Quantifying shape and ecology in avian pedal claws: The relationship between the bony core and keratinous sheath

Abstract

Terrestrial tetrapods use their claws to interact with their environments in a plethora of ways. Birds in particular have developed a diversity of claw shapes since they are often not bound to terrestrial locomotion and have heterogeneous body masses ranging several orders of magnitude. Numerous previous studies have hypothesized a connection between pedal claw shape and ecological mode in birds, yet have generated conflicting results, spanning from clear ecological groupings based on claw shape to a complete overlap of ecological modes. The majority of these studies have relied on traditional morphometric arc measurements of keratinous sheaths and have variably accounted for likely confounding factors such as body mass and phylogenetic relatedness. To better address the hypothesized relationship between ecology and claw shape in birds, we collected 580 radiographs allowing visualization of the bony core and keratinous sheath shape in 21 avian orders. Geometric morphometrics was used to quantify bony core and keratinous sheath shape and was compared to results using traditional arc measurements. Neither approach significantly separates bird claws into coarse ecological categories after integrating body size and phylogenetic relatedness; however, some separation between ecological groups is evident and we find a gradual shift from the claw shape of ground-dwelling birds to those of predatory birds. Further, the bony claw core and keratinous sheath are significantly correlated, and the degree of functional integration does not differ across ecological groups. Therefore, it is likely possible to compare fossil bony cores with extant keratinous sheaths after applying corrections. Finally, traditional metrics and geometric morphometric shape are significantly, yet loosely correlated. Based on these results, future workers are encouraged to use geometric morphometric approaches to study claw geometry and account for confounding factors such as body size, phylogeny, and individual variation prior to predicting ecology in fossil taxa.

Keywords: claw; individual variation; morphometrics; phylogenetic comparative methods.

Figure 1

Representative third pedal unguals showing typical claws for each of the three ecological groups. Flying taxa: (a) Tauraco porphyreolophus (Purple‐crested Tauraco) and (b) Psarocolius montezuma (Black Oropendola); Predatory taxa: (c) Harpagus bidentatus (Double‐toothed Kite) and (d) Aviceda leuphotes (Pacific Baza); Cursorial taxa: (e) Dendragapus canadensis (Blue Grouse) and (f) Meleagris gallopavo (Wild Turkey)

Figure 2

Traditional arc measurements taken for the (a) bony core and (b) keratinous sheath. (c) Landmark configuration with numbered landmarks and semilandmark curves for the bony core (yellow) and keratinous sheath (green). Landmark definitions in Table 1

Figure 3

Geometric morphometric claw shape data. (a) Principal component analysis of total claw shape showing separation between predatory and ground birds with flying birds spreading across morphospace. Blue = predatory, red = flying, yellow = ground. (b) Thin‐plate spline (TPS) representations of the positive and negative extremes of PC1 and PC2. (c) Allometric analysis of the common allometric component of shape and log‐transformed centroid size. TPS grids show representations of small (left) and large (right) claw shape

Figure 4

Range of variation for (a) traditional measurements and (b) geometric morphometric data of the combined keratinous sheath and bony core. Orange—Dendragapus canadensis (n = 20); Purple—Milvus migrans (n = 19); Pink—Puffinus griseus (n = 21); Dark blue—Tinamus major (n = 25). Variation in traditional morphometric metrics for (c) Dendragapus canadensis, (d) Milvus migrans, (e) Puffinus griseus, and (f) Tinamus major. For traditional metrics, the variation of the ratio of the bony core to keratinous sheath is displayed to the left of the vertical line using the left x‐axis and the variation of the angle of the bony core and keratinous sheath are displayed to the right of the vertical line using the right x‐axis for all taxa together (a) and individual taxa (c–f)

Figure 5

Measures of integration between the bony core and keratinous sheath for both (a) traditional morphometric data using phylogenetic general least squares regression (blue = predatory, red = flying, yellow = ground) and (b) geometric morphometric data using two‐block partial least squares analysis. TPS grids show differences in shape along each shape block. (c) Assessment of modularity showing the observed covariance ratio (CR) against a null distribution. The observed CR is not significantly lower than the distribution and so modularity is not supported

Figure 6

Two‐block partial least squares analysis of geometric morphometric data against traditional morphometric data. Inset TPS grids show shapes at the positive and negative ends of the geometric morphometric block. Blue = predatory, red = flying, yellow = ground