Data analyses

ANOVA tests and post hoc Tukey tests were used to assess statistically significant differences among MP and Bd treatments for the following continuous independent variables after being log10(x + 1)-transformed to reach normality: ‘time to reach Gosner stage 42–43’ and ‘time to reach Gosner 45’, ‘mass’, ‘body condition’ and ‘fluorescence levels at Gosner 45’, and ‘Bd loads at Gosner 42–43’. In all cases, MP concentration and Bd exposure were used as fixed factors, with their interaction included. ‘Tadpole body condition’, calculated following Peig and Green⁵¹, was introduced as a covariate in the ANOVA tests for ‘Bd load’ and ‘time to reach Gosner stage 42–43 and Gosner 45’. ‘Tadpole mass when exposed to Bd’ was introduced as a covariate in the ANOVA test for ‘mass at Gosner 45’. ‘Bd load at Gosner 42–43’ was introduced as a covariate in the ANOVA test for ‘fluorescence levels at Gosner 45’.

Generalized linear models with binomial errors were used for the binomial independent variables ‘Bd status’, ‘mortality’ and ‘reached metamorphosis’ at the latest stages of the experiment to assess statistically significant differences among MP and Bd treatments. MP concentration and Bd exposure were used as fixed factors, with their interaction included, except for ‘Bd status at the end of the experiment’ because none of the non-Bd exposed individual tested positive for Bd. ‘Tadpole body condition’, and ‘Bd loads at Gosner 42–43’ were introduced as covariates to asses if MP and Bd treatments had any effect on whether an individual undergoes metamorphosis, or died at the end of the metamorphosis, respectively.

‘Development stage’ was not included as a covariate in the statistical models because preliminary analyses discarded any significant contribution of this variable in any case. All statistical analyses were performed with JMP Pro 12 (SAS Inc.).