Phylogeny of the dagger moths (Lepidoptera: Noctuidae: Acronictinae: Acronicta) and the evolution of caterpillar defenses

Monday, November 17, 2014: 11:36 AM
Portland Ballroom 251 (Oregon Convention Center)
Brigette Zacharczenko , Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT
David Wagner , Department of Ecology & Evolutionary Biology, University of Connecticut, Storrs, CT
Jadranka Rota , Laboratory of Genetics, University of Turku, Turku, Finland
Niklas Wahlberg , Laboratory of Genetics, University of Turku, Turku, Finland
Highly variable larvae in the genus Acronicta prompted A.R. Grote to proclaim "There would seem to be no genus which offers a more interesting field to the biologist for exploration" (Grote, 1895). Along with related genera in the subfamily Acronictinae, Acronicta underwent a whirlwind of taxonomic revision in its infancy, rife with synonyms and rivalries between lepidopterists. Controversies have not lessened over time; splitting and lumping within Acronictinae has led to taxonomic instability and discordance between naming systems around the world. There are no clear synapomorphies for Acronictinae or Acronicta, making it difficult to sort the species based on morphology. To address this issue we inferred a phylogeny of Acronicta based on one mitochondrial gene (COI) and seven nuclear gene regions (EF-1α, RpS5, CAD, MDH, GAPDH, IDH and wingless) from over 70 specimens from North America, Europe, and Japan. Six genera were synonymized within Acronicta, as their continued recognition would render the genus paraphyletic. Results from maximum likelihood and Bayesian analyses were used to explore morphological, behavioral, and life history evolution in Acronicta larvae. The caterpillars sport an array of hair types, color patterns, resting postures, pupation habits, defensive behaviors, and host utilization patterns. We hypothesize that some of these characters, such as hair types and defensive behaviors, are evolving together in response to predation pressures. Characters were scored for each species that we were able to rear and analyzed using BayesTraits.