Biology of Tropical Birds

INTRODUCTION

The importance of classifying individuals to discrete age and sex categories in studies of avian ecology cannot be overstated because behavior, habitat use, and fitness can vary dramatically between sexes and different life stages (Vitz and Rodewald 2006, Cox et al. 2014, Székely et al. 2014, Pyle et al. 2020). Age determination in most birds relies on evaluation of age-based differences in molt patterns and plumages, but molt and plumage data are unknown in many species and families, especially in tropical regions (Johnson and Wolfe 2017, Rueda-Hernández et al. 2018).

Hummingbirds, for example, are an ubiquitous component of the Neotropical avifauna, and although numerous studies have focused on their foraging ecology and status as key pollinators (e.g., Temeles et al. 2000, Leimberger et al. 2022), their movements (e.g., Volpe et al. 2014), their cognitive abilities (e.g., González-Gómez et al. 2014, Gónzalez-Gómez and Araya-Salas 2019), and their breeding behaviors, especially in lekking species (e.g., González and Ornelas 2009, Martínez-García et al. 2013), the majority of these studies have paid, at best, just cursory attention to age and potential age-related differences within species. Only a few studies have evaluated molt patterns and age criteria in Neotropical species (e.g., Wolfe et al. 2009, Pyle et al. 2015, Johnson and Wolfe 2017, Carnes et al. 2021a), and molt descriptions and age criteria exist for just over 10% of species within Trochilidae.

Although early studies offer differing and sometimes contradictory interpretations of hummingbird molt strategies, recent work focusing on hummingbird molts has shown that the majority of species in the family follow the complex basic strategy (Sieburth and Pyle 2018, Carnes et al. 2021a, Pyle 2022a), with a preformative molt inserted within the first annual cycle, just as in the more basal Apodiformes such as swifts, although some migratory hummingbirds also undergo a prealternate molt (Dittmann and Cardiff 2009, Sieburth and Pyle 2018). Similarly, prior descriptions of hummingbird molt extents assumed that the majority of species underwent a complete preformative molt, with the exception of some basal species (Wolfe et al. 2009, Pyle et al. 2015, Johnson and Wolfe 2017), but new analyses indicate that complete preformative molts are the exception rather than the rule in this family, with most species showing a limited or partial extent to this molt (Sieburth and Pyle 2018, Carnes et al. 2021a, Pyle 2022a). Limited and partial extents to the preformative molt allow for the identification of individuals in their first annual molt cycle from older individuals by means of recognizing molt limits, the boundaries between the juvenile and formative feather generations.

One confounding factor in the study of hummingbird molts in northern Central America is that the timing of molt in hummingbird species of this region is concentrated in the period from February to May, but some individuals are reported to replace flight feathers outside of the typical season (Dickey and van Rossem 1938, Wagner 1957). The ability to identify molt and plumage stages relies in part on an understanding of when molts occur, and although previous authors have attempted to describe North American hummingbird molts and their timing through the use of banding data and specimen collections, banding data from the Neotropics is limited and museum collections often lack specimens in active molt (Rohwer et al. 2005). However, recent advancement in the use of digital photography to study molt (Viera et al. 2017, Panter and Amar 2021), paired with the creation of large citizen science projects such as eBird (Sullivan et al. 2009) and online photo repositories such as WikiAves and the Cornell Lab of Ornithology’s Macaulay Library, has resulted in an exponential increase in the number of “digital specimens” that can supplement the use of traditional specimens in molt study (Pyle 2022a).

In addition to molt patterns, a second dependable age-related characteristic in hummingbirds is the extent of corrugations along the culmen (Ortiz-Crespo 1972, Baltosser 1987, Yanega et al. 1997, Pyle 2022b). These corrugations run the length of the bill in newly fledged juveniles, and, as the bill hardens and wears, the corrugations smooth from tip to base, generally by at least 50% within nine months (Ortiz-Crespo 1972, Yanega et al. 1997). By one year of age, North American species show corrugations on a maximum of 10% of the bill (Yanega et al. 1997, Pyle 2022b). There has been relatively little study on the extent of bill corrugations in Neotropical hummingbirds, with somewhat contradictory findings, as Johnson and Wolfe (2017) suggest corrugations on 30% of the bill as a cutoff for indicating that hummingbird species in the central Amazon basin are less than nine months old, whereas Stiles and Wolf (1974) found that corrugations in Long-billed Hermit (Phaethornis longirostris) were reduced to 0% by 6 months of age in Costa Rica, and Pyle et al. (2015) found that Giant Hummingbird (Patagona gigas) and Green-backed Firecrown (Sephanoides sephanoides) showed maximums of 15 and 10% corrugations, respectively, at over one year of age. Carnes et al. (2021a) found that bill corrugations reduced to less than 10% of bill length by the time of the second prebasic molt in eight Amazonian species.

Over 60% of hummingbird species exhibit sexually dichromatic characteristics (Billerman et al. 2022), yet the age at which males attain definitive appearance is unknown in most species. Carnes et al. (2021a) found evidence of delayed plumage maturation in males of four Amazonian hummingbird species that showed either strong sexual dichromatism or sex-specific characteristics in the remiges and rectrices. The age at which males obtain a definitive appearance is unknown for Central American hummingbirds, but some genera show sex-specific characters in the flight feathers; for example, males in Pampa, Campylopterus, and Phaeochroa have thicker and more flattened rachides in the outermost primaries than females (Johnson and Wolfe 2017, Billerman et al. 2022), and other genera such as Anthracothorax and Thalurania exhibit strong sexual dichromatism (Billerman et al. 2022). Additionally, mandible coloration appears to be linked with age and sex in some hummingbird species, with older males showing larger and brighter areas of red coloration than younger males and females (Graves 2009, Stiles et al. 2017).

In order to expand knowledge of ageing and sexing criteria in hummingbirds and allow for more detailed future behavioral and ecological work, we document molt extent; age-specific molt patterns, plumages, and percent of the bill with corrugations; the presence or absence of delayed maturation in plumage and bill color of males; and molt timing in the 19 resident species of Belize, many of which range widely throughout Mexico, Central America, and northern South America (see Table 1 for species list). Our assessment includes data from banding stations and imagery from the Cornell Laboratory of Ornithology’s Macaulay Library.

