Although 498 mantid species have been recorded in the Neotropical region (Anderson 2025), the natural history of most taxa remains poorly documented or simply lacking. Sustained advances in Neotropical Mantodea taxonomy and systematics over the last 25 years, together with more recent regional faunistic surveys (e.g., Lombardo and Agabiti 2001, Rivera and Vergara-Cobián 2017, Schwarz et al. 2020, Moulin and Roy 2020, Trillo et al. 2021, Moulin et al. 2021), have reinvigorated interest in documenting the biology and behavior of praying mantises (Rivera and Svenson 2020). As a result, natural history accounts are becoming increasingly common, typically incorporated as complementary narratives within taxonomic treatments or derived from structured and semi-structured observations of both wild and captive individuals (e.g., Ariza 2011, Lanna et al. 2021, Schwarz and Glaw 2021, Scherrer and Aguiar 2022, Castillo-Carrillo et al. 2022, Ferraz et al. 2025). The systematic documentation of such natural history data is essential for interpreting morphological diversity, informing evolutionary hypotheses, and contextualizing species-level adaptations within their ecological and behavioral frameworks.

Within Neotropical Mantodea, the family Thespidae stands out as the most diverse and ecologically diversified assemblage of microhabitat specialists (Rivera and Svenson 2016). Despite this diversity, the group has received comparatively little attention from researchers, even though it offers exceptional potential for uncovering biological and behavioral novelties (e.g., Rivera and Callohuari 2020). This limited attention can be partly attributable to their cryptic lifestyles and small body size, making individuals difficult to detect and study in the field, as well as challenging to maintain or breed under laboratory conditions. Although the general ecological preferences of major thespid lineages, such as their typical habitat associations and ecomorphological specializations, are reasonably well characterized (Rivera and Svenson 2020), the specific environmental and biological traits of most genera and species remain poorly documented, with existing studies concerning the ant-mimicking genus Mantillica Westwood, 1889 (Agudelo and Rafael 2014) and cryptic, substrate-masquerading Pseudomiopteryx Saussure, 1870 (Favacho and da Silva Carvalho-Filho 2023). Among the various thespid ecomorphs, lichen- and moss-mimicking taxa constitute a further exception in terms of biological knowledge, having received comparatively greater attention. Life-history studies of these forms have revealed striking, and in some cases unique, behavioral adaptations related to predation, including chromatic polyphenism and highly specialized prey-capture strategies (Rivera et al. 2011, Blanco and Salazar-E 2015, Falcón and Padrón 2019, Rivera and Callohuari 2020), underscoring both the evolutionary significance of microhabitat specialization in Thespidae and the substantial gaps that remain in our understanding of the group’s natural history.

These gaps persist even for relatively widespread species such as Musonia margheritae Battiston & Picciau, 2008. A member of the tribe Thespini, this small and inconspicuously colored mantis (Fig. 1A, B) ranges from central Ecuador to northern Chile (Rivera and Svenson 2020), occupying arid to semiarid environments along the Pacific coast and the western slopes of the Andes, where it is relatively common. Musonia margheritae also persists in landscapes extensively modified by human activity, including agricultural areas, secondary vegetation, and urban environments. Across much of its distribution, it is effectively the only thespid species present and virtually the only praying mantis thriving in the harsh conditions characteristic of the Atacama–Sechura Desert system, one of the most extreme arid environments in South America.

A distinctive attribute of the family Thespidae is the chromatic variation exhibited by species such as Pseudopogonogaster hebardi (see fig. 1 in Falcón and Padrón 2019). Chromatic polyphenism, in which a single genotype produces multiple color phenotypes (Nijhout 2003, Goldstein and Ehrenreich 2021), is among the most striking forms of phenotypic plasticity in insects and facilitates responses to spatially and temporally variable environments. By allowing individuals to track changes in substrate coloration or seasonal shifts, chromatic polyphenism can enhance camouflage efficiency, particularly in sedentary ambush predators such as mantises. Although chromatic polyphenism and background-dependent color change have been investigated in several Old World mantids, such as Mantis religiosa Linaeus, 1758, Miomantis caffra Saussure, 1871, and Creobroter gemmatus (Stoll, 1813) (Battiston and Fontana 2010, Burke and Holwell 2024, Liu and Ma 2025), comparable documentation remains scarce for Neotropical taxa. In this region, chromatic polyphenism has only recently begun to receive attention, with documented cases including Pseudopogonogaster hebardi and Metaphotina spp. (Falcón and Padrón 2019, Ferraz et al. 2025). Given its marked color variation, broad environmental tolerance, and accessibility in both natural and human-modified habitats, M. margheritae represents a suitable system for examining how chromatic variation is structured across ecological contexts.

Accordingly, the objectives of this study are to (i) provide the first comprehensive natural history account of Musonia margheritae, integrating novel ecological, behavioral, morphological, developmental, and distributional data, and (ii) document and analyze chromatic polyphenism in apterous females and juveniles using complementary observational and analytical approaches in order to examine how color variation relates to microhabitat use and environmental heterogeneity and to frame testable eco-evolutionary hypotheses regarding the role of camouflage and phenotypic plasticity in sedentary mantids inhabiting arid and seasonally variable environments.

