Migratory insectivores as seasonal agents of selection on wing pigmentation: one strong test, many abiotic substitutes

Question: How do seasonal polyphenisms in wing pigmentation across Odonata and other insects track spatiotemporal variation in predation risk from migratory insectivores?

Answer

Taken together, the retrieved papers support a narrow but strong version of the hypothesis: in exactly one system — the rubyspot damselfly Hetaerina titia — seasonal wing pigmentation has been shown to covary with spatiotemporal variation in predation risk that is itself linked to the local abundance of migratory insectivorous birds, establishing an eco-evolutionary link between tropical and temperate assemblages [1]. Across the broader insect literature, however, seasonal wing-pigmentation polyphenisms are overwhelmingly documented as responses to abiotic cues (temperature, humidity, photoperiod) that predict the adult environment, with cue identity locally adapted to whatever variable reliably forecasts that environment [4] [6] [5], and with thermoregulation and desiccation stress providing well-supported alternative or co-selected functions for the same pigment shifts [8] [13] [3]. The defensible position is therefore that pigmentation polyphenisms track predator risk only indirectly, through seasonally reliable abiotic proxies, and that the direct-predation-tracking mechanism remains demonstrated in a single Odonate species rather than established as a general pattern.

Only one Odonate system directly links seasonal pigmentation to migratory insectivore abundance

Hetaerina titia shows a marked seasonal polyphenism across much of North America, with summer-emerging individuals bearing conspicuous dark wings [1]. A continentally distributed predation experiment found that predation risk covaries with spatiotemporal variation in wing pigmentation and is linked to shifts in the local abundance of migratory birds, while wing phenotypes outside the summer season appear strongly constrained [1]. The authors frame seasonal shifts in migratory insectivorous bird abundance as an important but previously overlooked source of variation in selection on insect conspicuous traits [1]. No other retrieved paper tests this link in Odonata.

The pigmented wing is simultaneously a sexual signal, so the polyphenism reflects fluctuating selection rather than pure predator avoidance

Conspicuous traits impose fitness costs on their bearers, and the relative costs and benefits of conspicuousness can vary through space and time, driving variation in selection [1]. In H. titia the seasonal dark-wing morph is described as a conspicuous signal, and the observed pattern of constraint outside summer is interpreted as a seasonal shift in the relative costs and benefits of pigmentation rather than as unidirectional predator avoidance [1].

Butterfly wing-pattern polyphenisms are cued by abiotic variables that forecast the adult environment, with predation acting as the downstream function

Satyrine dry-season morphs have small or no ventral eyespots and wet-season morphs large marginal eyespots that deflect predator attacks [4]. In Mycalesis mineus, humidity directly determines eyespot size, whereas in Melanitis leda humidity acts indirectly through host-plant quality, and the humidity-sensitive phase lies in the late larval and prepupal stages, temporally closest to the adult environment [4]. In Melanitis leda populations where temperature has the highest intra-annual variation, temperature rather than humidity governs eyespot size [6]. In Colias, plasticity of wing melanization is photoperiod-cued, and its evolution in California populations is consistent with avoiding thermal stress during warming springs [5]. The cue in every case is abiotic; predation risk enters only as the function the phenotype serves.

Cue identity is locally adapted, so no single seasonal variable substitutes for predation risk across a species' range

Three Indian populations of Melanitis leda respond to different cues: humidity where humidity differentiates seasons, humidity where both temperature and humidity are unreliable, and temperature where temperature varies most intra-annually [6]. Two sympatric satyrines experiencing similar ecological pressures achieve the same eyespot plasticity through different mechanisms — direct humidity sensing versus humidity-mediated host-plant quality [4]. Photoperiod-cued wing-melanization plasticity has evolved in California Colias but not in Colorado populations that experienced stronger increases in climate means relative to extremes [5]. Reaction norms, not a fixed cue, are the evolved unit [6].

Thermoregulation and desiccation explain the same seasonal colour shifts without invoking predators

Spring and summer forms of Araschnia levana differ in both colour and wing loading; morphology and colouration jointly govern thoracic heating and cooling, and after accounting for morphological differences the spring form's thorax heats and cools faster, consistent with form-specific thermal optimisation in a temperate climate [8]. Across 483 British butterfly assemblages, body colour is darker early and late in the season, mirroring spatial patterns and indicating thermoregulatory alignment with seasonal conditions [13]. In Mycalesis mineus, dry-condition adults both develop smaller eyespots and survive desiccation stress longer, with a negative correlation between eyespot size and desiccation tolerance suggesting shared developmental pathways [3]. Seasonal pigment variation is thus multiply determined, and predation need not be invoked to explain its seasonality.

Predator communities are functionally heterogeneous, so "predation risk" is not a scalar a reaction norm can track

In Peruvian aviary trials with wild avian predators, crypsis, aposematism and evasiveness produced largely similar mortality rates, but predator life stage and predator family strongly changed defence effectiveness, with skilled insectivores such as Vireonidae achieving higher capture success against defended prey [10]. In Ecuadorian forest, attacks by 43 bird species on Morpho helenor models differed by foraging guild: avoidance of conspicuous prey was prevalent in specialist flight-feeding invertivores but weak in opportunistic treehunter and frugivorous guilds, and cryptic ventral versus conspicuous dorsal surfaces experienced contrasting selection from different bird assemblages [11]. Risk from migratory insectivores is therefore guild- and surface-specific, not a single seasonal variable.