METHODS

We captured hummingbirds while banding under the Monitereo de Sobreviviencia Invernal (MoSI) protocol (DeSante et al. 2005), a standardized mist-netting effort for better understanding spatial-related and habitat-related variation in the overwintering physical condition and survivorship of migratory landbirds and resident Neotropical landbirds. We collected plumage and bill data from five MoSI field sites in Belize. Black Rock Lodge (17.043°N, 89.059°W, 100 meters above sea level) included riparian forest along the Macal River, Cockscomb Basin Wildlife Sanctuary (16.781°N, 88.460°W, 67 m.a.s.l.) included advanced second-growth broadleaf forest, Shipstern Nature Reserve (18.307°N, 88.188°W, 10 m.a.s.l.) included seasonally flooded coastal woodland, Freshwater Creek Forest Reserve (18.006°N, 88.359°W, 20 m.a.s.l.) included young second-growth tropical dry forest within a matrix of active and abandoned agricultural fields, and two sites at Runaway Creek Nature Reserve (17.293°N, 88.454°W, 15 m.a.s.l.; 17.300°N, 88.449°W, 15 m.a.s.l.) included, respectively, pine savanna bordering broadleaf forest on karst hills, and broadleaf forest on karst hills bordering a seasonal wetland.

For all captured individuals, we examined for the presence of actively molting feathers in the body, tail, and wings, and documented the presence and extent of molt limits and bill corrugations. We determined sex when possible by using plumage and bill characteristics as described in Fagan and Komar (2016) and Billerman et al. (2022). Known sex characteristics are summarized in Appendix 1, as well as previously undescribed sex-related plumage traits.

When weather, capture volume, and bird condition allowed, we photographed the right wing, tail, and body of captured individuals. We recommend that hummingbirds be photographed in-hand by holding the bird low over an even, relatively dark background and out of direct sunlight so as to minimize the amount of light reflecting off of and through the wing. Direct glare and reflected light can affect the appearance of the wing feathers so that higher quality basic feathers can appear loosely textured and washed out in color like juvenile feathers, and lead to assigning an incorrect plumage class to a bird. When spreading the wing to photograph it, do so with minimal obscuring of the feathers on the upperwing surface; this often can be accomplished by holding the wing open with a finger on the underside. Cameras should be set to an image size of at least 2000 x 1500 pixels with no flash, as flash produces similar glare effects as direct sunlight.

Additionally, we followed protocols from Pyle (2022a) to review images of hummingbirds from the Cornell Lab of Ornithology’s Macaulay Library. Our analysis included Macaulay Library images of the appropriate subspecies in these 19 species from Mexico, Guatemala, and Belize that had been uploaded through January 2022. For seven species with small photographic sample sizes from this geographic area (e.g., Band-tailed Barbthroat [Threnetes ruckeri ventosus], Black-crested Coquette [Lophornis helenae]), we included all images of appropriate subspecies south to Costa Rica, and for a partially migratory species with a pattern of vagrancy (i.e., Green-breasted Mango [Anthracothorax prevostii prevostii]) we included all images from the United States. We excluded misidentified images (< 5% of the ML catalogue for each species) from our analysis and flagged these images for the eBird review process.

We only recorded data from images in which individuals could be assessed for both age and molt status, in which all remiges were visible or in active molt. We excluded images of the same individual from within the same month based on molt and plumage status, date, location, eBird checklist data, appearance, and bill pattern. Individuals that were recorded over multiple months at a single location were assessed in each month of occurrence. Because banding hummingbirds in the field was not feasible, and the photos we reviewed from the Macaulay Library did not include marked birds, we consider each photograph used in analysis as separate samples rather than separate individuals, with the exception of a few well-documented individuals.

We used these photographs in combination with previously described age characteristics of hummingbirds, such as bill corrugation extent and the buff or pale tips found on juvenile feathers in many species, to both determine age characteristics by species and verify the ages of birds in our sample. As in some other studies of molt in Neotropical birds (e.g., Ruíz-Sánchez et al. 2012), determining basic plumage characteristics for separating birds in their first molt cycle from later molt cycles does not require a large initial sample size.

We categorized age classes using the WRP system (Wolfe et al. 2010) as modified by Johnson et al. (2011) and Pyle et al. (2021). To age a bird in this context means to establish a plumage class that correlates, sometimes loosely, to the actual age of an individual. The progression of hummingbird molts, their approximate extents, and the resulting plumage classes’ corresponding WRP codes is shown in Fig. 1. Molt and plumage terminology follows Humphrey and Parkes (1959) as modified by Howell et al. (2003). Our descriptions of molt extent follow Pyle (2022b), in which “limited” refers to replacement of some but not all body feathers and no flight feathers, “partial” refers to replacement of some to all body feathers and wing coverts, but not flight feathers, and “complete” refers to replacement of all feathers.

After assigning age classes to the hummingbirds in our sample, we analyzed photographs of individuals in first-cycle formative plumage and individuals that had just begun the second prebasic molt but only replaced a few feathers, with the goal of describing the range of feathers replaced in the preformative molt within the remiges, greater coverts, primary coverts, and rectrices. We categorized replaced feathers as formative based primarily on sequence of replacement in the wing coverts and remiges as described by Pyle (2022), degree of feather wear, shape, and color/pattern.

We next compared the extent of bill corrugation by age class in order to determine the maximum corrugation amount by age class in each species, and examined the timing of molts in order to determine if any of these 19 hummingbird species showed differences in the timing of the second and definitive prebasic molts, as shown by some North American species. This examination of molt timing appeared to also show a pattern of different extents of the preformative molt between species with similar annual timing to the second and definitive prebasic molts versus those with advanced timing to the second prebasic molt. We used a generalized linear mixed effects model (Bolker et al. 2009) following a poisson distribution in r (version 4.2.3; R Core Team 2022) using the “glmer” function in the lme4 package (Bates et al. 2015) to assess the relationship between total feather replacement and molt timing. Species acted as a random effect to account for variance in feather replacement between species that is not attributable to molt timing.

RESULTS

All species showed limited to partial preformative molts, which included the replacement of at least all upperwing lesser coverts and some median coverts, and complete subsequent prebasic molts (Table S1). In Fig. 2, we present the frequency of feather replacement in the preformative molt. Bill corrugations had reduced to ≤ 10% of bill length by definitive cycle basic plumage in all 19 species (Table S1). In sexually dichromatic species, males did not attain full male-like plumage characteristics until the second prebasic molt, especially when those characteristics affected the remiges and rectrices or included large areas of iridescent plumage (Table S1). In all 5 species with sexually dichromatic coloration in the bill, males showed delayed maturation in bill color (Table S1). We describe age and sex criteria by species and provide a primer in Appendix 1. In at least two species (White-necked Jacobin [Florisuga mellivora] and Green-breasted Mango), some females in definitive-cycle basic plumage have a male-like plumage.

