Metamorphosis Reverses the Behavioral Phenotype in Rana arvalis Along a Latitudinal Gradient

Abstract

Understanding how demographic processes and environmental conditions shape behavioral variation across populations is pivotal in evolutionary ecology. However, the role that such processes play in the link between behavior and life-history traits across populations remains largely unclear. The moor frog (Rana arvalis) has colonized Sweden via two distinct routes: from the south via Denmark and from the north via Finland. We collected R. arvalis eggs from multiple populations along a 1,700 km latitudinal gradient across Northern Europe and raised tadpoles in a common garden experiment. We assessed developmental growth and proactivity levels in ca. 300 individuals at two key stages of anuran larval development: tadpoles (Gosner stage 32) and froglets (Gosner stage 46). We found strong behavioral differences along the latitudinal gradient and between developmental stages. Tadpoles from northernmost populations were bolder (shorter time to leave a shelter) and showed higher activity levels in an open field test compared to those from southern populations. However, these behavioral patterns reversed at the froglet stage, individuals from northern populations showing reduced proactivity compared to those from southern populations. Further analyses indicated significant associations between developmental growth and boldness, with contrasting patterns across developmental stages and colonization routes. These findings support recent revisitations of the pace-of-life syndrome theory, emphasizing a decoupling of correlations between behavior and life-history traits across ontogeny, likely reflecting adaptive responses to divergent ecological and demographic constraints along the latitudinal gradient rather than a single fast-slow continuum.

FIGURE 1

Study sampling. Geographic location of the five Rana arvalis populations sampled along a latitudinal gradient in northern Europe: Hanover (H), Skåne (S), Uppsala (U), Umeå (Um), and Luleå (L). Yellow dots represent populations belonging to the southern genetic cluster, while blue dots represent populations from the northern genetic cluster. The shaded area marks the approximate contact zone between the two postglacial colonization routes into Scandinavia.

FIGURE 2

Consistency of individual behavior. (A) Individual differences in measurements of time to emerge from the shelter taken during two independent trials performed to moor frog tadpoles (n = 30, dark green) and froglets (n = 33, light green) raised in a common garden experiment sampled across multiple regions of a 1700 km latitudinal gradient in northern Europe. (B) Distribution of adjusted repeatability estimates of an individual time to emerge from the shelter in bootstrapping tests using a linear mixed model framework after controlling for effects introduced by region (see Stoffel et al. 2017). Blue dot indicates mean adjusted repeatability with confidence intervals for time to emerge in tadpoles (top) and froglets (bottom).

FIGURE 3

Behavior of moor frogs across the latitudinal gradient. Boldness (A), activity (B) and exploration (C) measured in moor frog tadpoles (dark green) and froglets (light green) across a 1700 km latitudinal gradient and raised in a common garden experiment (South genetic cluster: Hanover (Germany), red dot; Skåne (Sweden), yellow triangle; Uppsala (Sweden), green square; North genetic cluster: Umeå (Sweden), blue cross; Luleå (Sweden), pink crossed square). Text indicates p‐values of significance tests comparing the difference in behavior between genetic clusters (see Table 2; ns: P > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001). Mean value and standard error of boldness (D), activity (E), and exploration (F) of tadpoles (dark green squares) and froglets (light green diamonds) for the five regions sampled. Average values for regions not sharing any letter are significantly different (p < 0.05) in post hoc contrasts of independent models assessing differences in behavior across regions for tadpoles (dark green letters) and froglets (light green letters; see Table S1).

FIGURE 4

Association between behavior and life history in moor frogs across the latitudinal gradient. Relationships between (A) boldness, (B) activity, and (C) exploration and developmental growth (mass/time at Gosner stage 42) in moor frog tadpoles (top row) and froglets (bottom row) sampled across a 1700 km latitudinal gradient and raised in a common garden experiment (South genetic cluster: Hanover (Germany), red dot; Skåne (Sweden), yellow triangle; Uppsala (Sweden), green square; North genetic cluster: Umeå (Sweden), blue cross; Luleå (Sweden), pink crossed square). Trend lines and confidence intervals are displayed for the relationship between each behavior and developmental growth for values at the North (blue) and South (yellow) genetic clusters. Text indicates estimated marginal means and standard error for each trend in post hoc contrasts of statistical models (p < 0.05 in bold; see Table S2).