Macroevolutionarily, antipredator wing traits track predator guild composition and are highly labile

Eudamine skipper morphometrics are best described by Ornstein-Uhlenbeck models indicating stabilising selection around fitness optima, with recurrent effects of hindwing tails on forewing shape implying trade-offs between flight performance and antipredator defence [7]. Hindwing tails evolved at least seven times and blue-green coloration at least fifteen times, with losses more frequent than gains and both traits associated with elevated speciation rates [2]. Across moon moths, bat predation pressure estimated from nearly 200 species distribution models is positively associated with tail length and body size, while environmental variables are negatively associated with them [14]. These studies show predator-abundance gradients can be reconstructed and mapped onto wing traits, but at phylogenetic rather than seasonal resolution.

Conflicts and gaps

The direction of the pigmentation–risk relationship in Odonata is unresolved: the Hetaerina study reports covariation between predation risk and pigmentation and a constraint on wing phenotypes outside summer, but does not indicate in the retrieved abstract whether dark wings are favoured or penalised when migrant bird abundance peaks, nor does it identify the developmental cue or sensitive stage [1]. A sign conflict runs across taxa: enlarged eyespots are interpreted as antipredator devices expressed when predators are present [4], whereas conspicuous dark odonate wings are treated as a predation-imposed cost [1] — the same directional change in pigment investment carries opposite inferred fitness signs, and no retrieved paper supplies a common currency.

The risk proxy itself is contested. Plasticine prey models may underestimate avian insectivore foraging activity, observer scoring of attack marks is subjective, and camera traps record bird presence rather than predation events [15]. Any continental-scale inference about predation risk inherits these measurement problems [1].

Finally, the correlation between seasonal cues and predator arrival is itself moving. Spring first-arrival dates of migratory birds have advanced and last-departure dates delayed, with the magnitude of advance depending on wintering distribution [16], and European and North American odonates show coupled range and phenology shifts, with about half of species shifting in both space and time and roughly 10% not shifting at all [12]. Photoperiodism evolves predictably along latitude while other plastic traits do not [9]. None of the retrieved papers tests whether pigmentation reaction norms are tracking or lagging these shifts, and the taxonomic base remains a single damselfly species.

What would change the answer

A rearing experiment on Hetaerina and other calopterygids that crosses simulated avian predation cues (insectivore call playback, mounted models, attack simulations) with photoperiod, temperature and diet, scoring wing melanin area and identifying the sensitive stage, would establish whether a predator-derived cue exists at all rather than being proxied by season. Replacing plasticine with live tethered prey or camera-verified attacks [15], paired with radar-derived insectivore and insect traffic rates of the kind used to resolve swift foraging responses to aerial insect abundance [17], would give a defensible continental measure of risk. Reciprocal common-garden transplants of multiple calopterygid species across the migratory flyway, with reaction norms regressed on measured bird abundance, would extend the single-species result geographically. Combining odonate range and phenology shift data [12] with bird arrival phenology [16] would test for phenological mismatch between pigment morph frequency and predator arrival. Finally, treating the presence and strength of pigmentation polyphenism as a comparative trait across Odonata, in the Ornstein-Uhlenbeck framework applied to skipper wing morphology [7], would test whether polyphenism predicts migratory insectivore flyway intensity the way bat abundance predicts moth tail length [14].

References

[1] Patterson, C., Grether, G., Soley, F., Clavel, J. P., Bonillas Monge, E., Mendoza Cuenca, L., Palin, R., Perez Madrigal, A., Saban-Sequen, E., Tonkinson, A., Drury, J. P.. "Predation by migratory birds shapes seasonally fluctuating selection on a conspicuous signal". bioRxiv. 2026-10-19T00:00:00.000Z. URL: https://doi.org/10.64898/2026.08.24.746656

[2] Matos-Maravi, P., Linke, D., Ribeiro, P., Lamas, G., Freitas, A. V., Nunez, R., Kawahara, A. Y., Toussaint, E. F., Bacon, C. D., Wahlberg, N., Antonelli, A.. "Environmental Variation Promotes Convergent Evolution and Rapid Diversification of Wing Shape and Color in Skipper Butterflies". bioRxiv. 2025-12-27T00:00:00.000Z. URL: https://doi.org/10.64898/2025.12.27.696669

[3] Sharma, B. B., Kodandaramaiah, U.. "Are seasonally plastic anti-predatory and desiccation tolerance traits developmentally linked?". bioRxiv. 2026-05-21T00:00:00.000Z. URL: https://doi.org/10.64898/2026.05.19.726136

[4] Yumnam, T., Molleman, F., Walczak, U., Kodandaramaiah, U.. "Contrasting mechanisms for using humidity as cue for seasonal polyphenism in two tropical butterflies". bioRxiv. 2025-10-15T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.14.682489

