Background/Question/Methods   A fundamental limitation in experimental climate change research is that tests represent relatively short-term “shock” experiments and so do not reflect the adaptive responses (either phenotypic plasticity or evolution) that may occur during the gradual process of climate change. Capturing this aspect of climate change effects in an experimental design requires comparing individuals from multigenerational lineages with different long-term climate histories, which few studies have accomplished. I incorporated adaptive responses in an experimental design by examining a food web comprised of herbaceous plants, grasshopper herbivores, and spider predators along a natural 4.8°C temperature gradient in the northeastern USA. Experimental warming strengthens the indirect effect of predators in this food web, suggesting the magnitude of top-down effects would vary across a range of local temperatures. I combined laboratory and field experiments to determine if adaptation to local climate altered the effect of temperature on this food web.

Results/Conclusions   Experimental warming caused all spiders to seek thermal refuge lower in the plant canopy, reducing spatial overlap with prey and increasing herbivore daily feeding time. However, spider tolerance to temperature was calibrated to spider origin such that spiders from warmer study sites tolerated higher temperatures than spiders from cooler study sites. As a consequence, the magnitude of the indirect effect of spiders did not differ along the temperature gradient. These results show that species function in food webs may be conserved during climate change by adaptive responses, and that short-term experiments may not accurately predict the effects of long-term chronic changes in climate.