Study area.—Fieldwork was conducted in southwestern Ecuador, specifically within the provinces of Azuay, El Oro, and Loja, from March 2022 to June 2023, across an elevational range of 10 to 250 m a.s.l. On a broader regional scale, western Ecuador and Peru encompass a mosaic of biomes, including tropical and subtropical moist broadleaf forests at higher elevations and on windward slopes, as well as tropical and subtropical dry broadleaf forests and deserts and xeric shrublands at lower elevations and in coastal areas, reflecting the pronounced environmental heterogeneity of the western Andean region (Olson et al. 2001).

Behavioral observations of Musonia margheritae were carried out in the Yunguilla Valley (3.2853761°S, 79.2748317°W; 1180 m a.s.l.) (Fig. 1C) in Azuay Province. Vegetation in this region is classified as inter-Andean dry forest (Aguirre et al. 2006) and is characterized by dominant plant species including Acacia macracantha Humb. and Bonpl. ex Willd., 1806, Anadenanthera colubrina (Vell.) Brenan. (Fabaceae), Echinopsis pachanoi Britton & Rose, 1920 (Cactaceae), Dodonaea viscosa Jacq., 1760 (Sapindaceae), and Cybistax antisyphilitica (Mart.) Mart., 1843 (Bignoniaceae) (Aguirre et al. 2006).

According to the global biome and ecoregion classification proposed by Olson et al. (2001), the study area falls within the tropical and subtropical dry broadleaf forest biome, which predominates in the inter-Andean valleys of southern Ecuador.

Specimen collection.—Specimens were collected manually during both diurnal and nocturnal surveys. Surveys were unstructured and opportunistic, and specimens were collected randomly while actively searching for individuals in the field. At each sampling locality, a fixed search effort of approximately two hours was allocated for specimen detection and collection, ensuring a comparable level of sampling effort among sites (Fig. 1D, E; Suppl. material 1). Several specimens were preserved in 96% ethanol to ensure high-quality tissue preservation for future DNA analyses; these were later processed and deposited in the Zoological Museum of the Universidad del Azuay (MZUA) in Cuenca, Ecuador. All collections were conducted under permit No. 208-2019-DPAA/MA issued by the Ministry of Environment, Water, and Ecological Transition of Ecuador.

Specimen data compilation.—We examined comparative material from the following institutions:

MZUA Zoological Museum of the Universidad del Azuay, Cuenca, Ecuador.

QCAZ Zoological Museum of the Pontificia Universidad Católica del Ecuador, Quito, Ecuador.

INABIO National Institute of Biodiversity, Quito, Ecuador.

MEKRB Klaus Raven Büller Entomology Museum, National Agrarian University La Molina, Lima, Peru.

ANSP Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania, USA.

MUSM Natural History Museum, National University of San Marcos, Lima, Peru.

MNHN Muséum national d’Histoire naturelle, Paris, France.

In addition, distributional records of M. margheritae from Ecuador, Peru, and Chile were obtained from the citizen-science platform iNaturalist (Mason et al. 2025), compiled up to January 2024 (iNaturalist 2024). Specimens deposited in MZUA were photographed for subsequent morphological analyses (see Coloration Analysis section).

Developmental stages and rearing methods.—To document development from egg to adult, adult females were maintained in captivity to obtain oothecae and newly hatched nymphs. Juvenile specimens collected in the field were subsequently maintained and reared under controlled laboratory conditions until adulthood. All oothecae and immature stages were studied in the Entomology Laboratory of the Universidad del Azuay following the rearing procedures described by Falcón and Padrón (2019). Individuals were reared separately in plastic containers under standardized laboratory conditions. For the first three developmental instars, containers measuring 10 cm in height and 5 cm in diameter were used, whereas individuals from the fourth instar onward and adults were housed in larger containers (14 cm in height and 10 cm in diameter). All rearing containers were placed on a metal shelving unit with sliding glass doors. Illumination was provided by an LED tube, and air circulation was ensured using computer fans to maintain fresh airflow and prevent excessive humidity condensation and overheating. Both lighting and ventilation systems were regulated using programmable timers to control on–off cycles.

Feeding and hydration protocols were adjusted according to developmental stage. Mantises were fed three times per week; protonymphs were provided with fruit flies (Drosophila spp.), whereas deuteronymphs and adults were fed immature crickets reared under laboratory conditions. Water was supplied once per week by manual misting.

Descriptions of post-embryonic development follow the semaphoront framework proposed by Scherrer and Aguiar (2022), which is based on external morphological transitions during mantid ontogeny. Under this framework, mantid development is described using character-defined semaphoronts rather than instar number alone. Following Scherrer and Aguiar (2022), the prenymph refers to the short-lived emergent form preceding the protonymph, or first instar. The deuteronymph includes all subsequent nymphal instars and is further subdivided into mesonymphs, representing early deuteronymphs, and metanymphs, representing late deuteronymphs. Adult individuals are treated as the final post-nymphal semaphoront.

