Botanical Nomenclature and Plant Fossils

TAXON 59 (1) • February 2010: 261–268 Cleal & Thomas • Botanical nomenclature and plant fossils

NOMENCLATURE

Edited by John McNeill, Anthony E. Orchard & John C. David

Botanical nomenclature and plant fossils

Christopher J. Cleal1 & Barry A. Thomas2

1 Department of Biodiversity & Systematic Biology, Amgueddfa Cymru – National Museum Wales, Cardiff CF10 3NP, U.K. 2 Institute of Biological, Environmental and Rural Sciences, University of Aberystwyth, Llanbadarn Fawr, Aberystwyth SY23 3AL, U.K. Author for correspondence: Christopher J. Cleal, [email protected]

Abstract The provisions in the Code for naming plant fossil taxa have changed substantially over the years. The history of these changes reflects the tension between palaeobotanists (including palynologists studying plant microfossils) who need a flexible set of regulations, and the tendency for the Code to include nomenclatural regulations that constrain taxonomic decisions. The current Vienna Code now provides for plant fossils to be named as fossil taxa, which is a flexible taxonomic concept that should suit the needs of most palaeobotanists. However, the Vienna Code also incorporates the more restrictive concept of morphotaxa and most palaeobotanists seem to be under the misapprehension that plant fossils can only be named as morphotaxa. In our view, the concept of morphotaxa is logically flawed and unnecessary in practice, and should be removed from the Code.

Keywords form-genera; fossil taxa; morphotaxa; organ-genera; palaeobotany; palynology; satellite taxa

INTRODUCTION at least partly lost through fossilisation. They are now biologi- cally inert objects, but originated from living organisms. Since the publication of the Vienna Rules (Briquet, 1906) Fossil plants, in contrast, were the plants from which these it has been accepted that taxonomic nomenclature of plant fossilised remains were derived. They were whole living or- fossils should follow essentially the same procedures as used ganisms, which in principle could be treated in the same way for living plants. However, it soon became evident that the as living plants. But there is the crucial difference that fossil fragmentation of plants prior to and during fossilisation, and plants no longer exist except in the minds of palaeobotanists, the different types of information provided by different modes a view echoed by Bateman & Hilton (2009); nowhere can you of preservation, made it impossible to name taxa of fossils in today see a fossil plant other than as an illustration or model. exactly the same way as taxa of living plants. Thus, various It is of course true that the larger living plants are usually different nomenclatural measures have been proposed over the typified by herbarium specimens representing only parts of years for inclusion within the Code to accommodate fossils, the organism, and these could be regarded as analogous to the most recent being the introduction of “morphotaxa” in the fossil fragments (Stafleu, 1967). Since the nomenclature of St Louis Code (Greuter & al., 2000a) and of “fossil taxa” in whole living plants is based around fragments preserved in the Vienna Code (McNeill & al., 2006). herbaria, why can we not similarly name fossil plants based Many palaeobotanists (including palynologists dealing on our fragmentary plant fossils? The reason is, in our view, with organic-walled plant and “algal” microfossils) remain because there is a fundamental difference between the two uncertain about the current rules for palaeobotanical taxo- situations. With living plants it is still usually possible to go nomic nomenclature and their implications, and we agree that to a whole living organism to check features that may not the position appears somewhat cryptic. However, the current be shown by the type, a procedure that is impossible with Code does in fact meet the needs of palaeobotanists, and in long-extinct fossil plants. In some cases, reproductive organs our view there need only be a few fairly modest changes to and foliar organs have been found in apparent attachment, the provisions covering plant fossils for the situation to be- but except for the smaller fossil plants (e.g., Ash & Tidwell, come clear. 1986: fig. 2) whole articulated organisms are rare, and those that are available do not preserve the full range of their tissue and organs. WHAT ARE WE NAMING? In the current Code (McNeill & al., 2006) the wording is somewhat ambiguous. In the footnote to Preamble 7, the This may seem self-evident, but in our view there has been implication is that the palaeobotanical provisions within the some confusion. Palaeobotanists deal with two concepts: plant Code are dealing with fossil taxa (“… the term ‘fossil’ is ap- fossils and fossil plants. Plant fossils are the physical objects plied to a taxon when its name is based on a fossil type …”) taken out of the ground that reveal evidence of long dead and the term “Fossil taxa” is explicitly used in Art. 1.2. This plants. These remains nearly always represent just fragments term can only be referring to taxa of [plant] fossils and not of of the plants in which aspects of the original tissue have been reconstructed (“whole”) extinct plants. However, in most of

