When genetic and phenotypic data do not agree: the conservation implications of ignoring inconvenient taxonomic evidence

Abstract

Taxonomy plays a central role in conservation programs of threatened New Zealand taxa. The role of taxonomy is especially relevant for highly vulnerable taxa, where the identification of distinct lineages is essential to define units of conservation and to appropriately allocate conservation resources. Taxonomy traditionally relied on phenotype, but in the past 30 years, the use of genetic data has become prominent in the field. While both phenotypic and genetic approaches to taxonomy have their own merit, they do not always agree. In such cases, favouring one type of data over the other when they are in conflict can have important implications for conservation management. We highlight this issue using several examples from the taxonomy of threatened New Zealand birds. We caution against biases in interpretation of each data type and advocate for a more integrative approach to taxonomy where the limitations of all approaches are carefully considered.

The importance of recognising and conserving biological diversity: a reply to Dussex et al. (2018)

“…In our view, the 'lines of evidence' presented in Rawlence et al's (2017) paper were neither conflicting, inconvenient nor ignored. Cygnus atratus and C. sumnerensis, like many close pairs of avian sister taxa (see below), are clearly supportable under both phylogenetic (Cracraft 1983) and diagnosable (Baum & Donoghue 1995;Helbig et al 2002;Patten & Unitt 2002;Issac et al 2004;Cicero & Johnson 2006;Gill et al 2010) species concepts, the latter designed to specifically avoid taxonomic overinflation (in addition to the integrative taxonomic approaches apparently advocated by Dussex et al (2018)). …”

“…Specifically, Dussex et al (2018) take issue with Rawlence et al's (2017) extinct New Zealand swan (Cygnus spp.) taxa, despite the fact that Rawlence et al's (2017) recommendations were based on congruent findings from both morphological and genetic data analyses (Fig.…”

“…In their recent article, Dussex et al (2018) draw attention to some current issues in New Zealand avian systematics, and particularly the difficulties that can potentially arise in cases where molecular and morphological data prove difficult to reconcile. These researchers call for integrative approaches to avian systematics, to overcome the limitations of singlemethod studies, and we wholeheartedly agree with this common-sense view.…”

“…1). Unfortunately, Dussex et al (2018) provide no data or analyses to substantiate their seemingly inaccurate allegations of incongruence ("When genetic and phenotypic data do not agree"). In our view, the 'lines of evidence' presented in Rawlence et al's (2017) paper were neither conflicting, inconvenient nor ignored.…”

“…It is possible that Dussex et al (2018) are primarily taking issue with the depth of mtDNA differentiation detected between the extant Australian C. atratus and extinct New Zealand C. sumnerensis, whereby Rawlence et al (2017) "proceed to declare a new species of swan based on 1.4% divergence". Apparently, in their view, 1.4% mtDNA divergence (regardless of any other biological differentiation) is simply too small to merit species-level recognition.…”