Content uploaded by Linda K. Medlin
Author content
All content in this area was uploaded by Linda K. Medlin on Apr 09, 2018
Content may be subject to copyright.
1
© 2015 J. Cramer in Gebr. Borntraeger Verlagsbuchhandlung, Stuttgart, www.borntraeger-cramer.de
Germany. DOI: 10.1127/nova_hedwigia/2015/0295 0029-5035/2015/0295 $ 5.50
Nova Hedwigia
published online September 2015 PrePub Article
C
Opinion: Can coalescent models explain deep divergences
in the diatoms and argue for the acceptance of paraphyletic
taxa at all taxonomic hierarchies?
Linda K. Medlin
Marine Biological Association of the UK, the Citadel, Plymouth PL1 2PB, UK
e-mail: lkm@mba.ac.uk
With 4 gures and 3 tables
Abstract: Although ancestral polymorphisms and incomplete lineage sorting are commonly used
at the population level, increasing reports of these models have been invoked and tested to explain
deep radiations. Hypotheses are put forward for ancestral polymorphisms being the likely reason for
paraphyletic taxa at the class level in the diatoms based on an ancient rapid radiation of the entire
groups. Models for ancestral deep coalescence are invoked to explain paraphyly and molecular
evolution at the class level in the diatoms. Other examples at more recent divergences are also
documented. Discussion as to whether or not paraphyletic groups seen in the diatoms at all taxonomic
levels should be recognized is provided. The continued use of the terms centric and pennate diatoms
is substantiated with additional evidence produced to support their use in diatoms both as descriptive
terms for both groups and as taxonomic groups for the latter because new morphological evidence
from the auxospores justifies the formal classification of the basal and core araphids as new subclasses
of pennate diatoms in the Class Bacillariophyceae. Keys for higher levels of the diatoms showing
how the terms centrics and araphid diatoms can be defined are provided.
Key words: Araphids, coalescence, Diatoms, phylogeny, population markers, systematics.
Introduction
Within a decade of the discovery of PCR, broad-brush phylogenetic algal surveys
appeared for the macrophytic green, red, and brown algae, the unicellular diatoms,
dinoflagellates and haptophytes (see references for all groups in Medlin et al. (2007).
Four new microalgal classes were also discovered/formally recognized – first with
molecular data and then supported by morphological data: the bolidophytes, the
pelagophytes, the prasinophytes, and the pinguinophytes (Guillou et al. 1999, Andersen
et al. 1993, Moestrup, 1991, Kawachi et al. 2002, respectively). These phylogenies
were soon followed by a flurry of surveys at all taxonomic levels within all of the major
algal divisions. With this came the discovery of many paraphyletic and cryptic taxa in
all groups and with that, the conflict between morphological and molecular data arose
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
2
in all groups. This has had a profound effect on classification systems because they
should ideally reflect evolutionary history of identity by descent through homology.
In some cases, where molecular data have revealed cryptic species and paraphyly,
taxonomic changes at various levels have been made (Marin & Melkonian 1999, 2010,
Daugjberg et al. 2000, Edvardsen et al. 2000, 2012, Medlin & Kaczmarska 2004,
Yoon et al. 2006, Kooistra et al. 2008), but "in most cases, the paraphyletic lineages
have remained, with investigators either willing to live with non-monophyletic taxa,
unable to find new characters to define the new monophyletic groups, or unwilling to
go against conventional wisdom that would lead to the demise of long-standing taxa"
(Medlin et al. 2007).
History of phylogenetic studies in the diatoms
Among the diatoms, this debate has been particularly discordant. Medlin & Kaczmarska
(2004) described/revised two new classes of centric diatoms and revised the pennate
diatom class. Using slightly different analytical methods and fewer outgroups, Theriot et
al. (2009, 2010) could, at first, not recover the same classes as in Medlin & Kaczmarska
(2004) and instead recovered a grade of centric diatom clades with a monophyletic
pennate diatom clade. This sparked off an intense debate among the diatomists as
to whether paraphyletic clades should be recognized taxonomically (Williams &
Kociolek 2007, Medlin 2010). This was based on the assumption that the grade of
centric clades obtained by Theriot et al. (2009, 2010) was the correct scenario and
the monophyletic classes obtained by Medlin & Kaczmarska (2004) were not. Later,
Theriot and co-workers (Ashworth et al. 2012) recovered a monophyletic bipolar
centric, Class Mediophyceae, as described by Medlin and Kaczmarska (2004) with
three genes but their analyses continued to recover a grade of radial centric diatoms
(Class Coscinodiscophyceae), whereas Medlin and co-workers continue to recover
three monophyletic classes using single genes (Bowler et al. 2008, Medlin 2012, 2014)
or multiple genes (Sato 2008; Medlin 2014) with multiple outgroups. Most recently,
using a different taxon sampling, the two centric classes have been recovered with
multiple genes and bolidomonads as outgroups (Li et al. submitted). The addition of
more araphid taxa (ca 50% of the taxa used) has undoubtedly influenced this result.
