Evolutionary Ecology最新文献

Iris patterns in the animal kingdom are incredibly variable, with anurans having some of the most diverse and intricate patterns. Although the shape and colouration of anuran eyes seem to be correlated with ecological factors, the evolution of iris patterns remains unexplored. We used a large-scale phylogenetic comparison with 960 anuran species to examine the evolutionary and ecological correlates of iris patterns. We classified iris patterns into four broad categories: Reticulated, Plain, Dotted, and Lined, and examined whether iris pattern was correlated with diel activity (diurnal, nocturnal, and cathemeral activity) and habit (aquatic, arboreal, terrestrial, and fossorial) or both. Our analysis suggests that reticulated irises are the most common pattern in anurans and are the most likely ancestral character. The evolution of iris patterns across the anuran phylogeny best matched Brownian expectations, with many transitions between the four pattern types. Iris patterns, however, were mostly uncorrelated with diel activity or habit. The only exception was an association between plain irises and diel activity. Specifically, anurans with plain irises were more likely to be diurnal and less likely to be nocturnal; and the evolution of plain irises seemed to have preceded the evolution of diel activity. Overall, iris patterns across anurans are mostly unrelated to ecological factors, suggesting that this trait may be important for other functions, such as inter- or intra-specific interactions, or that the incredible diversity has evolved through neutral processes. Our findings open avenues for further research, especially to understand the potential adaptive value of the striking ornamentation in iris patterns across taxonomic groups.

Aposematism (considered here as an association between conspicuous colour patterns and the presence of a harmful secondary defence) has long been recognized as an anti-predator strategy, with salient traits serving as a warning signal to ward off would-be predators. Here we review evidence for a potentially widespread yet under-explored third component of this defensive syndrome, namely capture tolerance (the ability of the signaller to survive being captured and handled by would-be predators). We begin by collating the (largely anecdotal) available evidence that aposematic species do indeed have more robust bodies than cryptic species which lack harmful secondary defences, and that they are better able to survive being captured. We then present a series of explanations as to why aposematism and capture tolerance may be associated. One explanation is that a high degree of capture tolerance facilitates the evolution of post detection (“secondary”) defences and associated warning signals. However perhaps a more likely scenario is that a high capture tolerance is selected for in defended species, especially if conspicuous, because if they can survive for long enough to reveal their defences then they may be released unharmed. Alternatively, both capture tolerance and secondary defences may arise through independent or joint selection, with both traits subsequently facilitating the evolution of conspicuous warning signals. Whatever its ultimate cause, the three-way association appears widespread and has several key implications, including inhibiting the evolution of automimicry and shaping the evolution of tactile mimicry. Finally, we present a range of research questions and describe the challenges that must be overcome in developing a more critical understanding of the role of capture tolerance in the evolution of anti-predator defences.

Hidden colors are a widespread phenomenon in the animal kingdom, particularly in anurans. In some cases, hidden colors are suddenly exposed during defensive displays to startle predators, others seemingly remain hidden—particularly from researchers. Amazonian species of Neotropical harlequin toads (genus Atelopus) show striking and consistent ventral sexual dichromatism where females show and males generally lack melanization. Inspired by these observations we undertook a deeper inquiry across this species-rich genus. We collected data on ventral sexual dichromatism in Atelopus species and scored expression of sex-specific ventral melanization (i.e. black, brown and/or grey coloration). Ventral sexual dichromatism was present throughout the entire range of the genus and in almost all phylogenetic groups. However, there was a clear geographic signal with this trait being most common and widespread in Amazonian Atelopus species. Ventral melanization was correlated with temperature and elevation. Focusing on the Amazonian species, we present hypotheses on potential functions of sexually dimorphic ventral patterns and sex-specific ventral melanization as a baseline to further investigate the dynamics of sexual and natural selection as potential drivers of these traits. Selective pressures on less exposed body parts, such as ventral sides, likely differ considerably from those on dorsal appearance. Given the amount of research on amphibian coloration, it is remarkable how little we know about the evolution, function and underlying mechanisms of ventral appearance. We hope our work will spark more interest in the flip side of amphibians, thereby broadening our understanding of animal coloration.

Assortative or disassortative aggression – when individuals display more aggression towards conspecifics with similar or different phenotypic characteristics – can either maintain polymorphisms or facilitate gene flow between populations depending on which direction the aggression is aimed. Deciphering which factors elicit or prevent aggression is crucial to improving our knowledge of the origin and maintenance of reproductive barriers and subsequent speciation. The Peruvian mimic poison frog, Ranitomeya imitator, is a monogamous and territorial species that has evolved into four distinct color-pattern morphs in a mimetic radiation. Here we use historical landscape genetic data and competition trials between male individuals sourced from different populations and color-pattern morphs to show that the level of aggression between individuals is not associated with color morph or body size but rather with source population. Individuals spent more time in combat with individuals from their own deme (genetically homogeneous population), irrespective of color morph or size. These findings indicate that genotypic similarity is correlated with increased aggression in R. imitator, though the mechanism by which R. imitator males identify conspecifics as territorial threats remains unclear. As body size and color morph were not significantly associated with aggression levels, this study emphasizes the necessity of further research to identify whether other phenotypic traits are influencing territorial behavior between male frogs, and if these factors play a role in increasing gene flow, or conversely, the formation of reproductive barriers between populations.

