items in a species’ stomach and their general trophic position. Prey items were assigned a trophic rank from 1 to 5 under the following coding scheme: 1 = producer; 2 = primary grazer (exclusively phytophagous consumer); 3 = omnivorous consumer (phytophagous and predator of primary grazers); 4 = predator of omnivorous consumers; and 5 = predator of vertebrate prey species. Prey items that were identified only as higher order taxonomic groups with the potential to contain species from multiple ranked categories were assigned a value intermediate between the two ranks. The weighted trophic rank was then calculated as follows: where i is a given prey item, n is the number of distinct prey types present in a species’ stomach, p_i is the proportion of the ith prey item in the stomach, and T_i is the corresponding trophic rank of the prey item. This calculation ensures that a species’ trophic score will be more strongly affected by the rank of prey items on which it concentrates heavily, similar to the process of isotopic assimilation wherein species’ tissues accumulate the isotopes (and more strongly reflect the trophic level, with predictable isotopic enrichment) of the prey they preferentially consume.

Results

We found a significant effect of species category on both δ¹³C and δ¹⁵N values (p < 0.0001 for both) (Table 1, Fig 1), indicating that species differ significantly in their position in isotopic space. There was also a significant effect of location on δ¹⁵N values (p < 0.0001), but not on δ¹³C values (p = 0.755) (Table 1). We found that δ¹⁵N values for several species collected at Yamarna were significantly lower than values for those same species collected at Lorna Glen; the ranges for both isotopes also differed by location (Figs 2 and 3, Tables 1 and 2). Assuming a δ¹⁵N enrichment of 2.5‰ for each trophic transfer (a meta-analytical study of isotopic enrichment [41] does not include lizards, but finds that tissue from birds yields values of trophic enrichment between 2–3‰, with bird liver specifically yielding measurements closer to 2‰), both communities appear to span four to five trophic levels (Fig 1). This was not consistent with estimates of trophic position from stomach content data (Table 3, Fig 4).

Download:

Table 1. ANOVA results of stable isotope analysis showing effects of species and location.

https://doi.org/10.1371/journal.pone.0172879.t001

Download:

Table 2. Means and ranges of δ¹⁵N and δ¹³C signatures for species from Yamarna and Lorna Glen.

https://doi.org/10.1371/journal.pone.0172879.t002

Download:

Table 3. Weighted trophic rank of sphenomorphine species based on stomach content analyses.

https://doi.org/10.1371/journal.pone.0172879.t003

Download:

Fig 1. Differentiation in isotopic space for scincid lizards from Lorna Glen and Yamarna communities.

δ¹³C and δ¹⁵N signatures measured in units per mil for individuals collected from the Lorna Glen (A) and Yamarna (B) regional communities; note clustering of individuals by taxon. Species are significantly differentiated in isotopic space (Table 1), suggesting that stable isotope profiles contain information about species-specific ecological traits.

https://doi.org/10.1371/journal.pone.0172879.g001

Download:

Fig 2. Regional differences in mean δ¹³C and δ¹⁵N signatures for species present in both communities.

(A) Species’ mean δ¹⁵N from Lorna Glen vs. mean δ¹⁵N from Yamarna. Signatures tend to be higher for species from Lorna Glen, and there is no significant correlation between communities (r = 0.68, p = 0.09). 1:1 relationship shown for reference as a dashed line. (B) Species’ mean δ¹³C from Lorna Glen vs. mean δ¹³C from Yamarna. There is no significant correlation between communities (r = 0.54, p = 0.21), but a near significant correlation emerges upon removal of the outlier point representing Ctenotus dux (r = 0.80, p = 0.057; r² = 0.548). 1:1 relationship shown for reference as a dashed line.

https://doi.org/10.1371/journal.pone.0172879.g002

Download:

Fig 3. Regional shifts in mean stable isotope signature.

Mean δ¹⁵N and δ¹³C signatures of the same species are significantly different depending on location (p = 0.0004 and p = 0.0267 respectively). There is no clear trend for the shift in δ¹³C signature, but δ¹⁵N signatures are lower in Yamarna for all species except C. inornatus.

https://doi.org/10.1371/journal.pone.0172879.g003

Download:

Fig 4. Trophic estimates from stomach contents do not correlate with δ¹⁵N ranking.

(A) Mean δ¹⁵N signature vs. weighted trophic rank based on stomach content analyses for the Lorna Glen regional community (based on gut contents for species collected in Yamarna). Lines represent δ¹⁵N range for each species. Isotopic signatures suggest that the focal taxa span 4–5 trophic levels (assumed enrichment 2.5‰ per level), while the corresponding range inferred from stomach contents is 2–3. There is no significant correlation between δ¹⁵N ranking and estimated trophic rank (r = -0.20, p = 0.70). (B) Corresponding analysis for the Yamarna regional community. Isotopic estimate of trophic level range is narrower (3) than for Lorna Glen, but more similar to estimates derived from stomach contents (2–3). There is no significant correlation between δ¹⁵N ranking and estimated trophic rank (r = 0.35, p = 0.39). Diet data were not available for all species present in the Yamarna dataset, and the full set appears to span 4–5 trophic levels, see Fig 1). Prey item images are numbered by assigned trophic rank (see text for details) and aligned along the y-axis according to estimated relative nitrogen content (as predicted from a 2–3‰ enrichment per trophic transfer [41] beginning at 0‰ for spinifex primary producers; but note Acacia vegetation is enriched in δ¹⁵N, and prey items associated with these plants may reflect this enrichment [42]).

https://doi.org/10.1371/journal.pone.0172879.g004

There was no phylogenetic signal for δ¹³C or δ¹⁵N values from the pooled dataset. For δ¹³C, the maximum likelihood estimate of Pagel's λ was approximately zero (λ = 5.75 x 10⁻⁵); using a likelihood-ratio test, this value is not significantly different from a null model with no phylogenetic signal (P = 1 for λ_free versus λ = 0). The maximum likelihood estimate of Pagel's λ for δ¹⁵N was also approximately zero (λ = 5.74 x 10⁻⁵), and not significantly different from the null model (P = 1 for λ_free versus λ = 0). Additionally, there was no phylogenetic signal for isotopic values when separated by community (Lorna Glen: λ = 7.235 x 10⁻⁵ for δ¹³C, λ = 7.235 x 10⁻⁵ for δ¹⁵N, P = 1 for λ_free versus λ = 0; Yamarna: λ = 5.92 x 10⁻⁵ for δ¹³C, λ = 5.92 x 10⁻⁵ for δ¹⁵N, P = 1 for λ_free versus λ = 0). We note that tests of phylogenetic signal may have relatively low power for small phylogenies [36]. However, we consider it unlikely that low statistical power is driving these relationships given that maximum likelihood estimates of λ converged on zero (e.g., λ < 0.001) for all analyses.

