Fig. 2: Divergent heat tolerance between urban and forest populations.
A Map of Puerto Rico and collection sites60. B Violin plots of heat tolerance (Critical Thermal Maximum, CTMAX) measured for wild-caught lizards from forest (green) and urban (gray) populations in four municipalities (Aguadilla: forest n = 16, urban n = 16; Arecibo: forest n = 21, urban n = 20; Mayagüez: forest n = 18, urban n = 18; San Juan: forest n = 16, urban n = 25). Wild-caught urban populations display significantly greater heat tolerance than their forest counterparts across municipalities (linear mixed-effects model: 0.8247 ± 0.1771, χ2 = 20.093, p < 0.001, effect size = 0.33760). Black dots and vertical lines in the center of each violin represent the mean and SD, respectively. C Comparison of CTMAX between urban (gray) and forest (green) lizards from Mayagüez born and raised in common garden (n = 16; forest: 25 °C n = 3, 32 °C = 4) (urban: 25 °C n = 5, 32 °C = 4) vs. wild-caught (n = 36, forest n = 18, urban n = 18). Dots and black bars represent mean ± 1 SE. Lines connecting treatments highlight the magnitude of mean CTMAX difference between common garden and native environments. No significant differences in CTMAX are observed between urban and forest lineages under common garden conditions, supporting plastic response to divergent native thermal environments as the primary source of observed in situ differences. Ancestral variation in this plastic response may be the target of selection in the novel thermal environments of urban heat islands.