261 Cleal & Thomas • Botanical nomenclature and plant fossils TAXON 59 (1) • February 2010: 261–268

the rest of the Code where palaeobotanical nomenclature is no palaeobotanists, and they had no draft proposals from the dealt with (Arts. 8.5, 11.1, 13.1(f), 36.3, 38.1, 38.2; Rec. 8A.3) Thomas committee. Moreover, the American botanist H.W. the term “fossil plants” is used; only Art. 7.9 mentions “plant Rickett stated that American palaeobotanists had not had an fossils”. This has led to the strange concept of “morphotaxa opportunity to consider the 1935 proposals (which contradicts of fossil plants” (Art. 11.1); if we had available an actual fossil the comments by Thomas 1935: 111 that the British proposals plant, we would surely name it in the same way as we would “have received the general approval of many American palaeo- a living plant rather than using morphotaxa. There can be no botanists”). Consequently, no palaeobotanical provisions were doubt that in most places in the Code where the term “fossil added to the published supplement to the Rules (T.A. Sprague, plants” is used, it really means “plant fossils”. 1950), nor had any been included in the earlier “unofficial Many palaeobotanists seem to focus exclusively on these rules” published in America (Camp & al., 1947). hypothetical “whole” plants, at least partly because they are The taxonomic nomenclature of plant fossils was finally considered necessary for incorporation in phylogenetic (“cla- addressed properly in 1950 at the Stockholm Congress, where distic”) analyses. However, as they no longer exist as realities the International Association for Plant Taxonomy was estab- there are very real practical problems in formally naming ex- lished (Cowan & Stafleu, 1982; Stafleu, 1988), and the Rules tinct plants reconstructed from the fossil record, problems that became the International code of botanical nomenclature. At remain unsolved: a problem noted by Chaloner (1986). Argu- this meeting, H.H. Thomas tabled a report from the Special ably the best practice is to not name fossil plants formally at all, Committee for Palaeobotany (Lanjouw, 1953) that integrated but to refer to them by names such as the “Lepidodendron-tree” the Thomas (1935) and Jongmans & al. (1935) proposals (the outside of a formally codified system of nomenclature. Plant report was published as “Appendix III” in the proceedings fossils, in contrast, are physical objects that can be examined of the Stockholm Congress – Lanjouw, 1953: 547). There is directly and thus more suitable to codified nomenclature. no record of the discussions that produced this compromise wording, nor were the proposals on form- and organ-genera discussed in detail at the Stockholm meeting itself. However, INTRODUCTION OF FORM- AND ORGAN- except for the proposed change to Article 20 (later 23) giving GENERA the starting date for palaeobotanical taxonomic nomenclature, the Thomas report was accepted “almost unanimously” by The Vienna Rules (Briquet, 1906) specifically stated in the Stockholm Congress (Lanjouw, 1953: 534), and became Art. 9 that it covered fossils but few provisions were devoted incorporated into the Code, initially as Appendix IV (Stock- to them. The first serious attempt to address the issue came holm Code; Lanjouw & al., 1952) and then Appendix II (Paris with a meeting of British palaeobotanists held in 1934 at the Code; Lanjouw & al., 1956). British Museum (Natural History), where a series of propos- The original organ-genus definition was flexible: organ- als to emend the Rules were drafted (Thomas, 1935). This genera were to reflect the different parts of the plant being draft was also seen by three palaeobotanists from continental preserved and modes of preservation, but nothing was said Europe (Wilhelmus Jongmans, Thore Halle, Walter Gothan) about the level to which they could be classified. The concept who subsequently published a modified version of the same of form-genus was created with the clear aim of stabilising proposals (Jongmans & al., 1935). Both versions were in es- the use of a limited set of taxa, especially those introduced by sential agreement about the key proposal that there should be Brongniart (1822) that, although being widely recognised as ar- separate taxonomic categories for naming plant fossils: fossils tificial, had a long history of use. Form-genera were regarded as would normally be referred to organ-genera. However, there a small subset of organ-genera: they were “organ genera which would be a subset of organ-genera that would be entirely arti- have been proved to contain species that are not generically ficial (“artificial genera” in the Thomas proposal, “form gen- related in the ordinary taxonomic sense” (Jongmans & al., 1935: era” in the Jongmans & al. proposal) being defined on strictly 4). So, for most purposes, plant fossils were to be assigned to morphological criteria with no implications being made that organ-genera unless one of the Brongniart taxa was to be used. they incorporated biologically related species. Both versions of these proposals were presented at the 1935 Amsterdam International Botanical Congress (Sprague, 1935a), CHANGES TO FORM- AND ORGAN-GENERA where they seem to have met with general approval (Sprague, 1935b). However, because two versions had been tabled, it The Montreal Code (Lanjouw & al., 1961) saw the removal proved difficult to agree on the exact words to be adopted. A of the Appendix for fossil plants and the incorporation of its committee of palaeobotanists was therefore created, with H. regulations within the main body of the Code. More signifi- Hamshaw Thomas as secretary (Sprague, 1935b), to prepare a cantly, the definitions of the concepts of form- and organ- compromise set of proposals, which would form the basis of an genera were substantially changed. Instead of having an es- Appendix to the Rules (M.L. Sprague & al., 1950: 10). sentially flexible concept of organ-genera to which all plant The outbreak of war in Europe in 1939 brought all progress fossils could be assigned, but which included a relatively small on this to a halt. After the war, in 1948, a small meeting was subset of artificial form-genera for certain mainly foliage fos- held in Utrecht to try to initiate work again on revising the sils, there were now two rigidly defined and mutually exclusive Rules (M.L. Sprague & al., 1950). However, the group included concepts: an organ-genus was for fossils that could be assigned

