The Index Nominum Supragenericorum Plantarum Vascularium
Project is a joint effort between the
International Association for Plant Taxonomy, the
University of Maryland, and now Cornell University.
The purpose of the project is to capture all valid and legitimate extant vascular plants names, as defined by the International Code of Botanical Nomenclature, proposed above the rank of genus. These data are dynamic and constantly being updated. At any one time, the listing of a name means only that it is the earliest, valid place of publication found to date.
Data are presented by genus, by APGII families, and by taxonomic rank. A listing of abbreviations for suprageneric names is also available.
The goal of the project is merely to find and evaluate names, not to make taxonomic or emotional judgments. Names at the ranks of subfamily, tribe and subtribe are governed by strict priority. Above the rank of family, priority is not an issue. At present, only family names can be conserved. Three aspects of the Code are critical. First, many names currently accepted as validly published were published with a misplaced rank; for example, names proposed
as sections above the rank of genus are not validly published (Art. 33.9).
Second, names not published in Latin are not validly published (Art. 18.4).
Third, names above the rank of family, after 1 Jan 1935, must have a diagnosis or description in Latin, or a reference to a diagnosis or description in Latin (Art. 36.1). Mere citation of a validly published name is not sufficient to validate a name above the rank of family.
Early suprageneric names proposed can be validated in numerous ways. Mere
citation of an author name can serve as a direct reference, whereas mere
mention of an author in a work can be an indirect reference; both are sufficient
for valid publication. Rank is determined by its use in the entire work;
thus, mere mention of a rank for a single name can establish a rank for all
names (Art. 35.5). However, on or after 1 January 1908 ranks of suprageneric
names can be determined by termination if not rank is stated (Art. 35.2).
While most monographers know their technical literature, few appreciate
that the majority of suprageneric names often were first validly published
in the more general botanical, biological and medical literature. Many names
were validated in reviews of recently published works, especially when the
orthography of the name in the original publication was in French, German or
some language other than Latin. Encyclopedias, text books and other general
interest works are frequently reported here as the first valid place of suprageneric names.
Additional information is available on suprageneric names via my Vascular Plant Family Nomenclature page.
In any effort of this size, help and comments are much appreciated. Anyone
aware of a name validly published before that reported here is urged to send
a copy of the appropriate pages for review. Just finding the literature takes time. Anything to expedite the effort will be appreciated.
A SHORT ESSAY ON SUPRAGENERIC NOMENCLATURE
Names of taxa above the rank of family have not been rigorously studied, and the review here is not complete. Provisions in the current International Code of Botanical Nomenclature (McNeill et al., Regnum Veg. 146. 2006) dealing with names above the rank of family have improved considerably over previous edition of the Code. Of critical importance to modern authors of suprageneric names is that names above the rank of family published after 1 Jan 1935 must be accompanied with a description or diagnosis in Latin or a reference to a validly published description or diagnosis in Latin. This has resolved many inconsistencies practiced over the years regarding when suprageneric nomenclature (as least for spermatophytes and pterophytes) began and the parenthetical use of authorships for suprageneric names.
The following articles are mentioned here because of the their importance to suprageneric names:
Art. 3. Ranks are discussed in Arts. 3, 4 and 5. The principal suprageneric ranks are kingdom, division or phylum, class, order and family (Art. 3).
Art. 4. The rank of tribe may be inserted between family and genus (Art. 4.1). The sanctioned additional ranks are formed typically by adding the prefix "-sub" to form subkingom, subdivision or subphylum, subclass, suborder, subfamily and subtribe (Art. 4.2). Additional suprageneric ranks may be established as long as they do not cause confusion (Art. 4.3). Many modern authors have adopted the use of the rank superorder, and there is a growing movement for the use of supertribe (especially in Poaceae). A small number of workers have adopted superkingdom. My own view is that the prefix "super-" should be restricted to suprageneric names.
Art. 5. The relative order of the ranks may not be altered (Art. 5). Thus, in their full extent at the moment, the ranks in use are: superkingdom, kingdom, subkingdom, superdivision, division, subdivision, superclass, class, subclass, superorder, order, suborder, superfamily, family, subfamily, supertribe, tribe and subtribe.
