The reality of taxonomic change

Abstract

In their recent TREE paper [1xTaxonomic inflation: its influence on macroecology and conservation. Isaac, N.J.B. et al. Trends Ecol. Evol. 2004; 19: 464–469Abstract | Full Text | Full Text PDF | PubMed | Scopus (320)See all References[1], Isaac et al. assert that species numbers are increasing rapidly owing to ‘taxonomic inflation’, where previously identified species are reclassified under new species concepts into smaller and more numerous new taxa. Using recent phylogenetic concepts (as opposed to classic concepts, notably the biological or isolation concept) will certainly lead to an apparent increase in species number [2xThe species problem and conservation: what are we protecting?. Rojas, M. Conserv. Biol. 1992; 6: 170–178CrossrefSee all References, 3xSpecies concepts in systematics and conservation biology – an ornithological viewpoint. Cracraft, J. : 325–339See all References, 4xSpecies of marine invertebrates: a comparison of the biological and phylogenetic species concepts. Knowlton, N. and Weigt, L.A. : 199–219See all References]. What is unsupported is that such inflation is the dominant force behind the current increase in species numbers. We argue instead that this increase reflects the underlying nature of species.Isaac et al. point out that researchers use different species concepts for different purposes, suggesting that this is an unrecognized dimension of the species problem, although several publications describe this exact issue (one of which they cite [5xUnderstanding and confronting species uncertainty in biology and conservation. Hey, J. et al. Trends Ecol. Evol. 2003; 18: 597–603Abstract | Full Text | Full Text PDF | Scopus (150)See all References[5], but also see [6xPluralism in species concepts: dividing nature at its diverse joints. Sluys, R. and Hazevoet, C.J. Species Div. 1999; 4: 243–256See all References[6]). They then offer two solutions: (i) an ideal world, in which the species concept used for describing taxonomy (and, hence, taxic biodiversity) is the same regardless of whether it is applied to ecological, macroecological or conservation purposes; or (ii) standardized species lists. In our opinion, both solutions are flawed.‘In an ideal world, each [ecological, macroecological, etc.] hypothesis would be tested using the most appropriate rank in the taxonomic hierarchy.’ However, assigning ranks to taxonomic groups above the species level is largely subjective. Thus, ranks are non-comparable and of little use for testing hypotheses. It is this arbitrariness that has provoked proposals for a taxonomy and nomenclature devoid of categorical rankings [7xImplementing the PhyloCode. Donoghue, M.J. and Gauthier, J. Trends Evol. Evol. 2004; 19: 281–282Abstract | Full Text | Full Text PDF | PubMed | Scopus (16)See all References, 8xThe PhyloCode: naming of biodiversity at a crossroads. Sluys, R. et al. Trends Ecol. Evol. 2004; 19: 280–281Abstract | Full Text | Full Text PDF | PubMed | Scopus (13)See all References]. Even with respect to the rank of species, the ideal world will never materialize. Phylogenetic evolutionists, ecologists and population biologists, for example, have different research goals, focusing on different aspects of biodiversity arising from different evolutionary processes. The phylogenetic evolutionist is interested in the evolution of taxa and, therefore, will use a taxic descriptor of species, such as the phylogenetic species concept (PSC), based on historical patterns that can be objectively recognized in nature [5xUnderstanding and confronting species uncertainty in biology and conservation. Hey, J. et al. Trends Ecol. Evol. 2003; 18: 597–603Abstract | Full Text | Full Text PDF | Scopus (150)See all References[5]. Ecologists and population biologists are more likely to take a contemporaneous ‘snapshot’, focusing on ongoing processes in restricted areas. It is unlikely that these different views will or can ever agree. Isaac et al. also assume that conservationists have only one goal: the preservation of species. But the existence of several species-level entities in nature, described by various species concepts, means that conservationists must also decide, for any given case, which aspect of biodiversity can be diagnosed and then form the focal level of preservation efforts.The authors' second proposal is to use standardised species lists [9xGuidelines for assigning species ranks. Helbig, A.J. et al. Ibis. 2002; 144: 518–525Crossref | Scopus (192)See all References[9], so that studies based on such lists can at least be consistent, stating that such lists have ‘many obvious virtues’ and would form ‘a useful reference point for all species-related issues’ [our emphasis]. This ignores huge practical and theoretical difficulties. As argued above, no single species concept can fulfil all of the different requirements posed by the various disciplines. Furthermore, it is doubtful whether any standardized list would be stable for any reasonable period of time. For even some well known taxa, taxonomic lists have been far from stable [10xDutch avifaunal list: species concepts, taxonomic instability, and taxonomic changes in 1977–1998. Sangster, G. et al. Ardea. 