We found a significant positive relationship between pairwise distances in isotopic and habitat space for species of Lorna Glen (r = 0.43, p = 0.01) (Fig 5B). We distinguished the contributions of carbon and nitrogen by performing separate Mantel tests for δ¹⁵N and δ¹³C. We found no relationship between species’ distance for δ¹³C values and species’ distance in habitat space, while distances for δ¹⁵N and habitat values result in an r = 0.44 with p = 0.01. The correlation between isotopic composition and habitat use is strengthened after removing the non-Ctenotus taxa, which are ecologically distinct from the Ctenotus species that comprise the majority of sphenomorphines in the community (r = 0.63, p = 0.01; N = 7). The significant positive relationship between isotopic signature and habitat use for all species persists after accounting for autocorrelation based on phylogenetic relatedness using a phylogenetically permuted Mantel test (r = 0.44, p = 0.048); however, the relationship for only Ctenotus species becomes nonsignificant (r = 0.63, p = 0.31; N = 9). There is no relationship between distances in isotopic and habitat space for Yamarna species (r = -0.35, p = 0.98) (Fig 5A), and these results hold when comparing only Ctenotus species and when conducting a phylogenetically permuted Mantel test (r = -0.40, p = 0.93; r = 0.11, p = 0.87).

Download:

Fig 5. Pairwise distances in isotopic and ecological space for species from Lorna Glen and Yamarna regional communities.

Euclidean distances in isotope vs. habitat space are uncorrelated for (A) Yamarna species (r = -0.35, p = 0.98) and are positively correlated for (B) Lorna Glen species (r = 0.43, p = 0.01). Euclidean distances in isotope vs. diet space are positively correlated for (C) Yamarna species (r = 0.37, p = 0.05) and (D) Lorna Glen species (r = 0.37, p = 0.05).

https://doi.org/10.1371/journal.pone.0172879.g005

We found a significant and positive relationship between species’ distances in isotopic and dietary space for both Lorna Glen (r = 0.37, p = 0.05) (Fig 5D) and Yamarna (r = 0.37, p = 0.05) (Fig 5C) communities, but these results were not significant when the null distribution was constructed using phylogenetically-informed permutations (r = 0.67, p = 0.14; r = 0.61, p = 0.13). However, we note that phylogenetically-permuted Mantel tests have been shown to exhibit low power [39]. Computing distance with termites as the sole prey category (proportions in diet ranging from 10–75% for species in a dataset from ERP, Table 3) revealed a significant correlation with distance in isotopic space for Yamarna species (r = 0.73, p = 0.02), b b but not Lorna Glen. Additionally, species’ ordering along the δ¹⁵N axis did not match predictions based on calculations of weighted trophic rank for either community (Lorna Glen: r = -0.20, p = 0.70; Yamarna: r = 0.35, p = 0.39), although trophic spread was relatively consistent between estimates based on stomach content and isotopic signature for Yamarna (Table 3, Fig 4).

We found no significant relationship between species pairwise distances in diet and habitat space (Mantel test: r = -0.26, p = 0.86; phylogenetically permuted Mantel test: r = 0.19, p = 0.6).

Discussion

Isotopic signatures demonstrate that sphenomorphine skinks of a diverse community from the western Australian arid zone are differentiated in a stable isotope space defined by δ¹³C and δ¹⁵N, implying that isotopic composition is tracking an underlying set of species-specific ecological attributes. We find that differences between species are correlated with habitat and unrelated to phylogeny at the scale of this study. Additionally, while empirical data on gut content analysis are shown to function as a moderate predictor of isotopic content, stable isotope analysis may suggest alternative dietary habits in some species in addition to variation among sites. Our study shows that stable isotope signatures could be a useful but imperfect surrogate for identifying ecological differentiation among lizard species in the Australian deserts. However, we did not find evidence that trophic position as inferred from stomach content analysis is correlated with isotopic estimates of trophic position based on δ¹⁵N signatures.

Stable isotopes and stomach content analysis

Previous studies using stomach content analyses have found separation in dietary habits among Ctenotus species [20,26]. Our results from stable isotope analysis confirm that species appear differentiated in diet space, and suggest that this differentiation persists through time, given the timescales over which isotopic assimilation and turnover occur in reptile tissue [15–16]. Stable isotope signatures of the species in this study are comparable to values reported for reptiles in existing analyses [15–16,43]. While δ¹⁵N and δ¹³C values for prey items in the GVD are not available for comparison to signatures of their consumers, previous studies on detrital and grazing food webs in desert and other terrestrial ecosystems show that isotope signatures provide reliable measures of trophic links in these communities, and are subsequently detectable in higher order consumers [43–45]. These studies also demonstrate fine scale variation among prey items of the same trophic order that represent differences in grazing among C₃ and C₄ plant types, and that this variation is reflected by the spread of intraguild isotopic signatures in predators, similar to patterns of differentiation along the δ¹³C axis detected by the present study (Fig 1).