262 TAXON 59 (1) • February 2010: 261–268 Cleal & Thomas • Botanical nomenclature and plant fossils

to a family, but fossils that could not be assigned to a family could not be assigned to a family. In order to circumvent the had to be placed in a form-genus. problem they followed Meyen (1978) in using the term “satel- The revised definition had the merit of clarity (Schopf, lite taxa” for genera that could be conditionally included in 1963) but its rigidity caused difficulties for many palaeobota- suprageneric taxa for storage. They suggested that such an nists (Faegri, 1963), especially as it resulted in nomenclatural approach to family organisation could solve the problem of procedure impinging unnecessarily on purely taxonomic mat- fitting isolated plant organs of uncertain affinities into a taxo- ters. Faegri (1963) proposed to make these taxonomic catego- nomic scheme under nomenclatural legislation that no longer ries once again more flexible but this was rejected (Mamay, recognised organ-genera and had form-genera unassignable to 1964). Over the following years, various other proposals were families. Their suggestion was that satellite taxa might be in- made to try to improve the situation (Potonié, 1964; Krassilov, terpreted as genera tentatively being considered for assignment 1968; Jansonius, 1974; Meyen, 1975; see also the informal to families, but awaiting conclusive data. They also suggested suggestions by Stafleu, 1967) but again none were accepted that satellite taxon status could also be used at order level to (Traverse, 1975). Eventually, a re-definition of form-genus include all the plant parts that cannot, with some degree of was proposed by the Committee for Fossil Plants (as reported confidence, be assigned to a family. To their approach to using by Voss, 1976) that provided a flexible nomenclatural tool: satellite taxa in the lycophytes, they suggested that reproduc- a form-genus may or may not be assignable to a family, and tive organs be used as the basis for defining families. Bateman so there was no need for organ-genera. Unfortunately, how- & Hilton (2009) have broadly agreed with the potential value ever, the Editorial Committee “improved” the wording that of the satellite taxa concept, although they have argued that a was eventually included in the Leningrad Code (Stafleu & more flexible approach was needed to decide what plant organs al., 1978) so that now a form-genus could not be assigned to should be used to determine whether a taxon belonged to a a family under any circumstance (see comments by Meyen family or was a satellite of a higher-ranked taxon. & Traverse, 1979). The result was that palaeobotanists were left with a rigidly and narrowly defined form-genus concept for the less well-understood fossils, and nowhere to place the INTRODUCTION OF MORPHOTAXA better-understood fossils – arguably the worst of both worlds. Meyen & Traverse (1979: 597) attempted to resolve the prob- The situation resulting from the 1978 Leningrad Code lem by proposing that the whole concept of form- and organ- remained virtually unaltered for nearly two decades and is genera should be removed from the Code, and that the genera usefully summarised by Chaloner (1999). The next signifi- used for fossils should be treated nomenclaturally in the same cant development was through the proposals by Fensome & way as genera of living plants, but an unattributed “official Skog (1997). Following Meyen & Traverse (1979) and Boulter opinion” given to the Committee for Fossil Plants was that this (1979), they recognised the need for flexibility in the defini- approach was too simplistic and it was rejected (Traverse, 1981). tion of the taxonomic concepts to be used for plant fossils, Boulter (1979) argued in essence for a modified version of the and proposed a revised definition for form-genera: “Because proposal reported by Voss (1976) but with a clearer wording, and fossil specimens may represent dispersed parts of an organ- the suggestion that “form-genera” should be replaced by “fossil- ism or a single stage in a life cycle …. they can be assigned to genera”. Again, however, the proposal did not meet favour with form-genera. It may also be necessary to define form-genera the unnamed “official referee” and it was also rejected (Traverse, on the basis of different preservational modes” (Fensome & 1981). Thereafter, discussions on the issue ceased for a while; Skog, 1997: 557). This has much in common with the spirit the impression is given that palaeobotanists found the problem of the original proposals of Thomas (1935), Jongmans & al. to be intractable. Many palaeobotanists either simply ignored (1935) and Boulter (1979), as well as those reported by Voss the Code, or pragmatically bent or even broke the regulations (1976), where no restrictions are placed on the level to which as necessity required: for example, Meyen (1982) described a a form-genus could be classified or of the extent of its cir- new family of Palaeozoic seed-plants, the Rufloriaceae, with a cumscription. basionym of the foliar form-genus Rufloria Meyen; and Watson Chaloner & al. (1998) recognised the great merit of the Fen- (1988) included the pollen form-genus Classopolis Pflug, and some & Skog proposals, but went further and argued that such the foliar form-genus Geintizia Endlicher within the Mesozoic “fossil taxa” should not be limited to the generic rank – that conifer family Cheirolepidiaceae Hirmer & Hörhammer. it should be possible to have “fossil taxa” at any rank, specifi- For a further discussion of the history of the taxonomic cally mentioning species and families. They were unhappy (rather than nomenclatural) concepts behind form- and organ- with the term form-taxa, partly because form-genera had been genera, see Bateman & Hilton (2009). used previously with such a wide range of meanings, and so proposed adopting a possible alternative, “parataxa” (a term that had been first introduced by Hawksworth & al., 1994). The SATELLITE TAXA latter was also unacceptable to the Committee for Fossil Plants (Skog, 1999) and was ultimately replaced by “morphotaxa”. When attempting to group both living and fossil lycops- The origin of this term has not been recorded in the literature, ids into families, Thomas & Brack-Hanes (1984) encountered but according to J. Skog (pers. comm., 2008) it was based on the absurd but legislated problem that fossils as form-genera a suggestion by Knut Faegri.

