There is an increasing shift in taxonomy toward modifying the directives on how to name organisms to reflect genealogical relationships. However, despite the fact that most systematists would agree that the hierarchical system of names in biology should be governed by phylogenetic hypotheses instead of by intuitive similarity, the increasing pressure of phylogenetic demands on the naming processes results in intrinsic conflicts in systematics. Phylogenetic analyses have not been conducted for most groups of organisms yet, and it is highly unlikely that many will be undertaken soon. Additionally, new species are rapidly discovered in large numbers, particularly among invertebrates. Since a taxonomic treatment of such new taxa, that is, formal descriptions in taxonomic revisions or even single taxon descriptions, is a prerequisite to any other scientific exploration, new taxa are more rapidly described and made available than phyloge-netic hypothesis are worked out. The taxonomic treatment of certain taxa, such as those that are less known, newly discovered, and or extraordinarily species rich, is a fundamental first step that opens up the possibility of continuing with subsequent phylogenetical, genetical, behavioral, and other studies.
In practical taxonomy, it is often necessary or at least recommendable initially to leave a comprehensive revision or a phylogenetic reconstruction aside from research projects and the resulting publications. In principle I agree with Mayr and Ashlock (1991 p 347) that "the isolated description of ... new species ..., divorced from revisional or monographic work, is the least desirable form of taxonomic publication.'' But Mayr and Ashlock correctly qualify their assertion themselves in stating that it does not hold "in well-known groups.'' A well-known group is usually a taxon which is not only well known in terms of scientific research but also which additionally receives extraordinary attention by both the scientific and the general public. It is thus not surprising that isolated descriptions of fossil and recent Primates and large mammals often make their way to the highest ranked scientific journals (e.g., the description of Homo floresiensis by Brown et al. 2004). Obviously, the publicity that results from such publications can be a strong motivation for any scientist to publish a rather isolated description immediately rather than to invest more time to obtain more data for a more comprehensive publication, due to the current system of scientists being under considerable pressure to publish and to compete for limited research money and, finally, jobs.
There are several more potential reasons to conduct descriptive taxonomy (Godfray 2002), which may mutually affect each other in most cases. Examples are: (1) it might be desirable to make a new discovery formally available for further studies, which can or should not be conducted at the time of discovery of the new taxon for varying reasons; (2) if a scientist has discovered a new species, and if he intends to describe it formally, this scientist might want to guarantee that the name of this taxon as proposed by him is the first formal description and, thus, has priority over any subsequent name. This might be considered as to imply a connotation of personal rather than true scientific motivation, similar to the argumentation as presented earlier. However, priority is a basic principle of zoological nomenclature, and although it is explicitly thought to be priority of publication without reference to the date of discovery, the publication date here serves as an objectifyable reference point. Hence, priority of publication just replaces the priority of discovery for practical reasons; (3) incomplete knowledge of data might prevent systematists for conducting a comprehensive revision or a cladistic analysis. This might be due to the lack of characters (e.g., in fossils or other incompletely preserved specimens) or material (e.g., modern techniques such as molecular systematics or scanning electron microscopy require fresh or specifically preserved material, but many rare but potentially informative species are known only from dry museum material, which cannot be fully examined).
Another topic concerns the role of biological systematics in the context of a global biodiversity assessment. The development of global species inventories is considered to be an urgent and vitally important task that is a primary step and fundamental activity for any kind of biodiversity research (Stork and Samways 1995; Purvis and Hector 2000; Wilson 2003). Although it seems to be clearly unrealistic to describe every species of organism on earth, not to speak of the monumental uncertainties as to how many species there are (Godfray 2002), any step toward a global inventory of selected "target taxa'' should be achieved as soon as possible.
Phylogenies are inherently hypothetical, simply because they portray historical processes, which cannot be inferred directly. Hypotheses, however, can be of different quality. The reliability of a hypothesis largely depends on the quality of the underlying data, which involves numerous theoretical and methodological aspects. Among these aspects, the completeness of the data set is of crucial importance. Completeness of characters is an illusion, since each single organism theoretically consists of an infinite number of characters. Completeness of taxa is, at least in principle, possible. With regard to taxon sampling, the perfect systematic study would include all species of a given taxon, perhaps even both still-living and extant. Practically, completeness of taxa is unlikely to be achieved in most groups, except perhaps for some exceptionally well-studied taxa such as birds. However, the reliability of a phylogenetic analysis increases with the increasing completeness of the taxon sampling. Conversely, gaps in the taxon coverage result in considerably less reliable phylogenetic hypotheses. In many cases, phylogenetic analysis should better wait for a more complete species inventory, which, particularly in taxa with many species or with a patchy distribution, can hardly be achieved in one step.
In summary, publications concerned with the taxonomy of a given group are most desirable if their classificational results are based on a comprehensive phylogenetic hypothesis. If at all possible, a systematist should try to interpret his taxonomic data with respect to the corresponding phylogeny rather than relying on an intuitive character evaluation as a basis for a classification. Incidentally, since the hierarchical structure of the organisms on earth is the result of a historically real, continuous sequence of species splittings, the hierarchy of organisms as reconstructed by phylogenetic methods is in the end the only objective base for a classification.
Was this article helpful?