Despite a significant correlation between similarity in isotopic signature and similarity in dietary habits for both communities, much of the variation remains unexplained (Fig 5C and 5D). Isotopic composition better predicts dietary habits for the Yamarna community, from which data for stomach content analyses were collected. This is congruent with our results suggesting that intraspecific dietary habits differ between sites (Fig 3). Isotopic signatures for Yamarna species are correlated with the proportion of termites in a species’ diet, and are better matched to the trophic spread predicted by stomach content analysis than Lorna Glen signatures (Fig 4). However, for some species, the weighted trophic rank was a poor predictor for isotopic composition in both communities. For example, stomach content analyses indicate that C. quattuordecimlineatus feeds mostly on first and second order consumers (weighted trophic rank between 2 and 3), while the high δ¹⁵N values of this species from both Yamarna and Lorna Glen suggest a diet of tertiary or higher consumers, 3–4 trophic transfers (at 2.54‰ [41]) from the base of the food chain. Likewise, mismatches between the ordering of species along the δ¹⁵N axis and their weighted trophic rank relative to other species occurs in both communities; for example, trophic positioning based on stomach content analysis implies that C. grandis feeds almost exclusively at one full trophic level above C. inornatus, but the mean δ¹⁵N signatures of these species in the Yamarna community imply that they feed at roughly the same trophic level (and the mean δ¹⁵N signature of C. grandis is in fact lower than that of C. inornatus) (Table 3, Fig 4B). However, while the ordering of species along the δ¹⁵N axis is rearranged compared to what might be predicted from gut content analyses, most species pairs differ by less than what is expected for a single trophic transfer (2.54‰, [41]). Therefore, omnivorous species could feed more preferentially on higher or lower trophic order prey items than expected from gut content analyses, and these differences are reflected by the intraguild variation of sphenomorphine species in isotopic space.

Discrepancies between isotopic composition and stomach contents may be considered a reflection of biases in gut content analysis, in the relationship between isotopic composition and diet, or both. Stomach content analysis typically provides a “snapshot” view of an organism’s diet for a particular point in time and space, and may include biases of digestion rate and identification accuracy [14,46–48]. Likewise, isotopic data may not accurately reflect dietary patterns due to sampling error (e.g., small sample sizes), fractionation differences among species or prey items, or external sources of variation [49–50]. Furthermore, stomach content data came from sites at Yamarna that excluded certain habitats common to both regions (e.g. Acacia-dominanted shrubland). Because signatures from these habitat types may be reflected in some of the isotopic variation, this incongruence may explain the lack of strong correlation between isotopic data and stomach contents. Additionally, temporal incongruences among the datasets may reflect variation introduced by environmental change. Wildfires in the arid Australian interior can cause major turnover in plant communities, and lizard diets from the Great Victoria Desert have been shown to fluctuate during fire succession cycles [20,51]; discrepancies between isotopic data and stomach content data collected for this study during different years may well be due to fine-scale ecological variation associated with successional history.

Stable isotopes and habitat use

For the species of Lorna Glen, variation in δ¹⁵N isotopic signature is correlated with divergence in habitat use, such that species closer in habitat space are also more similar in nitrogen isotopic composition. Because we lack isotopic data for prey items for the communities considered here, the present study cannot say definitively whether increased similarity in isotopic composition is directly attributable to increased similarity in diet. Correspondingly, all species could be differentiated in diet and isotopic similarity is potentially attributable to habitat-specific differences in baseline δ¹⁵N signatures, due to differential nitrogen-fixing abilities of vegetation. However, this explanation cannot account for the lack of association between δ¹³C and habitat. Moreover, discrepancies between habitat variables and isotopic signature may reflect successional changes between the time of collection for habitat and isotope data, as discussed above.

Despite the absence of a significant relationship between δ¹³C composition and habitat use, changes in isotopic composition with individual habitat variables reveal predictable patterns of variation. Sites in the study region are characterized by a mosaic of habitats with dominant vegetation types that differ in their use of C₃ and C₄ photosynthetic pathways; among these are Acacia groves (C₃), spinifex sandplains (C₄), and chenopod shrublands (C₄). Because plants that utilize a C₄ photosynthetic pathway discriminate less strongly against the heavy stable isotope of carbon (¹³C) than C₃ plants, C₄ plants tend to exhibit δ¹³C signatures between -17 to -9‰, while C₃ plants exhibit δ¹³C signatures between -32 to -22‰ [42]. Since carbon is not significantly enriched through trophic transfers [4,52], prey items found in different habitats will reflect the baseline signatures of primary producers [4]. Because most species’ means for δ¹³C are intermediate between the ranges expected for C₃ and C₄ plants, skinks of these communities seem to be feeding on prey items that utilize both types of vegetation, or on predators that utilize two types of prey with differing C₃/C₄ phytophagy preferences. Species strongly associated with spinifex grasses (C. pantherinus, C. calurus, C. inornatus, C. quattuordecimlineatus, C. grandis [1]), which utilize a C₄ photosynthetic pathway, exhibit higher δ¹³C signatures than species that are more strongly associated with Acacia woodlands (C. schomburgkii and C. leonhardii [1]), in which the trees utilize a C₃ photosynthetic pathway (Welch’s two sample t-test, p = 0.03). Additionally, Acacia plants are nitrogen-fixing, resulting in soil and litter that is enriched in δ¹⁵N [42]; and the two species associated with this habitat demonstrate some of the highest δ¹⁵N signatures (Table 2).

Variation between sampling sites

In addition to revealing how dietary differentiation persists through time, results from stable isotope analysis show that species separation is also consistent among sites; however, several species from Yamarna demonstrate a significantly lower signal for δ¹⁵N values than the same species collected at Lorna Glen. A greater diversity or abundance of predatory arthropods in the Lorna Glen community might contribute to the higher δ¹⁵N signatures exhibited by species from this site. Such hypotheses are speculative, although nearby regions have been observed by the authors to show some variation in macroinvertebrate communities. Alternatively, potential differences in successional history between the two regions could have resulted in different plant communities [20], which could in turn allow for variation in populations of lower trophic order prey items like termites (in addition to promoting variability in δ¹³C signatures among sites). While plant communities of the two study regions are broadly comparable, we lack quantitative data necessary to assess variation. Thus, the present study is unable to distinguish whether inconsistencies in signatures between sites reflect variation in baseline resource values of δ¹⁵N and δ¹³C, or a change in food web structure [53], and isotopic signatures of consumer prey species from each site are recommended for accurate comparisons of trophic position [53].

