In a recent paper in ZooKeys, Wiemers et al. (2018) provided an updated list of European butterfly names. In this list the authors follow gender agreement for species names, when interpreted as adjectival in derivation, in contrast to the common practice among most lepidopterists. Here we comment on this aspect of the paper, and voice our concern that this reversal does not benefit the stability of Lepidoptera names and is, indeed, inimical to their stability.

Modern zoological science needs the communities of taxonomists and users to agree on the names that are used to communicate information about the taxa we study and cherish. In this age, such collegiate acceptance is more important than ever, given that the number of users of scientific names has increased enormously. Agreement is particularly important when considering the numerous online databases, observation sites, Wikipedia, etc. Several global and local initiatives over the last several decades have begun to compile authoritative lists of taxonomic names to serve the community and build towards a greater stability, including Species 2000 / Catalogue of Life (Roskov et al. 2018; Roskov et al. 2019), Global Biodiversity Information Facility (GBIF Secretariat 2019) and Fauna Europaea (de Jong et al. 2014; Fauna Europaea 2017).

Unfortunately, the current (and likely future) funding situation for most of these projects is poor, to say the least, and populating these databases relies heavily on a diminishing number of taxonomists, who rarely receive recognition for their work. The Fauna Europaea database, which is of special importance as Europe’s main zoological taxonomic index, has suffered particularly, being an EU-supported project that was only funded by the European Commission between 2000 and 2004. Subsequently, updating was carried out at the Zoological Museum of Amsterdam (de Jong et al. 2014), first under the umbrella of the PESI project (PESI 2011; de Jong et al. 2015), then later without funding, until the Amsterdam museum was merged with Naturalis Biodiversity Center in Leiden in 2011. Since then, the Fauna Europaea database has been run by the Museum für Naturkunde, Leibniz-Institute for Research on Evolution and Biodiversity, Berlin, Germany. Recently, however, updating has come to a stand-still, very much to the frustration both of taxonomists who wish to update their lists and of users who need an up-to-date and authoritative nomenclature.

Given these circumstances, we enthusiastically applaud the initiative that several specialists of European butterflies have taken separately to publish an update for butterflies (superfamily Papilionoidea) in an open access journal, to produce a new list for the use of the scientific community (Wiemers et al. 2018).

The Preamble of the International Code of Zoological Nomenclature (the Code) states: The objects of the Code are to promote stability and universality in the scientific names of animals and to ensure that the name of each taxon is unique and distinct (International Commission on Zoological Nomenclature 1999).

It is generally accepted (though perhaps not always welcomed) that names of taxa change as a result of new taxonomic and phylogenetic research. Such changes are an increasingly frequent occurrence, especially when new data sources (such as nucleotide sequences, from DNA barcodes to entire genomes) are brought to bear on taxonomic, evolutionary and phylogenetic problems. We fully support such changes when they are implemented as a result of rigorous scientific study and have been accepted by the scientific community. Other changes, by contrast, are nomenclatural in nature and based on the application of rules from the international codes of nomenclature. Some follow from research into original descriptions, types, etc., often depending on new interpretations of old literature, while others are purely linguistic, and these usually cause the most controversy. In this paper, we discuss one of these issues – gender agreement – which, arguably, jeopardizes the very aim of name stability in Linnaean nomenclature.

The major problem that this new list of European butterflies (Wiemers et al. 2018) creates might seem trivial, but in its wider context threatens the stability of thousands of names of moths and butterflies globally, namely, the question as to whether a species name (or epitheton), when interpreted as adjectival in derivation, should agree in grammatical gender with the genus name or not. Although such agreement is still formally required by the Code, most lepidopterists ceased following the rule many years ago (e.g. Scoble 1999). The justification for ignoring the rule was described in detail and defended by Sommerer (2002), who argued strongly that the original spelling (orthography) be used. This recommendation was adopted at the General Meeting of the Societas Europaea Lepidopterologica (SEL) in a resolution proposed at the 13^th European Congress of Lepidopterology in Korsør (Denmark) on June 4, 2002 (the text of which immediately follows the paper by Sommerer 2002). Welter-Schultes (2012: 92) in his nomenclature textbook published by GBIF, also described “the Lepidoptera model”, where gender agreement never gained a strong position, and suggests (page 93): In such a situation it can only be recommended to maintain the names in the incorrect forms as they are, and not to change them.

Since 2002, but also earlier, almost all major Lepidoptera databases and catalogues have followed this principle of original orthography (summarised in Table 1). The managers of Fauna Europaea initially opposed the position (Sommerer 2002), but finally adopted it as a general principle for Lepidoptera in the first update and published it as such in the broad description of the project (de Jong et al. 2014). Apart from the catalogues listed in Table 1, major modern handbook series, such as “Microlepidoptera of Europe”, “Noctuidae Europaeae”, “Geometrid Moths of Europe”, “The Moths and Butterflies of Great Britain and Ireland”, “Moths of Borneo”, “Heterocera Sumatrana”, “Moths of Thailand” and “Monographs of Australian Lepidoptera”, all adhere to the use of original spellings. We could only find some six recent national catalogues and two global catalogues that follow the Code strictly (Table 1), in addition to several checklists where a mixture of correct gender agreement and original spelling was applied, a further indication that adhering to this principle is not easy, even for those familiar with the complexities of Latin and Greek grammar.