Male-like plumages are well-known in females of White-necked Jacobin (see Falk et al. 2021), but less well-documented in females of Green-breasted Mango. Our determination of male-like plumages in females of this species relied entirely upon photographs showing birds in a male-like plumage on nests because male hummingbirds do not participate in incubation and raising offspring. We acknowledge that this creates some uncertainties in regard to the determination of sex in this and other species as based solely on plumage characteristics, but believe that our results are still a useful starting point for studying the plumages of these species.

We examined a total of 15,715 images from the Macaulay Library catalogue of these 19 hummingbird species, which included images of sufficient quality to assess age category and molt status in 2242 individuals by month, and we also examined 522 in-hand photographs of 202 individuals by month from banding operations. This includes 505 total first-cycle individuals (prior to the second prebasic molt), 54 individuals undergoing preformative molt, 147 individuals undergoing the second prebasic molt, 1840 individuals in definitive basic plumage, and 137 individuals undergoing definitive prebasic molt. We were able to assess individuals from each species in at least 11 months of the year, with only White-necked Jacobin and Long-billed Hermit not represented in a month. Total individuals by month were highest in the November–March period, even when accounting for individuals from banding operations.

Individuals undergoing the preformative molt all showed ≥ 20% of the bill with corrugations, and 18 of 23 individuals showed ≥ 50% corrugations. Timing of the preformative molt in most species appeared to occur soon after fledging, although in some individuals this molt may initiate while still in the nest and overlap with prejuvenile molt (Fig. 3). Timing of the definitive prebasic molt in most species appears to occur either in the same period of the year as the preformative molt, or slightly afterward (Fig. 3). Nine species showed advanced timing to the second prebasic molt, while 5 species showed similar timing to the second and definitive prebasic molts (Fig. 3). Sample sizes of molting birds were too small in the remaining 5 species to determine the respective timing of molts.

Species with similar timing to the second and definitive prebasic molts replaced more greater coverts in the preformative molt (X = 0.958 greater coverts replaced, +/- 0.153) than did species with advanced timing to the second prebasic molt (X = 0.153 greater coverts replaced, +/- 0.040). We found this relationship between total feather replacement and molt timing to be significant (est. = 1.8586, std error = 0.5413, z value = 3.433, p-value = 0.0006).

Species accounts

White-necked Jacobin (N = 133; 16 first-cycle)

The preformative molt included a limited number of body feathers and 0 to 3 inner greater coverts (mean = 1.0; Fig. 2). Juvenile characteristics retained into first-cycle formative plumage include a variable number of turquoise-blue feathers on the head and part or all of the buff malar stripe (Fig. S1). The tail pattern differs by age and sex, with definitive-cycle basic males showing mostly white rectrices with thin blue edging, definitive-cycle basic “blue-morph” females showing white rectrices with thick blue edging, and first-cycle formative birds showing thick blue edging and gray centers to the central rectrices (Fig. S1). Sexual dimorphism in this species is well-documented (see Falk et al. 2021), but the age at which females attain their definitive color morphs is unknown. Two females undergoing the second prebasic molt were transitioning from male-like formative plumage to a “speckle-throated” second basic plumage (Fig. S1). Bill corrugations reduced to 0% by the time of the second prebasic molt. The second and definitive prebasic molts have similar timing (Fig. 3).

Band-tailed Barbthroat (N = 72; 20 first-cycle)

The preformative molt included 0 to 4 inner greater coverts (mean = 1.0; Fig. 2, 4A). Retained juvenile feathers have a pale tip and are dull in color, rather than metallic bronze-green, but caution is warranted because basic inner secondaries have narrow buff edging (Fig. 4A-B, S2). Juvenile rectrices are narrow and sharply pointed in comparison to basic rectrices (Fig. 5A-B). Bill corrugations reduced to 0% by the second prebasic molt.

Long-billed Hermit (N = 62; 27 first-cycle)

The preformative molt included 1 to 6 inner greater coverts (mean = 2.24; Fig. 2, 4C). Retained juvenile feathers had pale or buff tips and were dull in color rather than a metallic green (Fig. 4C-D, S3). Juvenile rectrices are slightly narrower and looser in texture than basic rectrices (Fig. 5C-D). Bill corrugations reduced to 0% by the second prebasic molt. The second and definitive prebasic molts have similar timing (Fig. 3).

Stripe-throated Hermit, Phaethornis striigularis (N = 68; 13 first-cycle)

The preformative molt included 0 to 6 greater coverts (mean = 3.3; Fig. 2). Retained juvenile feathers are duller in coloration than the greenish-brown formative feathers and had pale or buff tips, but caution is warranted as basic inner secondaries also have narrow pale tipping (Fig. S4). Juvenile rectrices are narrower and more pointed in shape than basic rectrices (Fig. 5E-F). With the exception of one individual with 5% corrugations, bill corrugations reduced to 0% by definitive-cycle basic plumage.

Brown Violetear, Colibri delphinae (N = 137; 43 first-cycle)

The preformative molt included 0 to 3 inner greater coverts (mean = 0.15; Fig. 2, S5). Retained juvenile wing coverts had a pale tip, especially visible on the greater coverts and sometimes in the inner secondaries (Fig. S5). Bill corrugations reduced to 0% by the second prebasic molt. The second and definitive prebasic molts have similar timing (Fig. 3).

Purple-crowned Fairy, Heliothryx barroti (N = 114; 19 first-cycle)

The preformative molt included 0 to 6 inner greater coverts (mean = 0.62; Fig. 2, 4E). Juvenile wing coverts are a comparatively dull bronze-green, rather than the metallic emerald green of basic feathers (Fig. 4E-F, S6). Rectrix shape and pattern differs by age and sex, with definitive-cycle basic males having a short, broad tail with the outer three rectrices white and the inner two black; females having a long, narrow tail with the outer three rectrices white with one black bar, and the inner two black; and first-cycle males having a tail of intermediate length with a color pattern matching that of definitive-cycle basic males (Fig. 5G-I). Bill corrugations reduced to 0% by the second prebasic molt. First-cycle formative males retain some brown juvenile feathers in the head and do not show the full male-like head pattern (Fig. S6). The definitive prebasic molt occurs May–July, while the second prebasic molt occurs between August–November (Fig. 3).