Recent advancements in stable isotope analysis have enabled the comparison of ecological groups using multivariate ellipse-based metrics in a Bayesian framework, eliminating uncertainty derived from small sample sizes and allowing for robust statistical comparison among communities by accounting for natural variability within the system [54]. These descriptive metrics, expanded from a previous study [5], can be used as measurements of niche structure and incorporated with other axes of ecological information to obtain more comprehensive estimates of niche hypervolume. However, the lack of strong structure exhibited by isotopic data along ecological axes of diet and habitat in the present study, in addition to the lack of coordination between datasets, could make inference of niche width from species’ arrangements in isotopic space somewhat unreliable. Obtaining isotopic signatures for the prey items available to Sphenomorphine skinks in the GVD would also be beneficial, and would allow researchers to exploit isotope mixing models that would provide more precise estimates of the relative contribution of each prey category to an individual’s diet [55].

Conclusion

This study demonstrates that stable isotope analysis is useful in revealing patterns of species separation in the hyperdiverse lizard communities of the western Australian arid zone. Despite intraspecific variation in resource use and interspecific similarities in prey preference, our results show that stable isotopes can reveal patterns of trophic differentiation within and between sites when combined with stomach content analysis. These methods are an attractive complement to high-resolution analyses of individual stomach contents (e.g., [20,24], etc.) as they potentially provide a temporally integrated index of trophic ecology in reptile communities. However, much variation remains unexplained in our study, highlighting the need for a more comprehensive isotopic analysis of both prey and consumer species. The trophic ecology of lizard communities of the Great Victoria Desert and adjacent regions is perhaps as well studied as any squamate reptile assemblage [21–24], yet the incongruences noted here suggest that much remains to be learned about this and other systems.

Supporting information

S1 File. Habitat data from Rabosky et al. (2011) American Naturalist.

This supplementary file is a previous publication containing habitat data used in the present study.

https://doi.org/10.1371/journal.pone.0172879.s001

(PDF)

S2 File. Stable isotope and stomach content data used in the present study.

https://doi.org/10.1371/journal.pone.0172879.s002

(ZIP)

Acknowledgments

We thank K. Sparks and staff at the Cornell University Stable Isotope Laboratory (COIL) for efficient isotopic analysis of the samples used in this study. We are also grateful to L. Wingate at the University of Michigan Stable Isotope Laboratory for preliminary analysis and advice supporting this work. We thank C. Dardia, P. Doughty, I. Kealley, and staff of the Kalgoorlie office of the WA Department of Parks and Wildlife for logistical assistance associated with fieldwork and sample procurement. This work was supported by National Science Foundation (NSF) grants OSIE-0612855 and DEB-0814277, and by the David and Lucile Packard Foundation. Field sampling at Matuwa/Lorna Glen Conservation Park was conducted under permit SF0004654 to DLR (Western Australia Department of Parks and Wildlife). Pianka thanks the National Geographic Society and the Denton A. Cooley Centennial Professorship in Zoology at UT Austin for financial support. NP thanks CONICET for the external postdoctoral fellowship to carry out research activities at the University of Texas at Austin.

Author Contributions

Conceptualization: MRG DLR.
Formal analysis: MRG DLR.
Funding acquisition: DLR ERP.
Investigation: MRG DLR ERP NP MAC.
Methodology: MRG DLR.
Resources: DLR ERP NP MAC.
Visualization: MRG DLR.
Writing – original draft: MRG.
Writing – review & editing: MRG DLR ERP NP MAC.

