An end to all things? — plants and their names
Peter F. StevensMissouri Botanical Garden, PO Box 299, St Louis, MO 63166-0299, USA. Email: peter.stevens@mobot.org
Australian Systematic Botany 19(2) 115-133 https://doi.org/10.1071/SB05011
Submitted: 6 May 2005 Accepted: 3 October 2005 Published: 28 April 2006
Abstract
Great advances in our understanding of phylogenetic relationships have been made over the last decade and a half. Major clades in many groups, including flowering plants, now show substantial stability both in terms of content and relationships. This makes possible the development of a system in which only monophyletic ( = holophyletic) entities are named, entities that represent all and only the descendants of a common ancestor. However, some argue that use of Linnaean ranked names is inappropriate in such circumstances; this argument is bolstered by appeals to history and philosophy. Those who doubt the wisdom and / or very possibility of naming only monophyletic groups also argue that their position follows from history, or that ancestors cannot be incorporated into a Linnaean-type classification and that ancestors are an integral part of monophyletic groups. However, I argue that most of the apparently more cosmic issues brought up in this debate are based on a combination of a misunderstanding of the nature and purpose of language, fallacious reasoning and dubious—and largely irrelevant—interpretations of history. A flagged hierarchy helps memory and communication. Binomials in particular simply represent the noun–adjective combinations of ordinary language in a Latinised form, and are too valuable a communication device to be discarded because rank has been demonised. However, hierarchies can be misinterpreted and cannot be made complex enough to cope with the much more detailed phylogenies being produced. Thinking of naming systems as conventions may help clarify what we should be doing, if we are not to squander both the time and the reputation of systematics. Time is in short supply and our reputation not what it might be; solving the less cosmic issues may involve a self-discipline that also seems in short supply in the systematic community.
Acknowledgments
I thank two reviewers for their comments, and Brent Mishler and Marc Sosef for raising some interesting issues. I am particularly grateful to Andrew Doust, Micah Dunthorn, Lucia Lohmann, Beto Vicentini and Felipe Zapata for discussion and help with the figures, and to participants at the 2003 annual conference of the Australian Systematic Botany Society, Oregon State University, the Systematics Discussion Group at the Missouri Botanical Garden, and the BioLunch at the University of Missouri, Saint Louis, for helping me clarify several points.
Aarsleff, H (1976). Wilkins, John. In ‘Dictionary of scientific biography, vol. XIV, Verrill-Zwelfer’. pp. 361–381. (Charles Scribner’s Sons: New York)
Anderberg AA,
Rydin C, Kallersjö M
(2002) Phylogenetic relationships in the order Ericales s.l.: analyses of molecular data from five genes from the plastid and mitochondrial genomes. American Journal of Botany 89, 677–687.
(verified 28 November 2005)
Stevens PF
(2002) Why do we name organisms? Some reminders from the past. Taxon 51, 11–26.
had used the term monophyly, and its redefinition by phylogenetic systematists did not sit well with some evolutionary systematists, who preferred to call Hennigian monophyly, holophyly (e.g. Ashlock 1971; Mayr and Ashlock 1991). Their definition of monophyly was broader: a monophyletic taxon was one whose members descended from another taxon of the same or lower rank (e.g. Simpson 1961; Mayr 1969; Mayr and Ashlock 1991). This would include paraphyletic taxa as recognised here.
2 Strictly speaking, Linnaean nomenclature refers to Linnaeus’s generic names and the polynomials he used for species, e.g. Linnaeus (1751).
3 3Of course, some vigorously reject the value or applicability of ideas of individuality in systematics (e.g. Nixon and Carpenter 2000 and refs) or at least classification (Rapini 2004), or that species are clades (Nixon and Wheeler 1990). Others have redefined ‘class’, removing some connotations that are perceived to be negative (e.g. Mahner and Bunge 1997; Boyd 1999; Keller et al. 2003).
4 This is connected to a confusing debate as to whether (and how) taxon names are defined (e.g. de Queiroz 1997; Stuessy 2001; Monsch 2003; Schuh 2003; Laurin and Anderson 2004). Of course, clades can be recognised only if there are apomorphic characters, whether morphological or molecular, supporting their existence. In PC apomorphy-based definitions such features are used to diagnose the name, and they can also be used in mL diagnoses or characterisations. It is unhelpful to equate apomorphies and essences.
5 The comparable numbers in de Candolle (1844) are 6000 generic names, 1000 epithets, and 80 000 species, but millions of species could be unambiguously named using those 7000 words.
6 Note that if more than one apomorphy is used in an apomorphy-based definition, the one denoting the least inclusive clade may be chosen (Pleijel and Rouse 2002). Parallelisms will not cause problems since the apomorphy is linked to its occurrence in a particular taxon, e.g. in Magnoliaceae, sheathing stipules as found in Magnolia virginiana (Pleijel and Rouse 2002; see also PCn for complex apomorphies). Nevertheless, assigning particular apomorphies to an unambiguous position on the tree can be difficult (e.g. Stevens 2001b), even given a stable hypothesis of phylogeny. Thus one possible apomorphy of angiosperms, triploid endosperm (see e.g. Nixon and Carpenter 2000), is currently of uncertain position (Williams and Friedman 2004; for similar problems with apomorphies of Magnoliales, cf. Sauquet et al. 2003).
7 There have been extensive arguments whether monotypic taxa with empty ranks are permissible in a Linnaean hierarchy (‘Gregg’s paradox’: Gregg 1950, 1968; Buck and Hull 1969; Hull and Snyder 1969). Within families, the arrangement of genera remains alphabetical.
8 Alternatively, one could hire a nomenclator, somebody who could whisper in your ear the full clade address whenever a uninomial was mentioned. Cicero would have understood.
9 A reorganisation of part of a larger herbarium (the Harvard University Herbaria at Cambridge, MA) along phylogenetic principles was carried out as some 1 400 000 specimens were moved in 1997.
