CONSERVATION BIOLOGY AND BIODIVERSITY Nectar Sources of Day-Flying Lepidoptera of Central Illinois 1 JOHN F. TOOKER, PETER F. REAGEL, AND LAWRENCE M. HANKS Department of Entomology, University of Illinois at Urbana-Champaign, Urbana, IL 61801 Ann. Entomol. Soc. Am. 95(1): 84Ð96 (2002) ABSTRACT We tabulated ßowering plant species visited by 98 species of Lepidoptera as extracted from records of C. Robertson, who observed Ͼ15,000 insect visitors of ßowering plants in central Illinois during a 33-yr period. Diversity of Lepidoptera was highest on plant species of the Asclepia- daceae and Verbenaceae, and the greatest number of lepidopteran species visited Verbena stricta Ventenat (Verbenaceae), Aster pilosus Willdenow (Asteraceae), Cephalanthus occidentalis L. (Ru- biaceae), and Pycnanthemum flexuosum (Walter) Britton, Sterns, Poggenberg (Lamiaceae). Lepi- dopteranswere particularly abundant on A. pilosus and C. occidentalis. The most polyphagous lepidopteran species were Colias philodice Godart (Pieridae), Danaus plexippus (L.) (Danaidae), Artogeia rapae (L.) [ϭPieris rapae (L.); Pieridae], Phyciodes tharos (Drury) (Nymphalidae), Polites themistocles (Latreille) (Hesperiidae), Pontia protodice (Boisduval & LeConte) (Pieridae), and Everes comyntas (Godart) (Lycaenidae), each of which visited ßowers of 50 or more plant species. Colias philodice, Phyciodes tharos, and Pontia protodice were the most common species. Most lepidopteran species evidently visited a limited range of nectar plants, which may have implications for species conservation and selecting “butterßy plants” for gardening. KEY WORDS moth, skipper, butterßy, ßower, prairie, pollination ADULT LEPIDOPTERANS VISIT ßowering plantsto feed on taring host species for each of the lepidopteran species nectar and, in some cases, pollen, resources that power that Robertson recorded, and by updating all scientiÞc ßight in locating matesand hostplantsfor oviposition names. We have also assessed the diversity and abun- and, for pollen-feeders, contribute at least in part to dance of lepidopteran species per plant species and egg production (Saunders1932, Norris1934, Gilbert families, differences between lepidopteran families in 1972, Dunlap-Pianka et al. 1977, Opler and Krizek preference for particular plant species, as well as the 1984, Karlsson 1994, Allen 1997). In fact, food sources level of polyphagy and relative abundance of lepidop- of adult lepidopteranscan be asimportant aslarval teran species. hosts in the conservation of some lepidopteran species (Opler and Krizek 1984). Nevertheless, host plant Materials and Methods associations of adult lepidopterans have received short shrift in the literature (e.g., Holland 1903, Saunders We updated lepidopteran speciesnamesandtaxo- 1932, Opler and Krizek 1984, Allen 1997, Opler 1998, nomic afÞliations using the Check List of Lepidoptera Pyle 1998), although hosts have been well docu- of America North of Mexico (Hodgeset al. 1983). mented for a few species(Lazriand Barrows1984, Plant speciesnameswereupdated with the latestedi- Jackson 1987). tion (Fernald 1978) of RobertsonÕs original reference In thisarticle, we updated and analyzed information (Robinson and Fernald 1908) and a more recent ref- on host plant associations of adult lepidopterans orig- erence (Kartesz 1994). inally presented by Robertson (1928) who docu- Our analysis of RobertsonÕs data were hindered by mented, over a 33-yr period, 15,172 insects visiting a lack of information on hismethods;however, we ßowersof 453 plant speciesinthe vicinity of Carlin- have been able to glean some details from the short ville, IL, “. for the purpose of ascertaining the dif- introduction of hisbook (Robertson1928), a biogra- ferent kinds of insect visitors.” (See Tooker and Hanks phy written after hisdeath (Parks1936), and a report [2000] for information on hymenopteran parasitoids addressing his data on bees (Marlin and LaBerge from this data set.) The value of RobertsonÕs data, 2001). Robertson worked from early spring to late fall, however, hasbeen limited becausehisrecordsare often concentrating on visitors to plant species that arranged by plant species with no index to insect bloomed for only a short period of time (Parks 1936). species, making it difÞcult to evaluate host ranges. We Each day he appearsto have collected from only a few have made these data more accessible by listing nec- speciesofplants(Marlin and LaBerge 2001). He does not appear to have collected exhaustively, but rather 1 E-mail: [email protected]. made note of all the visitors and collected selectively 0013-8746/02/0084Ð0096$02.00/0 ᭧ 2002 Entomological Society of America January 2002 TOOKER ET AL.: HOST PLANTS OF ADULT LEPIDOPTERANS 85 for a reference collection, concentrating on rarer spe- phology