Electronic Supplementary Material

The influence of iridescent coloration directionality on male tree swallows’ reproductive success at different breeding densities

Behavioral Ecology and Sociobiology

Sonia Van Wijk*, Audrey Bourret, Marc Bélisle, Dany Garant and Fanie Pelletier

*Corresponding author: Université de Sherbrooke, 2500 boulevard de l’Université, Sherbrooke, Qc, Canada, J1K 2R1. tel.: 819-821-8000 #63020, fax: 819 821-8049,

Fig. S1 Relationships and Pearson correlation coefficients among hue, brightness and saturation and their equivalent using the visual model described in Endler and Mielke (2005) and Stoddard and Prum (2008). Hue relates to a) θ and b) Φ, two variables that quantify the vector angular position of color in the tetrahedral color space. C) Brightness relates to luminance, the sum of quantum catches by cones. D) Saturation relates to r achieved, the relative length of the color vector in the tetrahedral color space suggested by Stoddard and Prum (2008). The visual model and tetrahedral color space were performed with pavo package (Maia et al. 2013), with quantum catches transformed according to Fechner’s law and von Kries correction applied as recommended by Endler and Mielke (2005), blue tit’s visual system described in Hart et al. (2000), a blue sky illuminator and an idealized background that does not affect the color perceived. The calculation method for hue, brightness and saturation is available in the Materials and Method section. Mean values for the three rump feathers of males from both years were used.

Fig. S2 Relationships between color measurements of rump feathers with a 2 mm diameter beam (rump measurement settings) and with a 1 mm diameter beam (crown measurement settings). A) Hue, H1, b) Saturation, S8, c) Brightness, B2, d) Angular breadth. The black dashed line corresponds to the 1:1 relationship. The red line corresponds to the regression slope. One rump feather from 30 different males were used. All paired t-test showed significant differences between the two different measurement settings, such that if rump feathers had been measured with crown settings, they would have been bluer, duller, more saturated with a larger angular breadth (less directional). Since results show that crown is greener and less saturated than rump (using 1 mm diameter for crown and 2 mm diameter for rump), these differences cannot be due to the difference in diameter beam. We can conclude that the color differences we observed between crown and rump are not simply due to differences in diameter beam. We suggest that the greener measurement with a 2 vs a 1 mm diameter is caused by the fact that the feather border is usually greener. Consequently, a 2 mm diameter captures a higher proportion of the feather’s iridescent surface, and thus proportionally more feather border. We considered it was more important to get a more representative measure of rump coloration with a 2 mm diameter beam than to avoid the risk of producing artificial differences between rump and crown with a 1 mm diameter beam.

Fig. S3 A social pair of tree swallows. Male erecting his crown feathers (left) towards a female SY (right).

Fig. S4 Decision chart for assignation to a genetic father following CERVUS analysis. In this study, nestlings receiving a status WPY (within-pair) or EPY (extra-pair) are successfully genotyped at a minimum of 4 loci and have a known and genotyped social father. A nestling obtaining a valid parental assignation by CERVUS is assigned with a candidate father at a 90% confidence level, either the social male (WPY status) or one of the males caught within a 15-km radius from the nest (EPY status). 75.3% of nestlings with a valid assignation by CERVUS showed no locus mismatched with its assigned father, 24.0% had a mismatch at one locus and 0.7% at two loci. The number of nestlings corresponding to each assignation situation appears in parentheses.

Fig. S5 Variation in plumage color variables measured on 214 male tree swallows breeding in southern Québec, Canada, 2013-2014. The first and second row contains crown and back color variables, respectively. Boxplots indicate 1st and 3rd quartiles (edges), median (line) and 1.5 times the interquartile range (whiskers). Notches not overlapping show strong evidence for a difference between years. Note that high values of angular breadth represent lower directionality.

Table S1 Mean and standard deviation of color variables and other individual and environmental variables in male tree swallows used to model the influence of iridescent coloration on the number of EPY, the number and proportion of WPY and annual reproductive success.

Variable / Mean / Standard Deviation
Crown Hue (nm) / 497.38 / 14.62
Crown Brightness (%) / 16.41 / 2.32
Crown Saturation / 0.85 / 0.14
Crown Angular Breadth (°) / 36.23 / 3.25
Back Hue (nm) / 472.33 / 14.10
Back Brightness (%) / 18.32 / 2.33
Back Saturation / 0.92 / 0.12
Back Angular Breadth (°) / 30.20 / 1.82
Other variables
Minimal Age (years) / 1.68 / 1.09
Mass (g) / 19.94 / 1.10
Wing Length (mm) / 118.42 / 2.99
Tarsus Length (mm) / 12.14 / 0.43
Density (Nb. of occupied nests in a 15-km radius) / 30.23 / 15.35
Laying Date (Julian Day) / 141.64 / 9.12

Fig. S6 Correlation matrix of explanatory variables included in models explaining reproductive success of 214 male tree swallows breeding in southern Québec, Canada, 2013-2014. More intense color and narrower ellipses on the upper panel represent stronger correlation. Positive and negative correlations are represented by blue and red, respectively. Correlation coefficients are reported on the lower panel. The package corrplot (Wei 2013) was used with R to draw the figure.

Fig. S7 Paired comparison between coloration of social and extra-pair males from the same nest, revealing differences in crown directionality. A) Histogram of estimated differences in crown directionality between the social and the extra-pair male tree swallows, from 1000 Monte Carlo simulations of paired t-tests. As multiple extra-pair sires can occur in a single nest, for each paired t-test, the extra-pair sire of one randomly-selected EPY per nest was used (N=150nests). Considering unscaled values, social males were 0.60 to 0.90° more directional than extra-pair sires in his own nest. B) Histogram of p-value from the 1000 paired t-tests.


Endler JA, Mielke PW (2005) Comparing entire colour patterns as birds see them. Biol J Linn Soc 86:405–431. doi: 10.1111/j.1095-8312.2005.00540.x

Hart NS, Partridge JC, Cuthill IC, Bennett ATD (2000) Visual pigments, oil droplets, ocular media and cone photoreceptor distribution in two species of passerine bird: the blue tit (Parus caeruleus L.) and the blackbird (Turdus merula L.). J Comp Physiol A 186:375–387. doi: 10.1007/s003590050437

Maia R, Eliason CM, Bitton P-P, et al (2013) pavo : an R package for the analysis, visualization and organization of spectral data. Methods Ecol Evol 4:906–913. doi: 10.1111/2041-210X.12069

Stoddard MC, Prum RO (2008) Evolution of avian plumage color in a tetrahedral color space: a phylogenetic analysis of new world buntings. Am Nat 171:755–76. doi: 10.1086/587526

Wei T (2013) corrplot: Visualization of a correlation matrix. R package version 0.73. http://cran.r-project.org/package=corrplot