[Taxacom] Describing genera without molecular phyolgies

[John Grehan]

Since I do that all the time I am in some sympathy with the extracted text
below of a fairly recent article. Thought it might be of interest to others
who may not be aware of it.

John Grehan

BARNA PÁLL-GERGELY 2017 Should we describe genera without molecular
phylogenies? Zootaxa 4232 (4): 593–596

For nearly a decade I have published species descriptions and revisions of
land snails in peer- reviewed journals. As a systematist, it is often
requisite to reclassify species into other, sometimes unnamed genera.
Although in most cases editors and reviewers have not commented on the
taxonomic changes I have made and the new taxa I described, I sometimes
received negative critiques when I described new genera unaided by
molecular phylogenetic support. I feel these critiques have become
increasingly more frequent, and am convinced that many fellow taxonomists
share this experience. Addressing this problem is particularly difficult
due to three reasons. First, it is impossible to support these observations
with statistical data (i.e. the frequency of similar reviews increasing or
not); second, the increasing number of published phylogenetic works reveals
more and more cases of polyphyletic genera, which might suggest that
morphology-based generic grouping is unreliable, and thus, should be
avoided; and third, no publications exist suggesting not to describe genera
using (still) reliable morphological foundations. Subsequently, the
unsatisfactory review process of taxonomic works is often exacerbated by
the biases of reviewers and their inobservance of valuable historic
convention in light of the current wave of molecular phylogenies. Moreover,
the future value of morphological descriptions for biodiversity assessments
(i.e. IUCN) is not even addressed. Here, I argue against prohibiting
descriptions of genera without phylogenetic support, and make suggestions
how editors and reviewers should handle "phylogeny-free" genus
descriptions. To illustrate my experience, I quote some text from previous
reviews and editorial comments that I have received: Editor of a taxonomic
journal, (07.02.2012). "(...) In particular, I need you to provide much
more information regarding your decision to erect new genera; as of now you
provide no rationale, and you must fully defend this approach. I frankly
would feel much more comfortable if you had corroborating DNA sequence
analyses to demonstrate that these are not simply subgenera or even simple
groups within already existing genera -- and you have not even minimally
done this. There is a strong burden of poof to undertake such splitting,
and I expect these decisions to be clearly reasoned and proved with
empirical data. If you cannot do this, then I strongly suggest that you
place your new taxa into already existing generic entities." Anonymous
reviewer (02.11.2016). "(...) As for the description of a new subgenus, I
am strongly opposed to this taxonomic proposal. Taxonomy serves three
possible purposes: (1) the ego of the author, (2) the human desire to
classify things, (3) the scientific purpose to conceptualize ideas about
the evolutionary relationships amongst organisms. The only purpose this new
subgenus serves is 1. (...) As for purpose 3, attempting to conceptualize
the evolutionary relationships of species, this decision is even worse.
Inevitably, you are making a statement about the evolution of this group
and you imply that one species is the sister group of all other species
combined based on one character. You have a one in six chance that you are
right because your treatment is not backed up by any phylogenetic
analysis." Anonymous reviewer (23.11.2016) "(...) The taxonomic description
of the new species is also good. However, in the absence of molecular
sequence data, I consider that there are insufficient grounds for erecting
a new genus. (...) As taxonomists living in the 21st century we have a
responsibility to investigate the phylogenetic basis of
historically-erected genera and sub-genera by the combined study of
morphological and molecular data. This is often not possible because of
limited funding, and so the best course when describing species new to
science is to work with existing generic and sub-generic names and to avoid
contributing to taxonomic instability by creating new generic level taxa
purely on the basis of a few shell characters." 594 · Zootaxa 4232 (4) ©
2017 Magnolia Press PÁLL-GERGELY In the first case, my proposal to erect
new pulmonate land snail genera was based on shell and genital anatomical
characters, whereas in the latter two cases, I attempted to describe new
(sub)genera based on differences of the breathing tubes, a traditional
approach used in the respective land snail groups (Kobelt, 1902). In other
words, 30–200 years ago the scientific community would not have questioned
the establishment of these new genera based on morphology. Naturally,
without knowing the background of the above examples, it is not possible to
form an opinion whether I was right or wrong when I tried to erect new
genera, although this is not my point here. Instead, I aim to show that the
preference of existing generic names over erecting new genera is strong.
Are the reviewers right? Strictly speaking, the reviewers, who disapprove
and forbid morphology-based genera may be right. "How do I know that they
are really monophyletic?" We cannot know whether a taxon is monophyletic
without a proper phylogeny. To investigate this question we have to go back
to the definition of the genus. If we define a genus as "a group of species
that are more closely related to one another than they are to any species
assigned to another genus" (Wood & Collard 1999: 201), then only an
appropriate phylogeny would provide satisfactory evidence, although a
hypothesis of monophyly based on morphology should be also acceptable. If
we define a genus as a group of species defined by apomorphic character
states, morphology is sufficient. Regardless of generic definitions or
criteria, which are usually not given in genus descriptions, monophyly of
genera is, or at least should be assumed (Ebach et al. 2006) (although
species are not necessarily monophyletic; Schluter & Nagel 1995, Nosil et
al. 2002). The level of provided supporting evidence should be up to the