Medlin (2014) has extensively tested the use of a single gene with multiple outgroups
and has shown that part of the problem with the paraphyletic classes is that not enough
distant outgroups have been used (Fig. 1), which Medlin & Kaczmarska (2004) pro-
posed a decade ago.
Williams & Kociolek (2010) extended the discussion as to whether or not paraphyletic
groups should be recognized in the diatoms to question further the use of the terms
centric and pennate when referring to major categories (not taxonomic groups) of
diatoms. Williams & Kociolek stated "we cannot continue to teach diatoms using the
terms centric and pennates". They claim that the term centric diatom is redundant and
serves no purpose. However, the term centric implies a diatom with oogamous sexual
reproduction and radial areolation from an annular process, be that diatom, a radial
or a bipolar centric (Medlin 2009b).
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
3
Morphological support for the diatom classes
The type of auxospore divides the centrics into radial or bipolar ones (Medlin &
Kaczmarska 2004, Medlin 2009b). The only contradictory part of this explanation
is that the radial Thalassiosirales are phylogenetically placed within bipolar centrics,
making the radial centrics, as defined by valve morphology alone, paraphyletic. It is
likely that the thalassiosiralean lineage retains the original gene(s)/operons to make
auxospores only with scales, and have lost the ability to make the bands characteristic
of the bipolar centrics. The Thalassiosirales are usually a terminal divergence in
phylogenetic trees, implying a loss of bands, which are found in taxa that are basal in
the lineage (Kaczmarska et al. 2006, Alverson et al. 2007). Theriot et al. (2009) used
auxospore characters from Medlin and Kaczmarska (2004) in a cladistic analysis and
showed that the Thalassiosirales grouped with the other radial centrics and discounted
the use of auxospore characters to define the three diatom classes.
However, the Thalassiosirales possess other morphological features outside the
type of auxospore that clearly place them inside the bipolar centric lineage, such
Fig. 1. Phylogenetic tree redrawn from Medlin (2014) showing the effect of a single outgroup
(Bolidophyta) (A) as compared to multiple outgroups from three crown groups taxa plus non-pigment
and pigmented heterokonts (B). Bootstrap values support the monophyly of the three classes.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
4
as the perinuclear Golgi arrangement, the possession of a process (tube) inside the
valve annulus and the internal placement of the cribrum in the loculate (chambered)
areolae. A cladistic analysis using these features recovers the Thalassiosirales in the
Mediophyceae (Medlin 2014). There are a few exceptions to the latter character in
the Mediophyceae but most loculate areolae have an internal siliceous membrane,
the cribrum, whereas in the Coscinodiscophyceae, all genera with loculate areolae
have external cribra. Coscinodiscoid genera with pseudoloculate areolae (Endictya
Ehrenberg and Stephanopyxis Ehrenberg) have internal cribra. Pseudoloculate and
loculate areolae are not considered to be homologous structures (Anonymous 1975).
Also in the Coscinodiscophyceae, in those genera that appear to have central processes,
their tubes are not formed or located in the same manner as the central tubes in the
Mediophyceae. For example, in the Coscinodiscophyceae, the central labiate processes
of Azpeitia M.Peragallo and Stellarima Hasle & Sims are not inside the annulus
(resting spores excepted) and the central labiate process of Rhizosolenia T.Brightwell
begins on the margin and migrates to a central location during valve morphogenesis
(van de Meene & Pickett-Heaps 2007). Sims & Hasle (1987) proposed that the central
labiate processes of Stellarima may first have been at the margins and later migrated
to the centre implicating a pre-coscinodiscoid ancestor. Many mediophycean bipolar
centrics have processes inside an elongated annulus, which produces bipolar valves
with processes in polar positions, e.g. Biddulphiopsis von Stosch & Simonsen and
Odontella C.Agardh, rather than in a central position but the processes are still inside
the annulus, which is one of the defining characters for the class. Toxarium Bailey
and its relatives have an elongated, distorted elliptical annulus as evidenced by the
non-perforated area along the junction of the valve face and mantle and it is enfilled
with parallel siliceous ribs (Kooistra et al. 2003, Medlin et al. 2008). The new araphid
genus Astrosyne Ashworth & Lobban is a circular diatom, which appears to be another
taxon that has lost the ability to form bands, and thus produces a round morphology
with areolae radiating from its annulus but its molecular signature places it in the
araphid lineage (Ashworth et al. 2012) further supporting that the ancestral state of
the diatom auxospore is circular in shape with only scales to cover the cell. A cladistic
analysis using valve shape, absence of a process inside the annulus, areolae orientation
and marginal processes would place Astrosyne in the radial centrics. Its placement
in the pennate lineage assumes that it does not have oogamous sexual reproduction
(Kaczmarska et al. 2013), although it could have a larger female gamete (egg) and
non-flagellated possibly amoeboid male cells (spermatium) as in Rhabdonema, but it
cannot be referred to as a centric diatom despite its valve symmetry.
