Macroevolutionary analysis of swallows revived the sight-line hypothesis

The evolution of conspicuous ornamentation is often thought to be the consequence of sexual selection, but this might not always be the case. One such candidate is contrasting pale-dark facial color patterns in front of eyes in insectivorous birds and other animals (lore-forehead borderlines, hereafter). A sight-line hypothesis suggested that the contrasting color between lore and forehead assists in tracking and capturing a fast-moving prey. However, this classic hypothesis have been criticized (and thus ignored) for several reasons including lack of formal statistical test controlling for phylogenetic inertia and confounding effect of dark facial color markings that are beneficial by reducing glare. Here, using a phylogenetic comparative approach, we tested the sight-line hypothesis and a widespread alternative explanation, the sexual selection hypothesis, in hirundines (Aves: Hirundinidae). We found no support for the sexual function of lore-forehead borderline in hirundines, because lore-forehead borderline was not positively related to indices of sexual selection (sexual plumage dimorphism and extrapair mating opportunity). In contrast, we found consistent support for the sight-line hypothesis. Species foraging on large prey items (i.e., fast prey) had higher degrees of lore-forehead borderline than others in this clade. Furthermore, an analysis of evolutionary pathways suggested inter-dependent evolution of lore-forehead borderline and prey size; transitions to the state with large prey and no lore-forehead borderline were less likely to occur than transitions from that state. These results remained significant when excluding species that lack dark lore, and thus, not mere presence of dark lores, but contrasting color patterns would be important. To my knowledge, the current study is the first macroevolutionary support for the sight-line hypothesis.