International Code of Phylogenetic Nomenclature (PhyloCode)

Preface

Version 6 is the first version published as a printed volume. Previous versions were solely electronic and are available at www.phylocode.org. The material in this Preface has been summarized from a variety of sources; see the History section for literature citations.

The development of the International Code of Phylogenetic Nomenclature (referred to here as the PhyloCode) grew out of the recognition that the current rank-based systems of nomenclature, as embodied in the current botanical, zoological, and bacteriological codes, are not well suited to govern the names of clades. Clades (along with species) are the entities that make up the tree of life, and for this reason they are among the most theoretically significant biological entities above the organism level. Consequently, clear communication and efficient storage and retrieval of biological information require names that explicitly and unambiguously refer to clades and do not change over time. The current rank-based codes fail to provide such names for clades. Supraspecific names are not always associated with clades under the rank-based codes, and even when they are, they often fail to retain their associations with particular clades because the names are implicitly defined in terms of ranks and types. A clade whose hypothesized composition and diagnostic characters have not changed may be given a different name under the rank-based codes based purely on considerations of rank. Such instability is particularly objectionable given the wide recognition that rank assignment is subjective and of dubious biological significance.

In contrast to the rank-based codes, the PhyloCode provides rules for the express purpose of naming clades through explicit reference to phylogeny. In doing so, the PhyloCode extends “tree-thinking” to biological nomenclature. This development parallels the extension of tree-thinking into taxonomy, as manifested in the concepts of species as lineage segments and supraspecific taxa as clades. These nomenclatural and taxonomic developments are complementary but independent. Clades can be named using the traditional rank-based systems of nomenclature (though with the problems noted above), and a nomenclatural system based on phylogenetic principles does not require equating supraspecific taxa with clades. The PhyloCode, however, is designed for the specific purpose of naming clades.

The objective of the PhyloCode is not to replace existing names but to provide an alternative system for governing the application of both existing and newly proposed names. In developing the PhyloCode, much thought has been given to minimizing disruption of the existing nomenclature. Thus, rules and recommendations have been included to ensure that most names will be applied in ways that approximate their current and/or historical use. However, names that apply to clades will be redefined in terms of phylogenetic relationships rather than taxonomic rank and therefore will not be subject to the subsequent changes that occur under the rank-based systems due to changes in rank. Because the taxon membership associated with particular names will sometimes differ between rank-based and phylogenetic systems, suggestions are provided for indicating which code governs a name when there is a possibility of confusion. Mechanisms are also provided to reduce certain types of nomenclatural divergence relative to the rank-based systems. For example, if a clade name is based on a genus name, the type of the genus under the appropriate rank-based code must be used as an internal specifier under the PhyloCode (Article 11.10, Examples 1 and 2).

The starting date of the PhyloCode coincides with the publication of Phylonyms, a volume that provides phylogenetic definitions for many widely used clade names and the names of many large clades (see below). Names for which phylogenetic definitions were published before that date, and not subsequently, are not considered established under this code.

Properties of Phylogenetic Nomenclature

The phylogenetic system of nomenclature embodied in the PhyloCode exhibits both similarities to and differences from the rank-based systems embodied in the traditional codes. Some of the most important similarities are as follows: (1) Both systems have the same fundamental goals of providing unambiguous methods for applying names to taxa, selecting a single accepted name for a taxon from among competing synonyms or homonyms, and promoting nomenclatural stability and continuity to the extent that doing so does not contradict new results and conclusions. (2) Neither system infringes upon the judgment of taxonomists with respect to inferring the composition of taxa or to assigning taxonomic ranks. (3) Both systems use precedence, a clear order of preference, to determine the correct name of a taxon when synonyms or homonyms exist. (4) Both systems use the date of publication (chronological priority) as the primary criterion for establishing precedence. (5) And both phylogenetic and rank-based systems have conservation mechanisms that allow a later-established name to be given precedence over an earlier name for the same taxon if using the earlier name would be contrary to the fundamental goal of promoting nomenclatural stability and continuity.

