Experimentally testing mate preference in an avian system with unidirectional bill color introgression

Abstract

Mating behavior can play a key role in speciation by inhibiting or facilitating gene flow between closely related taxa. Hybrid zones facilitate a direct examination of mating behavior and the traits involved in establishing species barriers. The long‐tailed finch (Poephila acuticauda) has two hybridizing subspecies that differ in bill color (red and yellow), and the yellow bill phenotype appears to have introgressed ~350 km eastward following secondary contact. To examine the role of mate choice on bill color introgression, we performed behavioral assays using natural and manipulated bill colors. We found an assortative female mating preference for males of their own subspecies when bill color was not manipulated. However, we did not find this assortative preference in trials based on artificially manipulated bill color. This could suggest that assortative preference is not fixed entirely on bill color and instead may be based on a different trait (e.g., song) or a combination of traits, or alternatively may be due to lower statistical power alongside the bill manipulations being unconvincing to the female choosers. Intriguingly, we find a bias in the inheritance of bill color in captive bred F1 hybrid females. Previous modeling suggests that assortative mate preference and this kind of partial dominance in the underlying genes may together contribute to introgression, making the genetic architecture of bill color in this system a priority for future research.

Genomic islands of speciation harbor genes underlying coloration differences in a pair of Neotropical seedeaters

“…It is therefore likely that these genes are involved in mediating the coloration differences between S. beltoni and S. plumbea , the most notable of which is in the beak. These differences in beak color may contribute to promoting prezygotic reproductive isolation, as has been found between subspecies of Long-tailed Finches ( Poephila acuticauda ) [47,48]. Within the genus Sporophila, species have either gray, black, yellow or orange bills [24].…”

“…The copyright holder for this preprint this version posted November 15, 2023. ; https://doi.org/10.1101/2023.11.14.567069 doi: bioRxiv preprint color may contribute to promoting prezygotic reproductive isolation, as has been found between subspecies of Long-tailed Finches (Poephila acuticauda) [47,48]. Within the genus Sporophila, species have either gray, black, yellow or orange bills [24].…”

Diversifying selection and adaptive introgression of carotenoid-processing genes underlie the evolution of bill color in the long-tailed finch

“…Bill color phenotyping. Bill color was measured via UV-vis reflectance spectrophotometry for wild-caught and captive-bred adult long-tailed finches following protocols published in prior studies (Griffith and Hooper 2017;McDiarmid et al, 2023). In brief, spectral reflectance of the upper mandible from three consecutive scans per individual were averaged and smoothed using the R package Pavo 2 (Maia et al, 2019).…”

“…Spectra were then normalized by their maximum and minimum reflectance values, and the colorimetric variable H3 was calculated for all samples. H3 is a measure of bill hue that represents the wavelength midway between the minimum and maximum reflectance of a surface, which we bounded between 400 and 700 nm (Maia et al, 2019) and has previously been shown to effectively differentiate the bill colors of the two long-tailed finch subspecies and their hybrids (McDiarmid et al, 2023). To supplement reflectance data first examined by Griffith and Hooper (2017) and McDiarmid et al (2023) we measured bill color for an additional 165 wild-caught and 137 captive-bred samples in this study for a total of 948 wild and 550 captive samples with reflectance data (N = 1498 total).…”

“…We fit a univariate linear mixed model to test for an association between phenotype and each SNP genotype. We included a relatedness matrix to correct for population structure and because hybrid sex (McDiarmid et al 2023) and genome-wide hybrid index are significantly correlated with bill hue (Spearman's r = 0.66, p < 2.2×10 -16 ) we also included these as covariates during modeling. Keeping only variants with a minor allele frequency above 2%, genome-wide data from a total of 17,566,649 variants were analyzed (mean of 17.3 SNPs / kbp).…”