263 Cleal & Thomas • Botanical nomenclature and plant fossils TAXON 59 (1) • February 2010: 261–268

Chaloner & al. (1998) emphasised that their proposals had Early Devonian zosterophyll Tarella that are preserved partly the merit of flexibility, and allowed the practicing palaeobota- as adpressions showing morphology and partly as pyrite pet- nist to determine the limits of parataxa (later morphotaxa) on rifactions showing internal anatomy? Plant fossils preserved a case by case basis. However, they then proceeded to shackle as authigenic mineralisations in sideritic nodules can show that freedom by introducing a new limit to the definition of features of morphology and are named as taxa typified by ad- morphotaxa, whereby they can only be used for fossils that pression fossils (e.g., Wittry, 2006) but also reveal evidence of represent “those parts, life-history stages, or preservation states anatomy and have been named as taxa with petrifaction types of organisms that are represented by the corresponding types” (e.g., Jennings & Millay, 1978). Which is correct? Or should (Chaloner & al., 1998: 909). It was not stated in the proposal there be a third set of morphotaxa just for use with authigenic why this limitation was introduced, but it may have originated mineralisations? from views expressed earlier by two of the authors in their The Fensome & Skog (1997) and Chaloner & al. (1998) preamble to the Draft BioCode (Greuter & Nicolson, 1996: proposals were referred to the Committee for Fossil Plants 348): “Palaeobotanists may find it desirable to prevent the use (Greuter & Hawksworth, 1999) who accepted the former and of names of organ taxa … for different fossil organs, once the rejected the latter (Skog, 1999). The subject was further dis- link is established. We doubt that palaeobotanists would want cussed during the nomenclature sessions at the St Louis Con- to take up, say, a name typified by pollen organs for a fossil leaf gress, where the potential problem of priority of competing or wood sample when both pertain to the same organism.” The names for different plant parts was again raised (Greuter & authority on which this perceived opinion of palaeobotanists al., 2000b: 45–51). During these sessions, there was an ad hoc was based was not given. The quoted example is, of course, an meeting of the members of the Committee for Fossil Plants that extreme situation, and it is difficult to imagine any scenario were present, which resulted in a compromise that appeared where palaeobotanists would want to unite taxa of pollen and to combine the Fensome & Skog (1997) and Chaloner & al. leaves in this way. However, if the wording of the example is (1998) proposals. It was not a real compromise, however, as the changed slightly, the situation is less clear-cut: “We doubt that resulting new Article 1.2 that appeared in the St Louis Code palaeobotanists would want to take up, say, a name typified by (Greuter & al., 2000a) was based almost exclusively on the stems for a fossil leaf when both pertain to the same organism.” Chaloner & al. (1998) proposals and the more flexible approach Surely, if the evidence pointed beyond reasonable doubt to all advocated by Fensome & Skog (1997) was largely abandoned. stems of a particular form having borne leaves of a particular form, and that those leaves were only found on that form of stem, what would be the point of having separate names? INTRODUCTION OF FOSSIL TAXA Another difficulty is that Chaloner & al. (1998) did not specify what they meant by “plant parts”. For a Palaeozoic In an attempt to clarify the conceptual confusion surround- pteridosperm frond, for instance, is the “plant part” an indi- ing morphotaxa, Chaloner (2004) proposed that the Code vidual pinnule, a pinna or the whole frond? All three have at should include two examples showing how such taxa should different times been used as the basis for the generic classifi- be used. This proposal generated “strong opposing opinions” cation of such foliage (Brongniart, 1822; Gothan, 1941; Cleal among the Committee for Fossil Plants (Skog, 2005) and, after & al., 1990). Bateman & Hilton (2009) have argued that how some discussion, the Committee formulated an alternative the plant disarticulates should help determine what plant parts proposal: a Note should be added to Article 1.2: “Any fossil (in their terminology “organs”) should determine the circum- taxon that is described as including more than one part, life- scription of such taxa. This very practical approach certainly history stage, or preservational state is not a morphotaxon” has merit but it still does not overcome all of the difficulties (McNeill & al., 2005). The Committee’s secretary (J.E. Skog) encountered by the palaeobotanist. In the Mississippian (early presented this proposal to the Nomenclatural Section during Carboniferous) pteridosperm ovulate structures, for instance, the 2005 Vienna International Botanical Congress and despite is the “plant part” the individual ovule, or the groups of ovules it having not been announced before the Congress (the pro- in their original cupulate structures? We know that both appear posal was only reported-on after the meeting – McNeill & al., to have been abscised from the plant and are often found as 2005: 1063) it was incorporated into the Vienna Code (McNeill fossils (e.g., Rothwell & Wight, 1989). Similar problems ex- & al., 2006). This would seem to leave us in the position that ist with the term “mode of preservation”. For instance fossils a fossil taxon that is not a morphotaxon because it includes preserved as compressions give different levels of informa- more than one part, life-history stage or preservational state, tion to those preserved as impressions, and could arguably is still a fossil taxon. be classified in different ways. However, it is common for But what is a fossil taxon? The expression was explicitly macrofloras and sometimes even individual fossils to show used for the first time in the St Louis Code (Greuter & al., both types of preservation, hence the introduction of the term 2000a) in Art. 1.2. However, it seems to have been used here “adpression” by Shute & Cleal (1986). Where are the divid- as a simple adjective-noun couplet, analogous to “plant taxon”, ing lines between compressions, adpressions and impressions rather than a defined taxonomic concept, analogous to form- that we would need to know to interpret the circumscription genus in the earlier Codes. The nearest we can find to a defini- of morphotaxa on such fossils? And what about specimens tion in the St Louis Code is the footnote to Preamble point 7, such as those described by Edwards & Kenrick (1986) as the which states that “the term ‘fossil’ is applied to a taxon when

