Vertical and temporal specialisation in Amazonian butterflies and evolution of their thermal traits.

Closely-related species living in sympatry are often partitioned into divergent ecological niches. Such specialization can be enabled by the evolution of divergent traits enhancing adaptation to different niches. In this study, we investigate the partitioning of closely-related butterfly species into different forest strata and daily activity time and test the effects of such spatio-temporal niches on the evolution of thermal traits. First, using experiments in the field in Amazonia, we precisely characterized the daily activity patterns of nine species of Morpho butterflies, therefore documenting extensive temporal segregation among species and observing significant variations in temperature between their respective niches. Using controlled experiments in the lab, we then tested the thermal tolerance of wild individuals to both hot and cold conditions. The vertical distribution of species (understory vs. canopy micro-habitats) had a significant effect on several thermal traits, even when controlling for the phylogenetic distances between species, suggesting that forest stratification may shaped thermal adaptation in these tropical butterflies. However, butterfly activity time did not correlate with any thermal traits measured. The extensive temporal segregation observed between these sympatric species might thus stem from ecological interactions observed between species rather than thermal factors. 1.Introduction Community assembly in the wild is strongly shaped by historical factors, but the effect of ecological specialization also play an important role in co-existence between species (Armitage et al. 1994). Such specialization might stem from the evolution of divergent traits enabling the colonization of new microhabitat, favored by competitive interactions (competition for resources or reproductive interference, Davies et al. (2007)). Disentangling the effect of biotic vs. abiotic factors shaping the evolution of traits enabling specialization into divergent ecological niches is thus a central question in evolutionary biology and community ecology. The specialization in different niches can be enabled by the evolution of new traits ( e.g . key innovations, like the jaw development in Cichlidae fish, granting access to novel food sources (Liem 1973)). In turn, adaptation to divergent niches can promote the evolution of new traits. Habitat divergence between species can indeed be followed by the evolution of different traits, like the claw shape in closely related allopatric Anolis lizards, which are more or less curved depending on the specific micro-habitat of the island they occupy (Yuan et al. 2020). Identifying feedbacks between phenotypic evolution and niche specialization is required to identify the mechanisms involved in species diversification and community assembly in the wild. Tropical forests concentrate the highest level of species richness on earth (Beccaloni and Gaston 1994, De Souza Amorim et al. 2022), and the Amazonian rainforest is a relevant example of geographic regions with tremendous animal biodiversity (Myers et al. 2000, Mittermeier et al. 2003). Yet, the factors shaping the evolution and the persistence of such local diversity are still largely unknown. Specialization into multiple micro-habitats (e.g. such as canopy vs. understory habitats for bird species (Marra and Remsen 1997, Walther 2002) appears as an important driver of persistence of multiple species within tropical forest communities. Diversification into different timings of reproductive activities throughout the day (referred to hourly niche hereafter) within spatial micro-habitats is likely also an important and yet understudied dimension of the niche in such environments. Evidence for partitioning of species into divergent hourly niches have been documented in tropical areas, especially in insects ( e.g. daily activity patterns in the Hesperiidae butterfly genus, (Devries et al. 2008)). The partitioning into different hourly niches may reduce competition and facilitate co-existence of closely related species in sympatry. For instance, differentiated activity timings can be observed between two Acomys mice species, but only when their occur in sympatry (Jones et al. 2001) and such differentiation in foraging time is likely due to competitive interactions. The evolution of circadian rhythm between closely-related species might thus be influenced by selection generated by antagonistic ecological interactions. Competition for resources might thus be an important driver of such specialisation into different hourly niches, notably when resources themselves are partitioned throughout the day. For instance, nectar production (Matile 2006) can oscillate throughout a day, and nectar-consuming insect species often display matching foraging time. Such correlations between insect activity and flower opening times have notably been shown in nocturnal moths (Funamoto and Ohashi 2017). Furthermore, sexual interactions between species, by generating substantial reproductive interference (Gröning and Hochkirch 2008) may also promote the evolution of divergent temporal niches in sympatric species. Mathematical models recently highlight how male-male competition can promote the evolution of divergent hourly niches (Van Doorn et al. 2024, Bouinier et al. 2025). Such divergence in daily temporal niches, as observed for instance in the moth Spodoptera frugiperda, can either participate to the speciation process or be promoted by reinforcement (see Taylor and Friesen (2017) for a review on allochronic speciation). Competition