Art. 13. Art. 13.1.a. This article establishes 4 Aug 1789, and A.L. Jussieu’s Genera plantarum, as the starting date for suprageneric names of spermatophytes and pteridiophytes.
Art. 16. Art. 16 was redrafted in the St. Louis Code making the provisions and nature of names above the rank of family much clearer. Art. 16.1 defines "automatically typified names" and "descriptive names." Examples are provided. New Art. 16.2 indicates that for automatically typified names, the typical subdivision of a higher rank is to be based on the type of the higher taxon. Using Magnoliophyta as an example, the names would be as follows:
Division Magnoliophyta Cronquist, Takht. & Zimmerm. ex Reveal
Subdivision Magnoliophytina Frohne & U. Jensen ex Reveal
Class Magnoliopsida Brongn.
Subclass Magnoliidae Novák ex Takht.
Superorder Magnolianae Takht.
Order Magnoliales Bromhead
Suborder Magnoliineae Engl.
Another, more complex example is the suprafamilial ranks associated with Solanaceae in some systems of classification:
Division Magnoliophyta Cronquist, Takht. & Zimmerm. ex Reveal
Subdivision Magnoliophytina Frohne & U. Jensen ex Reveal
Class Rosopsida Batsch
Subclass Lamiidae Takht. ex Reveal
Superorder Lamianae Takht.
Order Lamiales Bromhead
Suborder Solanineae Engl.
Note the terminations. All but the one for superorder ("-anae") are recommended by the Code when a name is based on a generic type (Rec. 16A). However, their use is not mandatory. Some terminations differ when the plant is an algal or a fungus. It is important to remember that while names above the rank of family are not subject to the principle of priority (Note 2), in choosing among typified names for such ranks, it is recommended authors follow the principle of priority (Rec. 16B.1).
Art. 16.3 makes it clear that names published with an incorrect Latin termination are to be corrected. For example, "Cactarieae" (Dumortier, 1829, based on Cactaceae) is to be corrected to Cactales Dumort. (Ex. 4). The articles states that suprageneric names published with a non-Latin ending are not to be accepted: "Acoroidées" (Kirschl., Fl. Alsace 2: 103. 1853-Jul 1857) is not to be accepted as "Acorales Kirschl.", because it has a French rather than a Latin termination (Ex. 5). At present, the Code assigns Acorales to Reveal (in Phytologia 79: 72. 1996), but other changes in the Code (e.g., Art. 18, Ex. 4) fundamentally changed the validation of hundreds of names so that Acorales Link (1835) is the first place of valid publication and a new example needs to be selected. Left vague in the Code is the situation when an author who uses French terminations for all other ranks uses the termination "-ales" for orders – is this Latin or French? The proposal submitted to the St. Louis nomenclature session to make clear that ordinal names published with the French "-ales" were to be considered as not validly published, but the question was left in limbo even in Vienna.
Art 17. This article is now reduced to two provisions. The first establishes the endings for names at the ranks of order ("-ales") and suborder {"-ineae"). The second permits to use of certain classical terms such as "cohors," "nixus," "alliance," and "Reihe" when proposed in a position equivalent to order to be considered as having been published at the rank of order.
Art. 18. This article deals with the establishment of family names. With the exception of a few alternative names (see Art. 18.5), all family names end in "-aceae" (Art. 18.1). Names of families proposed with their rank denoted by the term "order" or "natural order" (or an equivalent expression in another language) are to be considered as having been proposed at the rank of family (Art. 18.2).
Introduction of Art. 18, Ex. 4 has resulted in the sudden validations of a few hundred new family names, a change of authorship and place of publication for several hundred existing family names, and will have an as yet unknown impact upon infrafamilial names which were proposed based on names traditionally considered to have been published at the rank of family but are not validated at the rank of order (Art. 41.1). Every effort was made to correct App. IIB of the Vienna Code, but it is almost certain some places of publication were missed (see a list of known changes as of 2005).
