Data and code from: Habitat permanency shapes the temperature-size rule in tropical anurans

Material and methods Experimental set up We studied the pond breeder anuran species Agalychnis callidryas and Allophryne ruthveni, and the stream breeder anuran species Espadarana prosoblepon, and Boana picturata. Tadpoles used for this study were captive bred specimens, obtained as freshly laid eggs or early embryos from European captive-breeding facilities (Table S1). Tadpoles were reared in 24 × 24 × 30 cm aquaria (12 L water, GH 1–2) in a climate-controlled room (12 h light:12 h dark). Aquaria contained washed gravel, volcanic rocks (for bacterial colonization; Bagchi et al., 2014), a bio sponge filter, and an aquarium heater. Tanks were separated by Styrofoam to prevent temperature and behavioral interactions. Water temperature was maintained at 17°C (acclimation) and then adjusted to one of five target temperatures using aquarium heaters and thermometers. Each tank received cacao leaves and bacterial balls (“Pure Pond”) to stabilize water quality. All tanks were cycled for two weeks before starting the experiments (Giller & Malmqvist, 1998; Bagchi et al., 2014). Temperature treatments and maintenance Clutches were initially reared at 24°C. After hatching (Gosner stage 25), tadpoles were randomly assigned to one of five temperature treatments per species (20, 23, 26, 28, 30°C). Each aquarium represented a single temperature treatment. Tadpoles were checked daily and fed ad libitum with a mixture of commercial fish food (40%, Hikari Sinking Cichlid Gold) and boiled rucola/beet salad (60%). Water levels were topped up with demi water every two days, and key parameters (GH, pH, nitrate, nitrite, ammonium) were monitored weekly and maintained within optimal ranges (GH 1–4, pH 6–7, nitrate <15 mg/L, nitrite/ammonium ~0). Water was changed (50%) when needed to maintain water quality within these parameters. Developmental monitoring and data analysis The date of emergence (Gosner stage 42–44) and completion of tail absorption (stage 46) were recorded for each individual. Body mass was measured in milligrams (mg) after completion of stage 46, once the tail was fully absorbed. Each froglet was weighed in a petri dish; the mass of the empty dish was subtracted from the combined mass to determine individual body mass. Then, we analyzed the effects of temperature and breeding habitat on development time and body size at metamorphosis. For this purpose, we fitted linear models to predict development time and log10-transformed body mass as response of breeding group, species (nested within group), and a quadratic term for temperature, including its interaction with group. Partial residuals from fitted models were used to assess the effect of predictors and to further study the relationship between development time and body mass within each habitat type. Model assumptions were checked via diagnostic plots. All analyses were conducted in R (R-Core Team 2025).