Some of the most important differences between the phylogenetic system of the PhyloCode and the rank-based systems of the traditional codes are as follows: (1) The phylogenetic system is independent of taxonomic rank. Although clades form nested hierarchies, the assignment of taxonomic rank is not part of the naming process and has no bearing on the spelling or application of clade names. As a consequence, the phylogenetic system does not require ranked taxonomies. (2) All taxa named under this code are clades. Clades are products of evolution that have an objective existence regardless of whether they are named. As a consequence, once a clade is named and its name associated with a phylogenetic definition, its composition and diagnostic characters become questions to be decided by empirical evidence rather than by personal decisions. (3) In addition to applying names to nested and mutually exclusive taxa, as in traditional nomenclature, the phylogenetic system allows names to be applied to partially overlapping clades. This provision is necessary to accommodate situations involving clades of hybrid origin. (4) In contrast to the rank-based codes, which use (implicit) definitions based on ranks and types to determine the application of names, phylogenetic nomenclature uses explicit phylogenetic definitions. Species, specimens, and apomorphies cited within these definitions are called specifiers because they are used to specify the clade to which the name applies. These specifiers function analogously to the types of rank-based nomenclature in providing reference points that determine the application of a name; however, they differ from types in that they may either be included in or excluded from the taxon being named, and multiple specifiers may be used. (5) The fundamental difference between the phylogenetic and rank-based systems in how names are defined leads to differences in how synonyms and homonyms are determined in practice. For example, under the PhyloCode, synonyms are names whose phylogenetic definitions specify the same clade, regardless of prior associations with particular ranks; in contrast, under the rank-based codes, synonyms are names at the same rank whose types are included within a single taxon at that rank, regardless of prior associations with particular clades. (6) Another novel aspect of the PhyloCode is that it permits taxonomists to restrict the application of names with respect to clade composition. If a taxonomist wishes to ensure that a name refers to a clade that either includes or excludes particular taxa, this result may be achieved through the use of additional internal or external specifiers (beyond the minimal number needed to specify a clade), or the definition may contain a qualifying clause specifying conditions under which the name cannot be used. (7) The PhyloCode includes recommended naming conventions that promote an integrated system of names for crown and total clades. The resulting pairs of names (e.g., Testudines and Pan-Testudines for the turtle crown and total clades, respectively) enhance the cognitive efficiency of the system and provide hierarchical information. (8) Establishment of a name under the PhyloCode requires both publication and registration. The purpose of registration is to create a comprehensive database of established names (discussed below), which will reduce the frequency of accidental homonyms and facilitate the retrieval of nomenclatural information.

Advantages of Phylogenetic Nomenclature

Phylogenetic nomenclature has several advantages over the traditional systems. It eliminates a major source of instability under the rank-based codes—changes in clade names due solely to shifts in rank. It also facilitates the naming of new clades as they are discovered. Under the rank-based codes, it is often difficult to name clades one at a time, similar to the way that species are named, because the name of a taxon is affected by the taxon’s rank, which in turn depends on the ranks of more and less inclusive taxa. In a group in which the standard ranks are already in use, naming a newly discovered clade requires either the use of an unconventional intermediate rank (e.g., supersubfamily) or the shifting of less or more inclusive clades to lower or higher ranks, thus causing a cascade of name changes. This situation discourages systematists from naming clades until an entire classification is developed. In the meanwhile, well-supported clades are left unnamed, and taxonomy falls progressively farther behind knowledge of phylogeny. This is a particularly serious drawback at the present time, when advances in molecular and computational biology have led to a burst of new information about phylogeny, much of which is not being incorporated into taxonomy. The availability of the PhyloCode will permit researchers to name newly discovered clades much more easily than they can under the rank-based codes. For many researchers, naming clades is just as important as naming species. In this respect, the PhyloCode reflects a philosophical shift from naming species and subsequently classifying them (i.e., into higher taxa) to naming both species and clades. This does not mean, however, that all clades must be named. The decision to name a clade (or to link an existing name to it by publishing a phylogenetic definition) may be based on diverse criteria, including (but not restricted to), level of support, phenotypic distinctiveness, economic importance, and whether the clade has historically been named.

Another benefit of phylogenetic nomenclature is that it permits (though it does not require) the abandonment of categorical ranks, which would eliminate the most subjective aspect of traditional taxonomy. The arbitrary nature of ranking, though acknowledged by most taxonomists, is not widely appreciated by non-taxonomists. The existence of ranks encourages researchers to use taxonomies inappropriately, treating taxa at the same rank as though they were comparable in some biologically meaningful way—for example, when they count genera or families to study past and present patterns of biological diversity. A rankless system of taxonomy, which is permitted but not required by the PhyloCode, encourages the development of more appropriate uses of taxonomies in such studies, such as counting clades or species that possess properties relevant to the question of interest, or investigating the evolution of those properties on a phylogenetic tree.