Family names based on an illegitimate generic name are themselves illegitimate but may be validly published (Art. 18.3). The example, however, makes clear that while Caryophyllus Mill. and Wintera Murray are illegitimate, the family names Caryophyllaceae and Winteraceae are legitimate because they have been conserved (see App. IIB). Any name published with a non-Latin ending is not valid (Art. 18.4). This problem has largely disappeared since the Tokyo Code.
The Code allows certain classical family names to remain in use. Each is typified by the type of the name it is its alternate. For example, Brassicaceae Lindl. is typified on the genus Brassica. The classical name is Cruciferae Juss. Note that Brassicaceae is the alternative name, not i>Cruciferae. The classical names are: Palmae (Arecaceae), Compositae (Asteraceae), Cruciferae (Brassicaceae), Gramineae (Poaceae), Guttiferae (Clusiaceae), Labiatae (Lamiaceae), Leguminosae (Fabaceae) and Umbelliferae (Apiaceae). When the Fabaceae are circumscribed in a sense so as to exclude the Caesalpiniaceae and the Mimosaceae, the classical name Papilionaceae may be used. Be aware that Papilionaceae has priority over Leguminosae only when Caesalpiniaceae and Mimosaceae are excluded. In other words, Papilionaceae can not be used in a broad sense when the family is circumscribed to include Fabaceae, Caesalpiniaceae and Mimosaceae.
Art. 19. This article deals with names at the ranks of subfamily, tribe and subtribe. With one exception (see Art. 19.7), all names at the rank of subfamily are formed like those of family names (Art. 18.1) but end in "-oideae" instead of "-aceae" (Art. 19.1). In parallel to Art. 18.2, names intended as subfamilies but proposed as "suborders" are to be treated as if they were published as subfamilies.
The terminations for names at the ranks of tribe and subtribe are "-eae" and "-inae" respectively (Art. 19.3).
Art. 19.4 instructs authors that when they adopt a family names, they are to use names at the lower ranks (subfamily, tribe and subtribe) based on the same generic stem. Thus, the following example (with the correct authorships):
The type of the family name Rosaceae Juss. is Rosa L. and hence the subfamily and tribe which include Rosa are to be called Rosoideae Arn. and Roseae Lam. & DC. One could also add here Rosinae J. Presl to complete the sequence of available names and ranks.
Other authorship corrections to examples 2-4 are: Pooideae Benth. (Ex. 2); Ericoideae Arn., Ericeae DC. ex Duby, Rhododendroideae (Juss.) Sweet, Rhodoreae DC. ex Duby and Rhododendreae Colla (Ex. 3); and Asteroideae Cas., Asterinae Cass. Cichorioideae Chevall., Cichorieae Lam. & DC. and Cichoriinae Cass. ex Dumort. (Ex. 4).
Some authors have suggested recently that Art. 19.4 refers to the validity of names. I disagree. An example of this is as follows:
When Bentham (J. Proc. Linn. Soc., Bot. 5(Suppl. 2): 87. 1861) proposed Casearieae in the family Samydaceae Vent., he should have used Samydeae Dumort. as his Casearieae included both Casearia Jacq. and Samyda Jacq. Bentham's use of Casearieae is contrary to Art. 19.4, but this in and of itself does not render Casearieae invalid because, given the provisions in Art. 52.3, Casearieae is not nomenclaturally superfluous or illegitimate. Bentham's name was associated with a brief description in Latin and was properly published so that it is otherwise validly published. All suprageneric names are mononomials, not combinations (a term restricted to names below the rank of genus - see Art. 6.7), and thus for purposes of validly, Bentham's name is validly published, but in this particularly classification scheme (as a tribe of Samydaceae, it is not available.
Art. 19.5 is parallel to Art. 18.3 and to the missing provision that was to have been included in Art. 16. In short, as long as Caryophyllaceae, based on the illegitimate generic name Caryophyllus Mill. is conserved, one may adopt Caryophylloideae Arn. and Caryophylleae Lam. & DC.
Art. 19.6 makes clear that improper Latin terminations are to be corrected without alteration of the authorship or date of publication. However, names proposed with non-Latin terminations are not validly published. Finally, Art. 19.7 permits the use of an alternative name, Paplionoideae L. ex DC. when one adopts the alternative family name Papilionaceae Giseke. Both names are typified on Faba Mill., the type of Fabaceae Lindl.
