Discussion
Birds, in particular water birds, are protected from water penetrating to the skin by the diameter and spacing of the barbs as determined by the parameter (r + d )/r and the absolute value ofr (Baxter and Cassie 1945; Rijke and Jesser 2011). The smaller these values are, the greater the resistance to water penetration. Only darters are known to benefit from water reaching the skin in order to reduce buoyancy while stalking prey on the bottoms of shallow lakes and streams. Their contour feathers show unusually large values for the parameter and lack barbules all together. However, all other species, with the possible exception of the Flightless Cormorant (P. harrisi ), have a contour feather structure that optimizes water repellency and resistance to suit the specific requirements of their habitat and behavior.
Swimming birds, by their weights, exert a pressure on their surface area in contact with water that remains well below that required to force water through the barbs. This is particularly true for the most aquatic of families, but, as shown in Table 3, decreasingly so for the families less intimate with open water. Swimmers are subject to a more or less static equilibrium between the pressure exerted by the weight of the bird and the capability of the outer contour feathers to resist penetration. Once the pressure exceeds this resistance, the underlying layers of feathers will eventually be penetrated as well and will provide no further protection against wetting (Rijke, Jesser and Mahoney 1989).
Diving birds, on the other hand, are subject to much greater, albeit temporary, pressures. On immersion, their bodies will be quickly surrounded by water. Initially, some air will be expelled, but the remaining air within the plumage as well as in air sacs and airways, will be trapped and compressed by hydrostatic forces. As they dive deeper, the pressure difference across the water-feather interface will no longer increase, but balance out as the compliant feather coat further compresses the trapped air at greater depths, thereby decreasing the volume and the buoyancy of the bird.
For birds alighting on water, the pressure on impact will not be balanced by compressed trapped air, but will instead produce a pressure gradient with the atmospheric air in the plumage. It is not known if this gradient is large enough to force water through the barbs of a single contour feather or a stack of multiple feathers. The available experimental data, few as they are, seem to suggest that each additional feather layer adds another 50 percent increase to water resistance (Rijke, Jesser and Mahoney 1989). Experiments of this kind, in which water is forced through feathers, may well closely resemble the conditions of birds landing on water. However, water on impact could also reach the skin by the flexing and bending of stacked feathers. How much each of these two dynamic mechanisms contributes to water penetration, if at all, is unknown. It is likely, however, that stacked layers mostly serve to reduce the bending and flexing of vanes in diving and alighting birds - and thus aid in preventing water from reaching the skin - but not in swimming birds.
As the data in Table 1 show, the contour feathers of penguins have barbs that are much shorter and thicker, and therefore more resistant to bending than those of less aquatic species: thirty times more so than those of divers, grebes and cormorants, fifty times more so than those of finfoots, jacanas and storm petrels, and six hundred times more so than those of waders. Compared with those of land birds, these contour feathers are more resistant to bending by as much as three orders of magnitude. We posit that these differences in magnitude as well as the wide range of resistance to bending represent evolutionary adaptations to the forces of impact associated with specific feeding habits and habitats.
The families in each foraging niche share a similar behavior with respect to their feeding habits and interaction with water. This is evident for families in the Aquatic Dive foraging niche, but less so for the taxonomically more distant families in other niches. Penguins, divers, grebes and cormorants all pursue their prey in much the same way, but this holds less true for families in the other foraging niches. In parallel with this observation, we find that the values forl /r are small for species in the Aquatic Dive niche, but larger in the others. In other words, the most aquatic species have stiff and very similar vanes in their contour feathers that resist bending, providing increased protection against water reaching the skin, whereas species with less interaction with open water have, apart from more diverse feeding habits, more flexible and dissimilar vanes with no such protection.
For swimming birds, we have seen that water may ultimately reach the skin if the weight of the bird exceeds the pressure required to force water through the barbs of the outer contour feathers, but in plunging and diving birds or birds landing on water surfaces, water penetration may also be caused by bending of the vanes. Closely stacked contour feathers should impede bending, but to which extent is difficult to measure experimentally. One would expect the denser the feather coat and the more the feathers overlap the more restriction to bending is attained. However, our calculations have shown there are approximately 100,000 to 150,000 feathers per m2 for water birds weighing less than 1.2 kg regardless of group. Furthermore, the extent of overlapping amounts to about 10 to 15 feathers in a stack for birds in all groups with approximately double that number for heavy birds. Apparently, feather overlapping is the same for all water birds and, as a result, the restriction stacking provides to bending is also the same. Only for birds weighing more than 1.2 kg do we find an increase in feather density and overlap with weight: up to 250,000 per m2 and stacks of 18 for the pink-backed pelican (P. rufescens ). This observation is in line with expectation as impact forces are directly proportional to mass.
The above findings may be explained by any of two or both possibilities: 1) the feather density and number of feathers in a stack are sufficiently large to prevent feather bending regardless of behavioral pattern and 2) the barb stiffness and resistance to water penetration of the contour feathers of each species are large enough to prevent water reaching the skin on their own account and do not benefit from a further increase in feather density or stacking.
The results of phylogenetic ANOVA have demonstrated that regardless of the phylogenetic relationships between bird species in this study, there is a significant difference in feather microstructure between the water bird groups. That no such significant difference was found for the land bird feeding niches supports the hypothesis of this study that the contour feathers of water birds exhibit features that are advantageous for specific aqueous habitats and behavioral patterns such as diving, plunging and alighting.
In summary, we have observed that the length and diameter of the barbs of contour feathers vary considerably among water birds with their stiffness parameters covering an eight-fold range however evolutionarily adapted to a specific niche. By referring to the mechanical properties of materials in general, we were able to show that short and thick barbs are stiff and resist bending, whereas long and thin barbs are flexible which facilitates bending. The value for l/r and, in particular the deflection parameter (l/r )3. (r + d )/r , is small for penguins, the most aquatic of bird families, but increases by orders of magnitude for birds with less interaction with open water. The families in each of these groups are taxonomically different, but have in common their method of feeding. This is particularly true for the species in the Aquatic Dive niche, but less so for other niche representatives, which populate a wider range of habitats and have more diverse feeding habits. This effect was not observed among terrestrial birds, although other terrestrial traits may remain conserved due to the birds’ respective niches.