for resources and reproductive interference might thus be key ecological interactions driving the evolution of divergent hourly niches in sympatry. Divergence in hourly niches, together with specialization in different micro-habitats of tropical forests (shadow vs. open areas, canopy vs. understory), may change the abiotic environment and generate contrasted selective pressures on other traits. Temperature and humidity levels have been documented to differ between different spatio-temporal niches (Montejo-Kovacevich et al. 2020) and these divergent abiotic conditions might promote the evolution of thermal or drought tolerance in species specialized in different niches. In particular, temperature influences physiological processes and tolerance to extreme temperatures is thus likely to set boundaries on where and when individual from different species can be active (Franzén et al. 2022). The evolution of thermal tolerance in different species might therefore strongly shape community assembly, by constraining physiologically-demanding activities in different micro-habitats and at different time of the day. Here, we thus aim to characterize the diversification in hourly niches within an Amazonian community of closely-related butterflies living in sympatry, and investigate variations in the thermal tolerances within and among these species. We focus on the community of Morpho butterfly species in French Guiana. The Amazonian forest in French Guiana is characterized by stable mean temperatures throughout the year (Lavado Casimiro et al. 2013, Adde et al. 2016), with an average daily temperature around 26 °C (ranging from 25.8 to 27.1). Out of the twenty-nine species of the Morpho genus, nine can be observed within localities in French Guiana. Divergence into different temporal niches has already been documented between two sister species of Morpho in Amazonian Peru, where male patrolling hours have been characterized using a capture/mark/recapture experiment (Le Roy et al. 2021). Morpho males typically patrol along trails and rivers, and this behavior is probably linked to mate searching and territorial activity. It is also probably a metabolically costly behavior, and might thus be constrained by thermal tolerance (Mattila 2015). Strong variations in wing size and color, as well as body length, are observed among Morpho butterflies, and the evolution of these traits might also impact thermal tolerance. Finally, the flight height strongly differs between Morpho species with species observed mostly in the canopy while others are typically observed in the understory (DeVries et al. 2010, Le Roy et al. 2021). This difference in vertical strata might also change the selective regimes acting on the evolution of thermal tolerance. In this study, we thus aim to: (i) characterize the hourly niche of Morpho species in French Guiana by documenting their patrolling behavior throughout the day; (ii) characterize variations in resistance to cold (-20°C) and hot (50°C) temperatures within and among species, and their co-variations with morphological traits (wing size and wing color pattern) and (iii) identify the historical and ecological factors likely influencing the evolution of thermal tolerance, using comparative analyses. 2. Material and Methods 2.1. Sampling strategy Figure 1: Map of the density of Morpho species throughout Central and South America. Darker colors indicates less coexisting species than lighter ones for a given place. All the data was pooled from iNaturalist, and filtered for only research-grade observations. The sampling site is indicated with a star, displaying 9 coexisting species. Individuals were sampled in the Amazonian rainforest within a single site located in French Guiana (Figure 1), with GPS coordinates 4°33’ N, 52°11’ O. We sampled 91 males from nine species of the Morpho genus in July-November 2023 (see table 1 for more details on the number of samples per species). Each captured butterfly was identified, kept in a glassine envelope, and fed a mixture of water and honey every morning to keep it in good condition. We also recorded the activity time of each individual caught and/or observed in this location throughout 36 days, and calibrated this time with the sunrise time that slightly changed over the course of the field experiment. Finally, we recorded the temperature at this site every 15mn under a shaded area for 8 consecutive days using a thermometer (Mastrad Thermal Probe; Model: Classic, Precision: ±1°C). 2.3. Thermal tolerance test Thermal tolerance testing was conducted at the LEEISA facilities in Cayenne, French Guiana, using butterflies captured at the same sites as the Mark/recapture experiments described above. We tested all 9 species, but with various number of individuals tested per species (see table 1 for more details). Cold recovery experiment Individuals were first acclimatized for 30 min at 27°C, then were put inside a 15 × 15 × 10 cm solid plastic container at -20°C for 5 minutes. Individuals were then retrieved from the container and put back in the 27°C environment. The time it took them to recover wing motor skills was recorded, and the individuals were then put at rest. Each individual underwent this experiment three times, with a 30 min rest in between each experiment. Hot knock-out and recovery experiment Individuals were then acclimatized for 30 min at 27°C, and a 1L glass container was submerged in a hot water bath (Polystat; Model: 86602; Precision: ±0,05°C), and temperature inside the container was monitored with a thermometer (Mastrad Thermal Probe; Model: Classic, Precision: ±1°C). When the temperature reached 50°C, an