Green-breasted Mango (N = 224; 29 first-cycle)

The preformative molt included 0 to 2 inner greater coverts (mean = 0.20; Fig. 2, S7). Juvenile wing coverts are comparatively duller green than basic feathers, with less iridescence (Fig. S7). Males showed delayed plumage maturation, with iridescent body plumage not appearing until the second prebasic molt (Fig. S7). Four definitive-cycle basic females on nests showed male-like body plumage, but no individuals showed intermediate plumage (Fig. S7). Bill corrugations reduced to 0% by the second prebasic molt. The definitive prebasic molt occurs from June to November, but the second prebasic molt can occur throughout the year, likely resulting from the timing of when birds fledge and protracted individual molts (Fig. 3). One vagrant individual overwintering in the southeastern United States showed a protracted second prebasic molt with a duration ≥ 133 days.

Black-crested Coquette (N = 150; 15 first-cycle)

The preformative molt included 0 to 2 inner greater coverts (mean = 0.17; Fig. 2, S8). Juvenile feathers are comparatively duller green than basic feathers, with less iridescence (Fig. S8). Males showed delayed maturation in both plumage and bill color, with the upper mandible not becoming mostly red until during or after the second prebasic molt (Fig. S8). Bill corrugations reduced to 0% by definitive-cycle basic. The second and definitive prebasic molts occur in similar time periods, but the peaks of the molting periods are offset, with the second prebasic molt peaking later in the year (Fig. 3).

Canivet’s Emerald, Cynanthus canivetii (N = 141; 18 first-cycle)

All greater coverts were retained in the preformative molt, leading to molt limits between the median and greater coverts (Fig. 2, S9). Juvenile feathers are comparatively duller green than basic feathers, with less iridescence (Fig. S9). Males showed delayed maturation in both plumage and bill color, with the body plumage showing partial and patchy iridescence and the bill not becoming fully red until after the second prebasic molt (Fig. S9). Bill corrugations reduced to 0% by definitive-cycle basic. The definitive prebasic molt occurs from June to August, while the second prebasic molt occurs from July to December, ~7 months after the preformative molt (Fig. 3).

Wedge-tailed Sabrewing, Pampa curvipennis (N = 136; 41 first-cycle)

The preformative molt included 0 to 6 inner greater coverts (mean = 1.31; Fig. 2, 4G). Juvenile wing coverts are a dull green, while formative and basic wing coverts are a metallic green (Fig. 4G-H, S10). Tail pattern differs by age and sex, with definitive-cycle basic males having long, graduated green and black rectrices with a rounded tip, females with a more-rounded tail shape and pale gray tips to the outer two rectrices, and some first-cycle males with narrow, pointed rectrices showing variable amounts of white in the outer two rectrices (Fig. 5J-L). Males showed delayed plumage maturation, with first-cycle formative individuals not showing modified rachides to the outer primaries until the replacement of the juvenile primaries during the second prebasic molt, and showing intermediate or female-like tail patterns. Bill corrugations reduced to 0% by definitive-cycle basic.

Violet Sabrewing, Campylopterus hemileucurus (N = 105; 6 first-cycle)

The preformative molt included 0 to 6 inner greater coverts (mean = 1.44; Fig. 2, 4I-J). Juvenile wing coverts are a dull green, while formative and basic wing coverts are a bright metallic green (Fig. 4I-J, S11). Tail pattern differs by age and sex: definitive-cycle basic males have broad black rectrices with large white tips on the outer three, definitive-cycle basic females have broad green to black rectrices with broad white tips, first-cycle males have broad green to black rectrices with large white tips on the outer three, and first-cycle females have narrow, pointed green to black rectrices with white tipping on the outer four (Fig. 5M-N). Males showed delayed plumage maturation, with first-cycle formative individuals showing incomplete violet iridescence and lacking modified rachides to the outer primaries (Fig. S11). Bill corrugations reduced to 0% by the second prebasic molt. The second and definitive prebasic molts show similar timing (Fig. 3).

Stripe-tailed Hummingbird, Eupherusa eximia (N = 64; 8 first-cycle)

The preformative molt included 0 to 2 inner greater coverts (mean = 0.18; Fig. 2, S12). Juvenile wing coverts are comparatively duller green than basic feathers, with less iridescence, and juvenile secondaries are a washed-out rufous in comparison to the bright rufous coloration of basic secondaries (Fig. S12). Males showed delayed plumage maturation, with first-cycle formative individuals showing incomplete iridescence, especially in the gorget (Fig. S12). Bill corrugations reduced to 0% by the second prebasic molt.

Crowned Woodnymph, Thalurania colombica (N = 76; 9 first-cycle)

The preformative molt included 0 to 3 greater coverts (mean = 1.13; Fig. 2, S13). Juvenile wing coverts are a dull green in color, contrasting with the metallic green formative coverts of females and violet formative coverts of males (Fig. S13). Males showed delayed plumage maturation, with first-cycle formative individuals showing duller coloration and incomplete iridescence, especially within the gorget, than definitive-cycle basic individuals (Fig. S13). Bill corrugations reduced to 0% by definitive-cycle basic.

Scaly-breasted Hummingbird, Phaeochroa cuvierii (N = 84; 30 first-cycle)

The preformative molt included 0 to 3 inner greater coverts (mean = 0.45; Fig. 2, 6L). Retained juvenile wing coverts have pale tips (Fig. 6L-M, S14). Juvenile rectrices are narrow and loosely textured, with convex white tips, while basic rectrices are broad with a jagged pattern to the white tips (Fig. 5O-P). Males showed delayed plumage maturation, with first-cycle formative individuals not showing modified rachides in the outer primaries (Fig. 6L, S14). Bill corrugations reduced to 0% by definitive-cycle basic. The second and definitive prebasic molts showed similar timing (Fig. 3).

Azure-crowned Hummingbird, Saucerottia cyanocephala (N = 178; 27 first-cycle)

All greater coverts were retained in the preformative molt, leading to molt limits between the median and greater coverts (Fig. 2, S15). Juvenile feathers are comparatively duller green than basic feathers, with less iridescence (Fig. S15). Bill corrugations reduce to 0% by the second prebasic molt. The definitive prebasic molt occurs from January to June, while the second prebasic molt occurs between August and April, ~7–10 months following the preformative molt (Fig. 3).

Cinnamon Hummingbird, Amazilia rutila (N = 168; 39 first-cycle)

With the exception of one individual that replaced the innermost greater covert, all greater coverts were retained in the preformative molt, leading to molt limits between the median and greater coverts (Fig. 2, S16). Juvenile feathers are comparatively duller green than basic feathers, with less iridescence (Fig. S16). Males showed delayed maturation in bill color, with most first-cycle formative males showing mixed black and pink coloration to the upper mandible (Fig. S16). Bill corrugations reduced to 0% by the second prebasic molt. The definitive prebasic molt occurs from March to July, while the second prebasic molt occurs December–May (Fig. 3).

