The effect of inducible defenses in plants on herbivore populations will depend on the presence of density dependence in plant responses. Few studies have measured this density dependence, and none have done this for both performance of larvae and oviposition. In this study we used Solanum carolinense (Carolina horsenettle) and the specialist herbivore false potato beetle (Leptinotarsa juntca) to address these issues.  We examined the effect of herbivore density on oviposition preference and larval performance, whether these effects differed with plant genotype, and whether effects were mediated by changes in leaf area or changes in leaf quality (induced resistance).

Replicates of five S. carolinense clonal lines were subjected to damage by eight densities of L. juncta larvae (ranging from 0 to 13 larvae per plant) , and then L. juncta females were allowed 24 hours to lay eggs on either damaged or undamaged plants in paired choice tests.  After oviposition, we used bio- and chemical assays to evaluate potential differences in chemical defenses. In the bioassay, we removed eggs laid by females, placed five L. juncta larvae on plants, and examined larval performance after 48 hours, taking leaf quality and quantity into account. We also assayed leaves for proteinase inhibitors. 

Results/Conclusions

Female L. juncta preferred to oviposit on undamaged control plants over damaged plants.  The proportion of eggs laid on damaged plants decreased as initial herbivore density increased, and this relationship significantly differed among plant genotypes.  This effect was not driven by quantitative changes in the plant; female oviposition was not affected by the relative leaf area of damaged plants.  The relative growth rate (RGR) for larvae placed on less damaged plants was greater than the RGR of larvae on highly damaged plants, suggesting induced resistance. Because proteinase inhibitor levels did not differ significantly among genotypes or density of initial herbivores, reduced larval RGR on plants with more damage may be due to missing leaf area rather than a response to a chemical defense.  The results of this study suggest that there is density dependence in both larval performance and oviposition, that this is mediated by a change in plant quality rather than quantity, and that density dependence differs among plant genotypes.