individual was put in the glass container, and time until loss of motor control was recorded. As soon as the individual lost motor control, it was removed and put back in the 27°C environment. The time it took to recover wing motor skills was recorded and they were then out at rest. Each individual underwent this experiment only once because further testing frequently results in individual death. Table 1: Sample sizes for each of the nine Morpho species studied: (1) Number of observations used for activity timing characterization and (2) Number of individuals used for thermal tolerance tests. The phylogenetic relatedness between species, inferred in Chazot et al. (2016), is shown on the left-hand side of the table. Understory species are shown in green font, while canopy ones are in red font. 2.4. Morphological trait measurements Because wing size and coloration deeply differ between species and could potentially affect thermal tolerance, we estimated variations in the wing size and color pattern, as well as in the wing area among all individuals used in the thermal experiment, by separating the wing from the body of individuals, and taking pictures of each wing separately on their dorsal side. Pictures were taken in a closed box with controlled light emission using a Nikon D90 camera. A xRite Mini ColorChecker was also used to color-correct the images afterwards, by using the R package patternize (v0.0.5) (Van Belleghem et al. 2018). We then used the R package recolorize (v0.1.0) (Weller et al. 2024), allowing us to segment in our images into a limited sets of discrete color categories, and enable comparisons. To assess color variation between species and quantify the relative area covered by those colors, the ‘recolorize’ package was used to carry out a k-means clustering to a palette of 5 selected colors. This palette was handpicked and chosen to be representative of the main colors seen on each side for the nine Morpho species studied here. We finally assessed proximal colors by selecting the third of the wing that was closest from the body, which was shown to account for most of the variation in thoracic temperature (Wasserthal 1975). The particular multi-layered scales structure of iridescent wing patches could potentially affect heat transfer, and thus thermoregulation capabilities of individuals. In supplementary analyses, we thus analyzed the effect of iridescence wing scale on thermal traits by considering different iridescent phenotype. We first categorized the Morpho species studied here into three phenotypes : full blue iridescence ( M.eugenia, M.marcus, M.menelaus, M.rhetenor ), blue-banded iridescence ( M.helenor, M.deidamia, M.helenor ), and no iridescence ( M.hecuba and M.telemachus ) . The type of wing scales and arrangements contributing to the structural blue coloration also differ between Morpho species, we thus distinguished five categories according to the relative size and overlap of scales, as described in Giraldo et al. (2016) : G1 ( M.marcus, as well as M.eugenia, inferred from the phylogeny since it was not part of the studied species in Giraldo et al. (2016)), G2 ( M.helenor, M.achilles, M.deidamia, M.menelaus ), G4 ( M.rhetenor ), and None ( M.hecuba and M.telemachus, as they do not arbor any iridescent blue scales). 2.5. Statistical analysis To assess how much the neutral divergence among Morpho species affect the evolution of the traits measured in this study, we computed the so-called phylogenetic signal of our traits, using Pagel’s λ, estimated from phytools (Revell 2012). We used the phylogenetic relationships among Morpho species inferred in Chazot et al. (2016). Pagel’s λ represents the degree to which differences in trait values among species matches the values obtained from a Brownian motion model accounting for the phylogenetic distance between species. When λ = 0, no phylogenetic structuration of the evolution of the trait is detected, meaning that phylogenetic relationship between species do no predict similarities in trait values. A value of λ = 1 (strong phylogenetic signal) means that trait evolution among species is akin to the Brownian motion model, where trait similarities are likely due to shared evolutionary history between species ( e.g. phylogenetically close species are more likely to share similar trait values). P -values for λ are then provided to estimate how much the null hypothesis (trait distribution is akin to Brownian motion) is rejected. To test the correlations between thermal resistances, activity timings, micro-habitat and wing color/sizes, we first built linear models to assess the effect of ecological variables on thermal tolerances. We then built Phylogenetic Generalized Linear Mixed Models (PGLMM) from the phyr package (Daijang et al. 2020), inserting the phylogenetic tree as a random variable. This allowed to account for the effect of phylogenetic relatedness on trait variation observed between species. Differences in activity timings or micro-habitat could simply stem from neutral divergence between species, and controlling for phylogenetic effects allows to test for a putative role of natural selection regimes differing between ecological niches. To distinguish the relative contributions of ecological specialization and phylogenetic structure in the variation of thermal tolerances, we used partial R 2 s from the “rr2” R package (Ives and Li 2018). The partial R 2 lik for each factor was calculated by comparing the full model with reduced models in which a given factor was removed. The reduction in likelihood is then measured, as well as p -values from a X²-test using likelihood ratio test values. This allows to test