A phylogenetic comparative analysis reveals correlations between body feather structure and habitat

Abstract

Summary Body feathers ensure both waterproofing and insulation in waterbirds, but how natural variation in the morphological properties of these appendages relates to environmental constraints remains largely unexplored. Here, we test how habitat and thermal condition affect the morphology of body feathers, using a phylogenetic comparative analysis of five structural traits [i.e., total feather length, the lengths of the pennaceous (distal) and plumulaceous (proximal) sections, barb density, and pennaceous barbule density] from a sample of 194 European bird species. Body feather total length is shorter in aquatic than in terrestrial birds, and this difference between groups is due to the shorter plumulaceous feather section in aquatic birds. Indeed, a reduced plumulaceous section in feather length probably reflects the need to limit air trapped in the plumage to adjust the buoyancy of aquatic birds. In contrast, the high pennaceous barbule density of aquatic birds compared to their terrestrial counterparts reflects water resistance of the plumage in contact with water. Our results show that birds living in environments with low ambient temperature have long plumulaceous feather lengths, low barb density, and low pennaceous barbule density. Data also suggest that plumage probably has limited function in reducing the heat absorption of species living in hot environments. Our results have broad implications for understanding the suite of selection pressures driving the evolution of body feather functional morphology. It remains to be tested, however, how other feather traits, such as the density of plumage (feathers per unit area) and the relative number of different feather types, for example downy feathers, are distributed amongst birds with different water resistance and thermoinsulative needs. A http://onlinelibrary.wiley.com/doi/10.1111/1365-2435.12820/suppinfo is available for this article.

How feathered are birds? Environment predicts both the mass and density of body feathers

“…The density of all feather types was generally higher in aquatic compared to terrestrial birds, which indicates that habitat affects not only the density of water‐repellent contour feathers, but also the thermoinsulative downy layer. The water resistant outer plumage layer is composed of short, overlapping contour feathers in aquatic birds (Eadie & Ghosh, ; Grémillet et al., ; Pap et al., ; Rijke & Jesser, ), and the resistance of this layer is further enhanced by the high vane density of the distal pennaceous parts of contour feathers, which prevents the infiltration of water into deeper plumage layers (Pap et al., ; Rijke, ; Rijke & Jesser, ). This result, that aquatic birds possess a higher density of contour feathers compared with their terrestrial counterparts, corroborates the earlier finding of Pap et al.…”

“…Given that the thermoinsulatory function of semiplumes and downy feathers are probably similar, and because the number of the former in most species was very low, we pooled our data for these two feather types in all analyses. We measured the length of one contour feather from each individual on both sides of the body, collected from the middle of the breastbone in case of the ventral part and from the middle of the backbone in case of dorsal part, using a ruler, and these data were completed with additional measurements from an earlier work (Pap et al., ); thus, our sample sizes for feather mass and density, as well as contour feather length differ slightly (Osváth et al., ). Feather samples from different species and individuals were randomly assigned to one of the five investigators (JP, CIV, GO, PLP and TD).…”

“…We used the most recent 10 years of temperature data for this study (encompassing the period between 2006 and 2015, downloaded on 3rd November 2016), and calculated average monthly mean temperature over this period for each spatial grid cell. From these monthly mean data, we created a 12‐layer shape file for each grid cell, and by intersecting temperature and species distribution files, we calculated monthly temperatures for each species separately for wintering and breeding grounds (Pap et al., , ; Vincze, ). Distribution maps (shape files) were retrieved from http://www.birdlife.org/datazone/info/spcdownload (BirdLife International & NatureServe, , accessed on 27th October 2016).…”

“…This means that quantifying the densities of different feather types is essential if we are to understand how avian plumage achieves its unique thermoinsulatory and water‐repellent functions. The most numerous components of avian plumage are contour feathers which cover most of the body and have multiple functions (Butler et al., ; Pap et al., ); the pennaceous distal region of these feathers form an outer water‐repellent cover over the inner thermoinsulatory layer, defined by the proximal plumulaceous feather section (Pap et al., ; Prum & Brush, ; Rijke, ; Srinivasan et al., ; Stephenson & Andrews, ). The contour feathers of some species also have an auxiliary shaft called the afterfeather; it has been assumed that this shaft has a function in thermoinsulation, although its role in thermoregulation remains unclear (Williams et al., ).…”

Unexpected bird–feather mite associations revealed by DNA metabarcoding uncovers a dynamic ecoevolutionary scenario

“…Bird body mass is evolutionary conserved, so that closely related species tend to have similar body sizes (Smith & Lyons, ). In addition, host body size is correlated with morphological variables of feathers that may constrain feather mite successful establishment, such as the interbarb distance of feathers (Pap et al, ). If unexpected hosts are phylogenetically distant from the typical hosts but with a similar body size, this would suggest that body size imposes a constraint to host‐switching (Clayton et al, ; Smith & Lyons, ).…”