• plant compensation;
  • herbivory;
  • apical dominance;
  • biological control;
  • induced resistance;
  • cotton;
  • Aphis gossypii;
  • Gossypium hirsutum


Crop plant compensation for herbivory and the population dynamics of herbivores are two key elements in defining an herbivore's pest status. We studied the dynamics of natural, unmanipulated populations of the aphid Aphis gossypii on seedling plantings of cotton, Gossypium hirsutum and Gossypium barbadense, over a 4-year period in California's Central Valley. Aphid populations colonized all plantings, but reached densities in excess of 0.5 aphids/leaf during only one year (1991), when outbreaks occurred. Outbreak populations were, however, ephemeral; predation and parasitism suppressed aphid populations prior to the initiation of flower bud production, when cotton plant growth may become photosynthate-limited. Effective natural biological control was observed despite the action of hyperparasitoids and the heavy mortality of immature parasitoids that occurred when predators consumed parasitized aphids.

We conducted manipulative experiments during 1991 and 1992 to quantify the ability of pre-reproductive G. hirsutum to compensate for aphid herbivory. In 1991 aphid populations in the high-damage treatment reached densities as high as any observed naturally during the past 37 years. Damage symptoms were severe: leaf area was reduced by up to 58% and total above-ground plant biomass was reduced by 45%. By the time of crop harvest, however, plants had compensated fully for the early damage in each of the three traits that define cotton's economic value: the timing of crop maturation, the yield of cotton fiber, and the quality of cotton fiber. Aphid feeding damage did, however, produce some changes in plant architecture that persisted throughout the growing season, including a decrease in the number of vegetative branches. In 1992 aphid populations and associated damage were much lighter, but the qualitative responses to herbivory were consistent with those observed in 1991. Plant compensation for early damage was complete for economically significant measures, and vegetative branch production was again suppressed in mature cotton plants. There was no evidence for a change in the suitability of G. hirsutum as a host plant for A. gossypii as a result of early damage (‘induced resistance’).

We conclude that pre-reproductive G. hirsutum, which has not yet begun strong allocations to reproductive structures or established architectural complexity, has retained effective means of compensating for herbivory. In contrast to other systems exhibiting strong compensation, G. hirsutum appears to compensate in part by enhancing apical dominance. The recognition of early-season A. gossypii as non-pests is critical to the sustainability of cotton production, because it will allow growers to forego pesticide applications that accelerate the evolution of pesticide-resistance and disrupt natural communities of predators and parasitoids.