264 TAXON 59 (1) • February 2010: 261–268 Cleal & Thomas • Botanical nomenclature and plant fossils

its name is based on a fossil type” (this footnote in fact dates morphotaxa in perpetuity, perhaps compiled by a committee back to the Sydney Code [Voss & al., 1983] although the phrase of eminent palaeobotanists? This is surely where nomenclature “fossil taxon” itself is not explicitly used there). is again impinging on purely taxonomic matters. In the context of the St Louis Code, the issue is somewhat In the above example, for instance, if we accept Lyginop- academic as all fossil taxa were to be regarded as morphotaxa teris as a morphotaxon simply because its type is an anatomi- (McNeill & al., 2006: xi). However, the changes introduced cally preserved stem, then we need a separate generic name in the Vienna Code effectively permitted the existence of a for the foliage (assuming we did not merely subsume it into non-morphotaxon fossil taxon. The newly-introduced Glossary the highly “artificial” and systematically meaningless Sphe- at the end of the Vienna Code defines fossil taxa by in effect nopteris, as suggested by Chaloner, 2004). However, there is reiterating the footnote in the Preamble (there had been an nothing in the Code that prevents us accepting the emenda- earlier Glossary, published separately from the Code, but this tion of the diagnosis of Lyginopteris made by Gothan (1931) did not refer to the term “fossil taxa” – McVaugh & al., 1968). whereby he also incorporated the distinctive range of frond So, we can only conclude that all taxa that are based on a fossil adpressions believed to represent the foliage borne by the same type are to be regarded as fossil taxa and their nomenclature plants as these stems, thereby transforming Lyginopteris into a presumably has to follow the same rules as for other plant taxa non-morphogenus fossil genus. Another example is Sigillaria unless the Code states otherwise. Brongn., which is quoted in Ex. 28 of the Code as an example of a morphogenus. Its type is an adpression of stem “bark” of a Carboniferous lycopsid, but many palaeobotanists use SO DO WE NEED MORPHOTAXA? this generic name also for anatomically preserved fossils of the whole stem (e.g., Brongniart, 1839; Phillips & DiMichele, It is quite clear from Art. 1.2 of the Vienna Code that mor- 1992). All of these stems produced similar “bark”, and this photaxa are a subset of fossil taxa (“Fossil taxa … may be type of “bark” is only known to have been produced by stems treated as morphotaxa” – emphasis by the present authors). It with this type of anatomy. It makes little sense to maintain is also quite clearly stated in Art. 1.2 that a particular morpho- Sigillaria as a morphogenus in the latter’s current rigid defi- taxon can only comprise “…the one part, life-history stage, or nition, as it would merely require another generic name to be preservational state represented by the corresponding nomen- created for the stem petrifactions. clatural type.” However, although Art. 1.2 explains how you define a taxon such as Sphenopteris (Brongn.) Brongn., if it is to be accepted as a morphotaxon, nowhere does the Code THE USE OF FOSSIL TAXA explain what makes it a morphotaxon in the first place. It has been suggested that if any fossil taxon is based on a In our view, the best solution is the simplest and most flex- type consisting of just one plant part in a particular phase of ible one. We should remove the concept of morphotaxa from its life-history, and preserved in one mode, then it is by defi- the Code – all taxa that are based on a fossil type are just nition a morphotaxon. Chaloner (2004) for instance argued fossil taxa. As with all other taxa dealt with in the Code, how that the fossil genus Lyginopteris Potonié was a morphogenus widely or narrowly a fossil taxon is circumscribed is set out in simply because its type was an anatomically-preserved stem. its diagnosis. A fossil taxon could be defined on very narrow However, this is not what Art. 1.2 or any other part of the Code morphological criteria, such as Sphenopteris, whose original says (the interpretation is, in technical logic, a fallacy of the diagnosis only mentioned details of pinnule shape and venation undistributed middle – see Damer, 2006); nowhere does it (Brongniart, 1822) and is widely accepted as a highly “artifi- stipulate that a taxon based on one plant part must be regarded cial” but still useful taxon. Alternatively, a fossil taxon could as a morphotaxon, only that it may be. Perhaps, therefore, be defined on wider criteria, such as Lyginopteris, which now clarification would be achieved by emending Art. 1.2 to say includes stems and fronds that are preserved either as adpres- “must” instead of “may”? However, such a change would not sions or petrifactions, and which belongs to a distinct family, in fact achieve any useful result, unless we also introduce some the Lyginopteridaceae Potonié (Gothan, 1931). A fossil taxon restriction whereby the circumscription of a morphotaxon can- could even represent the remains of an almost complete plant, not subsequently be emended. A morphotaxon would have to albeit with some of its internal tissue structure lost through be “fixed in stone” as a morphotaxon as soon as it was erected, fossilisation, such as the zygopteridalean fern Nemejcopteris or else a subsequent author could simply emend its diagno- Barthel (1968). The diagnosis and thus circumscription of a sis, changing its circumscription so that it could include more fossil taxon can of course be emended by subsequent authors. than one plant part, life-history stage or preservation state, If there is a good taxonomic case for emending a taxon so that and so was no longer a morphotaxon. If morphotaxa could be it can include additional plant parts or preservation states to emended in this way, it would make meaningless the distinc- those given in the original diagnosis, the Code does not and tion between fossil taxa that were and were not morphotaxa. should not prevent this. As far as we can see, there is no useful And what about the older taxonomic names, such as those role for the concept of morphotaxa as a formally demarcated introduced by Brongniart (1822) and Sternberg (1820–1838) subset of fossil taxa. that are still widely used by palaeobotanists? Would there Fossil taxa are often diagnosed so that their mutual hierar- be a definitive list of fossil taxa that would have to remain chical relationships reflect the assumed biological relationships