An advantage of the PhyloCode over the rank-based codes is that it applies at all levels of the taxonomic hierarchy. In contrast, the zoological code does not extend its rank-based method of definition above the level of superfamily, and the botanical code extends that method of definition only to some names above the rank of family (automatically typified names) and the principle of priority is not mandatory for those names. Consequently, at higher levels in the hierarchy, the rank-based codes permit multiple names for the same taxon as well as alternative applications of the same name. Thus, as phylogenetic studies continue to reveal many deep clades, there is an increasing potential for nomenclatural chaos due to synonymy and homonymy. By imposing rules of precedence on clade names at all levels of the hierarchy, the PhyloCode will improve nomenclatural clarity at higher hierarchical levels.

History

The theoretical foundation of the PhyloCode was developed in a series of papers by de Queiroz and Gauthier (1990, 1992, 1994), which were foreshadowed by earlier suggestions that a taxon name could be defined by reference to a part of a phylogenetic tree (e.g., Ghiselin, 1984). The theory was in development for several years before the first of these theoretical papers was published, and related theoretical discussions (e.g., Rowe, 1987; de Queiroz, 1988; Gauthier et al., 1988; Estes et al., 1988) as well as explicit phylogenetic definitions (Gauthier, 1984, 1986; Gauthier and Padian, 1985; de Queiroz, 1985, 1987; Gauthier et al., 1988; Estes et al., 1988; Rowe, 1988) were published in some earlier papers. Several other papers contributed to the development of phylogenetic nomenclature prior to the Internet posting of the first version of the PhyloCode in 2000 (Rowe and Gauthier, 1992; Bryant, 1994, 1996, 1997; de Queiroz, 1992, 1994, 1997a, b; Sundberg and Pleijel, 1994; Christoffersen, 1995; Schander and Thollesson, 1995; Lee, 1996a, b, 1998a, b, 1999a, b; Wyss and Meng, 1996; Brochu, 1997; Cantino et al., 1997, 1999a, b; Kron, 1997; Baum et al., 1998; Cantino, 1998; Eriksson et al., 1998; Härlin, 1998, 1999; Hibbett and Donoghue, 1998; Moore, 1998; Schander, 1998a, b; Mishler, 1999; Pleijel, 1999; Sereno, 1999). Other papers during this period applied phylogenetic nomenclature to particular clades (e.g., Judd et al., 1993, 1994; Holtz, 1996; Roth, 1996; Alverson et al., 1999; Swann et al., 1999; Brochu, 1999; Bremer, 2000).

Three early symposia increased awareness of phylogenetic nomenclature. The first one, organized by Richard G. Olmstead and entitled “Translating Phylogenetic Analyses into Classifications,” took place at the 1995 annual meeting of the American Institute of Biological Sciences in San Diego, California, USA. The 1996 Southwestern Botanical Systematics Symposium at the Rancho Santa Ana Botanic Garden in Claremont, California, USA, organized by J. Mark Porter and entitled “The Linnean Hierarchy: Past, Present and Future,” focused in part on phylogenetic nomenclature. Philip Cantino and Torsten Eriksson organized a symposium at the XVI International Botanical Congress in St. Louis, Missouri, USA (1999), entitled “Overview and Practical Implications of Phylogenetic Nomenclature.” A few critiques of phylogenetic nomenclature (Lidén and Oxelman, 1996; Dominguez and Wheeler, 1997; Lidén et al., 1997) and responses (Lee, 1996a; de Queiroz, 1997b; Schander, 1998a) were also published during this period, but the debate became much more active after the posting of the first version of the PhyloCode (see below).