Buff-bellied Hummingbird, Amazilia yucatanensis (N = 59; 18 first-cycle)

The preformative molt included 0 to 3 inner greater coverts (mean = 0.2; Fig. 2, 6N). Juvenile wing coverts are comparatively dull green, rather than the bright metallic green of basic feathers (Fig. 6N-O, S17). Juvenile rectrices are narrow with dusky bronze-green tipping, and basic rectrices are broad, with well-defined bronze-green edging (Fig. 5Q-R). Males showed delayed maturation in bill color, with most first-cycle formative males showing mixed red and black coloration to the upper mandible (Fig. S17). Bill corrugations reduced to 0% by definitive-cycle basic. Most individuals underwent the definitive prebasic molt March–May, while the second prebasic molt peaked in June–September (Fig. 3).

Rufous-tailed Hummingbird, Amazilia tzacatl (N = 290; 81 first-cycle)

The preformative molt included 0 to 3 inner greater coverts (mean = 0.13; Fig. 2, 6P). Juvenile wing coverts are a dull green, rather than the metallic bronze-green of basic feathers (Fig. 6P-Q, S18). Juvenile rectrices have washed-out bronze edging, sometimes broken at the tip, while basic rectrices have comparatively broad bronze edging (Fig. S18). Males had delayed maturation in plumage and bill color, with the gorget often having a “frosted” appearance rather than the solidly iridescent green of definitive-cycle basic males, and the upper mandible regularly showing limited red coloration, rather than the extensive red of definitive-cycle basic males (Fig. S18). Bill corrugations typically reduced to 0% by definitive-cycle basic, but 2.67% of definitive-cycle basic individuals showed a maximum of 10% corrugations. The definitive prebasic molt occurred April–September, with a peak in June–July, while the second prebasic molt occurred March–Aug, with peaks in April–May and July (Fig. 3).

White-bellied Emerald, Chlorestes candida (N = 170; 46 first-cycle)

The preformative molt included 0 to 2 inner greater coverts (mean = 0.04; Fig. 2, 6R). Juvenile wing coverts are a washed-out dull green, while formative and basic wing coverts are a metallic green (Fig. 6R-S, S19). Juvenile rectrices are narrow and bronze-green with a narrow dark subterminal band and small white tip, while basic rectrices are broad and bronze in color, with a more well-defined subterminal band and larger white tip (Fig. 5S-T). Bill corrugations typically reduced to 0% by definitive-cycle basic, but 4.42% of definitive-cycle basic individuals showed a maximum of 10% corrugations. The definitive prebasic molt occurred May–August, while the second prebasic molt occurred July–February (Fig. 3).

DISCUSSION

We showed that these 19 Central American hummingbird species can be aged and sexed by molt and plumage patterns, bill coloration, and bill corrugations, much as other recent studies of hummingbirds in temperate and tropical regions have shown (Johnson and Wolfe 2017, Carnes et al. 2021a, Pyle 2022a). These hummingbird species show similarities in their molts to other species from their respective clades in Peru and North America, with differences in timing likely resulting from differences in breeding regimes as correlated with latitude, seasonality, and phenologies of flowering plants.

Early studies of hummingbird molt in the region noted multiple peaks of molting birds within the calendar year (Dickey and van Rossem 1938, Wagner 1957), but did not recognize that there could be differential timing to the second and definitive prebasic molts, as was the case for some hummingbirds studied by Pyle (2022a). There is little available information on the timing of the preformative molt in tropical hummingbirds. Many recently fledged hummingbirds appear to already be in first-cycle formative plumage, and the high percentage of bill corrugations on individuals undergoing the preformative molt suggests that this molt can overlap with the prejuvenile molt in some individuals, as occurs in some Parulidae species (e.g., Wilson’s Warbler, Cardellina pusilla; Pyle 2022b). Previous observers could have mistakenly assumed that these overlapping molts in hummingbirds were a single molt episode, similar to early descriptions of the prealternate molts in some migratory flycatchers (see Carnes et al. 2021b, Pyle and Carnes 2022), leading to the misinterpretation of the second prebasic molt several months later as a complete preformative molt.

The advanced timing to the second prebasic molt shown by some species may have evolved out of a need to breed early in life. Other recent studies of hummingbird molt have found an advanced timing to the second prebasic molt in species in which males produce mechanical sounds with the remiges and/or rectrices during aggressive interactions and courtship displays (Sieburth and Pyle 2018, Pyle 2022a). The shape and structure of basic flight feathers appears essential to the production of mechanical sounds in display dives (Clark et al. 2011, 2018), indicating that males with juvenile flight feathers have little or no reproductive success. In the Bee clade, males are much smaller than females, engage in display dives and highly aggressive behavior, and have lower annual survivorship and shorter lifespans than females (Calder et al. 1983, Mulvihill et al. 1992). A potential loss of productivity in a year of what is already a reduced lifespan has likely driven the evolution of the advanced molt timing in the Bee clade. Two of our study species with advanced timing to the second prebasic molt (Black-crested Coquette and Canivet’s Emerald) are from genera where males are known to engage in display flights (Stiles and Skutch 1989), and others, such as the four species from Group D of the Emerald clade, are known to be highly aggressive (Billerman et al. 2022). In contrast, males of all age classes in Long-billed Hermit, which has similar timing to the second and definitive prebasic molts, can have annual survivorship of 60% or higher (Stiles 1992), likely reducing the need for successful breeding in the first year of life. Other hummingbird species with similar timing to the second and definitive prebasic molts may show similar high annual survivorship.

Additionally, species with an advanced timing to the second prebasic molt replace fewer greater coverts in the preformative molt than do species with similar timing to the second and definitive prebasic molts. In most avian families, more extensive molts are correlated with increased solar exposure and habitat abrasiveness (Pyle 1998, Pyle 2008, Carnes et al. 2021b, Guallar et al. 2021, Pyle and Carnes 2022); however, hummingbird species in this study with similar timing to the second and definitive prebasic molts and more extensive preformative molts are regularly species of forest understory, with limited solar exposure, while species with an advanced second prebasic molt are all species of forest canopy, forest edge, and open habitats (Billerman et al. 2022), which have greater solar exposure. Because feathers are on the body for less than a calendar year in species with an advanced second prebasic molt, there may be less need for these species to replace juvenile greater coverts in the preformative molt than there is for species with similar timing to the second and definitive prebasic molts.

