Our primary goal has been to create a list of North American Lepidoptera ranges and their corresponding host plant genera. Lepidopteran taxonomical updates rely on recent peer-reviewed publications. Our generic names of Lepidoptera may not conform to all published opinions, but our core range and host plant data for each species remains consistent. All names can be found by a search using the species name only or an expanded name search that includes synonyms. It is not our intent to resolve disputed usage or to establish proper current nomenclature.

Our list is in alphabetical order. The list can be filtered by butterflies, skippers, macromoths, micromoths, family, and subfamily. If taxonomic order is desired, the user can re-order the list by using the Hodges or MONA numbers. The species names follow various authors; as a result, proper species gender or suffix is not always clear. Routinely searching by a truncated species name may be the most successful search method.

Complications can occur when revisions contract or split a species. Every effort has been made to follow data to the logical conclusion and clearly attach records of former species to current usage. Not all species designations are static; suggested and accepted usage has been changed several times for some groups. Additionally, more often than not, DNA analyses identify new species. Some species, long accepted as Holarctic, have been separated from their European cousins (Gilligan et al. 2020). In a very few cases, the resulting new species have not been properly named but are retained on our list as near (nr.). Taxonomical updates are clearly attributed to the author(s). We note that the format we use is flexible enough to easily incorporate changes in nomenclature.

The common names column is not limited to currently accepted nomenclature. This column indiscriminately includes many common names which can be useful when tracking older lists. As a result, some common names may be linked to more than one species.

When possible, our primary source of range information is the most recent comprehensive publication. We started with current records and worked backwards. However, many species lack recent work, and, in a few cases, we needed to go back to the original source. Information was established by citing published literature including generic revisions, reviews, state lists, regional lists, guides, articles, compilations, and record lists. The resulting reference list contains some 4450 sources (Appendix 1). Occasionally, we cite our own personal observations. Some ranges are clarified by citing records such as Symbiota Collections of Arthropods Network (https://scan-bugs.org/portal/collections) and Bold Systems (Ratnasingham and Hebert 2007). We avoided unpublished records and unsupported information. Miscellaneous information found on the internet was researched but not included if we could not trace it to a citable source.

Ranges of Lepidoptera were created from states and provinces citations. A summary of the range is to the right of the equal sign (=) or a notation if parts of range are inferred (= infer:). Ranges compiled from a few sources occasionally skip over states (e.g. = OH AR NC), while a stated range will encompass all logical inclusions (e.g. in mountains from New York to Georgia = infer: NY PA NJ MD WV VA SC NC GA). Logical ranges often are supported and supplemented by several other references. In cases of conflict, the latest authority has precedence and is noted as such (defer to). Some reports are stated unlikely or seem so; these records are listed as fugitive and will appear in an expanded search.

The comprehensive compilation of cited sources of ranges is contained in a single column in our spreadsheet. Another column contains easily filtered and searched state and province abbreviations. Two more columns offer the option to search regions within the contiguous U.S. Excel enables filtering for the same field for two discreet codes. This makes it possible to confine searches to a single state or two states, or to expand searches to one or two regions. A fourth column notes Holarctic species and a partial list of Lepidoptera with Neotropical records. A fifth column can sort between exotic and native species. For example, a user could search for native, Holarctic Lepidoptera in Wisconsin and Michigan.

Lepidoptera range searches will return records that include 1) residents (those that overwinter) and 2) migrants (those that regularly appear and reproduce) but not strays, which are sporadic or rare visitors that offer no eco-services beyond pollination, nor fugitives, which are records that are unverified or in doubt for described reasons. There is a column expanding searchable established ranges to include stray and fugitive records. The words stray and fugitive do not contain any state nor province abbreviation to allow for inclusion in the search field.

Exotic and native Lepidoptera species can be filtered in a column. The exotic designation does not differentiate between accidental introductions and biological control introductions. The exotic status of a species is also identified at the beginning of the comprehensive justification of range (e.g. <EXOTIC> or <EXOTIC>. BIOCONTROL). Lepidoptera are considered exotic if there is a valid citation indicating that the species was introduced into North America. An expanded range of a North American species into new ranges is not consider exotic, but this information is noted as well. Neotropical strays into southern states are generally not consider exotics.

Excluded species are those Lepidoptera that have at some point been erroneously included on North American lists. Many are former usages that are still valid elsewhere. Some records represent misidentifications of species or locations. The reason for exclusion is noted and cited. Unconfirmed species are Lepidoptera that appear on current North American lists, but we have been unable to establish or exclude a North American range. Unconfirmed species and excluded species will not appear on any range search but will create a return in the strays and fugitive column. Host plant records of unconfirmed species will create a return in a search. The possibility of the presence of unconfirmed species in North America remains; these were not excluded.

The list of known but undescribed species is by no means a comprehensive list. These records will only be returned in a generic search and, occasionally, a host plant search. Although unnamed species have known locations, they will not be included in the discreet range columns until their status is established. It is possible some of these species also appear as recently described species.