References

1. Rabosky DL, Cowan MA, Talaba AL, Lovette IJ. Species interactions mediate phylogenetic community structure in a hyperdiverse lizard assemblage from arid Australia. Am. Nat. 2011;178: 579–595. pmid:22030728
- View Article
- PubMed/NCBI
- Google Scholar
2. Webb CO. Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. Am. Nat. 2000;156(2): 145–155. pmid:10856198
- View Article
- PubMed/NCBI
- Google Scholar
3. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ. Phylogenies and community ecology. Annu. Rev. Ecol. Syst. 2002;33: 475–505.
- View Article
- Google Scholar
4. Ponsard S, Arditi R. What can stable isotopes (δ¹⁵N and δ¹³C) tell about the food web of soil macro-invertebrates? Ecology. 2000;81(3): 852–864.
- View Article
- Google Scholar
5. Layman CA, Arrington DA, Montaña CG, Post DM. Can stable isotope ratios provide for community-wide measures of trophic structure? Ecology. 2007;88: 42–48. pmid:17489452
- View Article
- PubMed/NCBI
- Google Scholar
6. DeNiro MJ, Epstein S. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta. 1978;42: 495–506.
- View Article
- Google Scholar
7. DeNiro MJ, Epstein S. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta. 1981;45: 341–351.
- View Article
- Google Scholar
8. Michener R, Lajtha K, editors. Stable isotopes in ecology and environmental science. John Wiley & Sons; 2008 Apr 15.
- View Article
- Google Scholar
9. Ingram T, Shurin JB. Trait-based assembly and phylogenetic structure in northeast Pacific rockfish assemblages. Ecology. 2009;90: 2444–2453. pmid:19769123
- View Article
- PubMed/NCBI
- Google Scholar
10. Wagner CE, McIntyre PB, Smith K, Michel E. Diet predicts intestine length in Lake Tanganyika’s cichlid fishes. Funct. Ecol. 2009;23: 1122–1131.
- View Article
- Google Scholar
11. Matthews B, Marchinko KB, Bolnick DI, Mazumder A. Specialization of trophic position and habitat use by stickelbacks in an adaptive radiation. Ecology. 2010;91: 1025–1034. pmid:20462117
- View Article
- PubMed/NCBI
- Google Scholar
12. Bourguignon T, Sobotnik J, Lepoint G, Martin J-M, Hardy OJ, Dejean A, et al. Feeding ecology and phylogenetic structure of a complex neotropical termite assemblage, revealed by nitrogen isotope ratios. Ecol. Entomol. 2011;36: 261–269.
- View Article
- Google Scholar
13. Muschick M, Indermaur A, Salzburger W. Convergent evolution within an adaptive radiation of cichlid fishes. Curr. Biol. 2012;22: 2362–2368. pmid:23159601
- View Article
- PubMed/NCBI
- Google Scholar
14. Dalerum F, Angerbjörn A. Resolving temporal variation in vertebrate diets using naturally occurring stable isotopes. Oecologia. 2005;144: 647–658. pmid:16041545
- View Article
- PubMed/NCBI
- Google Scholar
15. Seminoff JA, Bjorndal KA, Bolten AB. Stable carbon and nitrogen isotope discrimination and turnover in pond sliders Trachemys scripta: insights for trophic study of freshwater turtles. Copeia. 2007;3: 534–542.
- View Article
- Google Scholar
16. Warne RW, Gilman CA, Wolf BO. Tissue-carbon incorporation rates in lizards: implications for ecological studies using stable isotopes in terrestrial ectotherms. Physiol. Biochem. Zool. 2010;83(4): 608–617. pmid:20441446
- View Article
- PubMed/NCBI
- Google Scholar
17. Durso AM, Mullin SJ. Ontogenetic shifts in the diet of plains hog-nosed snakes (heterodon nasicus) revealed by stable isotope analysis. J. Zool. 2016;
- View Article
- Google Scholar
18. Skinner A, Hugall AF, Hutchinson MN. Lygosomine phylogeny and the origins of Australian scincid lizards. J. Biogeogr. 2011;38: 1044–1058.
- View Article
- Google Scholar
19. Rabosky DL, Donnellan ST, Grundler MC, Lovette IJ. Analysis and visualization of complex macroevolutionary dynamics: an example from Australian scincid lizards. Syst. Biol. 2014;63(4): 610–627. pmid:24682412
- View Article
- PubMed/NCBI
- Google Scholar
20. Goodyear SE, Pianka ER. Spatial and temporal variation in diets of sympatric lizards (genus Ctenotus) in the Great Victoria Desert, Western Australia. J Herpetol. 2011;45: 265–271.
- View Article
- Google Scholar
21. Pianka ER. Habitat specificity, speciation, and species density in Australian desert lizards. Ecology. 1969;50: 498–502.
- View Article
- Google Scholar
22. Pianka ER. Sympatry of desert lizards (Ctenotus) in western Australia. Ecology. 1969;50: 1012–1030.
- View Article
- Google Scholar
23. Pianka ER. Zoogeography and speciation of Australian desert lizards: an ecological perspective. Copeia. 1972 Mar 8: 127–145.
- View Article
- Google Scholar
24. Pianka ER. Ecology and natural history of desert lizards. Princeton University Press, Princeton, NJ; 1986.
25. Rabosky DL, Reid J, Cowan MA, Foulkes J. Overdispersion of body size in Australian desert lizard communities. Oecologia 2007;154: 561–570. pmid:17874134
- View Article
- PubMed/NCBI
- Google Scholar
26. James CD. Temporal variation in diets and trophic partitioning by coexisting lizards (Ctenotus, Scincidae) in central Australia. Oecologia (Berlin). 1991;85: 553–561.
- View Article
- Google Scholar
27. Rabosky DL, Hutchinson MN, Donnellan SC, Talaba AL, Lovette IJ. Phylogenetic disassembly of species boundaries in a widespread group of Australian skinks (Scincidae: Ctenotus). Mol. Phylogenet. Evol. 2014;77: 71–82 pmid:24732682
- View Article
- PubMed/NCBI
- Google Scholar
28. Arrington DA, Winemiller KO. Preservation effects of stable isotope analysis of fish muscle. T. Am. Fish. Soc. 2002;131(2): 337–342.
- View Article
- Google Scholar
29. Sarakinos HC, Johnson ML, Vander Zanden MJ. A synthesis of tissue-preservation effects on carbon and nitrogen stable isotope signatures. Can. J. Zool. 2002;80: 381–387.
- View Article
- Google Scholar
30. Feuchtmayr H, Grey J. Effect of preparation and preservation procedures on carbon and nitrogen stable isotope determinations from zooplankton. Rapid Commun. Mass Spectrom. 2003;17: 2605–2610. pmid:14648896
- View Article
- PubMed/NCBI
- Google Scholar
31. Sweeting CJ, Polunin NVC, Jennings S. Tissue and fixative dependent shifts of δ¹³C and δ¹⁵N in preserved ecological material. Rapid Comm. Mass Spectrom. 2004;18(21): 2587–2592.
- View Article
- Google Scholar
32. Pough FH. Lizard energetics and diet. Ecology. 1973;54(4): 837–844.
- View Article
- Google Scholar
33. Werner EE, Gilliam JF. The ontogenetic niche and species interactions in size-structured populations. Ann. Rev. Ecol. Syst. 1984;15: 393–425.
- View Article
- Google Scholar
34. Cooper WE, Vitt LJ. Distribution, extent, and evolution of plant consumption by lizards. J. Zool., Lond. 2002;257: 487–517.
- View Article
- Google Scholar
35. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. 2015. URL http://www.R-project.org/.
36. Freckleton RP, Harvey PH, Pagel M. Phylogenetic analysis and comparative data: a test and review of evidence. Am. Nat. 2002;160(6): 712–726. pmid:18707460
- View Article
- PubMed/NCBI
- Google Scholar
37. Revell LJ. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 2012;3: 217–223.
- View Article
- Google Scholar
38. Dray S, Dufour AB. The ade4 package: implementing the duality diagram for ecologists. J. Stat. Softw. 2007;22(4): 1–20.
- View Article
- Google Scholar
39. Harmon LJ, Glor RE. Poor statistical performance of the Mantel test in phylogenetic comparative analyses. Evolution. 2010;64(7): 2173–2178. pmid:20163450
- View Article
- PubMed/NCBI
- Google Scholar
40. Lapointe FJ, Garland T. A generalized permutation model for the analysis of cross-species data. J. Classif. 2001;18: 109–127.
- View Article
- Google Scholar
41. Vanderklift MA, Ponsard S. Sources of variation in consumer-diet and δ¹⁵N enrichment: a meta-analysis. Oecologia. 2003;136(2):169–182. pmid:12802678
- View Article
- PubMed/NCBI
- Google Scholar
42. Bowman DMJS, Boggs GS, Prior LD, Krull ES. Dynamics of Acacia aneura-Triodia boundaries using carbon (14C and δ13C) and nitrogen (δ15N) signatures in soil organic matter in central Australia. The Holocene. 2007;17(3): 311–318.
- View Article
- Google Scholar
43. Smith KF, Sharp ZD, Brown JH. Isotopic composition of carbon and oxygen in desert fauna: investigations into the effects of diet, physiology, and seasonality. J. Arid Environ. 2002;52: 419–430.
- View Article
- Google Scholar
44. McNabb DM, Halaj J, Wise DH. Inferring trophic positions of generalist predators and their linkage to the detrital food web in agroecosystems: a stable isotope analysis. Pedobiologia. 2001;45: 289–297.
- View Article
- Google Scholar
45. Halaj J, Peck RW, Niwa CG. Trophic structure of a macroarthropod litter food web in managed coniferous forest stands: a stable isotope analysis with δ¹⁵N and δ¹³C. Pedobiologia. 2005;49: 109–118.
- View Article
- Google Scholar
46. Inger R, Ruxton GD, Newton J, Colhoun K, Robinson JA, Jackson AL, et al. Temporal and intrapopulation variation in prey choice of wintering geese determined by stable isotope analysis. J Anim. Ecol. 2006;75(5): 1190–1200. pmid:16922855
- View Article
- PubMed/NCBI
- Google Scholar
47. Lugendo BR, Nagelkerken I, van der Velde G, Mgaya YD. The importance of mangroves, mud and sand flats, and seagrass beds as feeding areas for juvenile fishes in Chwaka Bay, Zanzibar: gut content and stable isotope analyses. J. Fish Biol. 2006;69: 1639–1661.
- View Article
- Google Scholar
48. Cherel Y, Hobson KA, Guinet C, Vanpe C. Stable isotopes document seasonal changes in trophic niches and winter foraging individual specialization in diving predators from the Southern Ocean. J. Anim. Ecol. 2007;76(4): 826–836. pmid:17584388
- View Article
- PubMed/NCBI
- Google Scholar
49. Hart EA, Lovvorn JR. Interpreting stable isotopes from macroinvertebrate foodwebs in saline wetlands. Limnol. Oceanogr. 2002;47(2): 580–584.
- View Article
- Google Scholar
50. Parnell AC, Inger R, Bearhop S, Jackson AL. Source partitioning using stable isotopes: Coping with too much variation. PLoS ONE. 2010;5: e9672. pmid:20300637
- View Article
- PubMed/NCBI
- Google Scholar
51. Pianka ER, Goodyear SE. Lizard responses to wildfire in arid interior Australia: long-term experimental data and commonalities with other studies. Austral Ecol. 2012;37: 1–11.
- View Article
- Google Scholar
52. Tayasu I, Hyodo F, Abe T, Inoue T, Spain AV. Nitrogen and carbon stable isotope ratios in the sympatric Australian termites, Amiterms laurensis and Drepanotermes rubriceps (Isoptera: Termitidae) in relation to their feeding habits and the quality of their food materials. Soil Biol. And Biochem. 2002;34(3): 297–301.
- View Article
- Google Scholar
53. Post DM. Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology. 2002;83: 703–718.
- View Article
- Google Scholar
54. Jackson AL, Inger R, Parnell AC, Bearhop S. Comparing isotopic niche widths among and within communities: SIBER—Stable Isotope Bayesian Ellipses in R. Anim. Ecol. 2011;80: 595–602.
- View Article
- Google Scholar
55. Newsome SD, Yeakel JD, Wheatley PV, Tinker MT. Tools for quantifying isotopic niche space and dietary variation at the individual population level. J. Mammal;93(2): 329–341.
- View Article
- Google Scholar

