�� 2003 The Netherlands Entomological Society Entomologia Experimentalis et Applicata 108 : 205���209, 2003 205 Blackwell Publishing Ltd. Oviposition preference and larval performance of monarch butterflies ( Danaus plexippus ) on two invasive swallow-wort species Antonio DiTommaso 1 & John E. Losey 2 1 Department of Crop and Soil Sciences, 2 Department of Entomology, Cornell University, Ithaca, NY 14853, USA Accepted: 9 July 2003 Key words: Asclepias syriaca , Danaus plexippus , host plant preference, invasive plants, monarch butterfly, swallow-wort, Vincetoxicum nigrum , Vincetoxicum rossicum , Cynanchum, oviposition, dog- strangler vine, Lepidoptera, Nymphalidae Abstract The potential of two invasive herbaceous vines Vincetoxicum nigrum (L.) Moench and Vincetoxicum rossicum (Kleopow) Barbar. (Asclepiadaceae) to reduce monarch butterfly ( Danaus plexippus L.) (Lepidoptera: Nymphalidae, Danainae) populations was investigated by evaluating oviposition selec- tion in adult monarch butterflies and larval feeding preference in choice tests comparing the native host plant of monarch butterflies, Asclepias syriaca L. (Asclepiadaceae) and the two non-indigenous Vincetoxicum species. In both choice and no-choice tests, no eggs were oviposited on either of the two Vincetoxicum species whereas over 66 eggs per female were oviposited on A. syriaca plants. All first instar larvae allowed to feed on A. syriaca for 48 h survived while a significantly lower proportion sur- vived on V. rossicum (44%) and V. nigrum (14%). Mean weight of larvae that did survive on the Vin- cetoxicum species was significantly lower than the mean weight of larvae that fed on A. syriaca . The mean weight of surviving larvae, however, did not differ between the two Vincetoxicum species. The mean proportion of leaves consumed by larvae feeding on A. syriaca was significantly greater than the mean proportion of leaves consumed by larvae feeding on either Vincetoxicum species. Findings from this research indicate that V. rossicum and V. nigrum are not viable hosts of monarch butterflies and are likely to pose little direct threat to their populations as oviposition sinks. The ability of these highly aggressive plants, however, to out-compete and displace the native host of monarchs, A. syriaca , may pose a more serious threat. The potential of monarch populations to adapt to the two Vincetoxi- cum species as host plants over the long-term is discussed. Introduction The invasion of exotic plant species is recognized as one of the major potential dangers facing many ecosystems (Mack et al., 2000 Pimentel et al., 2000 Wilcove et al., 2000). Invasive plants have been shown to have substantial influence on both plant (Higgins et al., 1999) and animal (Pimentel, 2002) communities. These changes can occur either through direct competition between native and introduced plants or indirectly through changes in the local microclimate (Dukes & Mooney, 2002). Two invasive plant species of particular concern throughout a large portion of the north-eastern United States are Vincetoxicum nigrum (L.) Moench (syn. = Cynanchum nigrum ) [black swallow-wort or black dog- strangler vine] and Vincetoxicum rossicum (Kleopow) Bar- bar. (syn. = Cynanchum rossicum ) [pale swallow-wort or dog-strangler vine] (Asclepiadaceae). These two European species are invasive perennial herbaceous vines of natural and seminatural areas in the north-eastern United States and adjacent south-eastern Canada. Vincetoxicum nigrum , a native of western Mediterranean regions, is more widely distributed than V. rossicum in North America, and is found from southern Ontario to southern Pennsylvania and from the north-eastern Atlantic coast to Missouri and Kansas to the west (Pringle, 1973 Gleason & Cronquist, 1991 Sheeley & Raynal, 1996). Vincetoxicum rossicum is native to the Ukraine and south-western European Russia and is currently most abundant in the lower Great Lakes Basin, but is also found from as far east as Maine and as far Correspondence: A. DiTommaso. Department of Crop and Soil Sciences, 903 Bradfield Hall, Cornell University, Ithaca, NY 14853, USA. E-mail: ad97@cornell.edu