Understanding animal movements is vital for gaining insights into species' ecological patterns, habitat preferences, and reproductive strategies. Studies in dendrobatid frogs have revealed that home range behaviours, particularly in species with male parental care, are closely linked to the distribution of reproductive resources. Here we focus on males of the poison frog Phyllobates vittatus, endemic to the southern Pacific of Costa Rica to determine males’ home range size and the degree of overlap between individuals of P. vittatus. Sixteen individuals were tracked using harmonic direction finder over 4–5 days each, revealing an average minimum convex polygon area of 55.7 m² and a 50% kernel density estimates area of 26.75 m². Overlapping areas indicated shared home ranges, possibly due to resource distribution. Contrary to prior expectations, we did not observe aggressive encounters between males tracked but we observed one aggressive event between two males carrying tadpoles. Notably, we observed a novel behaviour: a female apparently defending tadpoles from a perceived male intruder. These findings provide important insights into P. vittatus’ behaviour and space use, which are key to developing and implementing conservation strategies, especially considering its vulnerable status and the limited available data on this endemic species.

Homing is the ability to return to previously visited sites, often to the home range. Most studies have focused on the mechanisms used to home, but few have addressed the cost-benefit analysis of homing behavior, e.g., by testing for associations between homing performance and ecological factors. We aimed to study homing ability in males of the poison frog Oophaga histrionica, by testing the general hypothesis that homing performance depends upon potential indicators of territory quality or the risk of losing it. First, we tested whether return time was related to displacement distance, body size, number of courtships during the previous month, or distance to nearest neighbors. 38 out of the 39 displaced males homed, and time to return was inversely related to displacement distance, yet not related to any of the other ecological variables tested. In a second experiment, we tested whether males’ homing performance was affected by adding or removing acoustic cues, to simulate changes in the number, identity, and spatial distribution of neighbors. All of the 41 displaced males homed; 78% homed within six hours, and the remaining 22% returned after six hours. Among the former, males exposed to additional loudspeakers (e.g., neighbors) within their territories and to a loudspeaker broadcasting from their very calling perch (replacement treatment), were found significantly closer to the capture site than males in other treatments. Our results thus indicate that the homing performance of males is affected by the perceived risk of being displaced from their territories.

Intraspecific variation in body size, both among populations and between sexes, is an important factor influencing life-history strategies. This variation might be the response to different environmental conditions, as well as natural and sexual selection, and can result in differences in behavior and reproductive strategies among populations. Here, we use the dyeing poison frog (Dendrobates tinctorius) as a model to investigate how interpopulation variation in body size and sexual size dimorphism affects reproductive strategies. As body size increased, sexual size dimorphism also increased, i.e., females were larger than males, and more so in populations with overall larger frogs. This indicates that there is a stronger selection for body size in females than in males, likely as a response to divergent reproductive investment between the sexes. Females from larger-bodied populations produced larger clutches, but the overall number of froglets produced per clutch did not differ among populations. We discuss potential causes and mechanisms that might be responsible for the observed divergence in body size, sexual size dimorphism, and reproductive strategies among populations that likely represent local adaptations. Our findings demonstrate the importance of cross-population studies, cautioning against drawing general conclusions about a species’ ecology without accounting for intraspecific variation.

The poison frog family (Anura: Dendrobatidae) consists of species with conspicuous (e.g., warning coloration and toxicity or low palatability) and cryptic (e.g., palatable, and inconspicuous coloration) traits. Previous literature suggests that conspicuous, but not cryptic, species require diet specialization in prey high in alkaloids. To test for dietary preferences of poison frog species, we identified, to the lowest possible taxonomic rank, the diets of 21 Epipedobates darwinwallacei (conspicuous) and 22 Hyloxalus awa (cryptic) frogs living in syntopy in the Otongachi Forest in northwestern Ecuador. We then tested for differences in diet assemblage composition, and diet specialization, in these putatively conspicuous and cryptic frogs. Our analyses showed significant differences in the composition of arthropod assemblages consumed by both frog species, which translated into a narrow niche breadth and nine arthropod taxa (out of a total of 18) consumed by both species. Moreover, the index of relative importance, which measures frog’s diet specialization, suggested that E. darwinwallacei, and H. awa prefer specific arthropod taxa, where the former consumes preferentially springtails and mites, while the latter consumes mostly ants and Coleoptera larvae. Thus, contrary to expectations, diet specialization is not a unique characteristic of the species with conspicuous traits when living in syntopy.

The analysis of two-way interactions in linear models is common in the fields of ecology and evolution, being often present in allometric, macroevolutionary, and experimental studies, among others. However, the interpretation of significant interactions can be incomplete when limited to the examination of model coefficients and significance tests. The Johnson–Neyman technique represents a step forward in the interpretation of significant two-way interactions, allowing the user to examine how changes in the moderator variable, it being categorical or continuous, affect the significance of the relationship between the dependent variable and the predictor. Despite its implementation in several software since its initial development, the available options to perform the method lack certain functionality aspects, including the visualization of regions of non-significance when the moderator is categorical, the implementation of phylogenetic corrections, and more intuitive graphical outputs. Here I present the R package JNplots, which aims to fill gaps left by previous software regarding the calculation and visualization of regions of non-significance when fitting two-way interaction models. JNplots includes two basic functions which allow the user to investigate different types of interaction models, including cases where the moderator variable is categorical or continuous. The user can also specify whether the model to explore should be phylogenetically informed and choose a particular phylogenetic correlation structure to be used. Finally, the functions of JNplots produce plots that are largely customizable and allow a more intuitive interpretation of the interaction term. Here I provide a walkthrough on the use of JNplots using three different examples based on empirical data, each representing a different common scenario in which the package can be useful. Additionally, I present the different customization options for the graphical outputs of JNplots.