CHAPTER III. Nomenclature of taxa according to their rank
SECTION 2. Names of families and subfamilies, tribes and subtribes
The name of a family is a plural adjective used as a noun; it is formed from the genitive singular of a name of an included genus by replacing the genitive singular inflection (Latin -ae, -i, -us, -is; transcribed Greek -ou, -os, -es, -as, or -ous, and its equivalent -eos) with the termination -aceae (but see Art. 18.5). For generic names of non-classical origin, when analogy with classical names is insufficient to determine the genitive singular, -aceae is added to the full word. Likewise, when formation from the genitive singular of a generic name results in a homonym, -aceae may be added to the nominative singular. For generic names with alternative genitives the one implicitly used by the original author must be maintained, except that the genitive of names ending in -opsis is always -opsidis.
The generic name from which the name of a family is formed provides the type of the family name (Art. 10.6) but is not a basionym of that name (Art. 6.10; see Art. 41.2(a)).
Family names formed from a generic name of classical origin: Rosaceae (from Rosa,Rosae), Salicaceae (from Salix,Salicis), Plumbaginaceae (from Plumbago,Plumbaginis), Rhodophyllaceae (from Rhodophyllus,Rhodophylli), Rhodophyllidaceae (from Rhodophyllis,Rhodophyllidos), Sclerodermataceae (from Scleroderma,Sclerodermatos), Aextoxicaceae (from Aextoxicon,Aextoxicou), Potamogetonaceae (from Potamogeton,Potamogetonos).
Family names formed from a generic name of non-classical origin: Nelumbonaceae (from Nelumbo,Nelumbonis, declined by analogy with umbo,umbonis), Ginkgoaceae (from Ginkgo, indeclinable).
The name of a family may be formed from any validly published name of an included genus, even one that is unavailable for use, although the provisions of Art. 18.3 apply if the generic name is illegitimate.
Cactaceae Juss. (1789) formed from Cactus L. (1753), now rejected in favour of Mammillaria Haw. (1812).
Names intended as names of families, but published with their rank denoted by one of the terms “order” (ordo) or “natural order” (ordo naturalis) instead of “family”, are treated as having been published as names of families (see also Art. 19.2), unless this treatment would result in a taxonomic sequence with a misplaced rank-denoting term.
Cyperaceae Juss. (1789), Lobeliaceae Juss. (1813), and Xylomataceae Fr. (1820) were published as “ordo Cyperoideae”, “ordo naturalis Lobeliaceae”, and “ordo Xylomaceae”, respectively.
If the term “family” is simultaneously used to denote a rank different from “order” or “natural order”, a name published for a taxon at the latter rank cannot be considered to have been published as the name of a family.
Names published at the rank of order (“řad”) by Berchtold & Presl (O přirozenosti rostlin ... 1820) are not to be treated as having been published at the rank of family, since the term family (“čeleď”) was sometimes used to denote a rank below order.
A name of a family based on an illegitimate generic name is illegitimate unless and until it or the generic name upon which it is based is conserved.
Caryophyllaceae Juss., nom. cons. (from Caryophyllus Mill. non L.); Winteraceae R. Br. ex Lindl., nom. cons. (from Wintera Murray, an illegitimate replacement name for Drimys J. R. Forst. & G. Forst.).
Nartheciaceae Fr. ex Bjurzon (1846), based on Narthecium Huds., nom. cons. (1762), became legitimate when the generic name was conserved over its earlier homonym Narthecium Gérard (1761) (see App. III).
When a name of a family has been published with an improper Latin termination, the termination must be changed to conform with Art. 18.1, without change of the author citation or date (see Art. 32.2). However, if such a name is published with a non-Latin termination, it is not validly published.
“Coscinodisceae” (Kützing 1844), published to designate a family, is to be accepted as Coscinodiscaceae Kütz. 1844 and not attributed to De Toni, who first used the correct termination (in Notarisia 5: 915. 1890).
“Atherospermeae” (Brown 1814), published to designate a family, is to be accepted as Atherospermataceae R. Br. and not attributed to Airy Shaw (in Willis, Dict. Fl. Pl., ed. 7: 104. 1966), who first used the correct spelling, or to Lindley (Veg. Kingd.: 300. 1846), who used the spelling “Atherospermaceae”.
However, Tricholomées (Roze in Bull. Soc. Bot. France 23: 49. 1876), published to designate a family, is not to be accepted as “Tricholomataceae Roze”, as it has a French rather than a Latin termination. The name Tricholomataceae was validly published by Pouzar (1983; see App. IIA).
The following names, of long usage, are treated as validly published: Compositae (nom. alt.: Asteraceae; type: Aster L.); Cruciferae (nom. alt.: Brassicaceae; type: Brassica L.); Gramineae (nom. alt.: Poaceae; type: Poa L.); Guttiferae (nom. alt.: Clusiaceae; type: Clusia L.); Labiatae (nom. alt.: Lamiaceae; type: Lamium L.); Leguminosae (nom. alt.: Fabaceae; type: Faba Mill. [= Vicia L.]); Palmae (nom. alt.: Arecaceae; type: Areca L.); Papilionaceae (nom. alt.: Fabaceae; type: Faba Mill.); Umbelliferae (nom. alt.: Apiaceae; type: Apium L.). When the Papilionaceae are regarded as a family distinct from the remainder of the Leguminosae, the name Papilionaceae is conserved against Leguminosae.
The use, as alternatives, of the eight family names indicated as “nom. alt.” (nomen alternativum) in Art. 18.5 is authorized.