Coexisting ant species differ in ability to meet their intrinsic nutritional needs

Coexisting ant species differ in ability to meet their intrinsic nutritional needs | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 28 January 2026 V1 Latest version Share on Coexisting ant species differ in ability to meet their intrinsic nutritional needs Authors : Hannah Riskas 0009-0003-7687-3553 [email protected] , Jonathan Shik , Lily Leahy 0000-0002-0733-6792 , Ian Wright , and Heloise Gibb 0000-0001-7194-0620 Authors Info & Affiliations https://doi.org/10.22541/au.176962156.69288030/v1 259 views 103 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Abstract 1. Coexisting species are often assumed to reduce competition by foraging for different foods with different nutrients, but few studies test whether nutritional partitioning reflects innate differences in nutritional needs. 2. We tested for overlap in field estimates of realised nutritional intake (tissue isotopes: δ¹⁵N; elemental ratios: C:N) among 17 co-occurring Australian ant species, then measured innate nutritional selectivity (protein:carbohydrate intake targets; P:C IT) for the same species using competition-free laboratory feeding experiments to infer whether extrinsic constraints prevent nutritionally optimized diets. 3. Free-ranging ants showed dietary and nutritional partitioning, with worker δ¹⁵N spanning 7.1‰ among species and C:N varying 1.33-fold, whereas lab-maintained ants exhibited overlapping P:C ITs (mean 1.00 ± 0.06 SE; range 0.60–1.42). Lab P:C ITs did not predict species-mean δ¹⁵N or C:N in the field (H1); pairwise coupling between lab P:C IT and field metrics was weak at best (~5% variance; H2), and residual/compensatory divergence was unsupported (H3), consistent with natural constraints (e.g., competitive monopolisation of carbohydrates, brood-mediated demand, risk-sensitive foraging) decoupling innate targets from realised intake. 4. Synthesis: By integrating field-based proxies of realised intake with controlled lab measures of innate macronutrient selectivity across a co-occurring ant assemblage, we show that strong divergence in realised diets can arise despite narrow, overlapping nutritional selectivity, advancing ecological understanding of how extrinsic constraints can generate nutritional niche differences in communities. Introduction Extrinsic factors, including competition, influence whether free-ranging organisms can acquire foods that enable them to meet their innate nutritional needs (Van Valen, 1965; Raubenheimer et al., 2009). Accordingly, coexisting species are assumed to have distinct, fundamental nutrient niches that enable them to thrive while foraging in distinct, realised nutrient niches (Gause, 1932; MacArthur & Levins, 1967; Tilman, 1982; Chase & Leibold 2003). While there are many examples of nutritional partitioning among free-ranging organisms (e.g., Raubenheimer et al., 2014; Lee, 2007), fewer studies have tested whether these measures of realised nutritional intake (realised nutritional niche) match innate nutritional selectivity (fundamental nutritional niche). Estimating dietary overlap amongst co-occurring species has been achieved through several methods: surveys of food item size and type across habitats (woodlands: Fellers 1987; deserts: Ryti & Case 1984; tropical canopies: Kaspari & Yanoviak 2001), or competition-manipulation experiments that reveal rapid diet shifts (Sanders & Gordon 2003; Gibb 2005). Nonetheless, these traditional approaches that primarily catalogue food items are insufficient for testing the basic assumption that competing species can take in optimal blends of nutrients relative to their intrinsic needs. Therefore, it is necessary to compare the realised nutrient niche with the fundamental nutritional niche. That niche reflects an organism’s physiological nutrient requirements, typically assessed under conditions where direct interspecific competition is experimentally excluded or naturally absent (Shik & Dussutour 2020; Sperfeld et al. 2016; Matich et al. 2021). Here, we integrate field and laboratory studies to test whether coexisting species can still meet their physiological needs while partitioning the available nutritional landscape. We focus on ants (Formicidae) because they dominate terrestrial ecosystems, often in large, high-density colonies, and exploit an exceptionally broad diet, gained via diverse activities such as herbivory, predation, scavenging and mutualistic honeydew harvesting (Hölldobler & Wilson 1990; Gibb & Cunningham 2011). We use stable isotope signature (δ¹⁵N) and body tissue carbon to nitrogen ratio (C:N) of ant workers as proxies for realised nutritional intake and use protein to carbohydrate intake targets (P:C IT) derived from a competition-free laboratory feeding experiment as a proxy for innate nutritional selectivity. We test these hypotheses using two measures. Whole-body elemental C:N ratios provide a snapshot of assimilated nutrients, whereas stable-isotope signatures (δ¹³C, δ¹⁵N) integrate diet over tissue-turnover times and allow simultaneous inference of trophic position (via δ¹⁵N) and primary carbon origin (via δ¹³C) (Fagan et al. 2002; Peterson & Fry 1987; Post 2002). δ¹⁵N increases by ≈ 3–4 ‰ with each trophic transfer; to estimate trophic position, the δ¹⁵N value of local primary producers or first-order consumers is used as a site-specific baseline against which higher-order consumers are positioned (Post 2002; Tillberg et al. 2006; Vander Zanden & Rasmussen 1999). Together, C:N and δ¹⁵N yield complementary realised nutritional intake signatures of free-ranging foragers in competitive environments. In animal tissues, lower C:N indicates nitrogen-rich (protein-biased) composition, whereas higher C:N indicates carbon-rich (lipid/carbohydrate-biased) composition. The Nutritional Geometric Framework (NGF) quantitatively analyses how animals regulate multiple nutrients to maintain homeostasis (Simpson & Raubenheimer, 2012; Raubenheimer et al., 2022). It typically uses competition-free, chemically defined laboratory diets to quantify an organism’s fundamental nutritional niche via its innate protein:carbohydrate intake target (P:C IT) (Simpson & Raubenheimer, 1993; Raubenheimer, Simpson & Mayntz, 2009), and it is equally suited to field studies estimating the nutritional composition of foraged items and realised intake in free-ranging individuals (e.g., Felton et al., 2009; Crumière et al., 2020; Raubenheimer, Simpson & Mayntz, 2009). Here we test whether lab-measured P:C IT of 17 coexisting Australian ant species predicts field-measured δ¹⁵N and C:N. Figure 2 provides descriptive context by visualising δ¹⁵N against the component elements, %C and %N. Limits to predictive power are then used to link competitive dynamics to nutritional shortfalls relative to intrinsic needs. Because resource quality and availability vary across space and time, realised intakes need not match evolved targets, producing nutritional mismatch when animals are pulled from their preferred balance (Raubenheimer & Simpson, 2016; Hou et al., 2021; Krabbe et al., 2019). We first summarise variation in realised field proxies (Fig. 2), then formally test three linked hypotheses (H1–H3) connecting innate macronutrient selectivity (P:C IT) to realised