Prealighting host plant location in the Baltimore checkerspot butterfly, Euphydryas phaeton, and its implications for host range evolution in butterflies.

Abstract

In contrast to the common belief that the Baltimore checkerspot butterfly, Euphydryas phaeton, is a monophagous, habitat specific species, field investigations in Michigan, Ohio and Missouri show that populations vary significantly in their habitat and primary and secondary host plant choices. Most northeastern populations choose wetlands and Chelone as their host plant. Southwestern populations and others spread throughout the range choose dry woodlands and Aureolaria. Scattered populations choose dry, old fields and the introduced weed, Plantago lanceolata. In addition to these primary hosts, many secondary larval hosts are documented. A natural experiment in a northern Michigan population shows that, in the spring, larvae may defoliate the primary host and must complete development on other species. Because early instar larvae of butterflies are generally incapable of locating their own food plants, female host location behavior is an integral part of host plant ecology and host range evolution. A model of pre-alighting search developed for butterflies divides this phase into three stages, habitat, patch and individual host plant search. All of these levels are supported by butterfly behavior studies, including the present study on E. phaeton. A quantitative examination of pre-alighting host location behavior in the northern Michigan population shows that females use primarily vision while searching for potential host plants. Cluster analysis was used to assess visual similarity among dicot species. During searches, females show non-random alightment on two levels. (1) They land non-randomly on non-host species in the cluster containing Chelone glabra, most often Aster simplex and Solidago canadensis. Landings among these similar species are random. (2) Individual landing plants are significantly taller than the habitat mean but not different from the maximum vegetation height. The results of these studies suggest modifications of previous theories of host range evolution in butterflies. Visual cues, as well as chemical, should be considered important indicators of potential new hosts. Once chosen by a female, larvae must grow and survive on the new host in order to complete the host shift. The use of many secondary hosts by larvae may provide the selection pressure necessary to preadapt them for feeding on a newly selected primary host.