Mammalian Biology 81 (2016) 185–188

Contents lists available at ScienceDirect

Mammalian Biology
journal homepage: www.elsevier.com/locate/mambio

Editorial

Tree thinking and species delimitation: Guidelines for taxonomy and

phylogenetic terminology
Frank E. Zachos
Natural History Museum Vienna, Mammal Collection, Burgring 7, Vienna 1010 Austria

a r t i c l e i n f o a b s t r a c t

Article history: By tradition, phylogenetic trees are presented in such a way that species-poor taxa are placed on the
Available online 22 October 2015 left side and their more diverse sister taxa on the right (usually with humans on the far right). This
often leads to reading the tree as a “ladder of progress” from left (allegedly primitive, basal or even
Keywords: ancestral) to right (allegedly advanced, derived, descendant). Although biologically and logically wrong
Phylogenetic trees and often bemoaned, the evolutionary literature teems with language perpetuating this phylogenetic
“Primitive lineage fallacy”
misconception. Likewise, the splitting of one into two or more species based on different versions of
Scala naturae
the Phylogenetic Species Concept has recently found a growing number of adherents, resulting, in the
Species concepts
Species splitting
eyes of many, in a trivialization of the species category (“taxonomic inflation”). This editorial briefly
summarizes these phylogenetic and taxonomic issues and provides guidelines to authors submitting
their studies to Mammalian Biology in order to avoid errors in phylogenetic discussions and to do justice
to the fundamental nature of species in biology.
© 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.

Introduction
This editorial is aimed at authors submitting their papers to
Mammalian Biology. We receive quite a number of manuscripts
dealing with phylogenetic relationships among mammal taxa
and/or the splitting and description of (new) mammalian species.
Most of these papers are good studies, reporting methodologically
sound analyses and results and drawing appropriate conclu-
sions from them. There are, however, even in otherwise good
studies, often very specific shortcomings, errors or inconsisten-
cies. With respect to phylogenetics, they usually result from an
erroneous interpretation of trees as “ladders of progress” result-
ing in, subsequently, phylogenetically inappropriate terminology
(“higher”, “lower”, “basal” and “ancestral” groups). In taxonomy
and species splitting/description, matters are more complicated
and less straightforward, but using monophyly/paraphyly of gene
trees or mere diagnosability as species criteria will inevitably lead
to taxonomic inflation (Isaac et al. 2004). This kind of taxonomic
proliferation is strongly discouraged here, and at the very least,
authors should explicitly name their criteria for species status (i.e.
the species concept adopted). Both topics are too complex to be cov-
ered in detail in an editorial like this, and below only a few general
E-mail address: frank.zachos@nhm-wien.ac.at
points and guidelines will be given. None of the points mentioned
here are novel insights (for tree interpretation, see Baum et al. 2005;
Omland et al. 2008; there is even a very similar editorial to this one,
published for the same reasons: Krell and Cranston 2004; for the
species debate, the list of publications is all but infinite, cf. Heller
et al., 2013, 2014; Zachos et al. 2013a; Zachos 2015, in press; for
an opposing view, see Groves 2013; Cotterill et al. 2014). How-
ever, the same issues and misconceptions arise again and again and
are perpetuated in quite a number of papers, so that emphasis and
reinforcement of the guidelines below are still needed. Authors sub-
mitting phylogenetic and taxonomic papers to Mammalian Biology
should follow them whenever possible.
Phylogenetic trees and how not to read them
Phylogenetic trees depict evolutionary relationships among
taxa. In cladograms, the only information conveyed is the topol-
ogy itself, i.e. the succession of branching events resulting in a
nested pattern of relationships. Other kinds of phylogenetic trees
also contain information on evolutionary rates (depicted by branch
lengths, phylograms) or absolute time (chronograms). In any case,
it never makes a difference which taxa are drawn on the right side
of the tree and which on the left (the same holds for the upper
and lower sides in horizontal trees). Relationships (i.e. the recency
of common ancestors) are only represented by the succession of

http://dx.doi.org/10.1016/j.mambio.2015.10.002
1616-5047/© 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
186 Editorial / Mammalian Biology 81 (2016) 185–188

splitting events, and therefore, in a vertical tree, only the y-axis A)

