Hidden white and black feather layers enhance plumage coloration in tanagers and other songbirds
Data files
May 30, 2025 version files 101.95 MB
-
Final_folder_Dryad.zip
101.94 MB
-
README.md
7.18 KB
Abstract
Birds are renowned for their diverse and colorful plumage. Here, we demonstrate that vibrant plumage in the tanager genus Tangara is dramatically intensified by a “hidden” layer of achromatic (white or black) plumage concealed beneath the outermost colorful layer. Using hyperspectral imaging, multispectral photography, microspectrophotometry, reflectance spectrophotometry, and optical modeling, we show that hidden white and black feather layers are systematically distributed on the body to enhance the brightness and saturation of carotenoid-pigmented and structurally colored plumage, respectively, by increasing or decreasing the amount of backscattered light that interacts with pigments or nanostructures. We compare male and female coloration and show that sexual dichromatism in some Tangara carotenoid plumage stems primarily from white layers in males and black layers in females rather than from differences in carotenoid pigmentation. Finally, we find that white and black hidden feather layers are widespread in colorful passerines. Hidden feather layers likely play a previously overlooked but critical role in colorful plumage evolution in birds.
Dataset DOI: 10.5061/dryad.1c59zw46k
Description of the data and file structure
This dataset contains all data and code to reproduce the Main and Supplementary figures and analyses in Price-Waldman et al. (2025), which demonstrates that hidden layers of achromatic (white or black) plumage are systematically distributed on the body to enhance the color of carotenoid-pigmented and structurally colored plumage in Tangara tanagers and other songbirds. Broadly, this dataset contains three types of data: 1) Measurements of color collected with reflectance spectrophotometry, microspectrophotometry (MSP), hyperspectral imaging (HI), and multispectral photography (MP); 2) code and supporting files for optical modeling; and 3) measurements of morphological features of feathers.
Data from MSP, HI, and MP capture the experimental changes in reflectance of the colorful regions of individual feathers placed on white or black achromatic backgrounds. We additionally used HI to measure the reflectance of white and black hidden achromatic feather layers on a tanager specimen with the colorful tips removed, MP to measure the reflectance of the achromatic regions of individual feathers, and spectrophotometry to measure the reflectance of the synthetic achromatic backgrounds that we used in experiments.
Files and variables
File: Final_folder_Dryad.zip
Description:
Essential_code/ contains:
- An R script (LayeringAnalyses_final_Dryad.R) which contains all code to reproduce the Main and Supplemental figures and analyses in the manuscript.
- An R source file (model_functions.R) which contains functions for optical modeling.
Essential_data/ contains subdirectories for data collected with hyperspectral imaging, multispectral photography, multispectrophotometry, reflectance spectrophotometry, as well as a subdirectory with files relevant to optical modeling. Essential_data/ also contains a CSV with feather length measurements.
-
Hyperspectral/ contains three CSV files with reflectance spectra (400 - 700 nm)
achromatic_feather_spectra.csv contains reflectance spectra for the exposed white and black achromatic layers from a Tangara chilensis specimen with the outermost colorful layer of feather tips removed.
blue_feather_spectra.csv contains reflectance spectra for a blue structurally colored feather placed on the exposed white and black layers from the T. chilensis specimen.
red_feather_spectra.csv contains reflectance spectra for a red carotenoid pigmented feather placed on the exposed white and black layers from the T. chilensis specimen.
Each file contains 500 reflectance spectra for each condition (for example, 500 reflectance spectra for the blue feather on the white background) and the mean reflectance. Reflectance spectra are smoothed (using pavo::procspec with span set to 0.3) but otherwise unmodified.
-
Microspectrophotometry/ contains
all_msp_specs_visible.csv, which contains reflectance spectra (300 - 700 nm) for nine barbs from each female and male feather placed on white or black backgrounds, as well as the average reflectance across nine barbs. Reflectance spectra are smoothed (using pavo::procspec with span set to 0.3) but otherwise unmodified.
msp_transmission_measurements.csv, which contains transmission measurements through the white achromatic, black achromatic, and blue structurally colored regions of three stacked feathers. Reflectance spectra are smoothed (using pavo::procspec with span set to 0.3) and are the average of transmission spectra across nine barbs (for structurally colored region) or six barbs and six barbules (for achromatic regions).
-
Model/ contains
normalized_astaxanthin_spectra_smooth_JH.csv, the normalized and smoothed absorbance spectrum for astaxanthin.
tangara_velia_epaulet_spec.csv, a reflectance spectrum from the structurally colored epaulet of Tangara velia.
keratin_reflectance.csv, a reflectance spectrum from the exposed white achromatic layer of T. chilensis that we used to approximate the reflectance of unpigmented keratin. Note that this spectrum is normalized and rescaled to a maximum reflectance of 40%.
-
Multispectral/ contains
multispectral_achromatic_barbs_HSL.csv, measurements of hue, saturation, and lightness values from individual barbs of feather achromatic regions imaged on black backgrounds. We used these values to compare the reflectance of male and female feather achromatic regions for plumage patches where achromatic layers were sexually dichromatic.
multispectral_achromatic_HSL.csv, measurements of hue, saturation, and lightness values from feather achromatic regions imaged on contrasting achromatic backgrounds (for example, white achromatic feather regions on black backgrounds).
multispectral_colorful_HSL.csv, measurements of hue, saturation, and lightness values from feather colorful regions imaged on white and black synthetic achromatic backgrounds.
patch_mechanisms.csv, classifications of whether the colorful feather layer for each patch in the dataset is colored by carotenoid pigments or structural colors.
-
Spectrophotometry/ contains
all_patch_spectra.csv, reflectance spectra (300 - 700 nm) for all focal patches in this study collected with reflectance spectrophotometry from museum specimens. Note that these reflectance spectra capture the overall reflectance of each patch, including colorful, achromatic, and downy layers. Reflectance spectra in this file are the average of three repeated measurements for each patch and have been smoothed (using pavo::procspec with span set to 0.4).
Calibrite_background_spectra.csv, measurements of the white and black photography standards that we used as achromatic backgrounds in our experiments. Columns contain reflectance spectra for the white and black squares on a Calibrite Colorchecker Grayscale ("CCG") and the second-brightest white and black squares a Calibrite Colorchecker Classic ("CCC"). We also include reflectance spectra of the exposed white and black achromatic layers on a Tangara chilensis specimen for comparison.
-
feather_length_measurements.csv contains measurements (in mm) of the length of each feather region (colorful, achromatic, and downy) for every feather in our dataset (we sampled three feathers from 12 focal patches for males and females). Each measurement is the mean of two measurements made independently by different authors.
Figures/ contains empty subdirectories Main/ and Supplement/. Running the code in Essential_code/LayeringAnalyses_final_Dryad.R will populate these subdirectories with all the plots and tables in the Main Text and Supplement (note that figures were further modified in Illustrator and tables were slightly reorganized, but all data and results are identical).
Code/software
All data can be viewed with standard CSV viewing software (for example, Microsoft Excel). We wrote all code in R v. 4.2.0.
Datasets in this project include reflectance spectra collected using UV-vis reflectance spectrophotometry, hyperspectral imaging, and microspectrophotometry. Additional datasets include distributions of hue, saturation, and lightness values from regions-of-interest of feathers from multispectral images.