markers (δ¹⁵N, C:N) (Fig. 4). Because whole-body C:N is a ratio derived from %C and %N, Fig. 2 plots the elemental components separately for transparency, whereas hypothesis tests use C:N directly. H1 (absolute coupling) predicts species-mean coupling between P:C IT and trophic/stoichiometric proxies; accordingly, we fit two species-level regressions (P:C IT ~ δ¹⁵N; P:C IT ~ C:N), with expectations positive with δ¹⁵N and negative with C:N (full regression output in Supporting information (pairwise divergence) predicts covariation between pairwise divergence in P:C IT and divergence in realised markers, such that |ΔP:C IT| increases with |Δδ¹⁵N| and/or |ΔC:N| across species pairs. H3 (compensatory/residual divergence) tests whether |ΔP:C IT| predicts |Δresid δ¹⁵N| and/or |Δresid C:N| after accounting for H1, indicating compensatory partitioning beyond direct lab–field coupling; for interpretation, we additionally visualise signed mismatch between fundamental expectations and realised markers (fundamental − realised; positive = protein shortfall) as an exploratory diagnostic rather than a separate hypothesis. Study area Ants were sampled from the Nangak Tamboree Wildlife Sanctuary, a 28-hectare reserve at La Trobe University in Melbourne, southeastern Australia. The sanctuary has a woodland canopy consisting of revegetated and remnant river red gum (Eucalyptus camaldulensis) and a shrub-dominated understory, interspersed with wetlands and grasslands. We sampled 17 ant species across five subfamilies, with two independent source colonies per species. We collected from areas with sparse ground-layer vegetation and open, exposed ground, often beneath large rocks or fallen branches of river red gums. Per colony, we collected 30 foraging workers and transferred them to laboratory intake-target assays the same day (60 workers per species; 1,020 total). From the same colonies, additional workers were preserved at −20 °C for field-proxy analyses (stable isotopes, elemental stoichiometry); pooled sample size per colony was a median of 4 workers (range 1–15) to achieve ~1.5 mg dry tissue, yielding ~172 workers across all species (≈122–222 for 1.0–2.0 mg). We collected and observed ants from different nests and microhabitats throughout the sanctuary, recording activity between 08:00 and 19:00. Experimental set up For each species, we established two replicate worker cohorts under controlled lab conditions. Cohorts comprised only minor workers actively foraging outside the nest, selected haphazardly to represent typical behaviour. Each replicate came from a distinct colony and contained 30 workers. Immediately after collection, cohorts entered a constant-temperature room (25 °C; ~50–70% RH) with a 12 h light:12 h dark cycle; overhead fluorescents provided ~400–600 lux at bench height during the light phase. Each cohort occupied a 160 mm × 290 mm foraging arena with an egg carton and red cellophane shelter (Baker et al., 1985; Van Meer et al., 2002) (Figure 1). Inner walls were coated with Fluon (Dupont, Wilmington, DE, USA) to prevent escape, and cohort positions on shelves were randomised. Workers were starved for one day before six-day trials. Brood was excluded and only adult workers were assayed to estimate P:C intake targets (P:C IT) under competition-free conditions. Although larvae are the main protein consumers and can drive worker foraging via nutritional signalling (Cassill & Tschinkel, 1995; Creemers et al., 2003; Kawatsu, 2013), our aim was to quantify short-term foraging decisions of worker groups rather than colony growth. To minimise drift in worker nutrient state due to brood absence, we standardised timing: day 0 = field collection; day 1 = 24 h fast; days 2–7 = intake trials. Because all species experienced the same schedule and no interspecific competition, estimates are comparable across taxa; brood exclusion may conservatively bias protein demand downward, without affecting the relative P:C IT comparisons underlying our cross-species analyses. Diet preparation and presentation We formulated three nutritionally defined diets targeting protein:carbohydrate (P:C) ratios of ~3:1, ~1:1, and ~1:3 by mass (Dussutour & Simpson, 2012; Krabbe et al., 2019). Each was moulded into ~1 cm³ cubes and colour-coded solely as a researcher cue—P3 (high-protein; dark blue), CP (intermediate; green), and C3 (high-carbohydrate; light blue)—with subdued tones to minimise visual salience. Ants can detect green and, in some species, blue wavelengths (Yilmaz et al., 2017) but rely heavily on olfactory and gustatory cues; given identical recipes, colour is unlikely to have influenced diet selection. Diets were prepared from water, whole-egg protein, whey protein concentrate, calcium caseinate, sucrose, and agar (non-nutritive binder). Recipes were constructed from manufacturer nutrient specifications, with minor moisture and trace elements comprising the remainder (Supporting information). Macronutrient composition and realised P:C ratios for P3, CP, and C3 are summarised in Table 1; slight lipid differences reflect lipids inherent to protein sources and are negligible relative to the systematic variation in protein and carbohydrate across diets. We therefore retain the labels “3:1,” “1:1,” and “1:3” to denote target P:C proportions despite small deviations. Table 1. Macronutrient composition of the three experimental diets. Percentage composition reflects dry-mass values calculated from batch recipes and manufacturer nutrient specifications (Supporting information). Realised P:C ratios represent the final protein-to-carbohydrate proportions available to ants during assays. Diet codes correspond to the nominal protein:carbohydrate targets used throughout the study (P3 = high-protein; CP = intermediate; C3 = high-carbohydrate). P3 3:1 65.9 22.0 12.1 ~2.99:1 CP 1:1 44.9 43.9 11.2 ~1.02:1 C3 1:3 23.9 65.5 10.7 ~0.36:1 To determine the nutritional composition and standardise intake calculations, we computed total carbohydrate, protein, and lipid per batch from ingredient masses and manufacturer specifications (Supporting information). Each standardised batch yielded ~34 g dry macronutrient, from which multiple ~1 cm³ cubes were prepared. To express intake on a dry-mass basis and match the percentages above, we derived wet:dry conversion factors by weighing cube subsamples before and after drying, then applied these corrections to all cubes. Experimental procedure We ran a three-choice assay with the diet cubes placed simultaneously, each 100 mm from the shelter and from one another, and replaced them every 24 h; fully consumed cubes were promptly replaced to ensure ad lib availability. Intake was measured daily for six days. After each period, cubes were weighed wet, dried 48 h at 60 °C, and re-weighed to obtain dry mass. Behavioural observations were conducted twice daily—0800 and 1700 h—sampling morning and late-afternoon activity (not full-day coverage). Each ~30-min session recorded, for each cohort and cube: (i) workers actively feeding on the cube surface; (ii) workers transporting diet fragments; and (iii) the fate of transported fragments. Transport was scored as “consumed/harvested” when diet was eaten at the cube or carried to the egg-carton shelter and retained, and as “scattered” when fragments