We present the first record of plumage color aberration and the occurrence of slow-growing feather tissue in the Painted Bunting (Passerina ciris), a colorful songbird breeding in the southern United States during summer and migrating to Central America in winter. We took advantage of this rare sampling opportunity and performed hydrogen stable isotope analysis on 12 cross-sections of 2 primary feathers to compare isotopic values in normal and aberrant tissue sampled from the same individual. We used feather hydrogen stable isotope variation as an informative environmental marker to track the movements of the bird across its annual cycle. Spatial probability density estimates suggest further movements after this bunting arrived at molting stopover sites in northern coastal Mexico, either migrating toward southern coastal latitudes or toward higher elevations. Although we are unable to definitively assign a cause to the aberrant feather and its slow growth rate, we highlight how combining the observation of an atypical morphological trait with a modern molecular approach offers an opportunity to advance our understanding of avian movement behavior.
The study of the ecology, behavior, and demography of migratory songbirds is constrained by the logistical challenges of tracking individuals and populations across the full annual cycle (Marra et al. 2015). The growing application of forensic technologies, such as stable isotope analysis, may reduce the barriers inherent to complex movement patterns in small migratory species (Rubenstein and Hobson 2004, Zenzal et al. 2018). Because food and water containing different isotopic ratios of common elements are distributed across geographic gradients and environments, stable isotopes of carbon, nitrogen, and hydrogen from biological samples can provide insight into the diet and habitat used by animals at different points in time and space (Inger and Bearhop 2008). In particular, by exploiting a global latitudinal gradient of hydrogen, analysis of stable isotope ratios extracted from inert tissues can be used as a biological marker to reconstruct large-scale individual movements (e.g., Hobson et al. 2001, Kelly et al. 2002, Rubenstein et al. 2002, Clegg et al. 2003). Feathers are frequently used in the stable isotope analysis of migratory songbirds due to the general predictability of molt cycles, although the degree to which atypical or aberrant plumage patterns may diverge from expected ratios of stable isotopes is unknown.
Depending on the nature of a plumage aberration, slower plumage growth rates may affect the timing of incorporation of different ratios of isotopes that are transferred from food and water into inert tissues such as feathers or claws (Michener and Lajtha 2007). Ratios of stable isotopes found in aberrant plumage may therefore provide information about the movements of the individual displaying the plumage aberration.
The Painted Bunting (Passerina ciris) is one of the most colorful songbirds breeding in North America, with the Spanish name “colorín sietecolores,” or the “seven-colors bunting.” Because of the extensive bright red, blue, and green plumage of adult males, they are a target of the illegal pet trade (Iñigo-Elias et al 2002, Contina et al. 2019a). Two populations of Painted Bunting breed in the southern United States and northern Mexico. Although almost indistinguishable in appearance, the eastern population (breeding in the Carolinas, Georgia, and Florida) and western population (breeding primarily in Oklahoma and Texas, with pocket populations in Kansas, Arkansas, Louisiana, and northern Mexico) are genetically distinct (Herr et al. 2011; Contina et al. 2019a, 2019b) and exhibit different molt strategies: eastern Painted Buntings molt on their breeding grounds before migrating to wintering grounds in southern Florida and Cuba, while western Painted Buntings molt at stopover sites along the coastal regions of Mexico before arriving at their wintering grounds in southern Mexico and northern Central America (Sykes and Holzman 2005, Contina et al. 2013, Rohwer 2013).
In 2018, as part of a research effort to deploy geolocation devices on Painted Buntings across southeastern Oklahoma, we captured a bunting with a single white primary feather on its left wing that, to our knowledge, represents the first documented plumage aberration in this species (Fig. 1a). Upon recapture, we observed that the feather took longer to grow than adjacent primary feathers, and that it was replaced by another aberrant white feather after the first was plucked (Fig. 1b). We hypothesized that the aberrant feather may have integrated environmental hydrogen over a longer period of time during the annual cycle, possibly providing insights into the wintering movements and diet of this bird that exceed current information derived from stable isotope analysis of normal feathers (e.g., grown at a single molting site). Therefore, our aims were twofold: first, we documented the occurrence of a color aberration in a Painted Bunting primary feather and its exceptionally slow rate of growth; second, we reconstructed individual movements through stable isotope analysis of that aberrant feather using cross-sections that represented multiple time intervals of growth during the annual migration cycle.