The preparation of the PhyloCode began in the autumn of 1997, following a decision by Michael Donoghue, Philip Cantino, and Kevin de Queiroz to organize a workshop for this purpose. The workshop took place August 7–9, 1998 at the Harvard University Herbaria, Cambridge, Massachusetts, USA, and was attended by 27 people from five countries: William S. Alverson, Harold N. Bryant, David C. Cannatella, Philip D. Cantino, Julia Clarke, Peter R. Crane, Noel Cross, Kevin de Queiroz, Michael J. Donoghue, Torsten Eriksson, Jacques Gauthier, Kancheepuram Gandhi, Kenneth Halanych, David S. Hibbett, David M. Hillis, Kathleen A. Kron, Michael S. Y. Lee, Alessandro Minelli, Richard G. Olmstead, Fredrik Pleijel, J. Mark Porter, Heidi E. Robeck, Timothy Rowe, Christoffer Schander, Per Sundberg, Mikael Thollesson, and André R. Wyss. An initial draft of the code prepared by Cantino and de Queiroz was provided to the workshop participants in advance and was considerably revised by Cantino and de Queiroz as a result of decisions made at the meeting. The initial draft of Article 22 (Governance) was written by F. Pleijel, A. Minelli, and K. Kron and subsequently modified by M. Donoghue and P. Cantino. The initial draft of what is now Recommendation 11.10B was contributed by T. Rowe. An earlier draft of what is now Article 10.10 was written by Gerry Moore, who also provided Example 1. Article 8 and Appendix A (both of which concern registration) were written largely by T. Eriksson. William M. Owens provided the Latin terms in Article 9.2. Whenever possible, the writers of the PhyloCode used as a model the draft BioCode (Greuter et al., 1998), which attempted to unify the rank-based approach into a single code. Thus, the organization of the PhyloCode, some of its terminology, and the wording of certain rules are derived from the BioCode. Other rules are derived from one or more of the rank-based codes, particularly the versions of the botanical and zoological codes that were in effect at that time (Greuter et al., 1994, 2000; International Commission on Zoological Nomenclature, 1985, 1999). However, many rules in the PhyloCode have no counterpart in any code based on taxonomic ranks because of fundamental differences in the definitional foundations of the alternative systems.

The first public draft of the PhyloCode was posted on the Internet in April 2000. Its existence was broadly publicized in the systematic biology community, and readers were encouraged to submit comments and suggestions. All comments received were forwarded to the advisory group via a listserver, and many of them elicited discussion. Numerous commentaries about phylogenetic nomenclature have been published since the first public posting of the PhyloCode, some of them critical (Benton, 2000, 2007; Nixon and Carpenter, 2000; Stuessy, 2000, 2001; Forey, 2001, 2002; Lobl, 2001; Berry, 2002; Blackwell, 2002; Jørgensen, 2002, 2004; Carpenter, 2003; Janovec et al., 2003; Keller et al., 2003; Kojima, 2003; Moore, 2003; Nixon et al., 2003; Schuh, 2003; Barkley et al., 2004; Wenzel et al., 2004; Pickett, 2005; Polaszek and Wilson, 2005; Tang and Lu, 2005; Monsch, 2006; Rieppel, 2006; Stevens, 2006; Platnick, 2009, 2012, 2013), some supportive (Bremer, 2000; Cantino, 2000, 2004; de Queiroz, 2000, 2006; Brochu and Sumrall, 2001; de Queiroz and Cantino, 2001a, b; Ereshefsky, 2001; Laurin, 2001, 2005; Lee, 2001; Bryant and Cantino, 2002; Bertrand and Pleijel, 2003; Pleijel and Rouse, 2003; Donoghue and Gauthier, 2004; Laurin, 2005a, b, 2008; Laurin et al., 2005, 2006; Lee and Skinner, 2007; de Queiroz and Donoghue, 2011, 2013), and some pointing out both advantages and disadvantages (Langer, 2001; Stevens, 2002). Other publications since 2000 have discussed properties of different kinds of phylogenetic definitions ( Gauthier and de Queiroz, 2001), the application of widely used names to a particular category of clades (Anderson, 2002; Laurin, 2002; Joyce et al., 2004; Laurin and Anderson, 2004; Donoghue, 2005; Sereno, 2005), the conversion of rank-based names to phylogenetically defined names (Joyce et al., 2004), the choice of specifiers (Lee, 2005; Sereno, 2005; Wilkinson, 2006), the number of specifiers (Bertrand and Härlin, 2006), the subjective nature of Linnaean categories (Laurin, 2010), the application of phylogenetic nomenclature to species or least inclusive clades (Pleijel and Rouse, 2000, 2003; Artois, 2001; Hillis et al., 2001; Lee, 2002; Spangler, 2003; Dayrat et al., 2004; Dayrat, 2005; Dayrat and Gosliner, 2005; Fisher, 2006; Wolsan, 2007), the relevance of phylogenetic nomenclature to phyloinformatics (Donoghue, 2004; Hibbett et al., 2005), the logic and symbolic representation of phylogenetic definitions (Sereno, 2005), the philosophy of different approaches to phylogenetic nomenclature (Härlin, 2003a, b; Pleijel and Härlin, 2004), the use of phylogenetic nomenclature without a code (Sereno, 2005), guidelines for interpreting and establishing pre-PhyloCode phylogenetic definitions after the PhyloCode is implemented (Taylor, 2007), similarities between phylogenetic nomenclature and nomenclature as practiced by 18th and early 19th century naturalists (de Queiroz, 2005, 2012), the possibility of combining elements of phylogenetic and rank-based nomenclature (Kuntner and Agnarsson, 2006), and the development of an integrated approach to naming crown and total clades (de Queiroz, 2007). There have also been many applications of phylogenetic nomenclature to particular clades (some early examples, in addition to those cited in the first paragraph of this section, are Donoghue et al., 2001; Gauthier and de Queiroz, 2001; Maryanska et al., 2002; Modesto and Anderson, 2004; Smedmark and Eriksson, 2002; Wolfe et al., 2002; Stefanovic et al., 2003; clarke-2004">Clarke, 2004; Joyce et al., 2004; Sangster, 2005; Taylor and Naish, 2005; Cantino et al., 2007).