The stance taken by so many lepidopterists has been adopted by some other zoologists. For example, Welter-Schultes (2012), although providing rules on how to apply gender agreement or avoid its necessity, concluded (page 87) “A gender is nothing useful”. In ornithology, too, a discussion is taking place as to whether gender agreement is of any use (Schodde and Bock 2016; 2017). To quote from the first of these papers (page 167): The second point is the sheer mind-numbing, time-consuming complexity of determining gender for species-group names in zoological nomenclature. This is also the single greatest source of regulation-driven change in the spelling of species-group names, often disruptively so.

These are precisely the reasons why the vast majority of lepidopterists long ago decided to use original spellings. To apply gender agreement correctly, not only is it necessary to determine the correct gender of the generic name, which is far from straightforward in many cases, but often it is even more difficult to ascertain whether the species epithet is an adjective or participle, which both can be declined, or a (composite) noun. The number of errors caused by this latter problem is huge, which is understandable both because most modern taxonomists have no training in Latin or Greek, and because many names can be easily interpreted in different ways. Such difficulties are particularly prominent in Lepidoptera, given that many artificial names sometimes are not Latin adjectives but rather composite nouns, e.g. those ending in the suffixes -ella, -ellus, -ellum (Huemer 1988). The name Erebia aethiopellus (Hoffmansegg, 1806), among the names discussed here, belongs also in this category (Table 2). Certainly, in the current age of diminishing resources for taxonomy (both financial and human), and given the great need for taxonomy at a time of catastrophic biodiversity loss, it is indefensible to spend time on these kinds of disputes instead of doing real taxonomy, as it requires delving into ancient Latin and Greek linguistics and grammar that few specialists understand. In contrast, adopting original spellings means they have only to be checked once, which is a relatively easy process nowadays, especially as most databases include original names anyway, and as most old publications have become easily accessible through initiatives such as the Biodiversity Heritage Library (2005–2019).

Although we strongly adhere to the objects of the Code as expressed in our quote of its Preamble above, we consider that Article 34.2, which deals with gender agreement between genus and species names, contradicts that part of the Preamble.

Given that the overwhelming majority of authors of publications on Lepidoptera use original spelling rather than gender agreement, we conclude that Wiemers et al. (2018) did a disservice to nomenclatural stability as well as to the users of butterfly names, who are generally unaware of the intricacies of zoological nomenclature. Clearly, Wiemers et al. (2018: 15) were conscious of the practice adopted by most lepidopterists and the reasons for it as shown in this quote:

This includes the controversial article 34.2, which mandates that »the ending of a Latin or Latinised adjectival or participial species-group name must agree in gender with the generic name with which it is at any time combined«. Due to its linguistic complexity, this rule has led to many wrong or ambiguous decisions and causes additional instability of nomenclature each time a species name is transferred to another genus. Therefore a majority of lepidopterists, including the group editors of Fauna Europaea, have decided to ignore this rule and use the original spelling instead (de Jong et al. 2014).

Yet, curiously, Wiemers et al. (2018) gave no reasons for not following the “original spelling” convention that has become common practice among their lepidopterist colleagues. The published list (Wiemers et al. 2018) was presented as an update for Fauna Europaea, which, as noted above, has expressly adopted original spellings for Lepidoptera. Two of us (OK, EvN), as Group coordinators for Lepidoptera in Fauna Europaea, made critical comments on the draft list, and objected to the use of gender agreement at the time (e-mails late 2016).

We strongly recommend that users of the list ignore the changes to the 14 names affected, which are published by Wiemers et al. (2018) in table 7, page 7, and recommend that the names in the right hand column be used instead. We provide in this response these names for convenience (Table 2), and recommend most strongly that these original spellings are adopted by all users for purpose of nomenclatural stability.

Continuation of the current situation, in which workers on a particular group of animals elect to ignore a rule of the Code, will undoubtedly lead to new discussions in future. Thus, we also plead for a consistent change in the next version of the Code to allow for greater stability in the endings of names in general and for Lepidoptera in particular. Suggestions for abandoning gender agreement have been proposed before. One option, perhaps, could be to make a special dispensation to exempt Lepidoptera from the practice of following gender agreement given that this has led to relative stability of global Lepidoptera names (Welter-Schultes 2012: 92 and personal communication). However, we are aware that such special pleading could easily open the floodgates to numerous bespoke claims in other groups in other areas of nomenclature, leading to increased instability. The simplest and most stabilizing way forward would clearly be to do away with gender agreement altogether.

For constructive comments and discussion we are grateful to Thomas Pape (Copenhagen) and Francisco Welter-Schultes (Goettingen). We thank Alberto Zilli and David Lees for constructive remarks on the first version. IJK was supported by NERC grant number NE/P003915/1.