External morphological character measurements.—External morphological characters were measured for 29 males and 20 adult females. Character selection follows Brannoch et al. (2017). The following structures were measured: Head, length and width, length and width of the frons. Body, total body length, measured from the vertex of the head to the tip of the abdomen (excluding cerci). Pronotum, total length (including prozone and metazone) and maximum width. Wings, length and maximum width of both forewings and hindwings. Prothoracic (fore) legs, length of the forecoxae; length and width of the forefemora; length of the foretibiae; total length of the foretarsi; measurements of the anterior limbs; length of antefemoral spines; length of anteroventral spines on the forefemur and foretibia; length of discoidal spines; and length of posteroventral spines on the forefemur and foretibia. Mesothoracic and metathoracic legs, length of the meso- and metathoracic coxae, femora, and tibiae, together with the total length of the tarsi. Raptorial forelegs, length of the forecoxa, forefemur, and foretibia, together with the total length of the tarsus.

Measurements were recorded in millimeters. Mean and standard deviation (SD) were calculated using the mean function in R (base package). All measurements were obtained from high-resolution photographs of museum specimens captured with an Mshot MS60 camera mounted on a Nikon SMZ 745T stereomicroscope. A 1-cm scale bar included in each image was used to calibrate pixel-to-millimeter conversions in ImageJ (Rasband 1997–2018). In addition, the length and width of ten oothecae were measured using the same protocol.

Genitalia.—Ten males and one female from multiple localities across the known Ecuadorian distribution of M. margheritae (Fig. 1F), all deposited in the MZUA collection, were dissected for genitalia study. Genitalia extraction followed the protocol of Brannoch et al. (2017) with minor modifications. The terminal abdominal segments were removed and placed in test tubes containing hot 10% potassium hydroxide (KOH) for 12 minutes to clear soft tissues. The structures were then rinsed in distilled water and cleaned under a Nikon SMZ 745T stereomicroscope equipped with an Mshot MS60 camera. Following cleaning, the genitalia were transferred to 95% ethanol for two minutes and subsequently stored in glycerin-filled vials for detailed examination. Photographs were obtained using image stacking with a Canon 5D Mark III camera and microscope lenses at 10× magnification. Terminology used in genitalia descriptions follows Brannoch et al. (2017) and Rivera and Callohuari (2020).

Analysis of color variation.—To analyze color variation in adult Musonia margheritae, 20 males and 20 females from the MZUA collection were photographed using a Canon 5D Mark III DSLR equipped with a Canon 100 mm macro lens. All photographs were taken under standardized lighting conditions against a white background to ensure consistency and comparability across specimens.

To obtain quantitative color data and minimize subjectivity in identifying hues within the RGB system, the seven dominant hexadecimal (Hex) color codes were extracted from each image in descending order of frequency. Each Hex code consists of six alphanumeric characters, with each pair representing a color value ranging from 0 to 255 (Lilley 2022).

Dominant colors were extracted using the R packages magick, tidyverse, scales, and imager. Images were first resized to 500 px and reduced to 24 colors using the image_quantize function in the magick package. The resulting color palette for each specimen, including the corresponding Hex codes, was then plotted to visualize the full range of coloration present across individuals.

Potential distribution area ofMusonia margheritae.—The potential distribution area of Musonia margheritae was estimated for Ecuador, Peru, and Chile, the countries where the species has been recorded, using the Maximum Entropy algorithm (Maxent; Phillips et al. 2006). Georeferenced occurrence records were compiled from three sources: (i) field-collected specimens, (ii) museum collections, and (iii) validated iNaturalist observations. All coordinates were carefully inspected and corrected for potential georeferencing errors using QGIS (Fig. 2A), and records lacking geographic coordinates were excluded. The final dataset was standardized in decimal degrees (WGS84) and used as input for ecological niche modeling.

Environmental predictors consisted of 19 bioclimatic variables (BIO1–BIO19) obtained from the CHELSA climate database (CHELSA Climate 2022; Table 1). All raster layers were provided in GeoTIFF format and resampled to a common spatial resolution and extent. To restrict model projections to ecologically relevant regions, environmental layers were clipped to the study area encompassing Ecuador, Peru, and Chile.

To reduce multicollinearity among environmental predictors, environmental values were randomly sampled across the study area and evaluated using variance inflation factors (VIF). Variables with VIF values greater than 10 were excluded prior to model calibration.

Model selection followed the Akaike Information Criterion (AIC; Akaike 1974), with the parameterization showing the lowest AIC value selected as the best-supported model. The resulting continuous suitability raster was exported and projected in QGIS for visualization and further spatial analyses.

Post-model analyses were conducted to examine the relationship between elevation distribution range and latitude. The relationship between elevation and latitude was explored using scatter plots. Given the expected non-linear response, a generalized additive model (GAM) was fitted to assess and visualize altitudinal trends along the latitudinal gradient. Elevation was modeled as a smooth function of absolute latitude, allowing flexible estimation of non-linear patterns. A vertical reference line at 0° latitude was included in the graphical outputs to indicate the equatorial region.