(time) contains information. The x-axis does not contain directional platypus kangaroo anteater aardvark deer mouse human
information, i.e. whether a group is more basal than or ancestral
to another! (In horizontal trees, it is the other way around, but I
will only discuss vertical trees as shown in Fig. 1). Phylogenetic
trees consist of branching lineages, each node (unless it is a poly-
tomy) gives rise to two sister taxa. Sister taxa are, by definition, of
the same age, and none of them can be more basal than or even
ancestral to the other. Which of the two is depicted on the right
and which on the left side is irrelevant and just a matter of con-
2
vention or, perhaps more precisely, tradition. The latter has it that
species-poor lineages (or those falsely deemed more “primitive”)
are usually shown on the left side, the species-rich sister taxa on 1
the right (“ladderize right” trees). This invites a common miscon-
me axis No direconal informaon! (illusory noon of progress)
ception, namely to read or interpret trees from “left to right” and
to conclude that there is an element of progress from left to right
when there is not (the so-called “primitive lineage fallacy”, Omland
et al. 2008). All branches can be freely rotated around the nodes
without changing the topology or any other information content of B)
the tree. Consequently, all three trees in Fig. 1 are actually exactly platypus kangaroo human mouse deer aardvark anteater

the same tree, although we are clearly more used to seeing ver-
sion A than B or C. Interpreting trees as ladders of progress also
introduces another linguistic mistake: taxa do not branch off the
tree. This would imply a main trunk, i.e. a main direction of the tree
(as in the famous Tree of Life by Ernst Haeckel). Such a main trunk,
however, does not exist; all possible trajectories from the root of the
tree to any of its tips are equivalent! There is only branching (into
sister groups), no branching off. In the words of Omland et al. (2008, 2
p. 863): “Stating which is the lineage to ‘branch off’ requires a fixed
reference that defines where evolution is heading” – clearly a long-
outdated view of evolution as a predetermined (“orthogenetic”) 1
process.
As a consequence, no terminal taxon in a phylogenetic tree
(cladogram, phylogram, chronogram) can be basal (basal to what
anyway? It is just as “far away” from its ancestor as its sister
group). Basal taxa exist, but not at the tips of trees, they are inter-
C)
nal ancestral or stem species (see circles 1 and 2 in Fig. 1). The
mouse human deer anteater aardvark kangaroo platypus
first split in Fig. 1 is indeed more basal (older) than split 2, but
apart from this specific context there is no use for the term basal
when describing a tree. Nor can taxa as a whole be primitive,
plesiomorphic, advanced, progressive or apomorphic; these terms
apply to characters and their various states – species and other
taxa are always an amalgam of plesiomorphies and apomorphies.
The platypus, sometimes viewed as the archetypically primitive
mammal (after all, it lays eggs) is in point of fact highly derived
with respect to a number of characters (the bill with its electrore-
2
ception and the venom spur, to name just the two most obvious
ones) (see Omland et al. 2008 for a web search of the platypus as
1
being primitive). With the possible exception of surviving stem
species (a highly contentious issue in phylogenetics), no extant
taxon can be ancestral to another. As Omland et al. (2008) rightly
state, phylogenies of extant (perhaps better: terminal) species
“show ‘evolutionary cousin’ relationships and should not imply Fig. 1. Three equivalent depictions of the same phylogenetic tree of representatives
one species is more primitive, whereas another is more advanced” of major mammalian taxa: Monotremata (platypus), Marsupialia (kangaroo), and
different groups within the Placentalia (Xenarthra: anteater, Afrotheria: aardvark,
(p. 856; they illustrate this with a pedigree of the Kennedy fam-
Laurasiatheria: deer, Euarchontoglires: mouse and human). Whether xenarthrans
ily). and afrotherians are really sister taxa is still being debated, but this is irrelevant
Still, the evolutionary and phylogenetic literature teems with for the present purpose. Version A shows the traditional way of representation
these biologically incorrect expressions. Omland et al. (2008) with monotremes on the far left, followed by marsupials and placental mammals.
give examples of which mammal species they found to be Humans are usually shown on the far right, implying a “ladder of progress” from
“primitive” monotremes and “slightly less primitive” marsupials to more “derived”
called primitive (expectedly, the opossum, monotremes, shrews placental mammals with humans at the top. This, however, is not implied in any of
and tenrecs ranked first). And consider this quote from the the three trees, nor is it (bio-)logically correct. Only the y-axis, but not the x-axis,
