International Code of Zoological Nomenclature - Structure - Gender Agreement

In the species group gender agreement applies. The name of a species, in two parts, a binomen, say, Loxodonta africana, and of a subspecies, in three parts, a trinomen, say Canis lupus albus, is in the form of a Latin phrase, and must be grammatically correct Latin. If the second part, the specific name (or the third part, the subspecific name) is adjectival in nature, its ending must agree in gender with the name of the genus. If it is a noun, or an arbitrary combination of letters, this does not apply.

• For instance, the generic name Equus is masculine; in the name Equus africanus the specific name africanus is an adjective, and its ending follows the gender of the generic name.
• In Equus zebra the specific name zebra is a noun, it may not be "corrected" to "Equus zebrus".
• In Equus quagga burchellii the subspecific name burchellii is a noun in the genitive case ("of Burchell").

If a species is moved, therefore, the spelling of an ending may need to change. If Gryllus migratorius is moved to the genus Locusta, it becomes Locusta migratoria. Confusion over Latin grammar has led to many incorrectly formed names appearing in print. An automated search may fail to find all the variant spellings of a given name (e.g., the spellings atra and ater may refer to the same species).

Many laymen, and some scientists, object to continued adherence to this rule, especially those who work with butterflies and moths. This is for historical reasons. In 1758, Linnæus placed all butterflies in the genus Papilio, which, after a few decades, contained thousands of species. From the beginning, the gender of Papilio was unclear, undecided, and disputed. Some authors regarded it as masculine, others as feminine. Linnæus knew this problem and avoided any statement. All his 250 specific names in the genus Papilio were either nouns, indeclinable adjectives, or adjectives ending in -is (which can be masculine or feminine but not neuter). He did not use a single adjective ending in -us, -a, -um. P. Brown, Cramer, Fabricius, Fueßlin, Goeze, Poda and Schrank regarded Papilio as masculine, Ménétriés, Pontoppidan and most modern authors as feminine. In ICZN Opinion 278 from 1954, it was regarded as masculine. In many cases lepidopterists would not change the ending of a name as used by the author who established a name. So we find for example Papilio fuscus or Papilio macilentus, but also Papilio osmana and Papilio paradoxa. Only in a few cases are both versions found in the Web (an example is Papilio multicaudatus and Papilio multicaudata). This works also with other butterfly genera of which the gender is undisputed. Graphium appears neuter, but only the inconsistent versions Graphium angolanus and Graphium mandarinus are used, while Graphium sandawanum can only be found with a neuter specific name. Likewise, pairs are more frequently found in genera of which the gender is not obvious: Delias castaneus and Delias gigantea, Belenois albumaculatus and Belenois rubrosignata, Mylothris arabicus and Mylothris ruandana. Even in moths, such pairs occur: Xylophanes obscurus and Xylophanes turbata, Manduca boliviana and Manduca caribbeus, Sphinx caligineus and Sphinx formosana, Macroglossum albolineata and Macroglossum vicinum. It may also occur that a lepidopteran subspecies can have a different gender from the name of the species, as for example in Papilio multicaudata pusillus Austin & Emmel, 1998, and Papilio torquatus flavida Oberthür, 1879.