Authorship corrections to Rec. 19A, Ex. 2 are: Pyroloideae Kostel. and Pyroleae Dumort., Monotropoideae Arn. and Monotropeae Dumort., and Vaccinioideae Arn. and Vaccinieae Rchb.
Art. 32. Contrary to often accepted practice, all names regardless of rank must be accompanied by a validating description or diagnosis. Numerous modern authors simply use suprageneric names without concern for their validity (Art. 32.1).
After 1 Jan 1953 there must be a full and direct reference to the basionym or to any validating Latin description or diagnosis (Art. 32.4). Mere mention of an author's name, or citation of a reference in the work does not constitute and full and direct reference.
An aspect of the Code that nearly everyone finds difficult is how indirect can an indirect reference be and still be a reference. While Art. 32.4 provides a concise definition, the examples associated with Arts. 32.4 provide considerable guidance. Art. 32 Ex. 5 clearly states that an indirect reference can be as simply as an author's name, while Ex. 7 reminds us that an indirect reference may be nothing more than a statement in the text alluding to another work with effectively published descriptions or diagnoses. These are relatively straight forward but often require reviewing an entire work for evidence of validity and not just the place where the name appears.
In the case of many suprageneric names one finds the reference in the form of a name, typically a descriptive name associated with one based on a generic stem. An example is Altingiaceae Horan. (Tetractys: 25. 1843) who gave only the expression "(Altingiaceae (s. Balsamifluae))." To the uninformed this may appear to be a nom. nud., but in fact "Balsamifluae" refers to an invalidly published (Art. 18.1) Blume & J. Fischer (Fl. Javae 17-18: 3. 1829) name associated with an effectively published description (see Art. 32.1(c)). As such there is an indirect reference to a description or diagnosis.
Early authors who were particularly prolific, like Cassini, often evolved in both their use of ranks and in the use of Latinized names. Thus, when a subsequent author refers to '"Cass.'" for a name in a new rank, one can rarely be certain which of Cassini's many works they are indirectly citing. Cassini initially used sections within three families that he later considered to be tribes in one family. He also published his names in both Latin and French. Thus, it can make a difference as to whether the name, say, at the rank of subfamily is a new name or one in which rank is finally assigned to a Cassini name.
For example, Alismatoideae Lam. & DC. ex Arn. (Encycl. Brit., ed. 7, 5: 136. 1832) is accompanied by a description in English, so it is validate by this action alone. Nonetheless, he attributes the name to Lamarck and Candolle who proposed the unranked name Alismoideae Lam. & DC. (Syn. Pl. Fl. Gall.: 155. 1806) which is associated with a brief description in Latin. Accordingly, I have given the above authorship. A somewhat more complex example is the following: Pfeiffer validated the subtribe Athanasinae Lindl. ex Pfeiff., (Nomencl. Bot. 1(1): 323. 1872) by providing a reference to an invalidly (Art. 33.5) published A.P. de Candolle (in J. Lindley, Intr. Nat. Syst. Bot., ed. 2: 260. 1836, as Division "Athanasieae Less.”) name that might seem to be a nom. nud., but in fact is that there is a reference to the unranked name Athanasieae Less. (Syn. Gen. Compos.: 262. 1832) with a brief description in Latin which effectively validates the Pfeiffer name.
In dealing with suprageneric names, one must have a full understanding of the literature to evaluate the validity of these names.
Suprageneric names published with an incorrect Latin termination are to be corrected (Art. 32.7). Priority resides with who proposed the name, not with the person who first supplied the correct ending or even the correct orthography.
Autonyms (see Art. 6.8) are not operative for names at ranks above that of genus. Therefore, Art. 32.8 is not applicable to suprageneric names which are mononomials. This is an all-too-common error even in today’s botanical literature.
Art. 33. Suprageneric names are not combinations and therefore articles dealing with combination (e.g., Art. 33.1) are not applicable. However, suprageneric names frequently are “transferred” from one rank to another. The word “transfer” is not in the Code but the concept is implied when a name is moved from one rank to another. The important point here is that the use of the expression “comb. nov.” for suprageneric names in not only inappropriate it is incorrect.