265 Cleal & Thomas • Botanical nomenclature and plant fossils TAXON 59 (1) • February 2010: 261–268

of the parent fossil plants (e.g., Cleal & Shute, 2003). Fossil was first published prior to that of the flower taxon, would taxa can also have a role, however, in naming fossils of un- the dispersed pollen name take priority? Potentially it would, certain systematic position. Such “artificial” fossil taxa should but only in a very few cases. Since the type specimen of the continue to be used if they provide a useful means of clas- name of the pollen taxon will be a dispersed grain, we would sifying certain fossils, such as Sphenopteris. They need not need to be certain that that particular grain (not just some be assignable to the full hierarchy of higher ranked taxa as grains of similar morphology) was produced by members of would a taxon of living plants, but can be regarded as satellite that taxon of flower. This will rarely be possible in practice; taxa. However, there need not be a sharp delineation between there are very few cases where we can be reasonably confident such “artificial” fossil taxa and fossil taxa that more closely that there is a one-to-one relationship between a pollen taxon reflect the taxonomy of the parent fossil plants; a fossil taxon and a flower taxon – that all the pollen of that form (including might not be assignable to a family, but perhaps be regarded the type specimen) was only ever produced by that one form as a satellite taxon (sensu Meyen, 1978; Thomas & Brack- of flower. Even if all of the flowers that we ever find only Hanes, 1984) of a class or even an order. Moreover, it should contain this form of pollen, it is impossible to be certain that be possible for the diagnosis of a fossil taxon to be emended so a dispersed pollen grain did not come from a plant growing that the level of its “artificiality” changes to reflect improved in habitats not represented in the fossil record and which had palaeobotanical knowledge or the practical needs of palaeo- quite different flowers. Consequently, since we cannot be cer- botanists; such an emendation would be a taxonomic matter tain that the type specimen of the name of the dispersed pollen and should not be constrained by the rules in the Code. taxon fits within the circumscription of the revised flower The nomenclatural rules for fossil taxa given in the current taxon, their names will not in practice compete. Code do not differ substantially from those for non-fossil taxa. If we were able to be reasonably confident that the type Other than differences in the starting point for their nomencla- of the dispersed pollen taxon came from a particular taxon of ture (Art. 13.1), the selection of types (Art. 9.13), the language flower, and if the name of that dispersed pollen taxon pre-dated that is allowed for diagnoses (Art. 36.3) and the role of illustra- that of the flower, then there could potentially be a problem. tions in protologues (Art. 38), the only substantive difference However, because of the nature of dispersed palynology, such from taxonomic nomenclature of extant plants is that names of circumstances are going to be very rare, and any resulting fossil taxa are allowed to represent remains of isolated parts of destabilisation of the flower and pollen taxonomic nomencla- plants in different stages of their life histories, and preserved ture could easily be overcome by suitable application of the in different ways. For instance, it is possible to have inde- conservation provisions of Art. 14 of the ICBN. Attempting to pendent fossil taxa for leaves preserved as adpressions, stems avoid such very rare problems by shackling palaeobotanical preserved as casts, and anatomically-preserved reproductive taxonomy with restrictive nomenclatural regulations seems organs that were all produced by the same taxon of parent to us unproductive. plant. Moreover, a fossil taxon might represent just leaves, or perhaps leaves and stems connected together as leafy shoots. What plant parts, life-history stages and preservation states a CONCLUSIONS fossil taxon represents is part of that taxon’s circumscription and is thus defined in its diagnosis. Since there is nothing in Most palaeobotanists have been aware of the weaknesses the Code to prevent subsequent emendation of this part of a in the previous Codes in dealing with the naming of plant fossil taxon’s diagnosis provided it does not cause the exclusion fossils, and have either ignored them altogether in their work, of the type (unless by formal conservation of the name with a or bent the rules to suit their immediate purposes. Bateman different type) fossil taxa represent a flexible nomenclatural & Hilton (2009) have recently suggested that these problems context within which palaeobotanists can name and classify could be overcome by reverting to some of the traditional their fossils according to purely taxonomic criteria. concepts of form- and organ-taxa, as well as introducing a concept of autapo-species for organ-species that have at least one unique character-state. In our view, however, this increase COMPETING TAXONOMIC NAMES in the diversity of palaeobotanical taxonomic concepts should not result in an increase in the complexity of the nomenclatural A perceived problem is that of competing names of dif- provisions for naming them. Giving rigid nomenclatural defi- ferent parts of plants or preservation states in the context of nitions as to what are or are not form-, organ- or autapo-taxa priority; this was one of the reasons for having morphotaxa so merely introduces artificial barriers between taxa, which could rigidly defined. In our view, however, this problem has been hinder what should be purely taxonomic decisions. Surely it is overstated. The traditional view has been exemplified by the better to continue with the current Vienna Code, which now example of a hypothetical fossil flower. A taxon is established allows any plant fossil to be named as a fossil taxon, whose and named for that flower based exclusively on its morphology, circumscription in terms of plant parts, life history stages and but another palaeobotanist later emends the diagnosis of the preservation states that can be included is defined by its di- flower taxon to include details of the pollen that it contains. If agnosis. These can include fossil species, fossil genera, fossil the pollen of that flower taxon could be assigned to a taxon of families and even fossil orders (a fossil order of dinoflagellate dispersed pollen, and the name of that dispersed pollen taxon cysts already exists; R. Fensome, pers. comm., 2009); these