Type location is given for many species. Unfortunately, original formats weighed type as a location record resulting a designation that might include holotypes, paratypes, syntypes, allotype, and synonymic types. While not incorrect, we note the possibility of confusion here and continue to update.

Host-plant records can be filtered by plant family or genus. These records are listed in three columns with a fourth containing appropriate citations.

Host-plant data were originally recorded only at genus level. This decision was based on the best available information. To illustrate, oaks (Quercus) had all host plants recorded to species whenever possible. However, of the 1014 Lepidoptera recorded on Quercus, 728 include a genus-only (Quercus sp.) records, and 263 include only a genus-level record. For our purposes the abbreviation sp. is used for unspecified species (example: Quercus sp. = oaks, red oak family, Quercus ssp., Quercus sp., or unidentified oak species). In addition, about 150 plant genus records required research for proper assignment to the current new genus. Of the 2068 lepidopteran species (comprising 73 families) which have been completely researched for host plant species, over 68% contain a record of a genus with an unspecified plant species. Fairly late in the process, we began to routinely include species records when they existed. Partial species level information can be found in the host plant reference column. Nevertheless, host genus remains the primary metric.

Some host plant genus names are embedded in another genus name. There is a column that creates discreet codes by inserting a period before and after the host plant (.Pinus.), eliminating false returns such as Lupinus (lupines) in a search for Pinus (pines). Another column offers the opportunity to return former generic usage excluded from the first search (Symphyotrichum[=Aster]). We established current uses for plant species nomenclature by using BONAP or, occasionally, for non-North American, plants Kew Plants of the World Online (https://powo.science.kew.org). We include records from beyond North America to paint a broader picture of Lepidoptera host ranges. These records are generally noted as such.

We have made every effort to clarify host plant genera to current usage using BONAP by tracing records at the species level. We provide a list of reexamined plant genera and reassignment explanations (Appendix 2). Some North American genus records use non-North American names; they remain unresolved (examples: Acacia[unresolved], Aster[unresolved], and Eupatorium[unresolved]. If there is a logical alternative, we include the alternate native genus in the search field (example: .Symphyotrichum.[unassignedAster]). In this way the native host plant will be returned in a search. When there is no good alternative, the original host genus remains (example: .Eupatorium.[unassigned]). This treatment clearly maintains the progression of data and a user can choose to exclude unassigned genus.

Although we have corrected plant species names as described above, not all valid stand-alone generic host plant records are correct. For example, our records report 183 Lepidoptera with Pyrus (pear) as a host plant, but many of these citations pre-date reassignment of some Pyrus to Sorbus (mountain-ash), Aronia (chokeberry), and Malus (apple). Only 14 Lepidoptera with Pyrus records do not also contain a potential reassignment. Despite the potential for misreporting, records for Pyrus remain because we cannot confidently justify another assignment.

Efforts have been made to correct misinformation. Host plant records can create unresolvable conflicts. A Lepidoptera species may have host plant preferences that vary by region or subspecies (Tallamy et al. 2010). For example, one author may state that a Lepidoptera species is a specialist on one plant in Arkansas while another author claims it is a specialist on a different plant in Maryland. Both can be correct due to regional specialization. We avoid eliminating records unless we can cite a clear denial that covers all of North America. Occasionally, questionable records are noted but are still included in our records. Adult plant associations are not included in host plant data.

Host-plant genera are listed alphabetically within the search fields. Records for host plants do not include some redundant citations. Weight cannot be applied to more frequent citations, as authors could be repeating each other. The earliest version of our plant list relied heavily on Robinson et al. (2002). Host plants in it were listed in one field while citations without specifics were listed in another. In many cases, this system persists. We are confident the host-plant data we present are traceable to the sources in the host plant references field.

Notes and clarifications made by us are in brackets. Brackets contain clarifications such as limitations of stated ranges, the taxonomical nomenclature used in the citation, and attribution of taxonomical updates. Data were assembled objectively; however, not all authors agree. Very rarely, we needed to arbitrate differences. Notes justifying our decisions and discussions noting discrepancies are also in brackets. Our goal is to report data as transparently as possible.

Parting with accepted use, we use no abbreviations except for states and provinces. This eliminates possible confusion, especially with binomial names and enables more specific searches. Synonymic lists contain no abbreviations. Additionally, synonymic lists can contain former combinations. The original combination is in a discreet column.

The authors have declared that no competing interests exist.

No ethical statement was reported.

No use of AI was reported.

This research was supported, in part, by National Wildlife Federation and the U.S. Forest Service. We thank Chelsea Cox for help with transcribing data.

Conceptualization: DWT. Data curation: KJS. Funding acquisition: DWT. Investigation: KJS. Methodology: DWT, KJS. Project administration: KJS. Visualization: KJS. Writing - original draft: KJS, DWT. Writing - review and editing: KJS, DWT.

Kimberley J. Shropshire https://orcid.org/0000-0002-6236-370X

Douglas W. Tallamy https://orcid.org/0000-0002-9207-6293

All of the data that support the findings of this study are available in the main text or Supplementary Information.