206 DiTommaso & Losey west as Iowa. The earliest North American herbarium specimen of V. nigrum was collected in Ipswich, Essex County, Massachusetts, USA in 1854 and the first collec- tion of V. rossicum at Toronto Junction, Ontario, Canada in 1889 (Moore, 1959). The first collections of V. rossicum in the north-eastern USA were near Rochester, NY and on Long Island, NY in 1897 (Sheeley, 1992). It is not clear how these species were first introduced into North America, but label information from herbaria specimen suggest that these plants may have been initially grown in botanical gardens and spread thereafter (A. DiTommaso, unpubl.). The above-ground morphology of the two species is nearly identical except for the flower color from which their names are derived. Below ground, V. nigrum is a clonal species with a deep rhizome system (Lumer & Yost, 1995) while no rhizome connections between culms of V. rossicum have been found (Sheeley, 1992 Christensen, 1998 Lawlor, 2000). Both Vincetoxicum species are mem- bers of the Asclepiadaceae (milkweed) family and are associated with disturbed and waste areas such as quarries and transportation corridors. Once established, however, these aggressive species readily move into nearby, less dis- turbed forest understories, Christmas tree plantations, or reduced tillage agricultural fields, often forming monospe- cific stands and displacing resident vegetation including the native perennial herb Asclepias syriaca L. (common milkweed) (Christensen, 1998 Casagrande & Dacey, 2001 A. DiTommaso, pers. obs.) as well as plant and animal species that are considered threatened or endangered (Bonanno, 1999 Lawlor, 2000). Recently, there has been increasing concern over the deleterious effects of these highly invasive and aggressive species, not only because of their potential to displace and out-compete threatened or endangered native plant and animal species in natural and seminatural habitats (Bonanno, 1999), but also because of their potential to attract ovipositing monarch butterflies ( Danaus plexippus L.) (Lepidoptera: Nymphalidae, Danainae) and thus reduce their populations in regions of North America where the two invasive weeds are found (Haribal & Renwick, 1998 Casagrande & Dacey, 2001 Mattila & Otis, 2003). Although several studies have compared these two invasive species to their native host, A. syriaca , no study to date has examined oviposition selection and larval feeding preference of monarch butterflies in the presence of both species. As the geographic distribution of each of these invasive plants continues to expand, it is likely that their ranges will overlap and will result in the three species (including A. syriaca ), at least initially, occupying many of the same habitats. Thus, it is important to assess the relative attraction of monarch butterflies to these three plant species when all co-occur in the same habitat. The objectives of this study were: (1) to evaluate oviposition selection in adult monarch butterflies, and (2) to deter- mine larval feeding preference in choice tests comparing the native host plant of monarchs, A. syriaca , and the two Vincetoxicum species. Materials and methods Plant collection Asclepias syriaca and V. rossicum plants were collected in June 2002 from an old-field near Ithaca, NY, USA, and from a field-edge population near Aurora, NY, USA, respectively. Vincetoxicum nigrum rhizomes were collected from a Rhode Island, USA, population in August 2001. Intact rhizomes of V. nigrum were placed in 2 liter plastic pots containing potting soil (1 : 1 sphagnum peat : vermiculite) and grown under glasshouse conditions for 9 months prior to being placed outdoors in mid-May 2002. Plants were cut back to a height of 10 cm to encourage tillering before placing outdoors. Plants of A. syriaca and V. rossicum were potted as for V. nigrum and also cut back to a height of 15 and 10 cm, respectively, twice, once in mid-July and once in early August 2002. All plants were watered as needed and fertilised weekly with 500 ml of a 1.25 g l ��� 1 20-20-20 (N-P-K) fertiliser solution prior to the start of trials in early September 2002. Vincetoxicum plants were, on average, 30 cm tall and had three to four tillers at the start of the oviposition experiments, whereas A. syriaca plants were, on average, 25 cm tall and had two to three tillers. All flowers were removed from plants at the beginning of the trials. Monarch butterfly source Mated female monarch butterflies were obtained from a captive colony in Ithaca, NY, USA that had been reared on Asclepias curassavica L. plants. Females were exposed to males for 3 days beginning 4 days after emergence from pupae. Adult host preference tests Oviposition trials were initiated on 3 September 2002 and performed under glasshouse conditions at a constant temperature of 26 �� 4 �� C, 60 �� 10% r.h., and L14:D10. A single mated female butterfly was randomly assigned to one of seven screened cages measuring 0.6 �� 0.6 �� 0.6 m in size and containing a single potted A. syriaca , V. nigrum , and V. rossicum that were approximately the same height and leaf surface area. Care was taken not to have the test plants touch each other in the cages. Sugar water in small sponges was provided as an energy source. Adult butterflies were released into the cages for a period of 48 h, after which they were removed and the number of eggs laid on each of the three plant species was recorded. Following