and ranged from white to yellow to pink cies or those whose taxonomic status was uncertain (Table 3), including four species of Asclepidaceae, (Marlin and LaBerge 2001). Poor representation of three species of Asteraceae, two species of Fabaceae, moth familiesin hisrecordssuggeststhathe made his and single species of Boraginaceae, Lamiaceae, and observations primarily during the day. Robertson Apiaceae. (1928) provided a qualitative index of abundance by The number of lepidopteran species that Robertson indicating which lepidopteran species were “abun- recorded from plant species could have been a con- dant” or “frequent” on particular plant species. Rob- sequence of the abundance of plant species: he may ertson observed visitors making note of certain activ- have sampled very abundant plant species more in- ities(e.g., collecting pollen, suckingon nectar, in tensively than rarer plants. In fact, the six plant species copula; Robertson 1928) and the sex of the insect associated with the greatest diversity of visitors (V. species when possible. stricta, A. pilosus, C. occidentalis, P. flexuosum, Ascle- Using RobertsonÕs data, we ranked the following: pias syriaca L., and Monarda fistulosa L.; Table 3) all (1) species diversity of lepidopterans per plant spe- appear to have been common in Illinoisat the time cies, (2) abundance of lepidopteran species per plant that Robertson made his observations (Illinois Natural species, (3) polyphagy of lepidopterans species at the History Survey 1936, Jones 1945), and remain common speciesandfamilieslevels,and (4) relative abundance today (Mohlenbrock 1986). Nevertheless, other abun- of lepidopteran species. We determined whether lep- dant plant species were visited by only three or fewer idopteran familiesdiffered in the plant speciesthey species of Lepidoptera (e.g., Gnaphalium obtusifolium favored by ranking the proportion of their species that L., Aster lateriflorus [L.], and Euphorbia corollata L.; visited the 15 most preferred plant species and testing Table 2; IllinoisNatural HistorySurvey 1936, Jones differencesbetween theserankswith the nonpara- 1945). Therefore, foraging preferencesof lepidopter- metric Kruskal-Wallis test (PROC NPAR1WAY, SAS ans, rather than abundance of plant species, evidently Institute 1988) and pair-wise correlations (PROC are at least partly responsible for patterns of ßoral CORR, SAS Institute 1988). visitation that we report. Among plant families represented by at least Þve speciesofplantsin RobertsonÕsrecords,the number of Results and Discussion lepidopteran speciesperplant specieswashighestin Robertson recorded Lepidoptera from 244 species Asclepiadaceae, with an average of 16.6 lepidopteran of ßowering plantsof 53 families(Table 1). Of these species (Table 2), followed by Verbenaceae (8.9 spe- plants, 64 species (26.2%) were of the Asteraceae, 20 cies), Lamiaceae (7.8 species), Asteraceae (7.7 spe- of the Fabaceae (8.2%), 19 of the Lamiaceae (7.8%), cies), Apiaceae (5.9 species), Fabaceae (4.6 species), 13 of the Rosaceae (5.3%), and 11 of the Apiaceae and Rosaceae (4.3 species); plant species of the re- (4.5%); the remaining plant familieswere represented maining familieswere visitedby an average of one to by nine or fewer species (Ͻ4%). four lepidopteran species. These patterns in diversity Among the 98 lepidopteran species recorded from of lepidopterans are consistent with lepidopteran ßowers, hesperiids were best represented with 26 spe- preferencesin ßoral morphology and color: Asclepia- cies(Table 2), followed by nymphalids(16 species), daceae, Verbenaceae, Lamiaceae, Asteraceae, and noctuids (13 species), lycaenids (10 species), and Apiaceae typically have clustered ßowers with rela- pierids (eight species); the remaining lepidopteran tively short corollas that range in color from yellow to families were represented by six or fewer species, white to pink (Opler and Krizek 1984, Jerviset al. 1993, including single species of Danaidae, Libytheidae, and Zomlefer 1994). Zygaenidae. This pattern in diversity is consistent with Lepidopteran familiesdiffered signiÞcantlyin the contemporary assessments of the number of lepidop- proportion of their species that visited the 15 most teran speciesperfamily in central Illinois(Covell preferred plant species (Table 3; Kruskal-Wallis sta- 1984, Opler and Krizek 1984, Sedman and Hess 1985, tistic ϭ 36.4; P Ͻ 0.0001). These differences in pref- Bouseman and Sternburg 2001). erence were not explained by Þdelity to particular Plant species visited by the greatest number of lep- plant familiesor ßower colors,but could be related to idopteran species (29Ð30 species) included Verbena similarities in ßower height, ßower size, ßoral scent, or stricta Ventenat (Verbenaceae), Cephalanthus occi- the types