Data from: Form-function relationships within species are uncoupled from those across species in swimming and jumping performance in arboreal frogs
Data files
Mar 17, 2025 version files 292.02 KB
-
arboreal_all.csv
26.28 KB
-
cinerea_data_final.csv
3.57 KB
-
max.vals.2023-08-13.csv
16.62 KB
-
max.vals.2023-10-08_swimming.csv
15.09 KB
-
README.md
7.18 KB
-
TreePL-Rooted_Anura_bestTree_PortikEA2023.tre
223.28 KB
Abstract
Variation of form-function relationships within populations is the substrate for adaptation at higher levels. Therefore, assessing similarity in form-function relationships within and between species may help reveal the processes shaping functional diversity. Here we test such similarity across three levels of anuran phenotypic divergence: within a population, among species in a single family (Hylidae; ~60 myr), and across a much broader sample of all anuran species using a single microhabitat (arboreal; ~120 myr). We expected less interspecific divergence to show higher similarity of form-function relationships with the intraspecific level. We analyzed the relationships between locomotor performance (in both swimming and jumping) and several hindlimb traits across these three evolutionary levels. While we found a positive correlation between swimming and jumping velocity at both intra- and interspecific levels, relationships between performance and body form did not match across levels. We suggest that different strengths of functional constraints or trade-offs may have produced more variation in form-function relationships across species, decoupling them from within-species patterns. We conclude that performance landscapes are likely qualitatively different across the different evolutionary scales, potentially reflecting changes in the relative importance of different behaviors across all arboreal species.
https://doi.org/10.5061/dryad.sf7m0cghx
Description of the data and file structure
We measured all Hyla cinerea individuals for jumping and swimming performance, for hindlimb morphology (including muscle mass) and foot area.
We used previously collected data on jumping and swimming performance, hindlimb morphology and foot area of 42 species of arboreal frogs (Moen et al. 2013, Moen et al. 2021).
Files and variables
File: arboreal_all.csv
Description: Data set with all individuals of all arboreal frog species analyzed.
Variables
- ID: individual identifier as a collected specimen (DSM series; character variable).
- species: name of species according to AmphibiaWeb as 20204 (character variable).
- family: taxonomic classification at the family level (character variable).
- micro: microhabitat (character variable).
- sex: sex of individuals (M=male, F=female; character variable).
- SVL: snout-to-vent length (numerical variable; units = mm).
- LEG.MASS: sum of all muscle mass from hindlimbs (numerical variable; units = g).
- max.vel.jump: peak jumping velocity (numerical variable; units = m/s).
- max.vel.swim: peak swimming velocity (numerical variable; units = m/s).
- max.acc.jump: peak jumping acceleration (numerical variable; units = m/s2).
- max.acc.swim: peak swimming acceleration (numerical variable; units = m/s2)
- LEG.LENGTH: sum of femur, tibiofibula, tarsus, and foot lengths (numerical variable; units = mm).
- FOOT.AREA: area of foot and webbing (numerical variable; units = mm2).
File: TreePL-Rooted_Anura_bestTree_PortikEA2023.tre
Description: Most recent phylogeny of anuran species from Portik et al. (2023)
File: max.vals.2023-10-08_swimming.csv
Description: repeated measures of swimming performance values for all H. cinerea individuals.
Variables
- ID: individual identifier for performance (character variable).
- max.vel: peak swimming velocity (numerical variable; units = m/s)
- max.acc: peak swimming acceleration (numerical variable; units = m/s2)
- max.power: peak swimming power (product of velocity and acceleration; numerical variable; units = m2/s3)
- Day: day at which trial was performed (integer variable).
File: max.vals.2023-08-13.csv
Description: repeated measured of jumping performance values for all H. cinerea individuals.
Variables
- ID: individual identifier for performance (character variable).
- max.vel: peak jumping velocity (numerical variable; units = m/s)
- max.acc: peak jumping acceleration (numerical variable; units = m/s)
- max.power: peak jumping power (product of velocity and acceleration; numerical variable; units = m2/s3)
- Day: day at which trial was performed (integer variable).
File: cinerea_data_final.csv
Description: maximum values of jumping and swimming performance for all H. cinerea individuals.
Variables
- perfID: individual identifier for performance (character variable).
- ID: individual identifier as a collected specimen (DSM series; character variable).
- sex: sex of individuals (M=male, F=female; character variable).
- SVL: snout-to-vent length (numerical variable; units = mm).
- LEG.LENGTH: sum of femur, tibiofibula, tarsus, and foot lengths (numerical variable; units = mm).
- LEG.MASS: sum of all muscle mass from hindlimbs (numerical variable; units = g).
- FOOT.AREA: area of foot and webbing (numerical variable; units = mm2).
- max.vel.swim: peak swimming velocity (numerical variable; units = m/s).
- max.acc.swim: peak swimming acceleration (numerical variable; units = m/s2).
- max.vel.jump: peak jumping velocity (numerical variable; units = m/s).
- max.acc.jump: peak jumping acceleration (numerical variable; units = m/s2).
Code/software
All analyses were performed in the R programming environment:
R version 4.4.2 (2024-10-31 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 10 x64 (build 19045)
SCRIPT 1: ‘Rscript_all.analysis.R’:
Script that contains all intraspecific (using repeated measures of performance) and interspecific comparative analysis (using species’ averages).
