Understanding the circumstances under which insect herbivores will adopt a novel host plant is a longstanding question in basic and applied ecology. Though there has been ample study of regional differences in host preference across numerous insect herbivores, the extent to which intraspecific variation in plant species shape these patterns is relatively less well known. In parts of its range the Baltimore checkerspot Euphydryas phaeton uses English plantain Plantago lanceolata, a novel host plant, in addition to its native host, White turtlehead Chelone galbra. We offered female butterflies from each region the non-native host plant sourced from both regions and compared their oviposition behavior. The non-native host was near universally rejected by butterflies in the region where only the native plant is used. In the region where both hosts are used, plants from that same region were preferred relative to plants from the other region. This suggests that regional variation in host plant use by an insect depends on intraspecific variation in both species in the interaction.

Female butterflies in two parts of the Baltimore checkerspots' range (two field sites in Maryland and one site in Massachusetts) were offered White turtlehead to assess their gravidity in the field. Gravid females were offered English plantain from both regions randomly in sequence, in addition to some naturally growing plants, to understand how acceptance differs based on butterfly and plant origin. See the Methods of the associated manuscript and Fig. 1 for more details.

Description of the data and file structure

Observation of the 'plantain_trials.csv' represents an individual offering of English plantain to Baltimore checkerspot females.

Each column describes:

• 'Date': the date of the trial in MM/DD/YYYY format
• 'But State': the butterfly state (also where the trial took place)
• 'But ID': color-coded identifier for each butterfly
• 'Plant Type': species of plant used (in this case all plantain)
• 'Plant State': plant origin
• 'Plant Site': plant site origin
• 'Plant #': number identifier for unique plants from each site
• 'time (ovi.sec)': for butterflies which accepted, the time until oviposition acceptance
• 'Acc/Rej': where the plant was accepted or rejected

NAs exist in the 'Plant #' column for all ground plants since there were no numbered ground plants. Since plants were selected haphazardly on the ground, we did not mark every ground plant. NAs exist in the 'time (ovi.sec)' column because times were not recorded for those butterflies that rejected the host plant. As such, all rows where 'Acc/Rej' is 0 also has 'time (ovi.sec)' as being NA.

Observation of 'turtlehead_control.csv' represents an individual offerings of turtlehead to apparently mated female butterflies:

• 'Date': the date of the trial
• 'But State': the butterfly state (also where the trial took place)
• 'But Site': site where the trial took place
• 'Plant Species': plant site origin
• 'Plant Collect': how the plant was collected
• 'time(day)': the time at which the trial took place
• 'time (ovi.sec)': for butterflies which accepted, the time until oviposition acceptance
• 'Acc/Rej': where the plant was accepted or rejected

NAs exist in the 'time (ovi.sec)' column because times were not recorded for those butterflies that rejected the host plant. As such, all rows where 'Acc/Rej' is 0 also has 'time (ovi.sec)' as being NA.

Code/Software

R Code has been made available for this project. The file 'oviposition_trials_of_E_phaeton_on_P_lanceolata.R' includes all analyses run with this data that were represented in the manuscript. We recommend copying the information (for both the directory and the ) lines 36 and 37. The document has been commented for ease of use. Please refer to the file itself which has a header including more information.