Variation in rates of phenotypic plasticity
For one experiment the estimate of λE was
negative, and six experiments lacked information on acclimation
temperature (Table S2). These experiments were excluded from all
subsequent analyses. Additionally, the classes thecostraca and
turbellaria were represented by only one and two species, respectively
(Table S2), and were therefore also excluded from subsequent analyses.
Finally, for the first set of analyses where we allowed for the
inclusion of a body mass effect, an additional 10 experiments that
lacked such data (Table S2) were excluded. Thus, the dataset used for
these analyses that had an SE of λE of less than
0.01 consisted of 219 experiments from a total of 84 species (12
insects, 6 crustaceans, 39 amphibians, 9 reptiles and 18 fishes).
Species-specific mean (SD) estimated λE of this
data set was 0.0345 (0.0305) h-1, and ranged from
0.0009 to 0.1892 h-1. When comparing the fit of models
with and without a random phylogenetic effect, the evidence for such an
effect was weak (AICc = -899.15 and -901.05 for models with and without,
respectively). Thus, we proceeded with comparison of models with
different fixed structures while including the random effects of
species, study, and experiment. Variation in λEwas best explained by a model that included taxonomic class, acclimation
temperature, the slope of the estimated exponential decay function attn , and the type of thermal tolerance measure,
whereas evidence for an effect of body mass was weak (Table 1). We
therefore repeated the comparison of models with different fixed effects
while excluding body mass from the full model. This allowed us to
include experiments that lacked body size information, increasing the
total number of experiments to 229, representing 91 species (13 insects,
7 crustaceans, 44 amphibians, 9 reptiles and 18 fishes).
Species-specific mean (SD) estimated λE of this
data set was 0.0347 (0.0302) h-1, with the same range
as in the previous data set. Strong evidence was found for effects of
taxonomic class, acclimation temperature, and the slope of the estimated
exponential decay function at tn , whereas
removing the effect of type of thermal tolerance measure from the model
caused a slight decrease in AICc (Table 1).
Inspection of coefficients from the best ranked model showed that
plasticity rates were highest among amphibians and reptiles, lowest
among fishes and crustaceans, and intermediate in insects (Table 2).
This pattern persisted when examining distributions of observedλE among these classes without correcting for the
covariates fitted in the best ranked model (Fig. 2). We also observed
that plasticity rates increased with acclimation temperature (Table 2,
Fig. 3). Finally, plasticity rates were observed to be higher when the
slope of the estimated exponential decay function attn was shallower (Table 2). In other words,
experiments where complete acclimation was more likely to have been
obtained were associated with higher estimated plasticity rates. This
pattern was mainly driven by the amphibian data, which had a large
number of experiments with relatively steep slopes attn (Fig. S7). This is in the opposite direction
to what might be predicted if a relationship between this slope and
estimated acclimation rate is a statistical artefact (Fig. S2). Rather,
the causality of this relationship is likely in the opposite direction,
i.e. experiments on species that had a low acclimation rate were stoppedbefore complete acclimation to the new temperature had occurred.
Nevertheless, to evaluate if this anomaly could have influenced our
results, we repeated the model comparison summarized in Table 1 but
including only experiments where the slope of the estimated exponential
decay function at tn was larger than -0.002 and
-0.001, respectively. In both cases, strong evidence was found for
effects of taxonomic class, acclimation temperature, and the slope of
the estimated exponential decay function at tn(Table S3, S4). Parameter estimates for the different taxonomic classes
showed only minor quantitative changes in comparison to the main
analysis (Table S5, S6 vs. Table 2).