Errors of citation of basionyms does not, in and of itself, invalidate a name as long as the error occurred prior to 1 Jan 1953 (Art. 33.2). This is particularly true when dealing with an indirect reference. For example, one might attribute a name to one individual who did not validate the name but through that reference can get to another individual who did validate the name. Again, the example of Pfeiffer who validated the subtribe Athanasinae Lindl. ex Pfeiff., (Nomencl. Bot. 1(1): 323. 1872) by providing a reference to an invalidly (Art. 33.5) published A.P. de Candolle (in J. Lindley, Intr. Nat. Syst. Bot., ed. 2: 260. 1836, as Division "Athanasieae Less.”) name can be mentioned inasmuch as there is a reference (e.g., Lessing) in Lindley to the unranked name Athanasieae Less. (Syn. Gen. Compos.: 262. 1832) with a brief description in Latin which effectively validates the Pfeiffer name.
Art. 33.3 applies only to "new combinations or avowed substitutes." It allows errors in bibliographic citation to be corrected. No such provision exists specifically for suprageneric names except that "transfers" of rank (e.g., from tribe to subfamily) involving names above the rank of genus indeed are "avowed substitutes" and therefore Art. 33.3 is applicable.
Art. 33.9 appears to be largely unknown to most users of suprageneric names. A taxon proposed at a misplaced rank is not validly published. Numerous authors have attributed the rank of tribe to names actually proposed as sections. This has declined since the mid 1990s but still happens. Historically, it is not uncommon to find the rank family recognized within a natural order (=family, see Arts. 18.2 and 19.2) with such secondary names considered to be invalid. This is no longer the case because of Art. 18, Ex. 4 and has resulted in the sudden validation in 2005 of hundred of new names at the ranks of order and family and the changes in the bibliographic data for several hundred more. Therefore, it is now critical to ascertain the rank of a validating name if a family, or in particular a subfamily, tribe or subtribe name was based on a seemingly previously valid family name. If that family name is now, because of Art. 18, Ex. 4, at the rank of order, Art. 41.1 comes into play and such names are not valid.
Art. 33.11 resolves a long-standing problem by declaring that when the same rank-denoting term is used at more than one non-successive position in the taxonomic sequence the names are to be treated as unranked even though, in the case of Art. 33, Ex. 27, the rank of series was used. This does not impact the validity of the subsequent validation of such names at a different rank inasmuch as with suprageneric names a name is validated by the validating description of the basionym, not the name itself (e.g., suprageneric names are NOT combinations).
Art. 35. This article deals with indication of rank. This is particularly troublesome in the early literature. After 1 Jan 1953, a rank must be stated, but before that date, it is open to interpretation although after 1 Jan 1908, a rank can be determined solely by the termination (Art. 35.2). The termination must correspond with those listed in Rec. 16A1-3, Art. 17.1, 18.1, 19.1 and 19.3 and the position of the named rank can not be in conflict with the order of ranks established in Arts. 3-5. The two examples are useful:
Jussieu (Mém. Mus. Hist. Nat. 12: 497. 1827) proposed Zanthoxyleae without specifying the rank. Although he employed the present termination for tribe (-eae), that name, being published prior to 1908, is unranked.
Zanthoxyleae Dumort. (Anal. Fam. Pl.: 45. 1829), however, is a tribal name, as Dumortier specified its rank.
Nakai (Chosakuronbun Mokuroku [Ord. Fam. Trib. Nov.], 1943) validly published the names Parnassiales, Lophiolaceae, Ranzanioideae, and Urospatheae. He indicated the respective ranks of order, family, subfamily, and tribe, by virtue of their terminations, even though he did not mention these ranks explicitly.