A second workshop on phylogenetic nomenclature was held at Yale University, July 28–30, 2002, organized by Michael Donoghue, Jacques Gauthier, Philip Cantino, and Kevin de Queiroz. There were 20 attendees from five countries, four of whom were observers. The active (voting) participants were Christopher Brochu, Harold Bryant, Philip Cantino, Kevin de Queiroz, Michael Donoghue, Torsten Eriksson, Jacques Gauthier, David Hibbett, Michel Laurin, Brent Mishler, Gerry Moore, Fredrik Pleijel, J. Mark Porter, Greg Rouse, Christoffer Schander, and Mikael Thollesson. Sixteen proposed changes in the rules and recommendations were discussed, 11 of which were approved. (Many other minor wording changes had already been circulated by e-mail and approved in advance of the workshop.)

In addition to specific rule changes, the 2002 workshop focused on several larger issues, the most fundamental of which concerned the governance of species names. The first public draft of the PhyloCode covered only clade names. Among the advisory group members, there was a diversity of opinions on how species names should be handled, ranging from those who thought that species names should never be governed by the PhyloCode to those who argued that their inclusion is so essential that the PhyloCode should not be implemented until rules governing species names have been added. The majority held the intermediate view that species names should eventually be included in the PhyloCode but that the controversy surrounding species and species names, both within the advisory group and in the systematics community as a whole, should not be allowed to delay implementation of the rules for clade names. Thus, it was decided, first, that rules for clade names and rules for species names would be published in separate documents and, second, that the timing of implementation of the two documents would be independent; thus, the rules for clade names would likely be implemented before those for species names. (This decision was reconsidered in 2006, and a different approach to species names was adopted by the CPN in 2007; see below.)

A second major decision at the 2002 Yale workshop concerned the proposal of a companion volume, to be published simultaneously with the PhyloCode, that would define various clade names following the rules of the PhyloCode and serve as its starting point with regard to priority. (This companion volume was later named Phylonyms: A Companion to the PhyloCode.) As originally conceived, the companion volume would have included phylogenetic definitions of the most widely known names in most major groups of organisms. It was soon realized that several volumes would be needed, that producing these volumes would be an immense job, and that linking the starting date of the PhyloCode to their publication would greatly delay its implementation. For this reason, the participants in the second workshop decided to reduce the scope of the companion volume. Instead of attempting a comprehensive treatment of widely known clade names for all major groups of organisms, the companion volume would include only examples involving taxa for which there were systematists who could be recruited to contribute entries. A plan for a conference was conceived in which participants would apply phylogenetic nomenclature to clades that they study. The definitions from the papers presented at the conference would form the nucleus of the companion volume. Michel Laurin offered to organize the meeting, and Kevin de Queiroz and Jacques Gauthier were chosen to edit the companion volume (Philip Cantino was enlisted in 2004 as a third editor).