whether the difference between the full and reduced models is significant. We treated the time in seconds before knock-out and recovery as the response variables. Explanatory variables included time of capture, micro-habitat, total wing area, percentage of black coloration, and proximal coloration of black, blue or white. We used only the black pigmentation in these first models, assuming that these pigments are more likely to play a role in heat absorption, and included an interaction term between the wing area and the black pigmentation. We also tested if potential correlations between thermal tolerances existed using Pearson correlation tests. In supplementary analyses, we replaced black pigmentation by either the iridescent phenotype or scale configuration, and used post-hoc Tukey’s tests to observe if there was any differences between groups of species. All statistical analyses were performed in R (v4.4.0). 1. Hourly temporal niche evolution across sympatric Morpho species We recorded 1,270 instances of male patrolling activity for the 9 species of Morpho butterflies in French Guiana throughout 36 days of observation. M. menelaus males (242 observations, 19% of all observations) were the most abundantly observed species in the understory, while M. rhetenor males were the most frequently observed in the canopy (323 observations, 25% of all observations). Although pseudo-replication likely occur in these observations, the large sample size and the relatively abundant population size allowed us to confirm that individuals of each species consistently patrolled within the same time window across multiple days. As seen in Figure 2, Morpho species in French Guiana are strongly partitioned throughout the day. Using phytools to quantify phylogenetic signal in activity timings, we found a medium and non-significative value for both Pagel’s λ ( λ = 0.359, p-value = 0.361). We suspect that our sample size of species is too small to infer any significant phylogenetic signal. We also note temporally differentiated niches within the species M.telemachus, where two distinct color pattern morphs within populations are observed in French Guiana (Gayman et al. 2016): a grey/blue morph and an orange morph. The two morphs show slightly different activity timings, with the orange morph being active later in the day (Supporting information). Figure 2: Recorded activity times of Morpho species in French Guiana . We represent boxplots of the activity timings of the two main recognized clades of Morpho: (A) Understory species, where individuals fly at 2-5 meters above ground, and (B) canopy species, where individuals fly at 8-10 meters above ground. Significant differences between groups (using a HSD Tukey test) are indicated with letters on the right of boxplots. We observe highly differentiated temporal niches between understory species, and less so in canopy species, although the mean timings are still significantly differentiated between species. 2. Thermal tolerance between species When sampling temperature in 15 minutes intervals over 9 days at the same location, we found significant differences in average temperature encountered in the different time windows occupied by the different species, with a difference of almost 5°C between the temperature conditions encountered by flying males from the species M. eugenia vs . M. achilles (Figure 3). We then tested the thermal tolerance of 91 male individuals belonging to the 9 species (see Table 1 for more details on the sample sizes within each species). We found no significant correlation between the time before heat knock-out and heat recovery timings (Pearson correlation test: r = -0.17, t (70) = -1.46, p = 0.15, 95% CI [-0.39, 0.06]), heat knock-out and cold recovery timings after the first pass ( r = 0.18, t (77) = 1.65, p = 0.10, 95% CI [-0.04, 0.39]) or heat recovery and cold recovery timings after the first pass ( r = 0.06, t (67) = 0.51, p = 0.61, 95% CI [-0.18, 0.29]). Phylogenetic signals inferred from thermal tolerances were all non-significant, potentially due to our small sample size. We nonetheless observed high phylogenetic signal for the time before heat knock-out ( λ = 0.84, p = 0.20) and the time before cold recovery ( λ = 0.75, p = 0.14). Phylogenetic signal for heat recovery was practically zero ( λ < 0.001, p = 1). We also documented whether iridescent phenotype or scale configuration had an effect on thermal tolerances (see Supporting information), but did not find any significant result. Figure 3: Temperature encountered during patrolling window and thermal traits measured within understory and canopy species . The patrolling windows of each (A) understory and (B) canopy species is shown in colored bands, with a picture of an individual on top. The black line represents the average temperature at each moment of the day, with the 95% confidence interval in gray. The dotted line is extrapolated temperature data, as we did not sample temperatures this late in the day. Below are the recorded timings for heat knock-out, heat recovery and cold recovery respectively for each species. Dashed lines represent the mean thermal tolerance value between species. Heat knock-out and recovery We detected a significant effect of the micro-habitat ( β = -0.36 ± 0.09, Z = -4.15, p < 0.001), but not of the time of capture, on the time before heat knock-out (Supporting information): individuals living in the understory were knocked-out earlier than the canopy ones, consistent with a reduced tolerance to heat in understory species. The controlled variables, i.e. the level of black pigmentation ( β = 1.843 ± 0.67, Z = 2.74, p = 0.008), the wing area ( β = 0.01 ± 0.004, Z = 2.02, p = 0.047) and their interaction ( β = -0.02 ± 0.01, Z = -2.09, p = 0.040) also had a significany effect on the resistance to heat. To test whether the effect of micro-habitat on heat tolerance was driven by phylogenetic divergence between canopy vs . understory species we then used PGLMM, allowing to account for the level of phylogenetic relatedness between species (Supporting information). In this analysis, only the effect of micro-habitat remained significant ( β = -0.32 ± 0.11, Z = -2.69, p = 0.007), while the previously detected effects of other variables were no longer significant, probably because of the strong links between wing color and size variations and phylogenetic distances between species. Phylogenetic effects did not significantly explain the variation in knock-out timings across species ( partial R 2 lik = -4.79%, ΔlogLik = -1.83, p = 1), as most of the variation was explained by micro-habitat ( partial R 2 lik = 5.0%, ΔlogLik = 2.0, p = 0.045). For recovery times from heat knock-out, we did not detect any effect of the micro-habitat or the time of capture (Supporting information). The controlled variables, i.e. the level of black pigmentation ( β = -5.62 ± 2.43, Z = -2.31, p = 0.024), the wing area ( β = -0.04 ± 0.02, Z = -2.65, p = 0.010) and their interaction ( β = 0.10 ± 0.02, Z = -2.51, p = 0.014) had a significany effect on the recovery to heat shock. But all these effects became non significant when correcting for the level of phylogenetic relatedness between species using a PGLMM approach (Supporting information). Cold recovery We detected a significant effect of the micro-habitat ( β = -0. 65 ± 1.55, Z = -4.18, p < 0.001), but not of the time of capture, on the recovery times from the cold knock-out (Supporting information) : individuals living in the understory recovered from the cold shock more quickly than the canopy ones, consistent with an increased tolerance to cold in understory species. The amount of black pigmentation ( β = -5.68 ± 1.30, Z = -4.37, p < 0.001) and wing area ( β = -0.04 ± 0.008, Z = -4.85, p < 0.001) also had negative effects on recovery to cold shock, where individuals with either higher percentage of black or larger wing area recovered quicker. Finally, the interaction between black pigmentation and wing area was also significant ( β = 0.08 ± 0.02, Z = 3.86, p < 0.001), showing that the influence of black pigmentation on cold recovery was stronger when wing area increased and vice-versa. When controlling for the level of phylogenetic relatedness between species (Supporting information), only the effects of micro-habitat ( β = -1.06 ± 0.48, Z = -2.19, p = 0.028) and wing area ( β = -0.03 ± 0.01, Z = -2.97, p = 0.003) remained significant. Phylogeny also explained the major part of the variation in cold recovery timings across species ( partial R 2 lik = 15.5%, ΔlogLik = 20.2, p < 0.001), followed by wing area ( partial R 2 lik = 7.0%, ΔlogLik = 8.8, p < 0.001) and micro-habitat explaining a smaller part of the variation ( partial R 2 lik = 3.4%, ΔlogLik = 4.1, p = 0.004). We then also tested more precisely the spatial arrangement of coloration, by studying the wing colors proximal to the body (shown in Supporting information). While we found differences in proximal coloration between species, no significant link between any type of proximal coloration and thermal tolerance was found (Supporting information).

Vertical stratification in sympatric species shaping the evolution of thermal traits Our results suggest that the main factor shaping the variations in thermal traits (heat resistance and cold recovery) among individuals was the specialization in the canopy vs . understory observed between sympatric species. This effect remained significant even when correcting for the phylogenetic distances between species, consistent with an effect of the divergent selection on thermal tolerance in the canopy vs. understory microhabitat. In the neo-tropical genus Heliconius, thermal tolerance has been shown to differ between populations living at different altitudes within 10 several species, and common-garden experiments in H. erato revealed a contribution of both plastic and genetic components in the thermal traits measured (Montejo-Kovacevich et al. 2020). Behavioural assays testing the responses of butterfly facing or dodging sunrise show the effect of solar-heating on the timing of first take-off (Meyer and Sisk 2001), highlighting the plastic response of individuals submitted to microclimatic variations. In our study, the contribution of plastic response to the conditions encountered by the individuals living in different micro-habitats vs . the genetic divergence between species specialized in different micro-habitats cannot be estimated. However, the significant differences between species suggest that genetic divergence between species may contribute to the divergence in thermal traits. Similarly, in a study on 45 different species of Nymphalidae butterflies from Brazil, Silva et al. (2020) detected significant variation in temperature tolerance between species, with higher maximum temperature tolerated in species found in open as compared to shaded environments. The sensitivity to heat and cold conditions might thus partly depend on the genetic variation across species evolving in different temperature conditions. In our study, individuals from understory species were found less resistant to heat shock and recovered quicker to cold shock, as compared to individuals from canopy species. The canopy vs . understory micro-habitats have been documented to differ in temperature conditions: significantly higher