Names published without a rank, or merely a term (e.g., Gruppe) are rankless, but are usually effectively as well as validly published (Art. 35.3). For purposes of priority, this is fixed when a rank (Arts. 3-5) is attributed to a name. Art. 35.4 is often ignored. If a name is not given a rank on one page, but names are given ranks elsewhere in the same publication (or different parts of the work), then rank might be established. Many workers simply look at a name on a page not realizing that what may appear to be a rankless name may, in fact, not be. For example, all names in some works are rankless except in the discussion the author refers to a specific rank (e.g., all of Bartling's (1830) names below the rank of family are seemingly rankless; only in two places does he use a rank, e.g. tribe).
Often ranks are given but not in a consistent fashion. Lindley (1836) frequently referred to one of his families as a "tribe" but not in the sense of a rank. Below the rank of family he used tribe and section, yet in other instances no rank is given at all. In such cases, the names may be (a) valid with a rank, (b) invalid, or (c) rankless but valid providing all other appropriate provisions of the Code were fulfilled.
Art. 36. Contrary to often accepted practice, all names regardless of rank proposed after 1 Jan 1935 must be accompanied by a validating description or diagnosis in Latin or have a reference to an effectively published description or diagnosis in Latin.
Numerous modern authors have proposed suprageneric names without concern for this requirement. For example, when Cronquist, Takhtanjan and Zimmermann proposed "Magnoliophyta" in their 1966 Taxon article, they cited as the basis of their name a validly published 1862 name that was associated with a description in German rather than Latin. As a result their name was not validly published. It was formally validated several years later, but long after it had become widely accepted.
Unlike a new combination, the creation of new suprageneric names by transferring a name from one rank to another, names above the rank of family must have a validating description or diagnosis in Latin or have a reference to an effectively published description or diagnosis in Latin whereas at the ranks of family and below Art. 41.1 allows much names to be treated like a new combination. In this case, the only requirement is that the reference must be to a valid name at the rank of family or a subdivision of the family.
A significant aspect of Art. 36 when it comes to suprafamilial names is that the validating description or diagnosis can be a name at any rank. Without a provision in Art. 41 restricting validating ranks for suprafamilial names there is a growing tendency to validate a suprafamilial name on a generic description when the suprafamilial name is itself monogeneric and often even monospecific. An example is given below:
Eucommiales Nemejc ex Cronquist, Integr. Syst. Class. Fl. Pl.: 182. 10 Aug 1981. Validated by a full and direct reference to the Latin description associated with Eucommia W.B. Oliv. in Icon. Pl. 20: ad t. 1950. Oct 1890 nom. cons.
Art. 41. This article provides certain excepts to the rules in Arts. 32-45 when it comes to names at the rank of family or any subdivision of a family above the rank of genus. Unlike names above the rank of genus which must, after 1 Jan 1935, can only be validated by a description or diagnosis in Latin, for family and infrafamilial names a reference needs only be to a reference to a description or diagnosis. Furthermore, up to 1 Jan 1953, the reference can even be indirect.
Critical to determining what is a validating description or diagnosis in Art. 41, Ex. 2 for here it clearly states that the validating description must be associated with a name that (a) is at a stated rank and (b) that rank must be at the rank of family or a subdivision of a family. Given Art. 18, Ex. 4 which fundamentally changed nearly two centuries of tradition, this has had a substantial impact on validation of family and infrafamilial names. The full extent of this, especially below the rank of family, is unknown.
ACKNOWLEDGMENTS: Work on this project has been supported by the University of Maryland and the International Association for Plant Taxonomy, in cooperation with the National Agricultural Library. Continued support is now provided by Cornell University. I am grateful for the help provided by Drs. Richard Brummitt, Alexander Doweld, Kanchi Gandhi, Werner Greuter, the late Ruurd D. Hoogland, John McNeill, Nicholas Turland, and a host of others who have provided information on suprageneric names. This work would not be possible without the aid and assistance of numerous librarians and libraries, the most notable being those at Harvard University Herbaria, The Natural History Museum in London, The New York Botanical Garden, McKeldin Library at the University of Maryland, Cornell University Libraries, and the Smithsonian Institution in Washington, D.C.
jlr326@cornell.edu.
Posted: 15 Sep 1995; Revised and updated 15 Nov 1998; revised: 14 Apr 2000, 12 Aug 2004, 6 Feb 2008
