Fig 1.
Plasticity of Dendropsophus ebraccatus tadpoles raised with graded predator cues.
A) Geometric morphometrics utilized 14 landmarks (left) to digitize the size and shape of each tadpole. The tail spot was outlined by hand (right) in order to measure the size and color. B) Morphology was analyzed with a principal components analysis (PCA). The scatterplots (left) show the first two components from the PCA plotted against one another for each of four predator concentrations of fish or dragonfly predator cues (None, Low, Medium or High). Small points represent individual tadpoles, large points represent the mean response for each rearing treatment and shaded regions represent 95% confidence ellipses. Linear regressions show the relationship between PC1 (right, top) and PC2 (right, bottom) and increasing concentrations of fish or dragonfly predator cues. Thin-plate spline deformation grids are inset to show the change of tadpole shape along the main directional axis of change between the two predators, from large PC1 and small PC2 values (red box, dragonfly treatment) to small PC1 and large PC2 values (blue box, fish treatment). C) Tail spot coloration was also analyzed with a PCA. The scatterplots (left) show the first two components plotted against one another for each of the four predator cue concentrations. Linear regressions (right) showing the relationship between increasing concentrations of fish or dragonfly predation cues and relative tail spot area and the saturation, brightness and hue of the tail spot. In all regression plots, points have been jittered slightly to improve legibility.
Fig 2.
Plasticity of Dendropsophus ebraccatus tadpoles raised with mixed or single predator cues.
A) Morphology was analyzed with a principal components analysis (PCA). The scatterplot (left) shows the first two components from the PCA plotted against one another. Small points represent individual tadpoles, large points represent the mean response for each rearing treatment and shaded regions represent 95% confidence ellipses. Thin-plate spline deformation grids are inset to show the change of tadpole shape along the main directional axis of change between the two predators, from large PC1 and PC2 values (red box, dragonfly treatment) to small PC1 and PC2 values (blue box, fish treatment). Bar graphs show the estimated marginal mean ± standard error for PC1 (right, top) and PC2 (right, bottom). B) Tail spot coloration was also analyzed with a PCA. The scatterplot (left) shows the first two components from the PCA plotted against one another. Points and ellipses are as described above. Box-and-whisker plots (right) show how relative tail spot area and the saturation, brightness and hue of the tail spot varied across the four predator treatments. Plots show the median (thick horizontal line), interquartile range (top and bottom of the colored box) and either the most extreme values (ends of the whiskers) or 1.5 times the interquartile range and outliers (ends of the whiskers followed by points). In all plots, predator treatments are DD: dragonfly cues only, CC: predator-free control, DF: mixed dragonfly and fish cues, and FF: fish cues only.