266 TAXON 59 (1) • February 2010: 261–268 Cleal & Thomas • Botanical nomenclature and plant fossils

are not to be confused with orders of whole fossil organisms, Cambridge, 1930; Adopted and revised by the International Botani- the basionym of whose names may be a fossil genus, such as cal Congress of Amsterdam, 1935. Brittonia 6: 2–30. the Archaeopteridales. Their circumscription can be changed Chaloner, W.G. 1986. Reassembling the whole plant, and naming it. Pp. 67–78 in: Spicer, R.A. & Thomas, B.A. (eds), Systematic and as new information and ideas become available, and so fossil taxonomic approaches in palaeobotany. Systematics Association taxa provide the flexible nomenclatural tool for which palaeo- Special Volume 31 Oxford: Clarendon Press. botanists have been seeking for decades. Fossil taxa can be Chaloner, W. G. 1999. Taxonomic and nomenclatural alternatives. Pp. regarded as the successors of organ-genera in their original 179–183 in Jones, T.P. & Rowe, N.P. (eds.), Fossil plants and spores: concept introduced by Thomas (1935) and Jongmans & al. Modern techniques. London: Geological Society. Chaloner, W.G. 2004. (213–214) Proposals to clarify the application (1935), and form-genera as in the proposals reported by Voss of the term “morphotaxon” in Fossil plant nomenclature. Taxon (1976) and made by Fensome & Skog (1997) (the latter re- 53: 850–851. named as morphotaxa), but not as those taxonomic concepts Chaloner, W.G., Greuter, W., Nicolson, D.H. & Traverse, A. 1998. became reinterpreted within the various editions of the Code. (97–100) Proposals regarding the nomenclature of fossil plants. In our view the concept of morphotaxa for naming plant Taxon 47: 907–910. fossils is flawed logically and unnecessary in practice; it has Cleal, C.J. & Shute, C.H. 2003. Systematics of the Late Carbonifer- ous medullosalean pteridosperm Laveineopteris and its associated tended to confuse palaeobotanists and to obscure the concept Cyclopteris leaves. Palaeontology 46: 353–411. of fossil taxa that can be more meaningfully used to name Cleal, C.J., Shute, C.H. & Zodrow, E.L. 1990. A revised taxonomy plant fossils. Moreover, it runs counter to one of the main for Palaeozoic neuropterid foliage. Taxon 39: 486–492. principles of botanical nomenclature, that it should not sig- Cowan, R.S. & Stafleu, F.A. 1982. The origins and early history of nificantly influence taxonomic decisions (McNeill, 2000). IAPT. Taxon 31: 425–420. We will, therefore, be making formal proposals to remove Damer, T.E. 2008. Attacking faulty reasoning: A practical guide to fallacy-free arguments, 6th ed. Florence, Kentucky: Wadsworth. morphotaxa from the Code and to make the meaning of fossil Edwards, D. & Kenrick, P. 1986. A new zosterophyll from the Lower taxa in that context clearer. Devonian of Wales. Bot. J. Linn. Soc. 92: 269–283. Faegri, K. 1963. Organ and form genera: Significance and nomencla- tural treatment. Taxon 12: 20–28. ACKNOWLEDGEMENTS Fensome, R.A. & Skog, J. 1997. (04–05) Two proposals to amend pro- visions regarding form-genera of fossil plants. Taxon 46: 557–562. Gothan, W. 1931. Die Steinkohlenflora der westlichen paralischen We thank John McNeill and Judy Skog for advice and valuable Car bon reviere Deutschlands, Lieferung 2. Arbeiten Inst. Paläobot. discussions on this problem. We also thank Richard Bateman, Robert 1: 49–96. Fensome and three other anonymous referees for extremely construc- Gothan, W. 1941. Paläobotanische Mitteilungen. 5. Die Unterteilung tive reviews of the submitted manuscript; and the library staff of the der karbonischen Neuropteriden. Paläontol. Z. 22: 421–438. Greuter, W. & Hawksworth, D.L. 1999. Synopsis of Proposals on Natural History Museum (London) for assistance with obtaining some Botanical Nomenclature – St Louis 1999: A review of the propos- of the publications referred to in this contribution. als concerning the International code of botanical nomenclature submitted to the XVI International Botanical Congress. Taxon 48: 69–128. LITERATURE CITED Greuter, W., McNeill, J., Barrie, F.R., Burdet, H.M., Demoulin, V., Filgueiras, T.S., Nicolson, D.H., Silva, P.C., Skog, J.E., Trehane, Ash, S.R. & Tidwell, W.D. 1986. Arnoldia kuesii: A new juvenile P., Turland, N.J. & Hawksworth, D.L. 2000b. International code fernlike plant from the Lower Permian of New Mexico. Bot. Gaz. of botanical nomenclature (Saint Louis Code): Adopted by the 147: 236–242. Sixteenth International Botanical Congress; St Louis, Missouri, Barthel, M. 1968. “Pecopteris” feminaeformis (Schlotheim) Sterzel July-August 1999. Regnum Vegetabile 138. Königstein: Koeltz und “Araucarites” spiciformis Andrae in Germar – Coenopteri- Scientific Books. deen des Stephans und Unteren Perms. Pälaontol. Abh., Abt. B, Greuter, W., McNeill, J., Hawksworth, D.L. & Barrie. G.R. 2000b. Paläobot. 2(4): 727–742. Report on botanical nomenclature – Saint Louis 1999. XVI Inter- Bateman, R.M. & Hilton, J. 2009. Palaeobotanical systematics for national Botanical Congress, Saint Louis: Nomenclature Section, the phylogenetic age: Applying organ-species, form-species and 26 to 30 July 1999. Englera 20: 1–272. phylogenetic species concepts in a framework of reconstructed Greuter, W. & Nicolson, D.H. 1996. Introductory comments on the fossil and extant whole-plants. Taxon 58: 1254–1280. Draft BioCode, from a botanical point of view. Taxon 45: 343–348. Boulter, M.C. 1979. A proposal to emend Article 3.2 of ICBN. Taxon Hawksworth, D.L., McNeill, J., Sneath, P.H.A., Trehane, R.P. & 28: 598–600. Tubbs, P.K. 1994. Towards a harmonized bionomenclature for life Briquet, J. 1906. Règles internationales de la nomenclature botanique on earth. Biology International, Special Issue 30: 1–44. adoptées par le Congrès International de Botanique de Vienne Jansonius, J. 1974. Form-genera versus organ-genera: A proposal. 1905. Jena: Fischer. Taxon 23: 867–868. Brongniart, A. 1822. Sur la classification et la distribution des végé- Jennings, J.R. & Millay, M.A. 1978. A new permineralized marat- taux fossiles en général, et sur ceux des terrains de sediment supéri- tialean fern from the Pennsylvanian of Illinois. Palaeontology 21: eur en particulier. Mém. Mus. Natl. Hist. Nat. 8: 203–240, 297–348. 709–716. Brongniart, A. 1839. Observations sur la structure intérieur du Sigil- Jongmans, W.J., Halle, T.G. & Gothan, W. 1935. Proposed editions laria elegans compare à celle des Lepidodendron et des Stigmaria to the International Rules of Botanical Nomenclature adopted by at à celle des végétaux vivants. Arch. Mus. Hist. Nat. 1: 405–561. the fifth International Botanical Congress Cambridge 1930. Heer- Camp, W.H., Rickett, H.W. & Weatherby, C.A. 1947. International len: Geologisch Bureau. Rules of Botanical Nomenclature: Formulated by the Interna- Krassilov, V.A. 1968. 253. Proposal on the definition of the terms tional Botanical Congress of Vienna, 1905, Brussels, 1910, and form-genus and organ-genus. Taxon 17: 741.