Methods
Study area and sample collection
We studied a breeding population of Painted Buntings in southeastern Oklahoma during summer 2018. We established a banding site within the University of Oklahoma Biological Station property boundaries where the vegetation is characterized by mixed sedges and Bermuda and little bluestem grasses, with clusters of cedars, willows, and cottonwoods. Annual mean summer rainfall for the area is 390 mm, with a mean summer temperature of 26 °C (Brock et al. 1995, McPherson et al. 2007).
We captured the aberrant adult male Painted Bunting at 33°52′45.4″N, 96°48′49.0″W via mist-netting using species-specific vocalizations and determined its age and sex based on plumage characteristics (Thompson 1992, Pyle 1997). The capture site falls within the breeding range of the western population, such that all feather samples collected from this bird during summer 2018 are assumed to have grown on its migration stopover site during fall 2017. We first collected the aberrant feather and a normal feather for comparison (left wing third primary feather [p3]; right wing first primary feather [p1], respectively) on 10 May 2018 (Fig. 1a). We took standard morphometrics, including nare-to-beak tip length (8.2 mm), tarsus length (17.3 mm), wing chord (76 mm), fat estimate (none), and mass (16.9 g). We marked the bird with an aluminum U.S. Fish and Wildlife Service band with the serial number 248094505. We recaptured the same individual on 25 June 2018 and noted that the normal p1 had completely regrown while the aberrant feather had only grown ∼1 cm (Fig. 1b). From this, a monthly growth rate of 0.7–1.0 cm was extrapolated to estimate the growing period of the entire feather (6–8 months).
Sample preparation
We prepared each feather for analysis of stable isotope of hydrogen (δ2H) by following the laboratory steps detailed in Chew et al. (2019). We washed the feathers in a 2:1 solution of chloroform–methanol to remove debris and possible oil contamination followed by an additional wash in a 30:1 deionized water–detergent solution. We dried the feathers at room temperature for 48 h and packaged 12 cross-sections (Fig. 2a) of about 200 µg (±10 µg) of sampling material into silver capsules. We stored our feather samples for 4 weeks to allow for equilibration of exchangeable hydrogen to the laboratory environment (Wassenaar and Hobson 2003). Hydrogen isotopic analyses were run using a Thermo Scientific TC/ EA high-temperature conversion elemental analyzer interfaced with a Thermo Scientific Delta V Advantage Isotope Ratio Mass Spectrometer via a Thermo Scientific ConFlo IV Continuous Flow Interface. We report hydrogen stable isotope values as mean ± SD in delta notation (δ2H) of part per mil (‰) from the standards (δDsample = [(Rsample/Rstandard)–1]×1,000) comparative to the Vienna standard mean ocean water [VSMOW]. We used Caribou Hoof Standard (δ2HCHS = –197.0‰) and Kudu Horn Standard (δ2HKHS = –54.1‰) as calibration reference (US Geological Survey, Reston Stable Isotope Laboratory, Virginia, USA).
Figure 2
(a) Normal (right p1) and aberrant (left p3) primary feather dimensions, estimated growth rates, and sectioning scheme of adult male Painted Bunting left wing feather captured in southern Oklahoma in 2018. (b) Line graph of heavy hydrogen depletion in part per mil (ppm) across 12 cross-sections of the aberrant (dashed line) and control (solid line) feathers. Analogous feather sections represent different estimated time steps across total growth period (aberrant ∼6–8 months vs. normal ∼4–6 weeks). Subset boxplot shows intra-feather variation in heavy hydrogen richness across sections. (c) Posterior probability density estimation of molt locations in Central America for the aberrant (left) and normal (right) feathers, with estimated growing periods. Top 10% posterior probability density estimates are plotted in red indicating high likelihood of molting areas.
Stable isotope analysis
We inferred the spatial origin for each section of the aberrant and control feather by modeling the probability density surfaces of hydrogen stable isotope values following the procedure proposed by Wunder (2010). Briefly, we obtained a published isoscape precipitation model (δ2H) for North America (Contina 2019, http://isomap.org, job key = 73538), and rescaled the isotopic geostatistical inferences based on precipitation (Bowen et al. 2005) to account for differences between environmental and tissue hydrogen values. Because a species-specific feather isoscape model for δ2H is not currently available for the Painted Bunting, our rescaling step followed the equation δ2Hf =–14.47 + 1.4* δ2Hp adapted from Paxton et al. (2007). Rescaling isoscapes across species is a common practice whenever species-specific data are missing (see Hobson et al. 2012). We calculated the variance of the expected values inferred from the linear rescaling step described above (Wunder and Norris 2008, Wunder 2010), constrained our assignment probabilities to the molting and wintering species range, and normalized the posterior probabilities of origin in the R package raster (R Core Team 2018, Hijmans 2019). We performed an F-test for variance in stable hydrogen isotopic richness between the aberrant and normal feather and found significantly unequal variance (P = 0.046). This was therefore followed by a t-test assuming unequal variance comparing stable hydrogen isotopic richness between the 2 feathers.