All statistical analyses and visualizations were performed in R using the ggplot2 package to generate publication-quality figures. Spatial analyses and map visualizations were conducted using R and QGIS.

Statistical analyses.—Morphological character measurements were summarized using measures of central tendency, specifically the mean and standard deviation (SD). All calculations were performed in R using the mean function from the base package and the SD function from the stats package.

Habitat in Ecuador.—Information from 228 specimens of M. margheritae was examined, including museum material, field-collected individuals, literature records, and specimens reared in captivity. These records originate from multiple localities in Ecuador, Peru, and Chile. Within Ecuador, M. margheritae is distributed in Azuay, Cañar, Guayas, El Oro, Esmeraldas, Loja, Santa Elena, Los Ríos, and Manabí provinces, confirming its broad distribution across the western slopes of the Andes and adjacent lowland regions.

Field observations revealed clear differences in habitat use among life stages and between sexes. Females and immature individuals were predominantly found on the ground, sheltered among herbaceous vegetation (Fig. 1D). In contrast, males were mainly encountered in the shrub layer or near artificial light sources during early morning hours or at night (Fig. 1E), suggesting greater mobility and a tendency to occupy higher vegetation strata. Individuals were observed at various times of day, indicating that activity is not restricted to a narrow diurnal window.

Potential distribution area ofMusonia margheritae.—Ecological niche modeling was conducted using 228 georeferenced occurrence records (Suppl. material 1). The Maxent model predicted areas of high environmental suitability for M. margheritae primarily along the Andean region of Ecuador and Peru, with suitability progressively declining toward southern latitudes in Chile (Fig. 2A). Model selection based on the Akaike Information Criterion corrected for small sample size (AICc) identified the best-supported model (AICc = 1051.04), which included eight precipitation-related variables (BIO12–BIO19). Variable importance analyses indicated that annual precipitation (BIO12) was the most influential predictor, contributing 45% to the model, followed by other precipitation-related variables. Jackknife tests further corroborated these results, showing that precipitation variables contained the highest amount of unique information, as their exclusion resulted in the greatest reduction in model gain, underscoring the dominant role of precipitation regimes in shaping the species’ environmental niche (Table 1).

The resulting potential distribution map revealed a pronounced concentration of suitable habitat along the western slopes of the Andes, encompassing the Ecuadorian provinces of Azuay, Cañar, Guayas, El Oro, Esmeraldas, Loja, Santa Elena, Los Ríos, and Manabí (Fig. 2A). Additional areas of predicted suitability were identified in northern Ecuador, along the eastern slopes of the Andes in Peru, and within adjacent coastal river valleys. Overall, the elevational range of environmentally suitable habitat extends from sea level to approximately 2,000 m.

To further examine how geographic variation influences the species’ realized niche, the altitudinal structure of predicted suitable areas was analyzed along a latitudinal gradient. Elevation was not included as a predictor in the species distribution model; instead, it was extracted post hoc from a digital elevation model to characterize elevational patterns associated with environmentally suitable conditions. When considering only areas with suitability values greater than 20%, the potential distribution of M. margheritae exhibited a broad altitudinal envelope near equatorial latitudes (Fig. 2B). Toward southern latitudes, both minimum and maximum elevations progressively declined, resulting in a marked contraction of the altitudinal range. This pattern indicates a latitudinal restriction of suitable elevational conditions toward the southern portion of the species’ distribution.

Elevation varied substantially across the examined latitudinal gradient, with the highest elevations concentrated near the Equator. Visual inspection of the scatter plot revealed a clear non-linear relationship between elevation and latitude, characterized by higher elevational values at low latitudes and a progressive decline toward higher southern latitudes. This pattern was further supported by the GAM, which detected a significant latitudinal effect. The smooth term displayed a pronounced non-linear shape, with a distinct peak in elevation near 0° latitude, followed by a marked decrease with increasing distance from the Equator. Confidence intervals around the smooth term were generally narrow across most of the latitudinal range, indicating high confidence in the estimated relationship (Fig. 2B). Collectively, these results suggest that the probability of occurrence at higher elevations increases toward equatorial regions and decreases toward higher latitudes, reflecting an altitudinal optimum associated with intermediate climatic conditions.

Development.—A total of 31 individuals were successfully reared to adulthood, originating from both captive-produced oothecae and juveniles collected in the field. The development of M. margheritae, consistent with the semaphoront-based framework proposed by Scherrer and Aguiar (2022), is described below and illustrated in Fig. 3.

Ootheca.—The ootheca (Fig. 3A) is deposited on low vegetation, typically among grasses, at heights below one meter. Attachment to the substrate occurs through a continuous layer of covering secretion that coats the entire structure, giving the surface a velvety, smooth texture; once removed, it exposes the pale, orangish-brown coloration of the external wall. The ootheca is compact, subrectangular, and slightly laterally compressed. It bears a short, filiform residual process directed upward and/or forward. The external wall distinctly reveals the individual egg chambers, which are prominently raised; each ootheca contains between 10 and 23 such chambers. The emergence area is located dorsally and consists of 10 to 17 exit openings. The lateral and transverse walls of these openings are elevated, with the transverse walls gradually decreasing in height toward the center, forming a shallow median groove. In the lateral view, the dorsal margin appears dentate or slightly serrated, with a subtle distal orientation.