[ref1] 1. Rabosky DL, Cowan MA, Talaba AL, Lovette IJ. Species interactions mediate phylogenetic community structure in a hyperdiverse lizard assemblage from arid Australia. Am. Nat. 2011;178: 579–595. pmid:22030728
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Webb CO. Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. Am. Nat. 2000;156(2): 145–155. pmid:10856198
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ. Phylogenies and community ecology. Annu. Rev. Ecol. Syst. 2002;33: 475–505.
View Article
Google Scholar

[10] View Article

[11] Google Scholar

[ref4] 4. Ponsard S, Arditi R. What can stable isotopes (δ¹⁵N and δ¹³C) tell about the food web of soil macro-invertebrates? Ecology. 2000;81(3): 852–864.
View Article
Google Scholar

[13] View Article

[14] Google Scholar

[ref5] 5. Layman CA, Arrington DA, Montaña CG, Post DM. Can stable isotope ratios provide for community-wide measures of trophic structure? Ecology. 2007;88: 42–48. pmid:17489452
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref6] 6. DeNiro MJ, Epstein S. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta. 1978;42: 495–506.
View Article
Google Scholar

[20] View Article

[21] Google Scholar

[ref7] 7. DeNiro MJ, Epstein S. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta. 1981;45: 341–351.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref8] 8. Michener R, Lajtha K, editors. Stable isotopes in ecology and environmental science. John Wiley & Sons; 2008 Apr 15.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref9] 9. Ingram T, Shurin JB. Trait-based assembly and phylogenetic structure in northeast Pacific rockfish assemblages. Ecology. 2009;90: 2444–2453. pmid:19769123
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref10] 10. Wagner CE, McIntyre PB, Smith K, Michel E. Diet predicts intestine length in Lake Tanganyika’s cichlid fishes. Funct. Ecol. 2009;23: 1122–1131.
View Article
Google Scholar