Bill corrugations were reduced or absent after first-cycle formative plumage in all 19 species, similar to temperate North American species and other tropical species (Stiles and Wolf 1974, Yanega et al. 1997, Carnes et al. 2021a, Pyle 2022b), providing a useful age criteria in situations where molt limits are difficult to detect, such as photographs in which harsh lighting washes out the iridescent quality of replaced feathers. Assuming consistent rates of bill smoothing over a period of 5–9 months in most species, the presence of corrugations in photos of some individuals undergoing the second prebasic molt in species with an advanced timing to this molt suggests that extensive corrugations on definitive-cycle basic individuals may indicate that the bird is in second-cycle basic plumage. It is also possible that some definitive-cycle basic individuals may keep a small degree of corrugations for the entire lifespan, as in some North American species (Yanega et al. 1997, Pyle 2022b). Similarly, in species with sexual dichromatism to the bill, definitive-cycle basic males with less-extensive areas of bright color likely indicate a bird in second-cycle basic plumage, as in some species of the southwestern United States such as in the genera Cynanthus, Basilenna, Leucolia, and Amazilia (Pyle 2022a).

Males of all species with sexual dichromatism showed delayed maturation in their plumage and/or bill characteristics and did not achieve full male-like characteristics until at least the second prebasic molt, similar to the delayed maturation of male hummingbirds in other regions (Carnes et al. 2021a, Pyle 2022a). In addition to the dramatic degree of sexual dichromatism present in some of our study species (e.g., Black-crested Coquette, Violet Sabrewing, Crowned Woodnymph), others (e.g., Buff-bellied Hummingbird, Rufous-tailed Hummingbird) show a more cryptic degree of dichromatism, with differences present in the bill coloration and patterning, and the plumage of the head and throat. Minor sex-based differences in coloration and patterning should be looked for in hummingbird species that are currently thought to be monochromatic. In White-necked Jacobin and Green-breasted Mango, which both have significant sexual dichromatism, some definitive-cycle basic females exhibited male-like body plumage. True polymorphism appears to be rare in female hummingbirds, with definitive evidence for this trait only in White-necked Jacobin and rare individuals in Anthracothorax (Clark et al. 2022). The first-cycle juvenile plumage of White-necked Jacobin is similar to the definitive-cycle basic plumage in males, and first-cycle females and definitive-cycle basic females of the male-like “blue” morph are found to experience less social aggression than definitive-cycle basic females of the “speckle-throated” morph; plumage type in this species appears to remain set after the second prebasic molt (Bleiweiss 1985, Falk et al. 2021). Unlike White-necked Jacobin, which shows a male-like plumage in juvenile and first-cycle formative plumages, Green-breasted Mango has a more female-like plumage in these plumages, meaning that only older females obtain male-like plumage, similar to older females in other strongly sexually dichromatic tropical birds such as manakins (Doucet et al. 2007, Scholer et al. 2022), which may be the result of altered hormone levels (Kimball and Ligon 1999). Different selective pressures likely affect the development of male-like plumage in females of White-necked Jacobin and Green-breasted Mango.

Photographic collections tend to be biased toward unusual and visually appealing individuals (Zbyryt et al. 2021, Pyle 2022a), and we encourage those contributing images to the Macaulay Library and eBird to include birds of all age and sex classes, and of birds in molt or worn plumages. Definitive-cycle basic males and females with male-like plumage, which are more visually appealing to many photographers, made up 56.1% of the images in our study species with easily discernable sexual dichromatism, with one species (Violet Sabrewing) as high as 80.2%. Also, overall sample sizes were small for species that do not frequent feeders and/or prefer dimly-lit forest understory habitats, and relatively few photographs were available of birds in active molt. These biases for sex, behavior, and appearance, along with higher image totals by month for the November–March period, which is the busiest portion of the year for tourism in the study region, suggests that Macaulay Library imagery from tropical regions may currently be biased by patterns of bird watcher behavior. This could be offset by encouraging photographic contributions from local birders throughout the annual cycle. Additionally, researchers using imagery from photographic libraries such as the Macaulay Library must be aware of limitations, including at least a small proportion of misidentified photos, numerous low-quality images, and relatively few photos showing birds in positions that allow for analysis.

Mark-recapture programs using appropriate-sized bands could further elucidate the progression of plumages on an individual basis, show the exact timing of delayed maturation, and reveal rates of females with male-like plumage characteristics. Additional banding studies will likely show minor differences with our data, especially in regard to bill corrugations by age and molt extent in species that rarely replace greater coverts, and could show if there are any variations in these characteristics within species with large geographic ranges. Most species in our study were not undergoing molt during the MoSI time frame, and banding efforts occurring in other portions of the annual cycle could obtain larger samples of molt and draw a better picture of molt timing in relation to breeding.

Further efforts are necessary to determine if males of species with an advanced timing to the second prebasic molt produce mechanical sounds or have reduced survivorship in comparison to females, as in North American species of the Bee clade (Calder et al. 1983, Mulvihill et al. 1992). Additionally, although photographs of females on nests are difficult to evaluate, all such individuals that we were able to age appeared to be in definitive-cycle basic plumage, implying that these hummingbird species do not breed prior to the second prebasic molt. Pyle (2022a) suggested that in one species with an advanced second prebasic molt (Violet-crowned Hummingbird, Leucolia violiceps), second-cycle basic individuals could attempt breeding after the breeding period of older individuals. Further study is needed to determine age at first breeding in these and other hummingbird species, and if there is differential individual timing to breeding according to age.

LITERATURE CITED

Baltosser, W. H. 1987. Age, species, and sex determination of four North American Hummingbirds. North American Bird Bander 12:151-166.

Bates, D., M. Mächler, B. Bolker, and S. Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1-48. https://doi.org/10.18637/jss.v067.i01

Billerman, S. M., B. K. Keeney, P. G. Rodewald, and T. S. Schulenberg, editors. 2022. Birds of the world. Cornell Laboratory of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bow

Bleiweiss, R. 1985. Iridescent polychromatism in a female hummingbird: is it related to feeding strategies? Auk 102:701-713.