were dropped away from the cube (see Diet classification). These counts quantify scattering, hoarding (retention under shelter), and direct feeding for each diet type. Figure 1. Experimental Setup for Ant Feeding Trials with Protein Ratios. Colony fragments of 30 workers were simultaneously offered three diet cubes with distinct protein:carbohydrate (P:C) ratios (3:1, dark blue; 1:1, green; 1:3, light blue), each in a separate petri dish. Diets were placed equidistantly within a foraging arena containing an egg-carton shelter under red cellophane. The macronutrient composition (g) for each diet type, as illustrated, is detailed in Supporting information. Trait measurements Body mass We quantified worker body mass so that we could scale food intake to both per-capita and mass-specific units, and test whether species with larger workers consume more food per day and differ in mass-specific intake rates. For each colony, we measured the wet and dry mass of 10 individual workers by weighing each ant separately, drying them for 48 h at 60 °C, and then re-weighing. We used the mean dry mass of these 10 workers as our colony-level estimate of worker body mass, which was then averaged to the species level for analyses of intake–mass scaling and for visualising body-size differences among species. Mass measurements were taken using a Mettler Toledo XS3DU microbalance with a readability of 0.1 µg, ensuring precise quantification of individual body masses. δ¹⁵N trophic position and C:N stoichiometry We quantified worker δ¹⁵N and C:N as proxies for realised trophic position and whole-body stoichiometry to test H2–H3 (Tillberg et al., 2006; Feldhaar et al., 2010; Tiunov, 2007). Workers were frozen at –18 °C and oven-dried at 50 °C for 48 h. For each feeding-experiment colony, pooled dried workers provided sufficient mass per replicate. Samples (≈1–2 mg; weighed to 3 µg precision) were analysed for %C, %N, and δ¹⁵N using an EA–IRMS (Isoprime; coupled to a Carlo Erba CE1100) at the ANU Research School of Biology Isotope Facility. Instrument calibration and drift corrections used laboratory standards (cane sugar, beet sugar, cysteine, glycine) spanning 3.06–7.8‰ δ¹⁵N; in-house software (SecondRat) assessed performance and normalised values. δ¹⁵N is reported in ‰ relative to AIR. The same standards, across a range of masses, informed %C and %N slope/offset corrections. This follows established EA–IRMS protocols for ants and other soil animals, and for inferring trophic structure and stoichiometry from δ¹⁵N and C:N (Tillberg et al., 2006; Menke et al., 2010; Roeder & Kaspari, 2017; Feldhaar et al., 2010; Tiunov, 2007). Data analysis P:C IT analysis We used a bivariate nutritional geometry approach to examine and compare protein:carbohydrate intake patterns across the 17 ant species from five subfamilies. For each colony replicate, we first calculated daily intake of carbohydrate and protein over the six-day feeding period. On each day, intake for a given nutrient (mg dry mass·worker⁻¹·day⁻¹) was obtained by converting diet disappearance to nutrient mass using the analysed composition of each diet and dividing by the number of workers alive at the end of that day. We then averaged these daily per-capita intakes across the six days to obtain mean protein and carbohydrate intake rates per worker for each colony. The cumulative P:C intake target (P:C IT) for each colony was defined as the ratio of mean protein intake to mean carbohydrate intake across the six days. Full details of diet mass corrections and survivorship adjustments, including cohort and worker counts, are provided in Supporting information. Statistical modelling Trait values were averaged at the species level (n = 17), using colony means as replicates for intake and isotope/stoichiometry traits (typically two colonies per species; see Experimental design). Analyses used R 4.4.2 (R Core Team, stats). We assessed significance at α = 0.05 and checked assumptions via residual plots and Shapiro–Wilk tests. Phylogenetic signal (phytools::phylosig, Pagel’s λ; Pagel, 1994; Revell, 2025) occurred for δ¹⁵N (λ = 0.853, p = 0.010) but not C:N (λ = 0.108, p = 0.785) or P:C IT (λ = 0.215, p = 0.503) (Supporting information). For a single, comparable framework—and because P:C IT lacked signal—we present non-phylogenetic species-mean models in the main text. Phylogenetic-signal estimates (Supporting information) and phylogenetically informed distance-based analyses (MRM) of |ΔP:C IT| against |Δδ¹⁵N| and |ΔC:N|, with and without phylogenetic distance, are in Supporting information. Thus, main predictions use uncorrected species means; robustness to phylogenetic non-independence follows those checks. “Absolute coupling” denotes associations between species-mean P:C IT and realised proxies (δ¹⁵N, C:N) on raw means, without conditioning on other variables. Pairwise distance analyses additionally included phylogenetic distances as covariates via MRM (see Supporting information). We first tested whether higher trophic position (δ¹⁵N) or lower body C:N predicts more protein-biased P:C IT via two OLS models: P:C IT ~ δ¹⁵N and P:C IT ~ C:N. For each regression we report slope (± s.e.), p-value, and R². We focused on δ¹⁵N and C:N because they provide complementary proxies for realised trophic position and body nutrient allocation, not algebraic re-expressions of laboratory P:C ratios. We next asked whether pairwise divergence in fundamental macronutrient strategy covaries with divergence in realised markers. For the 136 species pairs, we computed |ΔP:C IT|, |Δδ¹⁵N|, and |ΔC:N| and analysed the distance matrices using multiple regression on distance matrices (ecodist::MRM; 9,999 row–column permutations; no rank transformation), which accounts for non-independence of shared species (Smouse et al., 1986; Goslee & Urban, 2007). For each realised metric, we fitted an MRM with |ΔP:C IT| as the response and |Δδ¹⁵N| or |ΔC:N| as the predictor (with phylogenetic distance added as a covariate in robustness checks; see Supporting information). Models were compared using AICc; we interpreted ΔAICc as ≈0–2 (essentially equivalent), ≈2–7 (some support), >10 (little support) (Symonds & Moussalli, 2011; Burnham, Anderson & Huyvaert, 2011). Permutation pseudo-R² quantified variance explained. Because δ¹⁵N and C:N are moderately correlated (Spearman ρ = 0.50, p = 0.045), we evaluated trophic and stoichiometric divergences in separate MRMs to avoid collinearity. To assess compensatory partitioning (see Introduction)—i.e., residual divergence in realised diets after removing any direct P:C IT–field proxy link—we fit δ¹⁵N ~ P:C IT and C:N ~ P:C IT at the species level, extracted residuals (variation not explained by P:C IT), and tested whether pairwise divergence in these residuals increased with |ΔP:C IT|. We computed |Δ resid δ¹⁵N| and |Δ resid C:N| and compared them with the |Δ P:C IT| matrix. Associations were evaluated three ways: (1) Spearman correlations on vectorised upper-triangular entries with permutation-based inference; (2) Mantel tests on full matrices (vegan::mantel, Pearson r, 9,999 permutations; Oksanen et al., 2001; Mantel, 1967); and (3) MRM (ecodist::MRM; raw distances; 9,999 permutations) regressing |Δ resid δ¹⁵N| or |Δ resid C:N| on |Δ P:C IT|. We report Pearson r