Adult females produced between one and four oothecae, with size variation among successive depositions. Intervals between depositions ranged from one to two weeks and were influenced by food availability. Larger oothecae yielded more hatchlings, indicating a positive correlation between ootheca size and offspring number. Under laboratory conditions, development from ootheca deposition to hatching lasted 43–48 days (mean = 45 days; n = 3). Depending on size and deposition sequence, each ootheca produced between 15 and 35 hatchlings. In the Peruvian coast, wild egg cases are attacked by the chalcidoid microwasps Podagrion brasiliense (Torymidae, Podagrionini) (Rivera 2003) and Anastatus sp. (Eupelmidae) (J. Rivera pers. obs.).

Protonymph (first instar).—The protonymph (Fig. 3A, B) stage lasted 38 days. Protonymphs exhibit a distinctive morphology that sets them apart from later instars, including strongly contrasting banded patterns on all legs and reduced, weakly sclerotized spines on the forefemur and foretibia. Male wing pads were not externally evident at this stage. Genital segments are visible in the abdominal region but show no external sexual differentiation. Protonymphs have been observed to rely heavily on crypsis, using coloration and body ornamentation to remain inconspicuous, and they typically remain motionless for extended periods to avoid predation.

Deuteronymph (mesonymph and metanymph stages).—The combined duration of the deuteronymph (Fig. 3A, B) phase was 109 days. Although a few individuals reached the second instar, observations were limited due to high mortality, preventing adequate documentation of this stage. The mesonymph stage (third instar) lasted 27 days, during which individuals developed more pronounced foreleg spines and increased sclerotization, particularly on the proximal discoidal spine. Male wing pads expanded slightly relative to the protonymph stage. Genital segments became more evident but still lacked reliable sexual differentiation. Crypsis continued to be important, with individuals maintaining coloration and ornamentation that allowed them to blend with dry substrates.

The metanymph stage comprised the fourth and fifth instars, which lasted 33 and 49 days, respectively. During these instars, morphology became more robust and diagnostic. Foreleg spination became strongly sclerotized, and male wing pads increased substantially in size, indicating imminent development of functional wings. Genital segmentation grew increasingly prominent, and some individuals showed early external cues of sexual differentiation. Body ornamentation persisted and appeared more elaborate, enhancing camouflage within the typical dry and heterogeneous substrates of the species’ habitat.

Adult.—Adults of M. margheritae (Fig. 3A, B) show fully developed morphological structures characteristic of the species, with the foreleg spines reaching their complete form, including the sharply defined and fully sclerotized proximal discoidal spine. In males, the wings are fully formed and functional, supporting active dispersal. Genital structures are entirely differentiated and permit straightforward sex identification. The cryptic coloration and surface textures characteristic of earlier instars persist into adulthood, facilitating camouflage among leaf litter, dry soil, and vegetation. Adult longevity differed markedly between sexes. Females lived approximately two to three months under laboratory conditions, whereas males survived 30–38 days. As with all mantids, adults do not molt further, and the molt lines prominent in immature stages are absent.

Measurements of morphological characters.—Measurements of morphological characters of males, females, and oothecae are detailed in Tables 2, 3.

Genitalia description.—The configuration of the male genitalia is characteristic of Musonia (Rivera and Svenson 2020, Schwarz and Roy 2019) (Fig. 4). The species is distinguished by the shape of the distal margin of the left phallomere, which bears two dll (one rounded and the other truncated) that are divergent and separated by a moderately deep, rounded gap. Within Musonia, the relative size, position, and shape of these lobes constitute informative characters for species delimitation. The left phallomere has a curved and pointed afa, which is accompanied by a large loa. The ventral phallomere with a strongly sclerotized, almond-shaped region that is prominently visible distally along its left margin (part of the pda).

In females, the CG8 sclerite is large and exhibits irregular anterior and posterior margins. The posterior margin bears three triangular lobes of similar size, with the median lobe slightly shorter, whereas the anterior margin presents a prominent median triangular lobe flanked by two narrower and slightly curved lobes. The CX8 sclerites are narrow and weakly sclerotized and are only loosely articulated with the CG8 sclerite. The ovipositor is relatively elongated.

Coloration and chromatic polyphenism.—Seven dominant Hex colors were obtained from 40 adults (Fig. 5). The first color corresponded to the white background on which the specimens were photographed; therefore, it was removed, and the analysis considered the Hex codes starting from the second extracted color (a total of six colors), with the first being the most dominant. The color codes for each analyzed individual are presented in Fig. 5. A clear dominance of light to dark brown tones (#A18262) is evident. Greater variability in coloration patterns is also noticeable in females compared to males, resulting in pronounced chromatic polyphenism (Fig. 5B).