Bolker, B. M., M. E. Brooks, C. J. Clark, S. W. Geange, J. R. Poulsen, M. H. H. Stevens, and J.-S. S. White. 2009. Generalized linear mixed models: a practical guide for ecology and evolution. Trends in Ecology & Evolution 24:127-135. https://doi.org/10.1016/j.tree.2008.10.008

Calder, W. A. III, N. M. Waser, S. M. Hiebert, D. W. Inouye, and S. Miller. 1983. Site-fidelity, longevity, and population dynamics of Broad-tailed Hummingbirds: a ten year study. Oecologia 56:359-364. https://doi.org/10.1007/BF00379713

Carnes, B. H., M. Antezana Aponte, R. Huayanca Munarriz, and A. E. Martínez. 2021a. Preformative molt extant and ageing and sexing criteria for eight hummingbirds in southeastern Peru. Ornitología Neotropical 32:106-115. https://doi.org/10.58843/ornneo.v32i2.867

Carnes, B. H., C. M. Godwin, K. R. Foster, and P. Pyle. 2021b. Clarification of molt strategies in three Empidonax flycatchers. Wilson Journal of Ornithology 133:34-42. https://doi.org/10.1676/20-00109

Clark, C. J., D. O. Elias, and R. O. Prum. 2011. Aeroelastic flutter produces hummingbird feather songs. Science 333:1430-1433. https://doi.org/10.1126/science.1205222

Clark, C. J., J. A. McGuire, E. Bonaccorso, J. S. Berv, and R. O. Prum. 2018. Complex coevolution of wing, tail, and vocal sounds of courting male bee hummingbirds. Evolution 72:630-646. https://doi.org/10.1111/evo.13432

Clark, C. J., B. Robinson, and J. V. Remsen, Jr. 2022. Female plumage polymorphism is rare in hummingbirds. Journal of Ornithology 163:735-748. https://doi.org/10.1007/s10336-022-01975-4

Cox, A. W., F. R. Thompson III, A. S. Cox, and J. Faaborg. 2014. Post‐fledging survival in passerine birds and the value of post‐fledging studies to conservation. Journal of Wildlife Management 78:183-193. https://doi.org/10.1002/jwmg.670

DeSante, D. F., T. S. Sillett, R. B. Siegel, J. F. Saracco, C. A. Romo de Vivar Alvarez, S. Morales, A. Cerezo, D. Kaschube, and B. Milá. 2005. MoSI (Monitoreo Sobrevivencia Invernal): Assessing habitat-specific overwintering survival of neotropical migratory landbirds. Pages 926-936 in C. J. Ralph and T. D. Rich, editors. Bird conservation implementation and integration in the Americas. General Technical Report PSW-GTR-191. U.S. Forest Service, Pacific Southwest Research Station, Albany, California, USA

Dickey, D. R., and A. J. van Rossem. 1938. The birds of El Salvador. Field Museum of Natural History Zoological Series, Chicago, Illinois, USA. https://doi.org/10.5962/bhl.title.3068

Dittmann, D. L., and S. W. Cardiff. 2009. The alternate plumage of the Ruby-throated Hummingbird. Birding 41:32-35.

Doucet, S. M., D. B. McDonald, M. S. Foster, and R. P. Clay. 2007. Plumage development and molt in Long-tailed Manakins (Chiroxiphia linearis): variation according to sex and age. Auk 124:29-43. https://doi.org/10.1093/auk/124.1.29

Fagan, J., and O. Komar. 2016. Field guide to birds of Northern Central America. Houghton Mifflin Harcourt, New York, New York, USA.

Falk, J. J., M. S. Webster, and D. R. Rubenstein. 2021. Male-like ornamentation in female hummingbirds results from social harassment rather than sexual selection. Current Biology 31:4381-4387. https://doi.org/10.1016/j.cub.2021.07.043

González, C., and J. F. Ornelas. 2009. Song variations and persistence of song neighborhoods in a lekking hummingbird. Condor 111:633-640. https://doi.org/10.1525/cond.2009.090029

González-Gómez, P. L., and M. Araya-Salas. 2019. Perspectives on the study of field hummingbird cognition in the neotropics. Pages 199-212 in J. C. Reboreda, V. D. Fiorini, and D. T. Tuero, editors. Behavioral ecology of Neotropical birds. Springer, Cham, Switzerland. https://doi.org/10.1007/978-3-030-14280-3_10

González-Gómez, P. L., N. Madrid-Lopez, J. E. Salazar, R. Suárez, P. Razeto-Barry, J. Mpodozis, F. Bozinovic, and R. A. Vásquez. 2014. Cognitive ecology in hummingbirds: the role of sexual dimorphism and its anatomical correlates on memory. PLoS ONE 9:e90165. https://doi.org/10.1371/journal.pone.0114671

Graves, G. R. 2009. Ontogeny of bill color in male streamertail hummingbirds (Trochilus). Journal of Caribbean Ornithology 22:44-47.

Guallar, S., R. Rueda-Hernández, and P. Pyle. 2021. Evolution of the preformative molt in Cardinalidae correlates with transitions from forest to open habitats. Ornithology 138(1):ukaa070. https://doi.org/10.1093/ornithology/ukaa070

Howell, S. N. G., C. Corben, P. Pyle, and D. I. Rogers. 2003. The first basic problem: a review of molt and plumage homologies. Condor 105:635-653. https://doi.org/10.1093/condor/105.4.635

Humphrey, P. S., and K. C. Parkes. 1959. An approach to the study of molts and plumages. Auk 76:1-31. https://doi.org/10.2307/4081839

Johnson, E. I., and J. D. Wolfe. 2017. Molt in Neotropical birds: life history and aging criteria. CRC, Boca Raton, Florida, USA. https://doi.org/10.4324/9781315119755

Johnson, E. I., J. D. Wolfe, T. B. Ryder, and P. Pyle. 2011. Modifications to a molt-based ageing system proposed by Wolfe et al. (2010). Journal of Field Ornithology 82:422-424. https://doi.org/10.1111/j.1557-9263.2011.00345.x

Kimball, R. T., and J. D. Ligon. 1999. Evolution of avian plumage dichromatism from a proximate perspective. American Naturalist 154:182-193. https://doi.org/10.1086/303228

Leimberger, K. G., B. Dalsgaard, J. A. Tobias, C. Wolf, and M. G. Betts. 2022. The evolution, ecology, and conservation of hummingbirds and their interactions with flowering plants. Biological Reviews 97:923-959. https://doi.org/10.1111/brv.12828

Martínez-García, V., C. Lara, and J. F. Ornelas. 2013. Lek characteristics and the static male song of the Green Violet-Ear (Colibri thalassinus) during a 3-year study in a temperate forest of central Mexico. Ornitología Neotropical 24:183-200.

Mulvihill, R. S., R. C. Leberman, and D. S. Wood. 1992. A possible relationship between reversed sexual size dimorphism and reduced male survivorship in the Ruby-throated Hummingbird. Condor 94:480-489. https://doi.org/10.2307/1369220

Ortiz-Crespo, F. I. 1972. A new method to separate immature and adult hummingbirds. Auk 89:851-857. https://doi.org/10.2307/4084114

Panter C. T., and A. Amar. 2021. Sex and age differences in the diet of the Eurasian Sparrowhawk (Accipiter nisus) using web‐sourced photographs: exploring the feasibility of a new citizen science approach. Ibis 163:928-947. https://doi.org/10.1111/ibi.12918

Pyle, P. 1998. Eccentric first year molt patterns in certain Tyrannid flycatchers. Western Birds 29:29-35.