[5] Buckley, L., Kingsolver, J.. "Functional resurveys and models reveal the interplay of plasticity and evolution of Pierid butterflies in response to recent climate change". bioRxiv. 2025-09-25T00:00:00.000Z. URL: https://doi.org/10.1101/2025.09.23.678128

[6] Prasannakumar, I., Molleman, F., Walczak, U., Kodandaramaiah, U.. "Local adaptations in wing-pattern and life history trait plasticity in a butterfly: humidity as a cue where temperature is unreliable". bioRxiv. 2025-10-15T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.15.682538

[7] Linke, D., Debat, V., Matos-Maravi, P.. "Association of iridescence and tails with flight proxies suggests that predation drives convergence in skipper butterflies". bioRxiv. 2025-06-27T00:00:00.000Z. URL: https://doi.org/10.1101/2025.06.23.661062

[8] Linke, D., Okrouhlik, J., Bartonova, A. S., Re Jorge, L., Matos-Maravi, P., Kleckova, I.. "Heated debate: Is the seasonal polyphenism of Araschnia levana (Lepidoptera: Nymphalidae) driven by thermoregulation?". bioRxiv. 2026-02-09T00:00:00.000Z. URL: https://doi.org/10.64898/2026.02.08.704656

[9] Gotthard, K., Berger, D., Rohner, P. T.. "Life history evolution of insects in response to climate variation: seasonal timing versus thermal physiology". bioRxiv. 2025-07-08T00:00:00.000Z. URL: https://doi.org/10.1101/2025.02.05.636251

[10] Linke, D., Mejia, J. H., Navarro, V. N. P. E., Gohil, P., Garcia, C. R., Salinas, L., Elias, M., Matos-Maravi, P.. "Quantifying the success of prey crypsis, aposematism and evasiveness in avoiding predators attack". bioRxiv. 2025-01-03T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.31.621260

[11] Paez, E., Cadena-Ortiz, H., Ocana, E., Elias, M., Llaurens, V.. "Do tropical birds avoid evasive prey? An experimental study in Ecuadorian avian communities". bioRxiv. 2026-06-10T00:00:00.000Z. URL: https://doi.org/10.64898/2026.05.29.728647

[12] Sirois-Delisle, C., Gordon, S. C. C., Kerr, J.. "Range geography and temperature variability explain cross-continental convergence in range and phenology shifts in a model insect taxon". bioRxiv. 2025-08-14T00:00:00.000Z. URL: https://doi.org/10.1101/2024.07.22.604548

[13] Novella-Fernandez, R., Brandl, R., Chalmandrier, L., Pinkert, S., Talavera, G., Zeuss, D., Hof, C.. "Size- and colour-based mechanisms shape the phenological structure of butterfly communities". bioRxiv. 2026-03-27T00:00:00.000Z. URL: https://doi.org/10.64898/2026.03.26.713911

[14] Rubin, J., Campbell, C. J., Carvalho, A. P. S., St. Laurent, R. A., Crespo, G. I., Pierson, T. L., Guralnick, R. P., Kawahara, A. Y.. "Stronger bat predation and weaker environmental constraints predict longer moth tails". bioRxiv. 2024-12-15T00:00:00.000Z. URL: https://doi.org/10.1101/2024.05.05.590753

[15] SCHILLE, L., PLAT, N., BARBARO, L., JACTEL, H., RASPAIL, F., RIVOAL, J.-B., CASTAGNEYROL, B., MRAZOVA, A.. "Camera traps unable to determine whether plasticine models of caterpillars reliably measure bird predation". bioRxiv. 2024-11-15T00:00:00.000Z. URL: https://doi.org/10.1101/2024.10.01.616075

[16] Miyamoto, T., Kondoh, M.. "Shifts in spring migration phenology are greater in northern wintering species". bioRxiv. 2025-03-24T00:00:00.000Z. URL: https://doi.org/10.1101/2025.03.24.644893

[17] Bloch, I., Troupin, D., Toledo, S., Nathan, R., Sapir, N.. "Combining radio-telemetry and radar measurements to test optimal foraging in an aerial insectivore bird". bioRxiv. 2024-12-24T00:00:00.000Z. URL: https://doi.org/10.1101/2024.02.18.580922

How this survey was made

This survey answers one question: “How do seasonal polyphenisms in wing pigmentation across Odonata and other insects track spatiotemporal variation in predation risk from migratory insectivores?”

We keep our own collection of recent preprints and refresh it every day from bioRxiv, medRxiv, arXiv, PsyArXiv, SocArXiv and ChemRxiv. For each survey, we search that collection for recent work and keep the strongest matches: papers that use the same terms as the question, and papers on the same topic in different words.

The text was written from the titles and abstracts of those papers; the full texts were not read. Statements that draw on a particular paper are marked with a number in brackets. The reference list just above is the 17 papers whose work this text draws on, chosen from the 114 papers those searches returned; the rest of the shortlist is not printed. It is not a complete review of everything published on the question.

Treat this as a starting point and a reading list, not a peer-reviewed review. The searches behind this survey ran on October 21, 2026 and covered work from the previous two years.