Clear advantages for fall armyworm larvae from feeding on maize relative to its ancestor Balsas teosinte may not be reflected in their mother's host choice

Consistent with an increasing number of comparisons between crop plants and their wild ancestors, a previous study showed that in the field, maize [Zea mays ssp. mays L. (Poaceae)] suffers more herbivory by larvae of Spodoptera frugiperda JE Smith (Lepidoptera: Noctuidae) than its ancestor Balsas teosinte (Zea mays ssp. parviglumis Iltis & Doebley). This study addressed plausible causes of the differing herbivory levels between maize and Balsas teosinte, specifically whether host plant selection by adult females is biased for either host plant depending on larval performance-i.e., the preference-performance hypothesis-or risk of mortality by parasitoids and predators-i.e., the natural enemy-free space hypothesis. Field observations showed that performance of S. frugiperda larvae was superior on maize compared with Balsas teosinte, a result partially explainable by the greater toughness of teosinte compared with maize tissue, but not by a difference in inhibition of protein digestion by larvae. Additional field observations showed that the mortality risk of S. frugiperda larvae is higher on Balsas teosinte, as indicated by higher parasitism (ca. four-fold) and predation (ca. three-fold) rates on the teosinte compared with maize. However, laboratory observations showed that S. frugiperda females did not discriminate between maize and Balsas teosinte for oviposition. Overall, the study's results were consistent with prior observations that direct and indirect defenses of maize against S. frugiperda larvae were weakened with domestication. In contrast, the results were inconsistent with predictions of the preference-performance and natural enemy-free space hypotheses, because selection of maize or Balsas teosinte plants by S. frugiperda females was independent of their offspring's performance and risk of mortality by natural enemies. On the one hand, this study's results partially explained differing herbivory levels between a crop and its wild ancestor, and on the other hand they suggested that host selection by S. frugiperda may be mediated by larval dispersal behavior and host choices.