Pyle, P. 2008. Identification guide to North American birds: Part 2. Slate Creek Press, Bolinas, California, USA.

Pyle, P. 2022a. Examination of images from Macaulay Library images to determine avian molt strategies: a case study on hummingbirds. Wilson Journal of Ornithology 134:52-65. https://doi.org/10.1676/21-00012

Pyle, P. 2022b. Identification guide to North American birds, Part 1, Second edition. Slate Creek Press, Forest Knoll, California, USA.

Pyle, P., and B. H. Carnes. 2022. Molt strategies by age and subspecies in the Willow Flycatcher. Western Birds 53:216-231. https://doi.org/10.21199/WB53.3.4

Pyle, P., A. Engilis Jr., and D. A. Kelt. 2015. Manual for ageing and sexing birds of Bosque Fray Jorge National Park and northcentral Chile, with notes on range and breeding seasonality. Special Publication of the Occasional Papers of the Museum of Natural Science Louisiana State University, Baton Rouge, Louisiana, USA.

Pyle, P., K. R. Foster, C. M. Godwin, D. R. Kaschube, and J. F. Saracco. 2020. Yearling proportion correlates with habitat structure in a boreal forest landbird community. PeerJ 8:e8898. https://doi.org/10.7717/peerj.8898

Pyle, P., M. Gahbauer, E. I. Johnson, T. B. Ryder, and J. D. Wolfe. 2021. Application of a global age-coding system (“WRP”), based on molts and plumages, for use in demographic and other studies of birds. Ornithology 139(1):ukab063. https://doi.org/10.1093/ornithology/ukab063

R Core Team. 2022. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Rohwer, S., L. K. Butler, and D. Froelich. 2005. Ecology and demography of east-west differences in molt scheduling of Neotropical migrant passerines. Pages 87-105 in R. Greenberg and P. P. Marra, editors. Birds of two worlds: the ecology and evolution of migration. Johns Hopkins University Press, Baltimore, Maryland, USA.

Rueda-Hernández, R., S. Guallar, A. Ruiz-Sánchez, and P. Pyle. 2018. The molt issue: Where do we go from now? Ornitología Neotropical 29:S1-S2. https://doi.org/10.58843/ornneo.v29i2.390

Ruíz-Sánchez, A., R. Rueda-Hernández, S. Guallar, and P. Pyle. 2012. Age determination of the Spot-breasted Wren and White-breasted Wood-Wren using molt limits. North American Bird Bander 37:93-100.

Scholer, M. N., J. C. Kennedy, B. H. Carnes, and J. E. Jankowski. 2022. Preformative molt, plumage maturation, and age criteria for 7 species of manakins. Wilson Journal of Ornithology 133:379:389. https://doi.org/10.1676/20-00125

Sieburth, D., and P. Pyle. 2018. Evidence for a prealternate molt-migration in the Rufous Hummingbird and its implications for the evolution of molts in Apodiformes. Auk 135:495-505. https://doi.org/10.1642/AUK-17-231.1

Stiles, F. G. 1992. Effects of a severe drought on the population biology of a tropical hummingbird. Ecology 73:1375-1390. https://doi.org/10.2307/1940683

Stiles, F. G., J. V. Remsen, Jr., and J. A. McGuire. 2017. The generic classification of the Trochilini (Aves: Trochilidae): reconciling taxonomy with phylogeny. Zootaxa 4353:401-424. https://doi.org/10.11646/zootaxa.4353.3.1

Stiles, F. G., and A. F. Skutch. 1989. A guide to the birds of Costa Rica. Cornell University Press, Ithaca, New York, USA.

Stiles, F. G., and L. L. Wolf. 1974. A possible circannual molt rhythm in a tropical hummingbird. American Naturalist 108:341-354. https://doi.org/10.1086/282912

Sullivan, B. L., C. L. Wood, M. J. Illif, R. E. Bonney, D. Fink, and S. Kelling. 2009. eBird: A citizen-based bird observation network in the biological sciences. Biological Conservation 142:2282-2292. https://doi.org/10.1016/j.biocon.2009.05.006

Székely, T., A. Liker, R. P. Freckleton, C. Fichtel, and P. M. Kappeler. 2014. Sex-biased survival predicts adult sex ratio variation in wild birds. Proceedings of the Royal Society B: Biological Sciences 281:20140342. https://doi.org/10.1098/rspb.2014.0342

Temeles, E. J., I. L. Pan, J. L. Brennan, and J. N. Horwitt. 2000. Evidence for ecological causation of sexual dimorphism in a hummingbird. Science 289:441-443. https://doi.org/10.1126/science.289.5478.441

Viera, B. P., R. W. Furness, and R. G. Nager. 2017. Using field photography to study avian moult. Ibis 159:443-448. https://doi.org/10.1111/ibi.12445

Vitz, A. C., and A. D. Rodewald. 2006. Can regenerating clearcuts benefit mature-forest songbirds? An examination of postbreeding ecology. Biological Conservation 127:477-486. https://doi.org/10.1016/j.biocon.2005.09.011

Volpe, N. L., A. S. Hadley, W. D. Robinson, and M. G. Betts. 2014. Functional connectivity experiments reflect routine movement behavior of a tropical hummingbird species. Ecological Applications 24:2122-2131. https://doi.org/10.1890/13-2168.1

Wagner, H. O. 1957. The molting periods of Mexican hummingbirds. Auk 74:251-257. https://doi.org/10.2307/4081717

Wolfe, J. D., P. Pyle, and C. J. Ralph. 2009. Breeding seasons, molt patterns, and gender and age criteria for selected northeastern Costa Rican resident landbirds. Wilson Journal of Ornithology 121:556-567. https://doi.org/10.1676/08-111.1

Wolfe, J. D., T. B. Ryder, and P. Pyle. 2010. Using molt cycles to categorize the age of tropical birds: an integrative new system. Journal of Field Ornithology 81:186-194. https://doi.org/10.1111/j.1557-9263.2010.00276.x

Yanega, G. M., P. Pyle, and G. R. Geupel. 1997. The timing and reliability of bill corrugations for aging hummingbirds. Western Birds 28:13-18.

Zbyryt, A., P. Mikula, M. Ciach, F. Morelli, and P. Tryjanowski. 2021. A large-scale survey of bird plumage colour aberrations reveals a collection bias in internet-mined photographs. Ibis 163:566-578. https://doi.org/10.1111/ibi.12872