for Mantel tests because Mantel correlations are computed as correlations between distance matrices (Pearson is the default in vegan::mantel), whereas Spearman correlations on vectorised upper-triangular entries provide a rank-based robustness check; all p-values are permutation-based. Primary analyses omitted an explicit phylogenetic model for interpretability; robustness was checked by adding cophenetic phylogenetic distances in partial Mantel tests and MRMs. Full phylogenetic-signal results (Pagel’s λ) appear in Supporting information. Results Estimates of nutritional intake in the field (field proxies: C:N, δ¹⁵N) Across the 17 species, trophic position (δ¹⁵N) spanned almost a fourfold range from 2.40 to 9.49 ‰ (mean ± SD = 6.60 ± 2.13 ‰), whereas worker body C:N varied more modestly, by ~1.3-fold from 3.48 to 4.63 (mean ± SD = 3.95 ± 0.31). This indicates that the assemblage includes species occupying relatively low to relatively high trophic positions and differing detectably, though less dramatically, in body carbon–nitrogen balance. Subfamilies differed in δ¹⁵N (one-way ANOVA F₄,₁₂ ≈ 4.95, p ≈ 0.014), with Dolichoderinae and Ectatomminae showing higher mean δ¹⁵N than Formicinae and Myrmicinae, and Myrmeciinae intermediate. Body C:N ratios varied more modestly and showed weaker among-subfamily differences (F₄,₁₂ ≈ 1.74, p ≈ 0.21), with broadly overlapping ranges. Across species, δ¹⁵N covaried with elemental composition in a panel-specific way (%C and %N; Fig. 2): δ¹⁵N increased with body %C (OLS: β = 0.3699 ± 0.1647, F₁,₁₅ = 5.041, R² = 0.2515, p = 0.0403, n = 17) but showed no relationship with body %N (OLS: β = 0.0842 ± 0.6759, F₁,₁₅ = 0.0155, R² = 0.0010, p = 0.9025, n = 17), indicating modest coupling with %C but not %N despite the substantial spread in δ¹⁵N. Full regression statistics for all OLS models are provided in Supporting information. These relationships are descriptive context for the field proxies; hypothesis tests linking innate selectivity to realised proxies are presented below. Figure 2. Worker trophic position (δ¹⁵N) plotted against body carbon and nitrogen content across 17 ant species . Each point is a species (n = 17), with colours indicating subfamily. Lines show ordinary least-squares fits with 95% confidence intervals. (a) Species‐mean body carbon content (%C) versus trophic position (δ¹⁵N). (b) Species‐mean body nitrogen content (%N) versus trophic position (δ¹⁵N). Panel (a) shows a modest positive association between δ¹⁵N and %C, whereas panel (b) shows no association between δ¹⁵N and %N. These relationships are provided as descriptive context only (not a test of H1–H3); hypothesis tests are in Fig. 4, and full OLS statistics are reported in Supporting information. Innate nutritional selectivity (P:C IT; lab) To provide the laboratory baseline for our tests of H1–H3, which evaluate how protein:carbohydrate intake targets (P:C IT) relate to realised trophic (δ¹⁵N) and stoichiometric (C:N) proxies, we first quantified how strongly species differ in intake and how precisely those means are estimated. Total daily intake per worker varied by about fivefold across species, from 0.005 mg worker⁻¹ day⁻¹ in Crematogaster laeviceps (Myrmicinae) to 0.024 mg worker⁻¹ day⁻¹ in Myrmecia piliventris (Myrmeciinae). On a mass-specific basis, intake ranged from 0.0011 to 0.1254 mg diet mg⁻¹ worker day⁻¹ (118.78-fold) and declined strongly with worker body mass (log–log OLS: β = −0.9047 ± 0.0641, F₁,₁₅ = 199.2, R² = 0.93, p = 4.57 × 10⁻¹⁰, n = 17). Within-species standard errors were small (< 0.001–0.002 mg mg worker⁻¹ day⁻¹; largest in Iridomyrmex splendens, smallest in C. laeviceps), indicating that the species-mean intake and P:C IT estimates used in the coupling analyses are precise. Protein:carbohydrate intake targets (P:C IT) varied approximately 2.4-fold among species, from 0.60 in Camponotus claripes gp. A (Formicinae) to 1.42 in Iridomyrmex septentrionalis (Dolichoderinae), the two extremes highlighted with red arrows in Figure 3. Thus, although the diet cubes presented a nominal nine-fold range in P:C ratios (from narrower fundamental P:C intake target (Fig. 3). Figure 3. Protein–carbohydrate intake targets across ant species (linear axes). Mean protein and carbohydrate intake per worker (mg) for 17 species, with ±1 SE shown as horizontal (protein) and vertical (carbohydrate) error bars. Lines from the origin indicate diet ratios (1:3, 1:1, 3:1), and the panel uses equal x–y scales so slope equals the P:C ratio. Shaded areas denote carbohydrate bias (above 1:1; pink) and protein bias (below 1:1; blue), with the 1:1 line as the bias threshold. Two-letter species codes are coloured by subfamily. Red arrows mark the species with the lowest and highest P:C intake targets. Linking innate nutritional selectivity to realised nutritional intake (H1 → H2 → H3) We next asked whether innate nutritional selectivity predicts realised nutritional intake in the field, using a three-step sequence of tests. First (H1), species‐mean P:C IT showed no relationship with either δ¹⁵N or C:N (Fig. 4a,b; δ¹⁵N: β = 0.040 ± 0.026, p = 0.144, R² = 0.137; C:N: β = 0.163 ± 0.190, p = 0.406, R² = 0.046), indicating that—at the species level—innate nutritional needs did not predict realised nutritional signatures. Second (H2), we tested whether species pairs that differed more in P:C IT also differed more in δ¹⁵N or C:N across 136 species pairs (Fig. 4c,d). Using multiple regression on distance matrices (MRM; 9,999 permutations), we regressed pairwise divergence in P:C IT on divergence in realised proxies and found weak, non-significant associations (|ΔP:C IT| ~ |Δδ¹⁵N| + phylogeny: pseudo-R² = 0.050, p = 0.097; |ΔP:C IT| ~ |ΔC:N| + phylogeny: pseudo-R² = 0.015, p = 0.507; results were similar without phylogeny; Supporting information). Mantel tests showed a shallow correlation for |Δδ¹⁵N| (r = 0.223, p = 0.0420; partial Mantel controlling phylogeny: r = 0.217, p = 0.0499) and none for |ΔC:N| (r = −0.108, p = 0.781), consistent with at most a weak link between divergence in trophic position and divergence in P:C IT (Supporting information). Third (H3), residual or compensatory patterns were unsupported: MRMs regressing |Δ residual δ¹⁵N| or |Δ residual C:N| on |ΔP:C IT| (with phylogeny) were non‐significant (pseudo-R² = 0.0167, p = 0.347; pseudo-R² = 0.0147, p = 0.500; Supporting information). Overall, we therefore interpret the coupling between innate and realised nutrition as weak, and use a protein‐dimension mismatch synthesis (Supporting information) to visualise where constraints may be strongest. Figure 4. Innate macronutrient targets show weak links to realised trophic signatures at species and pairwise levels. (a–b) Species means (n = 17): P:C intake target (P:C IT) versus δ¹⁵N and versus C:N. Points are open circles with ordinary least‐squares (OLS) lines and 95% confidence ribbons. Axes use observed ranges with padding. (c–d) Species pairs (n = 136): |ΔP:C IT| versus |Δδ¹⁵N| and |ΔC:N|, with OLS lines. Points are shaded by phylogenetic distance (lighter = closer relatives). These patterns mirror the main analyses: |Δδ¹⁵N| shows a shallow association with |ΔP:C IT|, whereas |ΔC:N| does not. Vertical gridlines are axis guides only and do not indicate thresholds. Panels with non‐significant