The “biotic interactions hypothesis” predicts that biotic interactions play a greater role in driving adaptation at lower latitudes, where abiotic selection pressure is weaker. This latitudinal shift in biotic interaction strength is hypothesized to be a major driver of the poorly understood latitudinal biodiversity gradient, because coevolution of interactions may lead to faster speciation rates at low latitudes. Herbivory is a particularly important biotic interaction because plant-herbivore coevolution may be a major contributor to community diversity patterns, yet there is controversy about whether herbivory or defense are stronger at lower latitudes.
We tested whether herbivore pressure is greater at lower latitudes along a gradient from 27°N to 42°N in pokeweed (Phytolacca americana, Phytolaccaceae). Pokeweed is an herbaceous, short-lived perennial pioneer species native to the eastern US. We measured herbivory in the field in two summers in 5-10 populations along the gradient. Because most studies of latitudinal patterns in herbivory measure standing damage on mature leaves, which may greatly underestimate herbivory, we compared damage on young and mature leaves. To evaluate biologically relevant defense, we paired field surveys with experimental palatability assays for 13 populations in the lab using a naïve generalist Lepidopteran (Spodoptera exigua).
For young leaves, field herbivore damage decreased with latitude in both years. Mature leaf damage showed only a weakly negative relationship with latitude, and they were consumed less than young leaves. Unexpectedly, herbivory showed a threshold rather than a gradient pattern, with a drop north of 35°N. Abundance of Lepidopteran larvae (the primary herbivores) was a stronger predictor of damage than latitude. In no-choice palatability trials, cumulative consumption was greater at higher latitudes for both young and mature leaves, indicating that lower-latitude leaves are better defended. Young leaves were less palatable than mature leaves (despite greater consumption in the field).
Together, our results are consistent with the biotic interactions hypothesis. Beyond this, non-linear herbivory patterns suggest that P. americana herbivores may be limited to lower latitudes by growing season length or a temperature threshold. Strong effects of leaf stage on herbivory and palatability suggest that young leaves are targeted by P. americana herbivores, but contain chemical defenses unpalatable to the generalist consumer in the lab. Our study of latitudinal patterns in herbivory combines three strengths that have not been incorporated in one system: data on both herbivory and defense, measures of herbivory on young leaves, and quantifying defense through bioassays.