1999; 87: 139–165See all References[10]. For example, in the group of European amphibians and reptiles, of the 63 species recognized in 1996, only 35 had the same taxonomic status in 1960 [11xLists of European species of amphibians and reptiles: will we soon be reaching ‘stability’?. Dubois, A. Amphib. Rep. 1998; 19: 1–28CrossrefSee all References[11]. If such a small and well-studied group is and continues to be unstable, stability is unlikely for more cryptic taxa. Alternatively, Isaac et al. suggest the use of Evolutionary Significant Units (ESUs), which they incorrectly consider to represent a different rank in the taxonomic hierarchy. In any event, the use of ESUs is thwarted by their apparent lack of applicability in the real world [12xConsidering evolutionary processes in conservation biology. Crandall, K.A. et al. Trends Ecol. Evol. 2000; 15: 290–295Abstract | Full Text | Full Text PDF | PubMed | Scopus (998)See all References, 13xQuestioning species realities. Hendry, A.P. et al. Conserv. Genet. 2002; 1: 67–76CrossrefSee all References].Isaac et al. assert that taxonomic inflation will lead to an increase in false positives in testing macroevolutionary hypotheses, ‘because extra data are effectively pseudoreplicates’. This is true only if these ‘extra’ species are unreal, echoes of the core data. Application of the PSC might lead to ‘splitting’, but it is far from clear that either biological or phylogenetic species are more objectively real than the other. Furthermore, because phylogenetic species are diagnosably different from each other, they are not mere replicates, at least as regards the diagnosed characters. To even call the rise in species numbers ‘inflation’ is to imply that it is somehow unreal or unnatural, lacking an ontological basis. This implicit assumption is exemplified by the authors' assertion that ‘an inflated taxonomy [i.e. one based on the PSC] will therefore have inflated values of endemism’, because phylogenetic species generally will have smaller ranges and, thus, hotspots might be more apparent than real. Although they cite Meijaard and Nijman [14xPrimate hotspots on Borneo: predictive value for general biodiversity and the effects of taxonomy. Meijaard, E. and Nijman, V. Conserv. Biol. 2003; 17: 725–732Crossref | Scopus (31)See all References[14] in support of this viewpoint, these workers did not qualify one kind of hotspot as being less ‘real’ than another (V. Nijman personal communication). Instead, Meijaard and Nijman used the terms ‘apparent’ and ‘real’ to announce their objective to investigate whether hotspot identification depended on taxonomic classification or on the choice of taxa, a meaning different to that presumed by Isaac et al.Asserting that patterns of global species richness are sensitive to taxonomic inflation, Isaac et al. imply that this provides a biased picture of biodiversity. The fact that ‘inflated’ patterns might be real aside, the only evidence presented concerns the changed patterns in primate species richness. It is debatable whether such a small fraction of global biodiversity in a charismatic and heavily studied group adequately supports such a wide-ranging statement. Certainly, as the PSC will generally diagnose taxa with smaller ranges, more finely grained and numerous hotspots will emerge (cf. [14xPrimate hotspots on Borneo: predictive value for general biodiversity and the effects of taxonomy. Meijaard, E. and Nijman, V. Conserv. Biol. 2003; 17: 725–732Crossref | Scopus (31)See all References[14]). However, on the broad scale, it is our impression that studies generally identify similar global hotspots of species richness.Species are complex things, their identity resting on the methodologies used to diagnose them and on the many different ways that organisms have of belonging to a species. We would do well to embrace plurality, using different species concepts in different circumstances [5xUnderstanding and confronting species uncertainty in biology and conservation. Hey, J. et al. Trends Ecol. Evol. 2003; 18: 597–603Abstract | Full Text | Full Text PDF | Scopus (150)See all References[5], rather than searching for a platonic ideal. It is time to stop over-generalizing and to ‘get over it’ [15xSystematic generalization, historical fate and the species problem. O'Hara, R.J. Syst. Biol. 1993; 42: 231–246See all References[15].