slopes are rendered with dotted trend lines (no directional inference). Discussion In ant communities, laboratory P:C intake targets provide a proxy for the fundamental macronutrient balance that colonies regulate under idealised conditions, whereas δ¹⁵N and whole-body C:N reflect realised nutrient assimilation in the field. Across the 17 co-occurring species studied here, these fundamental and realised proxies aligned only weakly. The stronger association between δ¹⁵N and %C than %N may reflect greater among-species variation in carbon-rich reserves or structural carbon investment, whereas bulk nitrogen content may be more constrained. Three key results emerged from our tests of the expected links between innate macronutrient selectivity and field-based trophic and stoichiometric signatures. First, neither realised nutrient proxy (δ¹⁵N or C:N) predicted P:C IT (Fig. 4a–b). Second, pairwise divergence analyses (H2) indicated that species farther apart in realised trophic position (δ¹⁵N) also tended to differ more in P:C intake targets (Fig. 4c–d). This effect was modest, explaining only about 5% of the variance in P:C IT distance. Consistent with this weak direct coupling, this association was not retained after removing direct coupling in residual tests (H3; Supporting information). Finally, C:N divergence also explained little of the variation in P:C IT divergence, which is consistent with our finding that δ¹⁵N showed phylogenetic signal whereas P:C IT and C:N did not; within this dataset, macronutrient regulation appears comparatively evolutionarily labile. Linking fundamental and realised nutritional niches We predicted that the fundamental and realised nutritional proxies would be coupled (H1) and that any relationship would persist across species pairs (H2). Linear models showed that trophic position (δ¹⁵N) did not predict the intake target (P:C IT), rejecting H1. Pairwise divergence tests (H2) uncovered a weak positive relationship between the absolute difference in trophic position (|Δδ¹⁵N|) and the absolute difference in P:C IT (|ΔP:C IT|) across species pairs. This association was modest, explaining approximately 5% of the variance in |ΔP:C IT|. In contrast to this direct, albeit weak, coupling, our test for compensatory nutritional-niche partitioning (H3), which examined whether divergence in fundamental P:C IT predicted the residual pairwise divergence in realised niches, found no significant relationship. These results suggest a nuanced picture of nutritional-niche partitioning. The weak positive trend under H2 indicates that species with more divergent realised trophic positions also show somewhat more different fundamental nutrient targets, offering limited support for the idea that nutrient-strategy differentiation contributes to coexistence (Dussutour & Simpson, 2009; Cook et al., 2010; Kaspari et al., 2012). By contrast, the lack of support for H1 (species-mean coupling) and especially H3 (compensatory divergence) implies that strong coupling or clear compensatory shifts between fundamental P:C IT and realised δ¹⁵N are not prominent in this assemblage. For example, Rhytidoponera tasmaniensis —high in trophic position—selected a carbohydrate-rich blend similar to low-trophic Camponotus claripes (this study; Supporting information), showing that higher trophic position need not entail a protein-rich diet. Conversely, Iridomyrmex septentrionalis —also high in trophic position—favoured one of the most protein-rich blends, and its large gap from C. claripes aligns with the modest H2 trend. Field and lab observations also indicate flexible sugar use: in our assays , R. tasmaniensis shifted toward carbohydrate-rich diets when competition and larvae were removed; predatory ants similarly switch to nectar under intensified competition (Blüthgen & Fiedler, 2004; Kaspari et al., 2012). Isotopes integrate long-term realised diet, whereas intake assays capture acute regulation under controlled conditions without competition or brood demand. Together they provide a coherent picture, but whether these strategies translate into clear nutritional-niche partitioning that facilitates coexistence in this community remains uncertain given the mixed hypothesis support (Feldhaar et al., 2010). This context-dependent sugar use is consistent with ideas that carbohydrate intake fuels activity and aggression, shaping competitive dynamics (Grover et al., 2007). Body C:N stoichiometry in adult workers showed no meaningful association with the P:C intake target (P:C IT), either across species means nor in pairwise divergence analyses, indicating that elemental composition and macronutrient regulation describe different aspects of nutritional ecology (Fig. 4b,d). There is the possibility that using brood-free colony fragments reduced protein demand relative to a full reproductive cycle (Ulrich et al. 2016), which could conceivably decouple short-term P:C IT from whole-body C:N; however, this effect is likely minimal given that assays were conducted over a brief 6-day period designed to limit brood-absence artefacts. Experimental manipulation of colony energy balance in Tetramorium caespitum demonstrated that worker and larval C:N ratios vary primarily with sucrose availability rather than protein intake, implicating storage-pool dynamics instead of acute diet as the proximate driver (Kay, Rostampour & Sterner 2006). Taken together with our results, this experimental evidence suggests that whole-body C:N reflects longer-term storage and task ecology, whereas P:C IT captures short-term regulatory targets—helping to explain their weak coupling in this assemblage. Linking ecological traits to nutritional niches Most species pairs followed the weak positive association described above: larger differences in δ¹⁵N (or C:N) tended to coincide with larger differences in P:C intake targets. A minority deviated: four pairs differed by >1 trophic level (>3‰ δ¹⁵N) yet had nearly identical P:C ratios (<0.10), whereas eleven pairs lay within 1‰ in δ¹⁵N but diverged by 0.30 in P:C IT (Supporting information). These exceptions suggest trophic position imposes only broad constraints on macronutrient strategy: colonies can reach similar nutrient balances via different resource mixes, or adopt contrasting targets despite similar trophic signatures, highlighting flexible foraging as an added niche axis. Within this species set, worker body mass and mass density were not strongly associated with P:C IT, δ¹⁵N, or C:N (|r| ≤ 0.37, n = 17, all p ≥ 0.16), and pairwise density differences did not distinguish “decoupled” pairs. Nevertheless, structural and colony-level traits—including body size, mass density, venom system, colony age, and brood demand—remain plausible contributors. Phylogenetic signal in P:C IT and C:N was weak in this small set, and sample size limits multi-trait models; resolving which traits predict decoupling will require a larger comparative dataset. Other niche axes can reduce direct competition among species converging on similar P:C targets. Temporal and micro-habitat partitioning in the Nangak Tamboree Wildlife Sanctuary (Buxton et al., 2021) and other woodlands (MacArthur-Waltz et al., 2021) add axes—thermal conditions and plant-associated foods—that separate realised diets in space and time. These