[33] View Article

[34] Google Scholar

[ref11] 11. Matthews B, Marchinko KB, Bolnick DI, Mazumder A. Specialization of trophic position and habitat use by stickelbacks in an adaptive radiation. Ecology. 2010;91: 1025–1034. pmid:20462117
View Article
PubMed/NCBI
Google Scholar

[36] View Article

[37] PubMed/NCBI

[38] Google Scholar

[ref12] 12. Bourguignon T, Sobotnik J, Lepoint G, Martin J-M, Hardy OJ, Dejean A, et al. Feeding ecology and phylogenetic structure of a complex neotropical termite assemblage, revealed by nitrogen isotope ratios. Ecol. Entomol. 2011;36: 261–269.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref13] 13. Muschick M, Indermaur A, Salzburger W. Convergent evolution within an adaptive radiation of cichlid fishes. Curr. Biol. 2012;22: 2362–2368. pmid:23159601
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref14] 14. Dalerum F, Angerbjörn A. Resolving temporal variation in vertebrate diets using naturally occurring stable isotopes. Oecologia. 2005;144: 647–658. pmid:16041545
View Article
PubMed/NCBI
Google Scholar

[47] View Article

[48] PubMed/NCBI

[49] Google Scholar

[ref15] 15. Seminoff JA, Bjorndal KA, Bolten AB. Stable carbon and nitrogen isotope discrimination and turnover in pond sliders Trachemys scripta: insights for trophic study of freshwater turtles. Copeia. 2007;3: 534–542.
View Article
Google Scholar

[51] View Article

[52] Google Scholar

[ref16] 16. Warne RW, Gilman CA, Wolf BO. Tissue-carbon incorporation rates in lizards: implications for ecological studies using stable isotopes in terrestrial ectotherms. Physiol. Biochem. Zool. 2010;83(4): 608–617. pmid:20441446
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref17] 17. Durso AM, Mullin SJ. Ontogenetic shifts in the diet of plains hog-nosed snakes (heterodon nasicus) revealed by stable isotope analysis. J. Zool. 2016;
View Article
Google Scholar

[58] View Article

[59] Google Scholar

[ref18] 18. Skinner A, Hugall AF, Hutchinson MN. Lygosomine phylogeny and the origins of Australian scincid lizards. J. Biogeogr. 2011;38: 1044–1058.
View Article
Google Scholar

[61] View Article

[62] Google Scholar

[ref19] 19. Rabosky DL, Donnellan ST, Grundler MC, Lovette IJ. Analysis and visualization of complex macroevolutionary dynamics: an example from Australian scincid lizards. Syst. Biol. 2014;63(4): 610–627. pmid:24682412
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref20] 20. Goodyear SE, Pianka ER. Spatial and temporal variation in diets of sympatric lizards (genus Ctenotus) in the Great Victoria Desert, Western Australia. J Herpetol. 2011;45: 265–271.
View Article
Google Scholar

[68] View Article

[69] Google Scholar

[ref21] 21. Pianka ER. Habitat specificity, speciation, and species density in Australian desert lizards. Ecology. 1969;50: 498–502.
View Article
Google Scholar

[71] View Article

[72] Google Scholar

[ref22] 22. Pianka ER. Sympatry of desert lizards (Ctenotus) in western Australia. Ecology. 1969;50: 1012–1030.
View Article
Google Scholar

[74] View Article

[75] Google Scholar

[ref23] 23. Pianka ER. Zoogeography and speciation of Australian desert lizards: an ecological perspective. Copeia. 1972 Mar 8: 127–145.
View Article
Google Scholar

[77] View Article

[78] Google Scholar

[ref24] 24. Pianka ER. Ecology and natural history of desert lizards. Princeton University Press, Princeton, NJ; 1986.

[ref25] 25. Rabosky DL, Reid J, Cowan MA, Foulkes J. Overdispersion of body size in Australian desert lizard communities. Oecologia 2007;154: 561–570. pmid:17874134
View Article
PubMed/NCBI
Google Scholar

[81] View Article

[82] PubMed/NCBI

[83] Google Scholar

[ref26] 26. James CD. Temporal variation in diets and trophic partitioning by coexisting lizards (Ctenotus, Scincidae) in central Australia. Oecologia (Berlin). 1991;85: 553–561.
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref27] 27. Rabosky DL, Hutchinson MN, Donnellan SC, Talaba AL, Lovette IJ. Phylogenetic disassembly of species boundaries in a widespread group of Australian skinks (Scincidae: Ctenotus). Mol. Phylogenet. Evol. 2014;77: 71–82 pmid:24732682
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref28] 28. Arrington DA, Winemiller KO. Preservation effects of stable isotope analysis of fish muscle. T. Am. Fish. Soc. 2002;131(2): 337–342.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref29] 29. Sarakinos HC, Johnson ML, Vander Zanden MJ. A synthesis of tissue-preservation effects on carbon and nitrogen stable isotope signatures. Can. J. Zool. 2002;80: 381–387.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref30] 30. Feuchtmayr H, Grey J. Effect of preparation and preservation procedures on carbon and nitrogen stable isotope determinations from zooplankton. Rapid Commun. Mass Spectrom. 2003;17: 2605–2610. pmid:14648896
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref31] 31. Sweeting CJ, Polunin NVC, Jennings S. Tissue and fixative dependent shifts of δ¹³C and δ¹⁵N in preserved ecological material. Rapid Comm. Mass Spectrom. 2004;18(21): 2587–2592.
View Article
Google Scholar

[102] View Article

[103] Google Scholar

[ref32] 32. Pough FH. Lizard energetics and diet. Ecology. 1973;54(4): 837–844.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref33] 33. Werner EE, Gilliam JF. The ontogenetic niche and species interactions in size-structured populations. Ann. Rev. Ecol. Syst. 1984;15: 393–425.
View Article
Google Scholar

[108] View Article

[109] Google Scholar

[ref34] 34. Cooper WE, Vitt LJ. Distribution, extent, and evolution of plant consumption by lizards. J. Zool., Lond. 2002;257: 487–517.
View Article
Google Scholar

[111] View Article

[112] Google Scholar

[ref35] 35. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. 2015. URL http://www.R-project.org/.