maximum temperature were measured in the canopy vs . understory habitats, in both high and lowland South American rainforests (Montejo-Kovacevich et al. 2020). This vertical partitioning in temperature conditions could promote divergent evolution of thermal traits in communities of butterflies inhabiting tropical rainforest. Field survey in temperate areas have detected that butterfly activity patterns ( e.g. in Euphydryas aurinia, Parnassius apollo, Phengaris arion ) is affected by ambient temperature (Franzén et al. 2022), with different range of thermal niches between species, leading to variation in seasonal and daily niches. In tropical rainforest, where the average daily temperature is much less variable throughout each year, especially when considering only the dry season (Lavado Casimiro et al. 2013, Checa et al. 2014), seasonal variation in temperature are much less marked. Therefore, specialization into vertical strata might be an important niche dimension in tropical rainforest with specific selection acting on traits depending on the microclimatic differences between canopy and understory environments. Drivers of temporal niche diversification in sympatric species In this study, we also showed that sympatric Morpho species exhibit pronounced temporal niche partitioning throughout the day, with little degree of time overlap between species sharing the same vertical strata within the forest. Ecological niche diversification can arise due to various factors, notably niche availability (Wellborn and Langerhans 2015) and divergent selection (Nosil and Sandoval 2008). Temporal partitioning seems to be common in many species (Taylor and Friesen 2017), especially when those species are closely related (Harris 1971, Shkolnik 1971). Temporal partitioning in particular appears to facilitate species coexistence by reducing competition between individuals from closely related species (Delaval et al. 2005), which are more likely to share resources or exhibit similar mating cues. Nevertheless, temporal segregation between species throughout the day could also stem from divergent selection exerted by temperature variation during the day, promoting different physiology between species. Here we thus tested the correlations between temporal niche and heat and cold thermal tolerances, but no significant effect of activity time on any of the thermal tolerances tested was detected. Our thermal tolerance assays investigated the effect of extreme temperature (50 °C vs . -20°C) that are not encountered on the wild, and might therefore not allow to detect behavioural or physiological differences exhibited by butterflies from different species when submitted to more subtle temperature variations encountered in the field. Nevertheless, the two Morpho species with the earliest activity times ( M. eugenia and M. marcus ) had poorer resistance to heat, suggesting potential variation in thermal tolerance between species with the largest differences in the temperature conditions encountered in their respective temporal niches. Overall, the tolerance to extreme temperatures investigated here does not strongly differ between species occupying different temporal niches. The selection potentially exerted by temperature variation across different temporal niches might not be the main driver of temporal partitioning of the Morpho species in sympatry, begging the question of the factors shaping the temporal niche partitioning in this community. Experiments carried out in the field evidence intensive heterospecific male-male competitive interactions in wild populations of M. helenor and M. achilles (Le Roy et al. 2021). Controlled experiments in cages also highlighted an extensive courtship of wild males towards heterospecific female dummies from these two species (Ledamoisel et al. 2025). Reproductive interference between closely-related Morpho species living in sympatry is thus likely very high in natural communities, especially between species sharing similar wing colour patterns. Such heterospecific interactions might favour the evolution of divergent temporal niches (Bouinier et al. 2025). The lack of correlation between thermal traits and temporal niches in sympatric species observed in our study likely discard the selection exerted by divergent temperature conditions as the main driver of the evolution of divergent temporal niches in this community. Our results are thus more consistent with reproductive interferences between species a major factor promoting temporal divergence across sympatric species.

Here, we document the significance of daily temporal niches and vertical stratification as important factors structuring natural communities of closely-related species in tropical rainforest. We found that sympatric species from the same butterfly genus are strongly partitioned in differentiated temporal niches, but that this temporal niche divergence is poorly associated with thermal tolerance. We instead highlight that this thermal tolerance is rather shaped by the specialization in different micro-habitats, probably because of strong divergent selection exerted by temperature conditions encountered in the canopy vs. understory micro-habitats. The evolution of wing morphological traits between species living in different micro-habitats might notably have contributed to the variations in thermal tolerance across species. Thus, while the specialization of different sympatric species into different spatial niches could have resulted in differentiated thermal traits, temporal niche specialization might rather result from ecological interactions such as heterospecific competition or reproductive interference.