English Wikipedia site on mammals (https://en.wikipedia.org/ carries directional information (time). The circles with numbers 1 and 2 denote
wiki/Mammal, accessed on 4 October 2015, italics added): “The the first (and indeed most basal) split in the tree, i.e. the stem species giving rise
to monotremes and therians, and the second (less basal) split of the therian stem
relationships between these three lineages is contentious, and
species giving rise to marsupials and placental mammals.
three different hypotheses have been proposed with respect to
which group is basal with respect to other placentals. These
 Editorial / Mammalian Biology 81 (2016) 185–188
hypotheses are Atlantogenata (basal Boreoeutheria), Epitheria
(basal Xenarthra), and Exafroplacentalia (basal Afrotheria).” The
whole entry “Basal (phylogenetics)” (https://en.wikipedia.org/
wiki/Basal (phylogenetics), accessed on 5 October 2015) is even
more revealing, betraying serious misconceptions of phyloge-
netic theory. The only directional information in trees is along
the time axis and along various character state trajectories
from plesiomorphic to apomorphic (if these states are plotted
on the tree), but the latter will form different zigzag lines in
the tree due to mosaic evolution rather than a single straight
line.
The reason that the “primitive lineage fallacy” and reading
trees as ladders of progress is so widespread is almost certainly
a relic of a concept once popular among natural philosophers,
the scala naturae or “great chain of being” (Lovejoy 1936), the
idea that everything that exists can be arranged in a hierarchi-
cal scheme from lowest (inanimate objects) through all kinds of
living beings to highest (God). Needless to say, humans rank high-
est among the earthly living beings. Language making reference
(probably often unconsciously) to this concept (terms like higher
and lower, basal, primitive, etc.) permeates the evolutionary lit-
erature (see above and Rigato and Minelli 2013). While there is
of course increasing complexity in many lineages (just as there
is decreasing complexity), progress is an illusion of the human
mind! More than 150 years after the Origin of Species it is, at
long last, time to abandon these pre-evolutionary notions and
purge our scientific language from any terminology perpetuating
them.
Species splitting and the description of new species
Species delimitation, particularly the splitting of one into two or
more species and the (often) underlying notion of the diagnosabil-
ity version of the Phylogenetic Species Concept (dPSC), has recently
been a much debated topic in mammalian taxonomy (see refer-
ences above; Zachos in press presents a synthesis and extensive
discussion). Apart from theoretical arguments, conservation corol-
laries (epitomized by the title “The species and the specious” of a
comment in Nature on the occasion of the tercentenary of Linnaeus,
Marris 2007) have also been discussed at length, naturally with dif-
fering conclusions (e.g. Agapow et al. 2004; Frankham et al. 2012;
Gippoliti and Groves 2012; Zachos et al. 2013b). The situation with
species delimitation is much more complex than the interpreta-
tion of phylogenetic trees, mostly because reading trees as ladders
of progress is simply wrong, whereas species delimitation is much
more of a grey area. Because all 27 or so species concepts are based
on biological realities (such as interfertility, diagnosability, mono-
phyly, genetic compatibility and isolation, ecological niches and
many more), none of them can be wrong in a logical way (Zachos in
press). It is much more a question of whether they are appropriate,
feasible and consistent, which of course leaves much more room
for subjective opinion and interpretation. Ultimately, all species
concepts denote population lineages at some level or other in the
Tree of Life, which is why the Evolutionary Species Concept or
related definitions are often viewed as the primary species con-
cept, whereas all the other so-called species concepts are rather
secondary species identification criteria. I will not go into further
detail here, but refer the readers to the literature cited and refer-
ences therein.
Against this background, introducing precise rules on when to
allow species splitting and/or the description of a new species
would border on editorial censorship. The guidelines below in as
far as they pertain to taxonomy should therefore rather be viewed
as recommendations, but serious ones as species splitting will be
subjected to close scrutiny by both referees and editors.
187
Practical guidelines for phylogenetic and taxonomic papers
in Mammalian Biology
• Do not present phylogenetic trees as “ladders of progress”! In par-
ticular, refrain from using terms like “higher” and “lower” when
referring to taxa.
• Avoid language that treats phylogenetic trees as having a “main