In addition to the primary chromatic patterns, we recorded a female exhibiting an atypical green coloration during the rainy season (April–July). This individual was collected on April 4, 2023. Brown-colored females were also collected during the same rainy period, indicating that both chromatic morphs occur concurrently during this season (Fig. 5C). This rare phenotype stands out from the brown-dominated color spectrum and suggests that pigmentation may exhibit environmentally induced plasticity. The occurrence of this green morph specifically during periods of increased humidity and vegetation cover raises the possibility that such coloration provides adaptive benefits through enhanced crypsis in greener microhabitats. Although based on a single observation, this finding highlights the potential role of seasonal cues in modulating developmental pathways related to coloration in M. margheritae and underscores the complexity of chromatic variation within the species.

This study presents the first integrative natural history account of M. margheritae, bringing together ecological observations, developmental data, morphological analyses, chromatic characterization, and ecological niche modeling. The results document pronounced variation across life stages, sexes, habitats, and geographic contexts, revealing a species that combines broad environmental tolerance with fine-scale phenotypic and ecological structuring. Rather than providing direct tests of adaptive mechanisms, the patterns described here offer an opportunity to explore how life-history traits, mobility, habitat use, and environmental heterogeneity may interact to shape phenotypic variation in a widespread Neotropical thespid.

In the sections that follow, we examine these patterns in the context of climate, development, morphology, and coloration and consider their potential ecological and evolutionary implications.

Potential distribution area ofMusonia margheritae.—Ecological niche modeling based on 228 georeferenced occurrence records revealed a pronounced concentration of suitable habitat along the western Andean slopes of Ecuador and Peru, with suitability progressively declining toward southern latitudes in Chile. In Ecuador, M. margheritae inhabits arid to semiarid environments along the western slopes of the Andes, occurring primarily in coastal ecosystems near the Pacific Ocean and in inter-Andean valleys. Its elevational range extends from sea level to approximately 2000 m, as inferred from the altitudinal envelope of areas exhibiting environmental suitability values greater than 20%. These environments largely correspond to coastal dry shrubland, coastal deciduous forest, and western montane forest ecosystems (Ron 2020).

Beyond Ecuador, the model predicts extensive suitable habitat along much of the Peruvian coast, particularly in Tumbes and the interior of Piura, with a more or less continuous band south of Piura up to Ica and becoming increasingly fragmented southward toward Tacna. Additional areas of predicted suitability were identified along the eastern slopes of the Andes in Peru and within lower coastal valleys, underscoring the role of Andean topography in structuring the species’ potential distribution.

The model identified climatically suitable areas in the northern inter-Andean valleys of Guayllabamba (Pichincha) and Chota (Imbabura), as well as near Zumba in southern Ecuador (Zamora Chinchipe). However, despite intensive sampling efforts, M. margheritae has not been recorded in these regions. Conversely, the model failed to predict suitable habitat in northern Chile, where the species has been recently confirmed (Rivera and Svenson 2020, De La Fuente 2023).

Discrepancies between predicted and observed distributions may reflect overprediction, a well-recognized limitation of presence-only ecological niche models, which infer species–environment relationships based on statistical correlations rather than mechanistic constraints (Phillips et al. 2006, Shcheglovitova and Anderson 2013, Radosavljevic and Anderson 2014). Alternatively, environmentally suitable but unoccupied areas may arise from dispersal limitations, historical or geographic barriers, or biotic interactions, all of which can constrain species ranges independently of climatic suitability (Holt 2009, Godsoe and Harmon 2012, Lomolino et al. 2012, Giannini et al. 2013).

Among the climatic predictors, precipitation-related variables dominated the best-supported model, which included eight precipitation variables (BIO12–BIO19) and was selected based on the AICc. Annual precipitation (BIO12) was the most influential predictor, contributing 45% to the model. Jackknife analyses further confirmed the dominant role of precipitation, as the exclusion of these variables resulted in the greatest reduction in model gain, indicating that they contained the highest amount of unique information. This pattern is consistent with the environmental conditions characterizing the known distribution of M. margheritae, which is associated with warm temperatures, low annual rainfall, and vegetation dominated by herbaceous and xerophytic plant communities (Ministerio del Ambiente del Ecuador 2012). Notably, much of the species’ predicted distribution within Ecuador overlaps the Tumbesian region, a recognized biodiversity hotspot with high levels of endemism (Cuesta et al. 2017) that is currently experiencing severe deforestation, habitat fragmentation, and conversion to agricultural and urban land uses (Mosandl et al. 2008, Rivas et al. 2021). Although M. margheritae appears to be relatively widespread and capable of persisting in modified landscapes, the overlap between its suitable habitat and rapidly transformed dry-forest ecosystems suggests that future conservation assessments should evaluate population connectivity, habitat loss, and regional trends before a formal IUCN category can be proposed.

On a broader geographic scale, the predicted distribution of M. margheritae reflects a strong association with climatically stable environments typical of Andean and sub-Andean regions. The overwhelming importance of precipitation variables suggests that moisture availability, rather than temperature alone, plays a central role in defining the species’ environmental niche, a pattern consistent with other Neotropical mantids and insects associated with seasonally dry ecosystems. The observed decline in habitat suitability toward southern latitudes likely reflects increasing climatic seasonality and decreasing mean temperatures, which may exceed the species’ physiological tolerances. These results highlight the importance of montane environments as potential climatic refugia, particularly under future climate change scenarios.