Packages:
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] car_3.1-3 carData_3.0-5 MCMCglmm_2.36 coda_0.19-4.1
[5] lme4_1.1-36 PerformanceAnalytics_2.0.8 xts_0.14.1 zoo_1.8-12
[9] cowplot_1.1.3 reshape2_1.4.4 lubridate_1.9.4 forcats_1.0.0
[13] stringr_1.5.1 dplyr_1.1.4 purrr_1.0.4 readr_2.1.5
[17] tidyr_1.3.1 tibble_3.2.1 ggplot2_3.5.1 tidyverse_2.0.0
[21] plot3D_1.4.1 treeio_1.30.0 ggtree_3.14.0 plyr_1.8.9
[25] geomorph_4.0.9 Matrix_1.7-1 rgl_1.3.17 RRPP_2.1.0
[29] mvMORPH_1.2.1 subplex_1.9 corpcor_1.6.10 phytools_2.4-4
[33] maps_3.4.2.1 ape_5.8-1
Wokflow:
- Calculating intraspecific performance correlations for H. cinerea.
- Calculating univariate performance repeatabilities using lmer.
- Using multivariate mixed models to calculate performance correlations in H. cinerea using MCMCglmm.
- Estimating form-function relationships in H. cinerea using MCMCglmm.
- Calculating evolutionary performance correlations using species average data (all arboreal and Hylidae).
- Calculating magnitudes of divergence for all arboreal and Hylidae.
- Comparing models of performance evolution in all arboreal species.
- Estimating form-function relationships using mvMORPH in all arboreal species and Hylidae.
SCRIPT 2: ‘e_pgls_script.R’:
Script that performs the E-PGLS analysis using all specimens measured.
Packages:
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] plyr_1.8.9 geomorph_4.0.9 Matrix_1.7-1 rgl_1.3.17 RRPP_2.1.0 subplex_1.9 corpcor_1.6.10
[9] phytools_2.4-4 maps_3.4.2.1 ape_5.8-1
Workflow:
- Updating names in data set and removing NAs.
- Estimating a species and sex effect on performance data using lm.rrpp function.
- Estimating form-function relationships across species taking intraspecific variation into account using lm.rrpp.ws function for all arboreal species and Hylidae.
Access information
Other publicly accessible locations of the data:
- Dryad: https://doi.org/10.5061/dryad.z08kprr9s
Data was derived from the following sources:
- Moen, Daniel; Ravelojaona, Rojo; Hutter, Carl; Wiens, John (2021b). Testing for adaptive radiation: a new approach applied to Madagascar frogs [Dataset]. Dryad. https://doi.org/10.5061/dryad.z08kprr9s
Specimens and species
We collected 40 individuals of Hyla cinerea (eight females and 32 males) at Red Slough Wildlife Management Area (Oklahoma, USA) on June 24, 2023. Hyla cinerea is a hylid frog generally found during the breeding season on emergent vegetation in large, still water bodies (Dodd, 2023). Males were identified by the presence of nuptial pads and a distended throat sac, whereas females lacked these features; sex was verified by internal inspection of the gonads after we completed all work with live animals. This research was conducted under Oklahoma State University IACUC protocol 20-09 and Oklahoma Department of Wildlife Conservation scientific collecting permit W2302.
The additional interspecific data from 38 arboreal and four semi-arboreal species (n = 177; 1–8 individuals per species) were obtained from previously published data (Moen et al., 2013, 2021a, b). These species have a common ancestor estimated at 114.9 million years old, and they belong to five different families: Mantellidae (n = 19 species), Hylidae (n = 14), Microhylidae (n = 5), Hyperoliidae (n = 2), and Rhacophoridae (n = 2).
Jumping and swimming trials
All specimens were measured for jumping and swimming performance by using a high-speed camera. All details of performance trials, video capture, and video processing followed Moen et al. (2013; 2021) and can be found in those papers. In brief, individuals performed both behaviors in several trials on different days, such that we could maximize our chance to collect the peak performance in both behaviors for each individual. Jumping take-off and burst swimming were recorded using high-speed cameras at 250 frames-per-second. We then extracted peak acceleration (m2/s) and peak velocity (m/s) for both performances per individual.
Morphological measurements
We collected five linear measurements on all preserved specimens using a digital caliper: femur length, tibiofibula length, tarsus length, and foot length on the left hindlimb, plus snout-to-vent length (SVL). We summed the hindlimb bone lengths to calculate total leg length, which is relevant to both jumping and swimming performance (Moen 2019). We also photographed the left foot pressed against a glass slide, then measured foot webbing areas between all digits and foot area (without webbing) in ImageJ (version 1.53e, Schneider et al., 2012). We summed these area measurements to calculate total foot area, which should be relevant to swimming performance (Gal & Blake, 1988; Nauwelaerts et al., 2001; Richards, 2008, 2010). Finally, we dissected out all upper- and lower-leg muscles of the left hindlimb, removed excess surface ethanol by patting them dry, and weighed them on a scale with 0.0001g precision. We then added all the muscle weights to calculate total leg muscle mass. These hindlimb muscles contain the major extensors relevant for jumping and swimming (Gillis & Biewener, 2000; James & Wilson, 2008; Nauwelaerts et al., 2007; Olson & Marsh, 1998; Prikryl et al., 2009), and their summed mass has been shown to strongly correlate with performance in both behaviors across frog species (Moen et al. 2013; Moen 2019).