trunk” – trees are just successions of branching events, and in
each such event two equivalent sister lineages arise. There is no
direction, nor is any of the two sister lineages more basal or more
advanced, primitive or progressive, etc. than the other! In other
words: there is branching, but no branching “off” of any alleged
main trunk of a phylogenetic tree.
• Do not treat any terminal taxon as ancestral to another.
• The terms plesiomorphic and apomorphic as well as related
expressions (primitive, derived, advanced, etc.) are applicable
only to characters/character states, not to taxa. Taxa always
combine plesiomorphic and apomorphic traits (except for
hypothetical stem species which, by definition, lack exclusive
apomorphies).
• When describing and/or splitting species make explicit refer-
ence to the species concept you are basing your decision on and
explain (based on your data) why you consider the level of
distinctness sufficient for taxonomic recognition at the species
level.
• Species are such fundamental and important units that they
should not be introduced carelessly. Species description and
splitting based on superficial data like simple morphometric
differences (including those that are statistically significant),
arbitrary values of genetic distance or phylogenetic relation-
ships derived from limited molecular datasets (single-locus
analyses, particularly mtDNA) is strongly discouraged. All
of these may serve to support conclusions derived from
more appropriate datasets, but are not sufficient on their
own.
• In the case of allopatric populations, divergence (e.g. genetic dis-
tance) similar to that known from sympatric sister species is not in
itself an argument for separate species as it erroneously assumes
identical evolutionary processes in sympatry and allopatry (see,
for example, Zachos in press and references therein).
References
Agapow, P.-M., Bininda-Emonds, O.R.P., Crandall, K.A., Gittleman, J.L., Mace, G.M.,
Marshall, J.C., Purvis, A., 2004. The impact of species concept on biodiversity
studies. Q. Rev. Biol. 79, 161–179.
Baum, D.A., Smith, S.D., Donovan, S.S.S., 2005. The tree-thinking challenge. Science
310, 979–980.
Cotterill, F.P.D., Taylor, P.J., Gippoliti, S., Bishop, J.M., Groves, C.P., 2014. Why one
century of phenetics is enough: response to ‘Are there really twice as many
bovid species as we thought?’. Syst. Biol. 63, 819–832.
Frankham, R., Ballou, J.D., Dudash, M.R., Eldridge, M.D.B., Fenster, C.B., Lacy, R.C.,
Mendelson III, J.R., Porton, I.J., Ralls, K., Ryder, O.A., 2012. Implications of different
species concepts for conserving biodiversity. Biol. Conserv. 153, 25–31.
Gippoliti, S., Groves, C.P., 2012. “Taxonomic inflation” in the historical context of
mammalogy and conservation. Hystrix It. J. Mammal. 23, 6–9.
Groves, C.P., 2013. The nature of species: a rejoinder to Zachos et al. Mammal. Biol.,
7–9.
Heller, R., Frandsen, P., Lorenzen, E.D., Siegismund, H.R., 2013. Are there really twice
as many bovid species as we thought? Syst. Biol. 62, 490–493.
Heller, R., Frandsen, P., Lorenzen, E.D., Siegismund, H.R., 2014. Is diagnosability an
indicator of speciation? Response to “Why one century of phenetics is enough”.
Syst. Biol. 63, 833–837.
Isaac, J.B., Mallet, J., Mace, G.M., 2004. Taxonomic inflation: its influence on macroe-
cology and conservation. Trends Ecol. Evol. 19, 464–469.
Krell, F.-T., Cranston, P.S., 2004. Which side of the tree is more basal? Syst. Entomol.
29, 279–281.
Lovejoy, A.O., 1936. The Great Chain of Being: A Study of the History of an Idea.
Harvard University Press, Cambridge, MA.
Marris, E., 2007. The species and the specious. Nature 446, 250–253.
Omland, K.E., Cook, L.G., Crisp, M.D., 2008. Tree thinking for all biology: the problem
with reading phylogenies as ladders of progress. BioEssays 30, 854–867.
188 Editorial / Mammalian Biology 81 (2016) 185–188
Rigato, E., Minelli, A., 2013. The great chain of being is still there. Evol. Educ. Outreach
6, 18.
Zachos, F.E., 2015. Taxonomic inflation, the Phylogenetic Species Concept and lin-
eages in the Tree of Life – a cautionary comment on species splitting. J. Zool.
Syst. Evol. Res. 53, 180–184.
Zachos, F.E., 2015. Species concepts and species delimitation in mammals. In: Zachos,
F.E., Asher, R. (Eds.), Mammalian Evolution, Diversity and Systematics. Hand-
book of Zoology, Mammalia Series. De Gruyter, Berlin, ISBN 978-3-11-027590-2
(in press).
Zachos, F.E., Apollonio, M., Bärmann, E.V., Festa-Bianchet, M., Göhlich, U., Habel, J.C.,
Haring, E., Kruckenhauser, L., Lovari, S., McDevitt, A.D., Pertoldi, C., Rössner, G.E.,
Sánchez-Villagra, M.R., Scandura, M., Suchentrunk, F., 2013a. Species inflation
and taxonomic artefacts — a critical comment on recent trends in mammalian
classification. Mammal. Biol. 78, 1–6.
Zachos, F.E., Clutton-Brock, T.H., Festa-Bianchet, M., Lovari, S., Macdonald, D.W.,
Schaller, G.B., 2013b. Species splitting puts conservation at risk. Nature 494,
35.