The elevational patterns observed across the species’ latitudinal range further support this interpretation. Although elevation was not included as a predictor in the species distribution model, it was extracted post hoc from a digital elevation model to characterize altitudinal trends associated with climatically suitable areas. When considering only areas with suitability values exceeding 20%, the potential distribution of M. margheritae exhibited a broad altitudinal envelope near equatorial latitudes that progressively contracted toward the south. Elevation varied substantially across the examined latitudinal gradient, with the highest elevations concentrated near the Equator.

Visual inspection of the scatter plot revealed a clear non-linear relationship between elevation and latitude, characterized by higher elevations at low latitudes and a progressive decline toward higher southern latitudes. This pattern was statistically supported by a GAM, which detected a significant latitudinal effect. The fitted smooth term displayed a pronounced peak in elevation near 0° latitude, followed by a marked decrease with increasing distance from the Equator. Confidence intervals around the smooth term were generally narrow across most of the latitudinal range, indicating high confidence in the estimated relationship. Collectively, these results suggest that the probability of occurrence at higher elevations increases toward equatorial regions and decreases toward higher latitudes, reflecting an altitudinal optimum associated with intermediate climatic conditions.

Although the analysis is based on occurrence data that may be affected by sampling bias, the consistency of these patterns indicates a robust biogeographic signal. Future studies integrating fine-scale climatic data, physiological tolerance experiments, and demographic or dispersal analyses would be valuable for elucidating the mechanisms underlying the observed altitudinal and latitudinal gradients in M. margheritae.

Developmental stages.—Under laboratory conditions, adult females lived, on average, twice as long as males. This difference is likely associated with the higher energetic demands of ootheca production, as suggested by our rearing observations: Both the size and number of oothecae depend on food availability, which ultimately influences the number of nymphs produced. In contrast, males exhibited shorter lifespans, likely reflecting their primary role in mate searching and reproduction. Similar sex-biased longevity patterns have been reported in species of Stagmomantis and Choeradodis (Maxwell 1998, Lücking et al. 2010). The dataset presented here represents the most comprehensive biological information currently available for any mantis species recorded in Ecuador and constitutes the first detailed account for the genus Musonia.

Genitalia.—Our analysis of male genital structures showed that they are consistent with Battiston and Picciau (2008), making species identification straightforward. Female genitalia, on the other hand, has not been documented for Musonia. Within Thespidae, female genitalia are known for only three species: Pseudomiopteryx guyanensis Chopard, 1911, P. spinifrons Saussure, 1870 (Pseudomiopteryginae) (Favacho and da Silva Carvalho-Filho 2023), and Thesprotiella simpira (Rivera and Callohuari 2020) (Thespinae) (Rivera and Callohuari 2020). In M. margheritae, the presence of a large CG8 loosely articulated to narrow, elongated CX8 sclerites mirrors the condition observed in T. simpira and contrasts sharply with that of Pseudomiopteryx spp., in which the CG8 sclerite is broadly connected to equally broad, prominent, and heavily sclerotized and texturized CX8. The ovipositor is relatively elongated, like T. simpira, contrasting with the relatively shorter and stockier ovipositor of Pseudomiopteryx spp.

Our observations suggest that structural features of the CG8 (caudogyne) and CX8 (coxa 8) sclerites may constitute an informative character system for distinguishing Pseudomiopteryginae from at least some representatives of Thespinae. At present, Thespidae classification relies primarily on male genital morphology for subfamily-level diagnosis (Schwarz and Roy 2019, Rivera and Svenson 2020), and no character systems derived from female genitalia have yet been identified that independently support subfamily delimitation within the family. The patterns observed here suggest that female genitalic morphology may represent such an overlooked source of diagnostic characters; however, broader comparative sampling across Thespidae will be necessary to evaluate the consistency and taxonomic utility of this character system.

Coloration.—With respect to coloration, males exhibited greater uniformity, whereas females displayed marked variability, demonstrating a pronounced chromatic polymorphism. A similar phenomenon has been reported in females of Pseudopogonogaster hebardi Terra, 1982 (Falcón and Padrón 2019). Such phenotypic plasticity is likely to be an adaptation to enhance crypsis in heterogeneous environments (Noor et al. 2008) and may be shaped and maintained by natural selection, particularly predation pressure from birds and reptiles (Simpson et al. 2011).

The apterism of the female likely plays a central role in the evolution of this polymorphism. Unlike males, who possess functional wings and can escape by flight, as observed in other Neotropical mantids (Crane 1952, Robinson 1969), females rely primarily on crypsis for predator avoidance. Reproductive demands may also contribute: Ootheca production requires high energetic investment (Matsura and Morooka 1983), forcing females to forage for longer periods and increasing predation risk (Balakrishnan 2012). Their longer lifespan further elevates cumulative exposure to predators. Additionally, the heterogeneous substrates these mantises inhabit do not favor uniform coloration. Because males typically occupy higher vegetation strata and disperse more effectively, they can move among substrates that better match their existing coloration (Edmunds 1972). This, in addition to their capacity to evade predators by flight, likely reduces selective pressure for the evolution of multipurpose color patterns in this sex.