Results
Mean δ2H values in the normal and aberrant feather were identical (x̄ = –48.6‰), although variation within the aberrant feather was higher than that of the normal feather (SD normal = ±3.09‰, SD aberrant = ±5.82‰) (Fig. 2b). Both the normal and aberrant feather exhibited a general trend toward more negative values over the course of their estimated growing periods (normal ∼0.5 months, aberrant 6–8 months assuming 0.7–1.0 cm growth per month), although the most recent sections of the aberrant feather showed substantially negative values: –45.5‰ to –61.8‰ between sections 11 and 12, a pattern that was not reflected by the normal feather (Fig. 2b). Probability density surfaces suggested similar molt initiation locations for the control and aberrant feather, spanning the coastal regions of Mexico, either northwestern Mexico near Sonora and Sinaloa, or the eastern coast of Mexico, including Tamaulipas and Veracruz (Fig. 2c). Probability surfaces also included the Yucatan Peninsula and northern Central America, although these southern areas are outside the known molting sites of the species and simply reflect noise of the stable isotope approach. The control feather likely completed its growth (Section 12; Fig. 2a) in the same geographic range in which molt was initiated. Conversely, the more negative hydrogen stable isotope values of the cross-sections of the aberrant feather suggest that its molt concluded elsewhere, possibly in the Mexican Plateau or at coastal southern latitudes (Fig. 2c).
Discussion
Plumage aberration
We documented an aberrant primary feather in an adult male Painted Bunting, the first reported for this species to the best of our knowledge. Plumage aberrations vary in their degree of environmental and genetic control over pigment distribution and integration (Guay et al. 2012, van Grouw 2013, Galván and Solano 2016). The most common plumage pigments, the melanins eumelanin (black) and phaemelanin (brown), are derived from dietary tyrosine, and oxidized in a reaction catalyzed by tyrosinase in melanocytes (Galván and Solano 2016). Albino birds cannot synthesize melanin, but carotenoid pigments are largely diet-derived, such that an albino bird may still present yellow, orange, and red plumage (van Grouw 2006). The single mutation responsible for albinism in birds is not uncommon, but the lack of melanin pigment in the eyes makes albino birds particularly susceptible to bright light, limiting fledgling survival (van Grouw 2013). This, in combination with the complete lack of melanogenesis in the melanocytes of albino birds, makes it unlikely that the plumage aberration observed in this bunting was an example of albinism.
Similarly, a leucistic bird may appear devoid of melanins in feathers and skin, but it differs from albinism in the mechanism by which absence of melanins is achieved: albino animals are unable to synthesize melanins in melanocytes due to absence of the oxidizing catalyst tyrosinase, while in leucistic animals, melanoblasts (the embryonic precursor to melanocytes) do not migrate to their designated feather tracts. Therefore, the distinction between albinism and leucism lies in the mechanism of plumage pigmentation. The former is caused by an inability to synthesize pigments, whereas the latter is caused by a disruption in the supply chain, with 2 important consequences: (1) leucistic birds retain melanin pigments in their eyes, and are therefore likely to survive longer, be observed, and documented; and (2) partial leucism is possible if the disruption of melanoblast migration is nonuniform. This nonuniformity may result in plumage aberrations that appear patchy, though bilaterally symmetrical. Therefore, leucism remains an unlikely explanation for the single feather aberration in this bunting, as melanocytes act as pigment distribution centers for tracts of feathers rather than individual feathers, with their dendritic extensions depositing pigment along those tracts.
Inconsistencies in yellow, orange, and red plumage coloration are attributable to dietary carotenoids (e.g., lutein, zeaxanthin), where such variation is caused by the availability of carotenoid-rich foods, or else by dietary carotenoids that have undergone metabolic conversion (e.g., astaxanthin, canary xanthophylls). In the latter case, variation is a consequence of metabolism and aberrations in plumage coloration may therefore have genetic roots. Although the molecular mechanisms by which dietary carotenoids are converted are poorly understood, converted carotenoids are ornamental signals of greater mate quality, especially with regard to parasite resistance and parental behavior (Weaver et al. 2018). The relative allocation of carotenoids in Painted Buntings has not been studied, however the lack of red, yellow, and orange coloration in normal Painted Bunting primary feathers suggests that their absence from an atypical primary feather should not be unexpected, nor necessarily indicative of a carotenoid-related aberration.