The First International Phylogenetic Nomenclature Meeting took place July 6–9, 2004, at the Muséum National d’Histoire Naturelle in Paris, organized by a 10-member committee chaired by Michel Laurin. The meeting was described in detail by Laurin and Cantino (2004), and the program and abstracts are available at www.phylocode.org and www.phylonames.org. Unlike the preceding workshops, this conference included research presentations and was open to anyone interested in attending. It was attended by 70 people from 11 countries, and 36 papers were presented. The Paris conference also served as the inaugural meeting of the International Society for Phylogenetic Nomenclature (ISPN), including the election of a governing council and officers and approval of the bylaws (available at the subsequently established ISPN website: www.phylonames.org). The ISPN includes an elected Committee on Phylogenetic Nomenclature, whose responsibilities include ratifying the first edition of the PhyloCode and approving any subsequent modifications (for full responsibilities, see Article 22).

Presentations were given at the Paris meeting on the theory and practice of phylogenetic nomenclature and its applications to a wide variety of groups. Besides the inauguration of the ISPN, there were several other important outcomes of the meeting: (1) A proposal by K. de Queiroz and J. Gauthier to adopt “an integrated system of phylogenetically defined names,” including the application of widely known names to crown clades and forming the names of the corresponding total clades by adding the prefix “Pan-” to the name of the crown (Lauterbach, 1989; Meier and Richter, 1992; Gauthier and de Queiroz, 2001; Joyce et al., 2004), was introduced and vigorously discussed. Some participants were reluctant to make these conventions mandatory because doing so would result in replacing some names that had already been explicitly defined as the names of total clades (e.g., replacing Synapsida by Pan-Mammalia). A compromise that made exceptions for such names was acceptable to the majority of the participants, and it served as the basis for the set of rules and recommendations that was eventually adopted by the CPN (Recommendation 10.1B and Articles 10.310.8 in version 3 of the PhyloCode, and after some subsequent modifications, Recommendation 10.1B and Articles 10.310.7 in the current version). (2) Benoît Dayrat proposed that phylogenetically defined species names consist of a single word (the second part of the binomen in the case of already existing names) plus the author of the name, year of publication, and (if necessary to ensure uniqueness) the page number where published (Dayrat et al., 2004). In practice, the name of a small clade (generally corresponding to a genus under the rank-based system) would likely be cited before the species name, but it would not be part of the species name. In conversation and in teaching, the name would likely be abbreviated to the single word (i.e., omitting the author and year) when doing so is unambiguous. Dayrat’s proposal was well received by conference participants. (3) Julia Clarke proposed a flexible way of defining species names that is applicable to the wide variety of entities that are called species. The definitions would take the form “the species that includes specimen X” (de Queiroz, 1992), and the author would be required to explain what he/she means by “species.” This approach is similar to the way species names are implicitly defined in rank-based nomenclature but differs in that the species category is not a rank, and the author is required to explain the kind of entity to which the name refers. (4) In a straw vote of meeting participants, it was decided that Clarke, Dayrat, Cantino, and de Queiroz would draft a code for species names based on the above-described proposals of Clarke and Dayrat. Consistent with the decision made at the 2002 Yale workshop, this code would be separate from, but compatible with, the code for clade names.

In the fall of 2004, Cantino and de Queiroz drafted a code for species names based on the proposals approved at the Paris meeting. After review of the draft by Dayrat and Clarke and e-mail discussion of unresolved issues, the four potential authors of the code met at the Smithsonian Institution on May 20–21, 2006. In the process of drafting the code, the seriousness of the drawbacks of extending the PhyloCode to species names using an epithet-based format had become more apparent. Most critically, species names would be different under rank-based and phylogenetic nomenclature (e.g., “Homo sapiens” vs. “sapiens Linnaeus 1758”), which would create confusion. Second, differences in the way types are handled under the zoological and botanical codes would complicate the development of a universal code governing species names. Third, establishing and registering reformatted names for every species known to science would be an immense job—and one of questionable value given that there would be no fundamental difference in the way that the names would be defined. What emerged from the May 2006 meeting was an entirely different (and much simpler) way to reconcile the incompatibilities between traditional binominal species names and phylogenetic nomenclature—including the mandatory genus category and the fact that many genera are not monophyletic. This approach was subsequently adopted by the CPN (May, 2007), described in detail by Dayrat et al. (2008), and incorporated into the code (as Article 21).