Many mantis species are generalist predators capable of occupying diverse habitats; however, their sit-and-wait hunting strategy reduces mobility, especially for males, during flightless developmental stages. Limited motility may confine individuals to small home ranges and strengthen attachment to natal habitats. Such environmental consistency may facilitate the evolution of developmental plasticity (Burke and Holwell 2024).

In addition to the predominant brown chromatic patterns, we documented a female exhibiting atypical green coloration during the rainy season. This rare phenotype stands apart from the typical color spectrum of the species and suggests that pigmentation in M. margheritae may be influenced by environmentally induced plasticity. Its occurrence coincides with periods of increased precipitation, humidity, and seasonal vegetation greening, conditions that characterize the northern portion of the species’ distribution in Ecuador and northern Peru. Comparable environmentally associated chromatic variation has been reported in other species such as Mantis religiosa Linnaeus, 1758 (Mantidae), where green morphs are more frequently observed under conditions of increased humidity and vegetation cover (Battiston and Fontana 2010), supporting the interpretation that color expression in mantids may be modulated by multiple environmental factors rather than by substrate alone. In these otherwise arid environments, vegetation responds rapidly and recurrently to annual rainy seasons, potentially favoring the expression and maintenance of green chromatic morphs. In contrast, more southerly regions of Peru experience more persistent aridity, with increased rainfall largely restricted to exceptional events such as El Niño, which may limit the occurrence of such phenotypes. This pattern suggests that green coloration may be structured along precipitation gradients and ecoregional transitions within the Tumbesian region (Olson et al. 2001), where greener morphs are associated with areas of higher seasonal vegetation cover. Although based on a limited number of observations, including records from citizen science platforms such as iNaturalist (https://www.inaturalist.org/observations/192224510, https://www.inaturalist.org/observations/173877095, https://www.inaturalist.org/observations/62968599, https://www.inaturalist.org/observations/260758293), the presence of green females in Ecuador and northern Peru, coupled with their absence in more arid southern localities, supports the hypothesis that this coloration confers adaptive advantages through enhanced crypsis in greener microhabitats. Similar green morphs reported in other Musonia species (e.g., M. surinama and M. remota) further suggest that this chromatic variant may represent a conserved, environmentally responsive strategy within the genus rather than an anomalous trait.

We therefore propose that these ecological and behavioral factors collectively contribute to the expression and persistence of chromatic polymorphism in females of M. margheritae, consistent with its role as a key defensive strategy in apterous life stages. This interpretation is further consistent with the coloration variability observed in immature stages (not reported here but noted during rearing), which also lack wings and are likely to rely heavily on crypsis.

This study provides the first comprehensive natural history account of M. margheritae, integrating ecological observations, developmental data, coloration analyses, morphological measurements, genitalia documentation, and species distribution modeling. Our observations reveal that M. margheritae is a remarkably adaptable thespid, capable of thriving across a broad environmental gradient, from hyper-arid coastal scrub to seasonally dry inter-Andean valleys, and persisting in landscapes heavily modified by human activities. The reconstructed developmental sequence, based on controlled rearing and field-collected juveniles, clarifies the morphology and duration of each semaphoront, offering a robust framework for comparative ontogenetic studies within Thespidae. Furthermore, the detailed characterization of chromatic polyphenism in apterous females and juveniles highlights its central role in crypsis within structurally heterogeneous habitats, consistent with a scenario in which limited mobility and prolonged exposure to variable substrates favor plastic, rather than fixed, coloration strategies in apterous life stages, underscoring the potential of M. margheritae as a model system for investigating phenotypic plasticity in Neotropical mantises.

Our results also expand the taxonomic and morphological foundations for the species. The high-resolution documentation of male and female genitalia, illustrated here for the first time, reveals structural attributes of potential diagnostic and taxonomic value, suggesting that female genital characters may represent an underexplored source of phylogenetic signal within Thespidae, complementing current efforts to evaluate the systematic relevance of female genital morphology within Thespidae. In parallel, the species distribution model, driven primarily by precipitation-related variables, delineates the environmental parameters shaping the range of M. margheritae and predicts areas of suitable habitat that may currently be occupied or remain uncolonized due to historical, biogeographic, or dispersal constraints, indicating that climatic suitability alone may not fully determine the realized distribution of the species.

Collectively, this work establishes a much-needed baseline for understanding the ecological breadth, life-history strategies, and phenotypic diversity of M. margheritae and for exploring how these dimensions interact within an ecological framework while also opening new avenues for future research. Further studies should examine the species’ feeding preferences under natural conditions and its functional role within the ecosystems it inhabits. Additionally, the chromatic polymorphism documented here provides a foundation for experimental approaches, including controlled laboratory studies aimed at elucidating the physiological and genetic mechanisms underlying color variation. Together, these perspectives position M. margheritae as a valuable focal species for broader investigations into the evolution, systematics, and diversification of Thespidae in the Neotropics.