[ref36] 36. Freckleton RP, Harvey PH, Pagel M. Phylogenetic analysis and comparative data: a test and review of evidence. Am. Nat. 2002;160(6): 712–726. pmid:18707460
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref37] 37. Revell LJ. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 2012;3: 217–223.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref38] 38. Dray S, Dufour AB. The ade4 package: implementing the duality diagram for ecologists. J. Stat. Softw. 2007;22(4): 1–20.
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref39] 39. Harmon LJ, Glor RE. Poor statistical performance of the Mantel test in phylogenetic comparative analyses. Evolution. 2010;64(7): 2173–2178. pmid:20163450
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref40] 40. Lapointe FJ, Garland T. A generalized permutation model for the analysis of cross-species data. J. Classif. 2001;18: 109–127.
View Article
Google Scholar

[129] View Article

[130] Google Scholar

[ref41] 41. Vanderklift MA, Ponsard S. Sources of variation in consumer-diet and δ¹⁵N enrichment: a meta-analysis. Oecologia. 2003;136(2):169–182. pmid:12802678
View Article
PubMed/NCBI
Google Scholar

[132] View Article

[133] PubMed/NCBI

[134] Google Scholar

[ref42] 42. Bowman DMJS, Boggs GS, Prior LD, Krull ES. Dynamics of Acacia aneura-Triodia boundaries using carbon (14C and δ13C) and nitrogen (δ15N) signatures in soil organic matter in central Australia. The Holocene. 2007;17(3): 311–318.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref43] 43. Smith KF, Sharp ZD, Brown JH. Isotopic composition of carbon and oxygen in desert fauna: investigations into the effects of diet, physiology, and seasonality. J. Arid Environ. 2002;52: 419–430.
View Article
Google Scholar

[139] View Article

[140] Google Scholar

[ref44] 44. McNabb DM, Halaj J, Wise DH. Inferring trophic positions of generalist predators and their linkage to the detrital food web in agroecosystems: a stable isotope analysis. Pedobiologia. 2001;45: 289–297.
View Article
Google Scholar

[142] View Article

[143] Google Scholar

[ref45] 45. Halaj J, Peck RW, Niwa CG. Trophic structure of a macroarthropod litter food web in managed coniferous forest stands: a stable isotope analysis with δ¹⁵N and δ¹³C. Pedobiologia. 2005;49: 109–118.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref46] 46. Inger R, Ruxton GD, Newton J, Colhoun K, Robinson JA, Jackson AL, et al. Temporal and intrapopulation variation in prey choice of wintering geese determined by stable isotope analysis. J Anim. Ecol. 2006;75(5): 1190–1200. pmid:16922855
View Article
PubMed/NCBI
Google Scholar

[148] View Article

[149] PubMed/NCBI

[150] Google Scholar

[ref47] 47. Lugendo BR, Nagelkerken I, van der Velde G, Mgaya YD. The importance of mangroves, mud and sand flats, and seagrass beds as feeding areas for juvenile fishes in Chwaka Bay, Zanzibar: gut content and stable isotope analyses. J. Fish Biol. 2006;69: 1639–1661.
View Article
Google Scholar

[152] View Article

[153] Google Scholar

[ref48] 48. Cherel Y, Hobson KA, Guinet C, Vanpe C. Stable isotopes document seasonal changes in trophic niches and winter foraging individual specialization in diving predators from the Southern Ocean. J. Anim. Ecol. 2007;76(4): 826–836. pmid:17584388
View Article
PubMed/NCBI
Google Scholar

[155] View Article

[156] PubMed/NCBI

[157] Google Scholar

[ref49] 49. Hart EA, Lovvorn JR. Interpreting stable isotopes from macroinvertebrate foodwebs in saline wetlands. Limnol. Oceanogr. 2002;47(2): 580–584.
View Article
Google Scholar

[159] View Article

[160] Google Scholar

[ref50] 50. Parnell AC, Inger R, Bearhop S, Jackson AL. Source partitioning using stable isotopes: Coping with too much variation. PLoS ONE. 2010;5: e9672. pmid:20300637
View Article
PubMed/NCBI
Google Scholar

[162] View Article

[163] PubMed/NCBI

[164] Google Scholar

[ref51] 51. Pianka ER, Goodyear SE. Lizard responses to wildfire in arid interior Australia: long-term experimental data and commonalities with other studies. Austral Ecol. 2012;37: 1–11.
View Article
Google Scholar

[166] View Article

[167] Google Scholar

[ref52] 52. Tayasu I, Hyodo F, Abe T, Inoue T, Spain AV. Nitrogen and carbon stable isotope ratios in the sympatric Australian termites, Amiterms laurensis and Drepanotermes rubriceps (Isoptera: Termitidae) in relation to their feeding habits and the quality of their food materials. Soil Biol. And Biochem. 2002;34(3): 297–301.
View Article
Google Scholar

[169] View Article

[170] Google Scholar

[ref53] 53. Post DM. Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology. 2002;83: 703–718.
View Article
Google Scholar

[172] View Article

[173] Google Scholar

[ref54] 54. Jackson AL, Inger R, Parnell AC, Bearhop S. Comparing isotopic niche widths among and within communities: SIBER—Stable Isotope Bayesian Ellipses in R. Anim. Ecol. 2011;80: 595–602.
View Article
Google Scholar

[175] View Article

[176] Google Scholar

[ref55] 55. Newsome SD, Yeakel JD, Wheatley PV, Tinker MT. Tools for quantifying isotopic niche space and dietary variation at the individual population level. J. Mammal;93(2): 329–341.
View Article
Google Scholar

[178] View Article

[179] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Isotopic analysis

Ecological data

Data analysis

Results

Discussion

Stable isotopes and stomach content analysis

Stable isotopes and habitat use

Variation between sampling sites

Conclusion

Supporting information

S1 File. Habitat data from Rabosky et al. (2011) American Naturalist.

S2 File. Stable isotope and stomach content data used in the present study.

Acknowledgments

Author Contributions

References