processes may help explain coexistence among species with similar P:C targets in our study, although we did not quantify recruitment or interference. Many species regulate intake to a narrow P:C IT range, yet fundamental niches are not identical; convergence reflects shared physiological constraints to balance amino acids and energy within tight limits (Simpson & Raubenheimer, 2012), not identical diets. Routes to that balance differ: myrmeciines often obtain protein via active predation (e.g. Myrmecia croslandi ; Jayatilaka et al., 2014), whereas several Camponotus favour carbohydrate-rich plant exudates (Cook & Davidson, 2005; Feldhaar et al., 2007) and typically scavenge for protein (Mrowka, 2007). Both groups nonetheless reach similar intake targets under distinct foraging regimes and specialisations (Blüthgen et al., 2003; Vaes & Detrain, 2022). Diversity in morphology, behaviour, and activity windows allows co-existing ants to minimise competition by differentiating along multiple axes (Albrecht & Gotelli, 2001; Cerdá & Retana, 1997). Accordingly, no single trait predicts nutrient intake in isolation; colony demands and environmental context interact to shape nutritional strategy (Dussutour & Simpson, 2009; Csata & Dussutour, 2019). While our results illuminate colony-level nutritional strategies, they rest on methodological constraints. Brood-free fragments likely underestimate total protein demand because larvae are strong protein sinks (Ulrich et al., 2016), so our assays probably capture conservative (lower-bound) protein requirements. Even so, short-term, adult-only feeding assays are a standard, ecologically meaningful way to probe workers’ fundamental macronutrient niche under controlled conditions (nutritional geometry; Dussutour & Simpson, 2009; Krabbe et al., 2019). Many colonies naturally experience extended broodless periods: north Florida Odontomachus brunneus lack larvae for ≈6 months annually (Hart & Tschinkel, 2012), and queens of several temperate Formicinae and Myrmicinae cease oviposition during winter diapause, leaving colonies larva-free until spring (Kipyatkov, 2001). Our design therefore quantifies short-term worker intake rules under brood-free conditions, not lifetime colony protein balance. That focus aligns with using worker body composition (C:N, density) as long-term integrators of diet and allocation, and implies any mismatch between innate targets and realised trophic or stoichiometric signatures is likely conservative. In supplementary analyses, we visualised a protein-dimension “mismatch” as the difference between each species’ innate protein bias inferred from its P:C intake target and its realised protein signature inferred from δ¹⁵N and body C:N . In social insects, mismatch is further shaped by carbohydrate competition and mortality risk, which alter resource access and state-dependent foraging (Grover et al., 2007; Wittman et al., 2018; Barbee & Pinter-Wollman, 2023). We interpret mismatch as a descriptive constraint indicator: when colonies cannot achieve their defended target in the field, realised protein falls below or above target. The sign denotes direction (protein shortfall vs excess); the magnitude reflects the strength of decoupling between fundamental selectivity and realised trophic signatures. Mechanisms consistent with this interpretation include (i) “rules of compromise,” over-harvesting imbalanced foods to approximate targets under restricted options (Krabbe et al., 2019); (ii) communal regulation linking protein demand to brood, altering processing and redistribution (Dussutour & Simpson, 2009); (iii) competitive monopolisation of carbohydrate-rich resources, with dominants (and mutualists) controlling sugars and forcing subordinates onto protein-biased diets (Kay et al., 2010; Wittman et al., 2018); and (iv) risk-sensitive foraging, where mortality-risk cues shift choices from those made without risk (Barbee & Pinter-Wollman, 2023). Accordingly, both positive and negative mismatch values should be read conservatively as signals that extrinsic factors—competition, mutualisms affecting carbohydrate access, temporal variability, and risk—can decouple fundamental selectivity from realised intake, even when intrinsic targets are narrow. Conclusions Using δ¹⁵N (trophic position) and whole-body C:N as long-term field proxies for assimilated intake, and P:C intake targets (P:C IT) from competition-free assays as a proxy for innate selectivity, we asked how closely free-ranging ants meet innate needs. Across 17 species, realised diets (δ¹⁵N, C:N) varied widely, yet most converged on a narrow range of P:C ITs. This divergence shows realised trophic and stoichiometric niches are only weakly coupled to innate macronutrient targets. Colonies persist, implying either δ¹⁵N and C:N capture only part of the nutrient environment, or that colonies sustain high fitness across broader realised blends than short-term intake targets imply. Species means showed P:C IT unrelated to δ¹⁵N or C:N. P:C IT and C:N showed no phylogenetic signal, whereas δ¹⁵N did. Including phylogenetic distances did not improve fit, with no evidence of compensatory divergence between innate P:C IT and residual realised-diet variation. Instead, colonies can reach similar P:C ITs via different foraging routes, implying high flexibility in nutrient regulation. This flexibility, plus partitioning along unmeasured dimensions, may maintain coexistence despite overlapping preferences. Our protein-mismatch analysis indicates which species match innate protein targets, suggesting competition and extrinsic constraints can limit access to preferred nutrients. These strategies influence predation, mutualisms and nutrient cycling in ant-dominated communities. Tracking colony ontogeny and foraging will clarify how nutritional ecology mediates competition and community structure in social insects. References Albrecht, M., & Gotelli, N. J. (2001). Spatial and temporal niche partitioning in grassland ants. Oecologia , 126(1), 134–141. https://doi.org/10.1007/s004420000497 Baker, T. C., Van Vorhis Key, S. E., & Gaston, L. K. (1985). Bait-preference tests for the Argentine ant (Hymenoptera: Formicidae). Journal of Economic Entomology, 78 (5), 1083–1088. https://doi.org/10.1093/jee/78.5.1083 Barbee, B., & Pinter-Wollman, N. (2023). Nutritional needs and mortality risk combine to shape foraging decisions in ants. Current Zoology, 69 (6), 747–755. https://doi.org/10.1093/cz/zoac089 Blüthgen, N., & Fiedler, K. (2004). Competition for composition: Lessons from nectar-feeding ant communities. Ecology, 85 (6), 1479–1485. https://doi.org/10.1890/03-0430 Blüthgen, N., Gebauer, G., & Fiedler, K. (2003). Disentangling a rainforest food web using stable isotopes: Dietary diversity in a species-rich ant community. Oecologia , 137(3), 426–435. https://doi.org/10.1007/s00442-003-1347-8 Burnham, K. P., Anderson, D. R., & Huyvaert, K. P. (2011). AIC model selection and multimodel inference in behavioral ecology: Some background, observations, and comparisons. Behavioral Ecology and Sociobiology, 65 (1), 23–35. https://doi.org/10.1007/s00265-010-1029-6 Buxton, J. T., Robert, K. A., Marshall, A. T., Dutka, T. L., & Gibb, H. (2021). A cross-species test of the function of cuticular traits in ants (Hymenoptera: Formicidae). Myrmecological News, 31 . https://doi.org/10.25849/myrmecol.news_031:031 Cassill, D. L., & Tschinkel, W. R. (1995). Allocation of liquid food to larvae via trophallaxis in colonies of the fire ant, Solenopsis invicta . Animal Behaviour, 50 (3), 801–813. https://doi.org/10.1016/0003-3472(95)80140-5 Cerdá, X., & Retana, J. (1997). Links between worker thermal tolerance, thermal preference and activity in Mediterranean ant communities. Journal of Animal Ecology , Chase, J. M., & Leibold, M. A. (2003). Ecological niches: Linking classical and contemporary approaches . University of Chicago Press. Cook, S. C., & Davidson, D. W. (2005). Nutritional and functional biology of exudate-feeding ants. Entomologia Experimentalis et Applicata, 118 (1), 1–10. https://doi.org/10.1111/j.1570-7458.2006.00374.x Cook, S. C., Eubanks, M. D., Gold, R. E., & Behmer, S. T. (2010). Colony-level macronutrient regulation in ants: Mechanisms, hoarding and associated costs. Animal Behaviour, 79 (2), 429–437. https://doi.org/10.1016/j.anbehav.2009.11.022 Creemers, B., Billen, J., & Gobin, B. (2003). Larval begging behaviour in the ant Myrmica rubra . Ethology Ecology & Evolution, 15 (3), 261–272. https://doi.org/10.1080/08927014.2003.9522671 Crumière, A. J. J., Stephenson, C. J., Nagel, M., & Shik, J. Z. (2020). Using nutritional geometry to explore how social insects navigate nutritional landscapes. Insects , 11(1), 53. https://doi.org/10.3390/insects11010053 PubMed Csata, E., & Dussutour, A. (2019). Nutritional regulation in ants (Hymenoptera: Formicidae): An integrative review. Myrmecological News , 29, 111–124. https://doi.org/10.25849/myrmecol.news_029:111 Dussutour, A., & Simpson, S. J. (2009). Communal nutrition in ants. Current Biology, 19 (9), 740–744. Dussutour, A., & Simpson, S. J. (2012). Ant workers die young and colonies collapse when fed a high-protein diet. Proceedings of the Royal Society B: Biological Sciences, 279 (1737), 2402–2408. https://doi.org/10.1098/rspb.2012.0051 Fagan, W. F., Siemann, E., Mitter, C., Denno, R. F., Huberty, A. F., Woods, H. A., & Elser, J. J. (2002). Nitrogen in insects: Implications for trophic complexity and species diversification. The American Naturalist, 160 (6), 784–802. https://doi.org/10.1086/343879 Feldhaar, H., Gebauer, G., & Blüthgen, N. (2010). Stable isotopes: Past and future in exposing secrets of ant nutrition (Hymenoptera: Formicidae). Myrmecological News, 13 , 3–13. https://doi.org/10.25849/myrmecol.news_013:003 Feldhaar, H., Straka, J., Krischke, M., Berthold, K., Stoll, S., Mueller, M. J., … Gross, R. (2007). Nutritional upgrading for omnivorous carpenter ants by the endosymbiont Blochmannia . BMC Biology, 5 (1). https://doi.org/10.1186/1741-7007-5-48 Fellers, J. H. (1987). Interference and exploitation in a guild of woodland ants. Ecology, 68 (5), 1466–1478. https://doi.org/10.2307/1939230 Felton, A. M., Felton, A., Raubenheimer, D., Simpson, S. J., Foley, W. J., Wood, J. T., Wallis, I. R., & Lindenmayer, D. B. (2009). Nutritional goals of wild primates. Functional Ecology , 23(1), 70–78. https://doi.org/10.1111/j.1365-2435.2008.01431.x Gause, G. F. (1932). Ecology of populations. The Quarterly Review of Biology, 7 (1), 27–46. Gibb, H. (2005). The effect of a dominant ant, Iridomyrmex purpureus , on resource use by ant assemblages depends on microhabitat and resource type. Austral Ecology , 30(8), 856–867. https://doi.org/10.1111/j.1442-9993.2005.01533.x Gibb, H., & Cunningham, S. A. (2011). Habitat contrasts reveal a shift in the trophic position of ant assemblages. Journal of Animal Ecology, 80 (1), 119–127. https://doi.org/10.1111/j.1365-2656.2010.01747.x Goslee, S. C., & Urban, D. L. (2007). The ecodist package for dissimilarity-based analysis of ecological data. Journal of Statistical Software, 22 , 1–19. https://doi.org/10.18637/jss.v022.i07 Grover, C. D., Kay, A. D., Monson, J. A., Marsh, T. R., & Holway, D. A. (2007). Linking nutrition and behavioural dominance: Carbohydrate scarcity limits aggression and activity in Argentine ants. Proceedings of the Royal Society B: Biological Sciences, 274 (1628), 2951–2957. https://doi.org/10.1098/rspb.2007.1065 Hart, L. M., & Tschinkel, W. R. (2012). A seasonal natural history of Odontomachus brunneus (Hymenoptera: Formicidae) in north Florida. Insectes Sociaux , 59(1), 45–54. https://doi.org/10.1007/s00040-011-0187-0 Hölldobler, B., & Wilson, E. O. (1990). The ants . Harvard University Press. Hou, R., Chapman, C. A., Rothman, J. M., Zhang, H., Huang, K., Guo, S., … Raubenheimer, D. (2021). The geometry of resource constraint: An empirical study of the golden snub-nosed monkey. Journal of Animal Ecology, 90 (3), 751–765. https://doi.org/10.1111/1365-2656.13408 Jayatilaka, P., Raderschall, C. A., Narendra, A., & Zeil, J. (2014). Individual foraging patterns of the jack jumper ant Myrmecia croslandi (Hymenoptera: Formicidae). Myrmecological News, 19 , 75–83. Kaspari, M., & Yanoviak, S. P. (2001). Bait use in tropical litter and canopy ants—evidence of differences in nutrient limitation. Biotropica, 33 (1), 207–211. https://doi.org/10.1646/0006-3606(2001)033[0207:BUITLA]2.0.CO;2 Kaspari, M., Donoso, D., Lucas, J. A., Zumbusch, T., & Kay, A. D. (2012). Using nutritional ecology to predict community structure: A field test in Neotropical ants. Ecosphere, 3 (11), 1–15. https://doi.org/10.1890/ES12-00136.1 Kawatsu, K., & Matsuura, K. (2013). Preadaptation for parthenogenetic colony foundation in subterranean termites Reticulitermes spp. (Isoptera: Rhinotermitidae). Journal of Ethology, 31 , 123–128. https://doi.org/10.1007/s10164-012-0356-7 Kay, A. D., Rostampour, S., & Sterner, R. W. (2006). Ant stoichiometry: Elemental homeostasis in stage-structured colonies. Functional Ecology, 20 (6), 1037–1044. https://doi.org/10.1111/j.1365-2435.2006.01187.x Kay, A. D., Zumbusch, T., Heinen, J. L., Marsh, T. C., & Holway, D. A. (2010). Nutrition and interference competition have interactive effects on the behavior and performance of Argentine ants. Ecology, 91 (1), 57–64. https://doi.org/10.1890/09-0908.1 Kipyatkov, V. E. (2001). Seasonal life cycles and the forms of dormancy in ants (Hymenoptera: Formicidae). In V. E. Kipyatkov (Ed.), Seasonal development of insects with diapause and quiescence (pp. 211–238). St. Petersburg University Press. Krabbe, B. A., Arnan, X., Lannes, P., Bergstedt, C. E., Larsen, R. S., Pedersen, J. S., … Shik, J. Z. (2019). Using nutritional geometry to define the fundamental macronutrient niche of the widespread invasive ant Monomorium pharaonis . PLoS ONE, 14 (6), e0218764. https://doi.org/10.1371/journal.pone.0218764 Lee, K. P. (2007). The interactive effects of protein quality and macronutrient imbalance on nutrient balancing in an insect herbivore. Journal of Experimental Biology, 210 (18), 3236–3244. https://doi.org/10.1242/jeb.008060 MacArthur, R., & Levins, R. (1967). The limiting similarity, convergence, and divergence of coexisting species. The American Naturalist, 101 (921), 377–385. MacArthur-Waltz, D. J., Nelson, R. A., Lee, G., & Gordon, D. M. (2021). Tree preference and temporal activity patterns for a native ant community in an urbanized California woodland. Journal of Insect Behavior, 34 (4), 211–222. https://doi.org/10.1007/s10905-021-09778-w Mantel, N. (1967). The detection of disease clustering and a generalized regression approach. Cancer Research, 27 (2, Part 1), 209–220. Matich, P., Bizzarro, J. J., & Shipley, O. N. (2021). Are stable isotope ratios suitable for describing niche partitioning and individual specialization? Ecological Applications, 31 (6), e02392. https://doi.org/10.1002/eap.2392 Menke, S. B., Suarez, A. V., Tillberg, C. V., Chou, C. T., & Holway, D. A. (2010). Trophic ecology of the invasive Argentine ant: Spatio-temporal variation in resource assimilation and isotopic enrichment. Mrowka, J. (2007). Brush piles. North Branch Restoration Project , Fall, 1–6. Oksanen, J., Simpson, G. L., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P. R., … Weedon, J. (2001). Vegan: Community ecology package. R package. Pagel, M. (1994). Detecting correlated evolution on phylogenies: A general method for the comparative analysis of discrete characters. Proceedings of the Royal Society of London. Series B: Biological Sciences, 255 (1342), 37–45. Peterson, B. J., & Fry, B. (1987). Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics , 293–320. https://www.jstor.org/stable/2097134 Post, D. M. (2002). Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology, 83 (3), 703–718. https://doi.org/10.1890/0012-9658(2002)083[0703:USITET]2.0.CO;2 Raubenheimer, D., & Simpson, S. J. (1993). The geometry of compensatory feeding in the locust. Animal Behaviour, 45 (5), 953–964. https://doi.org/10.1006/anbe.1993.1114 Raubenheimer, D., & Simpson, S. J. (2016). Nutritional ecology and human health. Annual Review of Nutrition, 36 (1), 603–626. https://doi.org/10.1146/annurev-nutr-071715-051118 Raubenheimer, D., Machovsky-Capuska, G. E., Chapman, C. A., & Rothman, J. M. (2014). Geometry of nutrition in field studies: An illustration using wild primates. Oecologia, 177 (1), 223–234. https://doi.org/10.1007/s00442-014-3142-0 Raubenheimer, D., Senior, A. M., Mirth, C., Cui, Z., Hou, R., Le Couteur, D. G., … Simpson, S. J. (2022). An integrative approach to dietary balance across the life course. iScience, 25 (5). https://doi.org/10.1016/j.isci.2022.104315 Raubenheimer, D., Simpson, S. J., & Mayntz, D. (2009). Nutrition, ecology and nutritional ecology: Toward an integrated framework. Functional Ecology , 4–16. https://www.jstor.org/stable/40205497 Revell, L. J. (2025). Ancestral state reconstruction of phenotypic characters. Evolutionary Biology , 1–25. https://doi.org/10.1007/s11692-025-09645-y Roeder, K. A., & Kaspari, M. (2017). From cryptic herbivore to predator: Stable isotopes reveal consistent variability in trophic levels in an ant population. Ecology, 98 (2), 297–303. https://doi.org/10.1002/ecy.1641 Ryti, R. T., & Case, T. J. (1984). Spatial arrangement and diet overlap between colonies of desert ants. Oecologia, 62 , 401–404. https://doi.org/10.1007/BF00384274 Sanders, N. J., & Gordon, D. M. (2003). Resource-dependent interactions and the organization of desert ant communities. Ecology, 84 (4), 1024–1031. https://doi.org/10.1890/0012-9658(2003)084[1024:RIATOO]2.0.CO;2 Shik, J. Z., & Dussutour, A. (2020). Nutritional dimensions of invasive success. Trends in Ecology & Evolution, 35 (8), 691–703. https://doi.org/10.1016/j.tree.2020.03.009 Simpson, S. J., & Raubenheimer, D. (2012). The nature of nutrition: A unifying framework. Australian Journal of Zoology, 59 (6), 350–368. https://doi.org/10.1071/ZO11068 Smouse, P. E., Long, J. C., & Sokal, R. R. (1986). Multiple regression and correlation extensions of the Mantel test of matrix correspondence. Systematic Zoology, 35 (4), 627–632. https://doi.org/10.2307/2413122 Sperfeld, E., Raubenheimer, D., & Wacker, A. (2016). Bridging factorial and gradient concepts of resource co-limitation: Towards a general framework applied to consumers. Ecology Letters, 19 (2), 201–215. https://doi.org/10.1111/ele.12554 Symonds, M. R., & Moussalli, A. (2011). A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterion. Behavioral Ecology and Sociobiology, 65 (1), 13–21. https://doi.org/10.1007/s00265-010-1037-6 Tillberg, C. V., McCarthy, D. P., Dolezal, A. G., & Suarez, A. V. (2006). Measuring the trophic ecology of ants using stable isotopes. Insectes Sociaux, 53 (1), 65–69. https://doi.org/10.1007/s00040-005-0836-7 Tilman, D., Kilham, S. S., & Kilham, P. (1982). Phytoplankton community ecology: The role of limiting nutrients. Annual Review of Ecology and Systematics, 13 , 349–372. https://www.jstor.org/stable/2097072 Tiunov, A. V. (2007). Stable isotopes of carbon and nitrogen in soil ecological studies. Biology Bulletin, 34 , 395–407. https://doi.org/10.1134/S1062359007040127 Ulrich, Y., Burns, D., Libbrecht, R., & Kronauer, D. J. (2016). Ant larvae regulate worker foraging behavior and ovarian activity in a dose-dependent manner. Behavioral Ecology and Sociobiology, 70 , 1011–1018. https://doi.org/10.1007/s00265-015-2046-2 Vaes, O., & Detrain, C. (2022). Colony specificity and starvation-driven changes in activity patterns of the red ant Myrmica rubra . PLoS ONE, 17 (8), e0273087. https://doi.org/10.1371/journal.pone.0273087 Van Meer, R., Slowik, T. J., & Thorvilson, H. G. (2002). Semiochemicals released by electrically stimulated red imported fire ants, Solenopsis invicta . Journal of Chemical Ecology, 28 , 2585–2600. https://doi.org/10.1023/A:1021448522147 Van Valen, L. (1965). Morphological variation and width of ecological niche. The American Naturalist, 99 (908), 377–390. Vander Zanden, M. J., & Rasmussen, J. B. (1999). Primary consumer δ13C and δ15N and the trophic position of aquatic consumers. Ecology , 1395–1404. https://www.jstor.org/stable/177083 Wittman, S. E., O’Dowd, D. J., & Green, P. T. (2018). Carbohydrate supply drives colony size, aggression, and impacts of an invasive ant. Ecosphere, 9 (9), e02403. https://doi.org/10.1002/ecs2.2403 Yilmaz, A., Dyer, A. G., Rössler, W., & Spaethe, J. (2017). Innate colour preference, individual learning and memory retention in the ant Camponotus blandus . Journal of Experimental Biology, 220 (18), 3315–3326. https://doi.org/10.1242/jeb.158501 Information & Authors Information Version history V1 Version 1 28 January 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords ecological strategies elemental c:n ratio nutritional ecology nutritional partitioning protein:carbohydrate ratio stable isotope analysis Authors Affiliations Hannah Riskas 0009-0003-7687-3553 [email protected] La Trobe University View all articles by this author Jonathan Shik University of Copenhagen View all articles by this author Lily Leahy 0000-0002-0733-6792 La Trobe University View all articles by this author Ian Wright Western Sydney University View all articles by this author Heloise Gibb 0000-0001-7194-0620 Deakin University View all articles by this author Metrics & Citations Metrics Article Usage 259 views 103 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Hannah Riskas, Jonathan Shik, Lily Leahy, et al. Coexisting ant species differ in ability to meet their intrinsic nutritional needs. Authorea . 28 January 2026. DOI: https://doi.org/10.22541/au.176962156.69288030/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176962156.69288030/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe1c1577c6e1b23',t:'MTc3OTE3OTU5MA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();