Another possible explanation is the presence of an internal injury that disrupted local melanin distribution in an otherwise healthy bird. Anecdotes about failed predation attempts describe damaged tissue with patches of white feather growth (Hatchisuka 1928, Brimley 1944), as do old instances of experimental plucking (Pearl and Boring 1914) and injury by misadventure (Phillips 1954, Sage 1962). We performed a visual inspection of the tissue surrounding the white feather and felt the area for physical abnormalities but detected none.
We deem the most likely cause of the white feather abnormality to be either an internal injury to the tissue surrounding the feather or a genetic mutation, either of which could disrupt melanin distribution. We speculate that if the white feather aberration exhibited by this individual was under genetic control, it would have to be a mutation that interrupted the distribution of melanins from a melanocyte to a specific feather within its associated feather tract rather than a mutation affecting the production of melanins (albinism) or the migration of melanocytes (leucism). It is impossible to rule out an internal injury without collecting the bird, and given the high breeding site fidelity of Painted Buntings, we decided that there was greater value in repeated observations of the same bird than there was in specimen collection.
Intra-feather stable isotope variation
Taking advantage of the slower growth rate of an aberrant feather, we used the intra-feather variability of plumage stable isotopes to reconstruct individual annual movement patterns. Intrafeather variability in stable isotope values may differ by species, molt strategy, and feather growth rate. Intra-feather variability can be affected by extrinsic factors (e.g., environmental stability), and is complicated by the limitations of the analytical tools available to perform stable isotope analysis on short (<5 cm) feathers (Gordo 2019). Few published examples of intra-feather variability in stable isotope richness exist, and most describe non-passerines (e.g., Smith et al. 2008, Grecian et al. 2015, but see Gordo 2019). Differences in isotopic enrichment between distal and proximal sections of non-passerine feathers are attributable either to fractionation or environmental variability in hydrogen isotopes (Smith et al. 2008). Fractionation is the separation of heavy and light isotopes, and may be dependent on tissue growth rate, such that rapidly grown tissue is more depleted of heavy isotopes. We deem fractionation an unlikely explanation for the difference in depletion in hydrogen isotope values between feather sections 1–12 due to the unusually slow growth of the aberrant feather and the small difference between the same sections of the faster growing normal feather. Although there is a paucity of literature describing intra-feather variability in passerines, Gordo (2019) found no differences in isotopic enrichment between proximal and distal sections of flight feathers in 4 species of songbird, consistent with our control feather.
The normal growth rate of the control feather captured stable hydrogen isotope values that generally aligned with known molt locations for western Painted Buntings. We therefore believe that the depletion of heavy hydrogen isotopes from the white Painted Bunting feather is likely a consequence of spatial variability in environmental hydrogen isotopes, caused by migratory movements of the bird through regions with different isotopic values (Bridge et al. 2011).
Aberrant plumage is inconsistently documented in birds, perhaps due to the limited number of studies involving captures and close plumage examination. Aberrant plumages are more likely to be observed and reported in human-dominated habitats, biasing our understanding of both the underlying spatial distribution and species-specific propensity for plumage aberrations. Yet, many specimens of atypical plumage are housed in museum collections on the basis of their rarity, and we urge researchers to view these specimens in light of their potential to address aspects of avian ecology that go beyond the intrinsic value of specimen rarity. Under conditions of normal plumage growth, stable isotope analysis constitutes a useful tool in forensically delineating molt and observation sites, and in doing so revealing avian migratory patterns. Although in our case repeatability is limited by the nature of the plumage aberration (a single feather from a single individual), intra-feather variability in stable isotope values in birds with slower-growing feathers represents an emerging avenue for linking species to their post-molt movements. Our unconventional application of cross-section analysis allowed us to explore the implications of an anecdotal instance of plumage aberration that resulted in intra-feather variability in hydrogen isotopes. In this way, we have demonstrated the utility of studying atypical plumage growth rate in relation to avian post-molt movements.
Acknowledgments
We thank the University of Oklahoma - Oklahoma Biological Station, G. Wellborn, D. Cathey, and M. Tucker. We are grateful to J. Kelly, M. Patten, and R. Weaver for providing valuable comments on the manuscript. Our research was approved by the Institutional Animal Care and Use Committee of the University of Oklahoma (R18-013), and conducted under state and federal banding permits (#23215). Our research was supported by the National Science Foundation (NSF) (DGE 1545261) as well as the NSF National Research Traineeship Program (NSF-NRT 1545261).
