Rapid adaptive evolution in a biological control insect colonizing a high-elevation environment in western Oregon

Background/Question/Methods

Classical biological control, the deliberate introduction of natural enemies to control invasive species, has been practiced for over 100 years. Yet few studies have been designed to detect genetic variation and adaptive evolution in biological control organisms colonizing new environments.  Procedures for screening candidate biological control organisms prior to introduction – such as climate matching and host specificity testing - usually ignore evolutionary potential.  We studied the phenology of the cinnabar moth, Tyria jacobaeae L. (Lepidoptera: Arctiidae), a univoltine insect introduced to North America as a biological control agent of ragwort Jacobaea vulgaris.  Anthropogenic spread of the cinnabar moth 1970-1990 from the Willamette Valley to the Cascades placed cinnabar moths in habitats where evolution could occur.  Our question was: has microevolution changed the phenology of the cinnabar moth since it colonized this new environment? We used (1) field observations and degree-day modeling to compare phenologies of Willamette Valley and Cascades populations, (2) a common garden experiment to test for genetic differentiation in juvenile development time, and (3) laboratory rearing of pedigreed offspring (from parental crosses within and between Willamette Valley and Cascade populations) to determine the pattern of inheritance and genetic differentiation in juvenile development time between populations.

Results/Conclusions

Three lines of evidence support the conclusion that shorter juvenile development time evolved after anthropogenic redistribution of the cinnabar moth from the Willamette Valley to the Cascade Mountains, where temperatures are cooler and growing seasons are shorter: (1) field observations showed the Cascades population has shorter juvenile development times corresponding to a shorter growing season, (2) our common garden experiment revealed genetic differentiation in juvenile development times between Willamette Valley and Cascade Mountain populations, and (3) our laboratory experiment rearing pedigreed offspring demonstrated genetic differentiation between populations in a quantitative trait (juvenile development times).  Results further show that Cascade populations likely evolved by directional selection imposed on phenological development of Willamette Valley populations that were redistributed to the Cascades.  Taken together, these results show that biological control organisms are capable of rapid evolution while colonizing and invading new environments, making the evaluation of evolutionary potential a necessary component of risk-benefit-cost analysis of candidate biological control organisms prior to release.