The Second Meeting of the ISPN took place June 28–July 2, 2006, at the Peabody Museum of Yale University, organized by an eight-member committee co-chaired by Nico Cellinese and Walter Joyce. Most of the papers were presented in three symposia: phylogenetic nomenclature of species (organized by David Baum and Benoît Dayrat), implementing phylogenetic nomenclature (organized by Philip Cantino), and phyloinformatics (organized by Michael Donoghue and Nico Cellinese). The meeting was described in detail by Laurin and Cantino (2006, 2007), and the program and abstracts are available at www.phylocode.org and www.phylonames.org. At this second ISPN meeting, more time was devoted to open-ended discussions of issues raised in the presentations. The new approach to species names that was developed at the May 2006 meeting in Washington (see previous paragraph) was presented in talks by Clarke and Dayrat and was well received in the subsequent discussion. They and their coauthors (Cantino and de Queiroz) were encouraged to continue work on a set of rules and recommendations that would implement this approach.

Another issue that generated a lot of discussion at the second ISPN meeting was the integrated system of crown- and total-clade names that was introduced at the 2004 Paris meeting and incorporated into PhyloCode version 3. Although the rules and recommendations promoting an integrated system in version 3 represented a compromise, there was still a lot of dissatisfaction on the part of some discussants. An alternative means of referring to total clades using “pan” as a function name was proposed by T. Michael Keesey. In the course of the discussion, it was suggested that the prefix “pan-” (lower case) be used to designate informal names for total clades that may or may not have a formal name. Because informal names do not compete with formal names for precedence, they can coexist without violating Principle 3 (that each taxon may have only one accepted name). Using this approach, a widely used name could be retained for a total clade and coexist with an informal name with the prefix “pan-”. For example, the total clade of mammals might have the formal name Synapsida and the informal name pan-Mammalia. This suggestion led to changes in Article 10 that were approved by the CPN in January 2007 and included in this version of the code.

The Third Meeting of the ISPN took place July 20–22, 2008, at Dalhousie University in Halifax, Nova Scotia, organized by a four-member committee chaired by Harold Bryant. It was coordinated with a joint meeting of the International Society of Protistologists and the International Society for Evolutionary Protistology. In addition to contributed papers (see Laurin and Bryant, 2009, for details), including a plenary talk by Sina Adl focusing on issues in protist nomenclature, the meeting focused on how to expedite completion of two critical projects that must reach fruition before implementation of the code: preparation of the companion volume (Phylonyms), to be published simultaneously with the code, and development of the registration database (RegNum).

Because registration is required for establishment of names under the PhyloCode, the registration database (which has come to be known as RegNum) had to be developed before the code could be implemented. Torsten Eriksson and Mikael Thollesson initially designed the database structure and reported on it at the 2002 Yale workshop. Further development of the database and web/user interface was subsequently carried out at Uppsala University by Jonas Ekstedt and M. Thollesson. An alpha test site for this version was announced at the 2004 Paris meeting. A prototype of RegNum was demonstrated at a meeting of the ISPN Registration Committee (Mikael Thollesson, Torsten Eriksson, and Nico Cellinese) and other interested persons at Yale University on November 2–3, 2005, and it was subsequently demonstrated at both the 2006 and 2008 ISPN meetings. In 2009, Nico Cellinese (University of Florida), who was chair of the Registration Committee at the time of this writing, started the development of a more comprehensive and flexible prototype. The new version of RegNum was conceived in line with other developments concerning biological name repositories and resolution services (e.g., Global Name Architecture). RegNum’s prototype was demonstrated at the 2009 Biodiversity Information Standards (TDWG) meeting in Montpellier, France, during a workshop that focused on phylogenetic nomenclature and future informatics development. The RegNum database currently satisfies the requirements of the PhyloCode, and plans are in place to integrate it with several tools and data repositories (e.g., TreeBASE) that will enhance its relevance for phylogenetic research.