267 Cleal & Thomas • Botanical nomenclature and plant fossils TAXON 59 (1) • February 2010: 261–268

Lanjouw, J. 1953. Nomenclature, NOM. Report. Pp. 457–550 in with ovules of the Hydrasperma tenuis-type. Rev. Palaeobot. Paly- Osvald, H. & Åberg, E. (eds), Proceedings of the Seventh Inter- nol. 60: 295–309. national Botanical Congress: Stockholm July 12–20, 1950. Stock- Schopf, J.M. 1963. Nomenclature for fossil plants. Taxon 12: 279–282. holm: Almqvist & Wiksell; Waltham Massachusetts: The Chronica Shute, C.H. & Cleal, C.J. 1986. Palaeobotany in museums. Geol. Botanica Co. Curator 4: 553–559. Lanjouw, J., Baehni, C., Merrill, E.D., Rickett, H.W., Robyns, W., Skog, J. 1999. Report of the Committee for Fossil Plants: 2. Taxon Sprague, T.A. & Stafleu, F.A. 1952. International code of botani- 48: 817–819. cal nomenclature: Adopted by the Seventh International Botanical Skog, J. 2005. Report on the Committee for Fossil Plants: 6. Taxon Congress; Stockholm, July 1950. Regnum Vegetabile 3. Utrecht: 54: 827. International Bureau for Plant Taxonomy of the International As- Sprague, M.L., Lanjouw, J. & Andreas, C.H. 1950. Minutes of the sociation for Plant Taxonomy. Utrecht Conference. Chron. Bot. 12: 9–53. Lanjouw, J., Baehni, C., Robyns, W., Rollins, R.C., Ross, R., Rous- Sprague, T.A. 1935a. Synopsis of proposals concerning nomenclature seau, J., Schulze, J.M., Smith, A.C., de Vilmorin, R. & Stafleu, submitted to the Sixth International Botanical Congress. Amster- F.A. 1956. International code of botanical nomenclature: Adopted dam 1935. Cambridge: The University Press. by the Eighth International Botanical Congress; Paris, July 1954. Sprague, T.A. 1935b. Preliminary opinions concerning nomenclature Regnum Vegetabile 8. Utrecht: International Bureau for Plant Tax- proposals submitted to the Sixth International Botanical Congress, onomy of the International Association for Plant Taxonomy. Amsterdam 1935. Cambridge: The University Press. Lanjouw, J., Baehni, C., Robyns, W., Ross, R., Rousseau, J., Schopf, Sprague, T.A. 1950. International Rules of Botanical Nomenclature: J.M., Schulze, J.M., Smith, A.C., de Vilmorin, R. & Stafleu, Supplement embodying the alterations made at the Sixth Interna- F.A. 1961. International code of botanical nomenclature: Adopted tional Congress, Amsterdam, 1935. Chron. Bot. 12: 63–77. by the Ninth International Botanical Congress; Montreal, August Stafleu, F.A. 1967. Palynology, nomenclature and terminology. Rev. 1959. Regnum Vegetabile 23. Utrecht: International Bureau for Plant Paleobot. Palynol. 3: 15–26. Taxonomy of the International Association for Plant Taxonomy. Stafleu, F.A. 1988. The prehistory and history of IAPT. Taxon 37: Mamay, S.H. 1964. Reports nomenclature committees: Committee for 791–800. Fossil Plants. Taxon 13: 133–135. Stafleu, F.A., Demoulin, V., Greuter, W., Hiepko, P., Linczevski, McNeill, J. 2000. Naming the groups: Developing a stable and efficient I.A., McVaugh, R., Meikle, R.D., Rollins, R.C., Ross, R., nomenclature. Taxon 49: 705–720. Schopf, J.M. & Voss, E.G. 1978. International code of botani- McNeill, J., Barrie, F. R., Burdet, H.M., Demoulin, V., Hawk- cal nomenclature: Adopted by the Twelfth International Botanical sworth, D.L., Marhold, J., Nicolson, D.H., Prado, J., Silva, P.C., Congress; Lenningrad, July 1975. Regnum Vegetabile 97. Bohn, Skog, J.E., Wiersema, J.H. & Turland, N.J. 2006. International Scheltema & Holkema, Utrecht. code of botanical nomenclature (Vienna Code): Adopted by the Sternberg, K.M. von 1820–1838. Versuch einer geognostisch-bot- Seventeenth International Botanical Congress; Vienna, Austria, anischen Darstellung der Flora der Vorwelt. Prague: Published July 2005. Regnum Vegetabile 146. Ruggell: Gantner. by the author. McNeill, J., Stuessy, T.F., Turland, N.J. & Hörandl, E. 2005. XVII Thomas, B.A. & Brack-Hanes, S.D. 1984. A new approach to family International Botanical Congress: Preliminary mail vote and re- groupings in the lycophytes. Taxon 33: 247–255. port of Congress action on nomenclature proposals. Taxon 54: Thomas, H.H. 1935. Proposed additions to the International Rules 1057–1064. of Botanical Nomenclature suggested by British palæobotanists. McVaugh, R., Ross, R. & Stafleu, F.A. 1968. An annotated glossary of J. Bot. 73: 111–113. botanical nomenclature. Regnum Vegetabile 56. Utrecht: Interna- Traverse, A. 1975. Committee for Fossil Plants: Pre-Leningrad report. tional Bureau for Plant Taxonomy of the International Association Taxon 24: 387–389. for Plant Taxonomy. Traverse, A. 1981. Report of the Committee for Fossil Plants. Taxon Meyen, S.V. 1975. Paleobotanical taxonomy and nomenclature: the 30: 473–475. need for a new approach. Taxon 24: 251–254. Voss, E.G. 1976. XII International Botanical Congress: Mail vote Meyen, S.V. 1978. Systematics, phylogeny and ecology of propterido- and final Congress action on nomenclature proposals. Taxon 25: phytes. Byull. Moskovsk. Obshch. Isp. Prir. Otd. Biol. 83: 72–84. 169–210. [In Russian.] Voss, E.G., Burdet, H.M., Chaloner, W.G., Demoulin, V., Hiepko, Meyen, S.V. 1982. The Carboniferous and Permian floras of Angara- P., Mcneill, J., Meikle, R.D., Nicholson, D.H., Rollins, R.C., land (a synthesis). Biol. Mem. 7: 1–110. Silva, P.C. & Greuter, W. 1983. International code of botanical Meyen, S.V. & Traverse, A. 1979. Proposal on Article 3: Remove nomenclature: Adopted by the Thirteenth International Botanical “form-genus” too! Taxon 28: 595–598. Congress; Sydney, August 1981. Regnum Vegetabile 111. Utrecht: Phillips, T.L. & DiMichele, W.A. 1992. Comparative ecology and Bohn, Scheltema & Holkema; The Hague: W. Junk. life-history biology of arborescent lycopsids in Late Carboniferous Watson, J. 1988. The Cheirolepidiaceae. Pp. 382–447 in Beck, C.G. swamps of North America. Ann. Missouri Bot. Gard. 79: 560–588. (ed.), Origin and evolution of gymnosperms. New York: Columbia Potonié, R. 1964. Palaeobotanical discussion of the Code 1961. Taxon University Press. 13: 126–128. Wittry, J. 2006. The Mazon Creek fossil flora. Downers Grove, Il- Rothwell, G.W. & Wight, D.C. 1989. Pullaritheca longii gen. nov. and linois: Esconi Associates, . Kerryia mattenii gen. et sp. nov.: Lower Carboniferous cupules

268