Adde, A., Roucou, P., Mangeas, M., Ardillon, V., Desenclos, J.-C., Rousset, D., Girod, R., Briolant, S., Quenel, P. and Flamand, C. 2016. Predicting Dengue Fever Outbreaks in French Guiana Using Climate Indicators. - PLoS Negl. Trop. Dis. 10: e0004681.Armitage, P. D., Cranston, P. S. and Pinder, L. C. V. 1994. Chironomidae : biology and ecology of non-biting midges. - Chapman and Hall.Beccaloni, G. W. and Gaston, K. J. 1994. Predicting the species richness of neotropical forest butterflies: Ithomiinae ( Lepidoptera: Nymphalidae ) as indicators. - Biol. Conserv. 71: 77–86.Bouinier, T., Brunaud, A., Smadi, C. and Llaurens, V. 2025. Evolution of divergent daily temporal niches shaped by male-male competition can generate sympatric speciation. bioRviv doi:10.1101/2024.07.31.601896Chazot, N., Panara, S., Zilbermann, N., Blandin, P., Le Poul, Y., Cornette, R., Elias, M. and Debat, V. 2016. Morpho morphometrics: Shared ancestry and selection drive the evolution of wing size and shape in Morpho butterflies. - Evolution 70: 181–194.Checa, M. F., Rodriguez, J., Willmott, K. R. and Liger, B. 2014. Microclimate Variability Significantly Affects the Composition, Abundance and Phenology of Butterfly Communities in a Highly Threatened Neotropical Dry Forest. - Fla. Entomol. 97: 1–13.Daijang, L., Dinnage, R., A. Nell, L., R. Helmus, M. and R. Ives, A. 2020. phyr : An r package for phylogenetic species-distribution modelling in ecological communities. - Methods Ecol. Evol. 11: 1455–1463.De Souza Amorim, D., Brown, B. V., Boscolo, D., Ale-Rocha, R., Alvarez-Garcia, D. M., Balbi, M. I. P. A., De Marco Barbosa, A., Capellari, R. S., De Carvalho, C. J. B., Couri, M. S., De Vilhena Perez Dios, R., Fachin, D. A., Ferro, G. B., Flores, H. F., Frare, L. M., Gudin, F. M., Hauser, M., Lamas, C. J. E., Lindsay, K. G., Marinho, M. A. T., Marques, D. W. A., Marshall, S. A., Mello-Patiu, C., Menezes, M. A., Morales, M. N., Nihei, S. S., Oliveira, S. S., Pirani, G., Ribeiro, G. C., Riccardi, P. R., De Santis, M. D., Santos, D., Dos Santos, J. R., Silva, V. C., Wood, E. M. and Rafael, J. A. 2022. Vertical stratification of insect abundance and species richness in an Amazonian tropical forest. - Sci. Rep. 12: 1734.Delaval, M., Henry, M. and Charles-dominique, P. 2005. Interspecific competition and niche partitioning: example of a neotropical rainforest bat community. - Rev. DÉcologie Terre Vie 60: 149–165.Devries, P. J., Austin, G. T. and Martin, N. H. 2008. Diel activity and reproductive isolation in a diverse assemblage of Neotropical skippers ( Lepidoptera: Hesperiidae ). - Biol. J. Linn. Soc. 94: 723–736.DeVries, P. J., Penz, C. M. and Hill, R. I. 2010. Vertical distribution, flight behaviour and evolution of wing morphology in Morpho butterflies. - J. Anim. Ecol. 79: 1077–1085.Franzén, M., Francioli, Y., Askling, J., Kindvall, O., Johansson, V. and Forsman, A. 2022. Differences in phenology, daily timing of activity, and associations of temperature utilization with survival in three threatened butterflies. - Sci. Rep. 12: 7534.Funamoto, D. and Ohashi, K. 2017. Hidden floral adaptation to nocturnal moths in an apparently bee‐pollinated flower, Adenophora triphylla var. japonica (Campanulaceae). - Plant Biol. 19: 767–774.Gayman, J.-M., Melier, F., Ouvaroff, J., Bénéluz, F., Lacomme, D. and Purser, B. 2016. Les Morpho – Distribution, diversification, comportement. - Association des lépidoptéristes de France.Giraldo, M. A., Yoshioka, S., Liu, C. and Stavenga, D. G. 2016. Coloration mechanisms and phylogeny of Morpho butterflies. - J. Exp. Biol. 219: 3936–3944.Gröning, J. and Hochkirch, A. 2008. Reproductive Interference Between Animal Species. - Q. Rev. Biol. 83: 257–282.Harris, T. L. 1971. Crepuscular Flight Periodicity of Trichoptera. - J. Kans. Entomol. Soc. 44: 295–301.Ives, A. and Li, D. 2018. rr2: An R package to calculate R 2 s for regression models. - J. Open Source Softw. 3: 1028.Jonathan Davies, T., Meiri, S., Barraclough, T. G. and Gittleman, J. L. 2007. Species co‐existence and character divergence across carnivores. - Ecol. Lett. 10: 146–152.Jones, M., Mandelik, Y. and Dayan, T. 2001. Coexistence of temporally partitioned spiny mice: roles of habitat structure and foraging behavior. - Ecol. Soc. Am. 82: 2164–2176.Lavado Casimiro, W. S., Labat, D., Ronchail, J., Espinoza, J. C. and Guyot, J. L. 2013. Trends in rainfall and temperature in the Peruvian Amazon–Andes basin over the last 40 years (1965–2007). - Hydrol. Process. 