In October 2011, the CPN received a proposal by Nico Cellinese, David Baum and Brent Mishler concerning the treatment of species in this code, the justification for which was published by Cellinese et al. (2012). Their fundamental premise was that the code is too strongly tied to a particular view on the nature of species, which is not accepted by everyone who would like to use phylogenetic nomenclature, and their proposed solution was to eliminate all mention of species in the code. The proposal was publicized on the ISPN website and stimulated more than a year of intermittent discussion on the CPN listserv. In the end, the CPN accepted the underlying premise of the proposal but rejected the proposed solution, which would have entailed radical changes in the code, including: eliminating the use of species as specifiers (thus only specimens would be specifiers); permitting the establishment of preexisting specific epithets as clade names; redefining the term “homonym” such that established clade names may be identical provided that the authors and publication years differ; and elimination of Article 21, which provides recommendations on how to use species names governed by the rank-based codes in conjunction with clade names governed by this code. Instead of eliminating all reference to species, the definition of “species” was broadened to accommodate the view that the species category is simply a rank in the Linnaean hierarchy, while continuing to accommodate the view that it is a kind of biological entity. In addition, the CPN discussion of the Cellinese et al. proposal led to other changes not proposed by those authors, especially changes in Article 11 related to using species versus type specimens as specifiers.

An important change approved by the CPN in August 2013 was the expansion of the treatment of phylogenetic definitions, which in many ways are the heart of this code. What were formerly a long and complicated note (9.3.1) and three related recommendations (9.3B, 9D, and 9E) were expanded into six articles (9.49.7 and 9.99.10). Following a proposal by de Queiroz (2013) that was motivated by a distinction emphasized by Martin et al. (2010), node-based and branch-based definitions were replaced with minimum-clade and maximum-clade definitions, respectively, throughout the code. The primary reason for this change was to employ definitions that are applicable in the context of both common interpretations of phylogenetic trees—one in which branches represent lineages and nodes represent ancestors at lineage-splitting events; the other in which branches represent relationships and nodes represent taxa (the term “branch-based” is inappropriate in the latter context, in which all phylogenetic definitions are effectively node-based).

Another important change in Article 9 is the explicit recognition of definition categories for the names of crown clades (Article 9.9) and total clades (Article 9.10), the variants of which (e.g., the maximum-crown-clade definition, which is roughly equivalent to what were previously called branch-modified node-based definition) are presented more systematically and thoroughly than in previous versions of the code. Additionally, a new note (9.5.1) was added describing a kind of minimum-clade definition that had no analog in previous versions of this code, the directly-specified-ancestor definition, weherein the ancestor in which the clade originated is identified by name rather than being specified indirectly through its descendants, and Note 9.9.2 was added to describe its use for defining the names of crown clades.

Readers interested in more information about the sequence of changes from one version of this code to the next are referred to www.phylocode.org, where all previous versions are available. The changes implemented in versions 3 and 4 are summarized in the preface of each, and changes implemented in version 5 are summarized in a separate document.

There is only one major modification in the current version of the code (version 6) relative to version 5. The rules on publication (Articles 4 and 5) have been revised to permit electronic publication, based on modifications proposed by Nico Cellinese and Richard Olmstead. Most other changes from version 5 are simply clarifications, but two new rules (Articles 10.7 and 14.5) have been added to ensure that accepted panclade names are always based on the names of the corresponding crown clades.

Acknowledgments

We thank current and past members of the Committee on Phylogenetic Nomenclature for reviewing and approving earlier drafts of the Code as well as proposed additions, deletions, and other modifications: Sina Adl, Frank (Andy) Anderson, Brian Andres, Tom Artois, Christopher Brochu, Harold Bryant, David Cannatella, Nico Cellinese, Julia Clarke, Benoît Dayrat, Jim Doyle, Micah Dunthorn, Jacques Gauthier, Sean Graham, John Hall, David Hibbett, David Hillis, Walter Joyce, Michael Keesey, Max Cardoso Langer, Michel Laurin, David Marjanović, Richard Olmstead, Kevin Padian, Fredrik Pleijel, Greg Rouse, George Sangster, David Tank, and Mieczyslaw Wolsan. We wish to give special thanks to David Marjanović for his thorough reviews and numerous constructive suggestions, Michael Keesey for design and maintenance of the PhyloCode website, and Michael Donoghue for the initial push to undertake this project. Chuck Crumly has tirelessly and patiently supported publication of this Code for many years. We also wish to thank Michele Dimont at CRC Press and Rachel Cook at Deanta Global for overseeing copy-editing and production of the book. Other people who have made important contributions to this project are mentioned in History section above.