journal's policy and the taxonomists' personal preference. In the vast
majority of species descriptions, no species criteria are specified. Unlike
genera, describing species without molecular support appears acceptable.
Species criteria (e.g. presence of reproductive barriers) are mostly
assumed rather than tested in species descriptions. On the contrary,
assuming monophyly is deemed to be insufficient by some members of the
taxonomic community. The reason for this double standard is puzzling.
Especially since one might encounter larger difficulties handling synonyms
of species than those of genera. Reviewers suggesting not to describe
phylogeny-free genera supposedly aim to minimize the subjectivity of
taxonomy caused by the never-ending battle of lumpers and splitters. This
goal is admirable. However, taxonomy is a science, and all new taxa that
are proposed are hypotheses that can be refuted and falsified (Evenhuis
2008). Not always a matter of money Although molecular phylogeny becomes an
everyday tool for some proportion of professional taxonomists, it is not
available for many researchers due to the lack of funds. Moreover, in
Europe, 60 % of new taxa are described by nonprofessionals (Fontaine et al.
2012), who have nearly no chance to support their taxonomic decisions with
molecular phylogeny. However, the lack of funds and limited access to
molecular laboratories explain only a fragment of cases when molecular
phylogeny could not be performed. In many cases it is not possible to
perform phylogenetic analysis due to a number of other reasons. For
example, the absence of specimens in museum collections potentially
containing DNA (e.g. empty mollusc shells and wrongly preserved historical
specimens; see Jaksch et al. 2016); type localities of species described
centuries ago are not precise enough to allow revisiting the original
sites; the original sites are destroyed and the targeted species might be
extinct in the wild; not to mention fossil taxa, which also need to be
placed in the system obviously without molecular support. Furthermore,
molecular studies that propose new taxa should also return to morphological
characters in light of the molecular phylogenetic trees. More repercussions
than merits Genera, irrespective of the fact that they are "old" or "new",
are testable hypotheses of monophyletic groups that predict the
distribution of characters (Platnick 1979; Wheeler 2004). In the absence of
molecular support, I cannot see a good scientific reason to prefer existing
genera over new genera. The action of placing a species into a genus
irrespective of being named or not, is a hypothesis of the species'
evolutionary history. Describing new, morphologically well-defined genera
is not "unnecessary splitting", but small steps towards understanding
evolutionary relationships of organisms. Forcing classification of new
species into already named genera could have more repercussions than
merits. It would, for example, result in unnecessarily "dumping" of species
into well-defined genera, the reduction of diagnostic apomorphic characters
in morphologically defined groups. Such practice will turn a well-defined
genus into a wastebasket taxon, which may, for instance, create artificial
biogeographic "connections" between areas. Taxonomy and systematics faces
several challenges in the 21st Century, and the naming of new genera
without phylogeny is certainly not the biggest of these. For instance,
incorrect identification of specimens used for molecular studies (Nilsson
et al. 2006; Groenenberg et al. 2011), taxonomic inflation (Isaac et al.
2004; Harris & Froufe 2005), and DESCRIBE GENERA WITHOUT MOLECULAR
PHYLOGENIES? Zootaxa 4232 (4) © 2017 Magnolia Press · 595 the increasing
gap between phylogeny and classification (Franz 2005) are much more serious
problems. On one hand, phylogeny provides us with a powerful tool to
hypothesise evolutionary relationships. On the other hand, since the
world's biodiversity is largely unknown (some 86% of non-marine eukaryotic
species are unknown; Mora et al. 2011) and the number of taxonomists is
decreasing (Bebber et al. 2014; Wheeler 2014), it is not yet the time to
regulate morphology-based grouping of species. Instead of criticising their
well-trained, professional expertise (gleaned through years of intense
study), we should invest more trust in the taxonomic evaluations of the
decreasing number of taxonomists as well as allow more freedom for
morphology-based grouping. Especially, given that if someone wants to
describe something, he/she can find a way to do so in local, small non
peer-reviewed journals or self-published books. One-sided critiques
emphasising only the taxonomic value of molecular assessment could well
result in the weakening trust of taxonomists (mostly the ones not dependent
on impact factors) in peer-reviewed journals, which is already a major
problem in today's taxonomy. Morphology is still what makes the organism a
tangible entity beyond its DNA. Necessary changes in the review process
What we need, is better editorial practice. To overcome this recent trend,
I suggest considering the following points: (1) Authors have to justify
their conclusions clearly with their data. Also, if the reason is other
than no access to molecular laboratory and funds, they should state the
reason why molecular phylogeny is not performed. (2) Editors need to
understand the hypothesis-driven nature of taxonomy, systematics and
phylogeny, and need to be able to ignore reviewers who do not engage the
subject. (3) Reviewers should refrain from rejecting the description of new
genera on the basis of the lack of molecular data. Instead, they should be
critical of poorly defined genera, no clear gaps in the morphological
continuum, overlapping character states across genus-group taxa, and genera
awaiting descriptions based on non-conventional morphological characters.
(4) Taxonomic journals could state whether they allow, discourage or
prohibit descriptions of taxa above species level without molecular
phylogenetic support. On one hand, this would allow the authors to prepare
to the nature of reviews, and choose journals where the lack of sequence
data will not result in changing the proposed taxonomic decisions. On the
other hand, this would create a rather harmful division in how journals
deal with taxonomic submissions.