27: 2944–2957.Le Roy, C., Roux, C., Authier, E., Parrinello, H., Bastide, H., Debat, V. and Llaurens, V. 2021. Convergent morphology and divergent phenology promote the coexistence of Morpho butterfly species. - Nat. Commun. 12: 7248.Ledamoisel, J., Buatois, B., Mauxion, R., Andraud, C., McClure, M., Debat, V. and Llaurens, V. 2025. Sending mixed signals: convergent iridescence and divergent chemical signals in sympatric sister-species of Amazonian butterflies. bioRviv doi: 10.1101/2024.12.12.627956Liem, K. F. 1973. Evolutionary Strategies and Morphological Innovations: Cichlid Pharyngeal Jaws. - Syst. Zool. 22: 425.Marra, P. and Remsen, J. V. 1997. In the neotropics: habitat selection and foraging behavior in understory birds of tropical and temperate forests. - Ornithol. Monogr. 48: 445–483.Matile, P. 2006. Circadian rhythmicity of nectar secretion in Hoya carnosa. - Bot. Helvetica 116: 1–7.Mattila, A. L. K. 2015. Thermal biology of flight in a butterfly: genotype, flight metabolism, and environmental conditions. - Ecol. Evol. 5: 5539–5551.Meyer, C. L. and Sisk, T. D. 2001. Butterfly Response to Microclimatic Conditions Following Ponderosa Pine Restoration. - Restor. Ecol. 9: 453–461.Mittermeier, R. A., Mittermeier, C. G., Brooks, T. M., Pilgrim, J. D., Konstant, W. R., Da Fonseca, G. A. B. and Kormos, C. 2003. Wilderness and biodiversity conservation. - Proc. Natl. Acad. Sci. 100: 10309–10313.Montejo-Kovacevich, G., Martin, S. H., Meier, J. I., Bacquet, C. N., Monllor, M., Jiggins, C. D. and Nadeau, N. J. 2020. Microclimate buffering and thermal tolerance across elevations in a tropical butterfly. - J. Exp. Biol.: jeb.220426.Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A. B. and Kent, J. 2000. Biodiversity hotspots for conservation priorities. - Nature 403: 853–858.Nosil, P. and Sandoval, C. P. 2008. Ecological Niche Dimensionality and the Evolutionary Diversification of Stick Insects. - PLoS ONE 3: e1907.Revell, L. J. 2012. phytools: an R package for phylogenetic comparative biology (and other things). - Methods Ecol. Evol. 3: 217–223.Shkolnik, A. 1971. Diurnal activity in a small desert rodent. - Int. J. Biometeorol. 15: 115–120.Silva, V. D. e, Beirão, M. V. and Cardoso, D. C. 2020. Thermal Tolerance of Fruit-Feeding Butterflies ( Lepidoptera: Nymphalidae ) in Contrasting Mountaintop Environments. - Insects 11: 278.Taylor, R. S. and Friesen, V. L. 2017. The role of allochrony in speciation. - Mol. Ecol. 26: 3330–3342.Van Belleghem, S. M., Papa, R., Ortiz‐Zuazaga, H., Hendrickx, F., Jiggins, C. D., Owen McMillan, W. and Counterman, B. A. 2018. patternize: An R package for quantifying colour pattern variation. - Methods Ecol. Evol. 9: 390–398.Van Doorn, G. S., Schepers, J., Hut, R. A. and Groot, A. T. 2024. Sex-specific expression of circadian rhythms enables allochronic speciation. - Evol. Lett.: qrae049.Walther, B. A. 2002. Vertical stratification and use of vegetation and light habitats by Neotropical forest birds. - J. Für Ornithol. 143: 64–81.Wasserthal, L. T. 1975. The rôle of butterfly wings in regulation of body temperature. - J. Insect Physiol. 21: 1921–1930.Wellborn, G. A. and Langerhans, R. B. 2015. Ecological opportunity and the adaptive diversification of lineages. - Ecol. Evol. 5: 176–195.Weller, H. I., Hiller, A. E., Lord, N. P. and Van Belleghem, S. M. 2024. recolorize: An R package for flexible colour segmentation of biological images. - Ecol. Lett. 27: e14378.Yuan, M. L., Jung, C., Wake, M. H. and Wang, I. J. 2020. Habitat use, interspecific competition and phylogenetic history shape the evolution of claw and toepad morphology in Lesser Antillean anoles. - Biol. J. Linn. Soc. 129: 630–643. Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License. Collection

Authors Metrics & Citations Metrics Article Usage 423views 378downloads Citations Download citation Titouan Bouinier, Charline Smadi, Violaine Llaurens. Vertical and temporal specialisation in Amazonian butterflies and evolution of their thermal traits.. Authorea. 07 April 2025. DOI: https://doi.org/10.22541/au.174402613.32892338/v1 DOI: https://doi.org/10.22541/au.174402613.32892338/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. Cited by - Genomic Signatures of Speciation in Butterflies, Systematic Biology, (2026).https://doi.org/10.1093/sysbio/syag029 Loading...