Ancestral polymorphisms explain paraphyly
The fact that these two distantly taxa share a valve structure like the radial centrics is
most likely the result of retention of an ancestral feature or polymorphism (Fig. 2), viz.,
the possession of scales or another organic components of the incunabula (Kaczmarska
et al. 2013) on the auxospore as the only structures protecting the zygote and the
incomplete lineage sorting of this feature.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
5
Fig. 2 is used to illustrate that, based on an assumption that scales and bands are the
expression of a single genes/operons, one can use a coalescence model to illustrate how
taxa deep in one lineage can share genetic/morphological features with taxa in another
lineage (= paraphyly), assuming that the morphological features are homologous. I
am using the morphological features as if they are controlled by genes whose allelic
distribution can be analyzed by coalescent models. Thus, for illustrative purposes only,
the gene lineages, shown in blue in Fig. 2: possess only auxospore scales, controlled
either by a single loci or multiple ones, which are present in all three of the modern
diatom lineages. This feature is the ancestral polymorphism.
The lineages shown in red contain the new feature (mutations) of bands added to the
auxospore. It only takes the loss of this feature or the absence of expression of this gene/
operon to resort to a radial morphology in the vegetative cell because the only gene
then present or expressed is that of the scales. The retention of ancestral polymorphism
of scales on the auxospore, regardless of how many genes/operons control this feature,
continues into the evolution of the bipolar centrics and pennate diatoms because nearly
all diatoms have scales or some organic incunabular components on the zygote. The
bipolar centrics have scales initially but add properizonial bands and other incunabular
components that squeeze the zygote into their bi- or multi-polar shape. It is these bands
that likely have been lost in the Thalassiosirales in the Mediophyceae and Astrosyne
in the Bacillariophyceae to form radial vegetative cells. As the time span between
divergences of the radial and bipolar centric diatoms increases, the chance also increases
that one of the alleles might be lost before the second splitting event occurs (Whitfield
& Lockhart 2007) but even after millions of years have passed and the alleles have still
not sorted between the lineages and the polymorphisms are retained, then we must
assume that the ancient polymorphism infers some adaptive advantage to the cell, such
as a protective covering over the developing zygote in its early stages in the diatoms.
Fig. 2. Schematic diagram showing the evolutionary trajectory of individuals possessing different
auxospore characters. It is assumed that the suxospore characters used here for illustration purposes
are behaving as single alleles. The ancestral polymorphism are the alleles shown in blue, which are
the scale bearing auxospore genes. A second allele, shown in red, represents bands that are added
to the auxospore but if this allele is lost, then the individual reverts back to auxospores only with
scales, the ancestral polymorphism. Modified from an original figure drawn by Dr. Frederik Leliaert.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
6
Ancestral polymorphisms and incomplete lineage sorting in deep phylogenies
Ancestral polymorphisms and incomplete lineage sorting are used in coalescent theory
at the population level to explain anomalies in gene trees at the species level in recent
time and at very shallow timescales. (Nordberg 2000, Degnan & Rosenberg 2009,
Sánchez-Gracia & Castresana 2012, Fujita et al. 2012, Leliaert et al. 2014). In species
trees, we are taking samples from the population rather than showing the genealogy
of the entire population (Nordberg 2000). The genealogy of a group of individuals
(samples from the population) may be generated by simply following lineages back in
time, generation by generation, keeping track of coalescences between lineages, until
eventually the most recent common ancestor (MRCA) is found (Nordberg 2000), even
though the MRCA in the diatom examples shown in Fig. 2 and Fig. 3A is hundreds of
millions of years ago rather than thousands of generations. Thus, incomplete lineage
sorting and ancestral polymorphisms can also occur in deep phylogenies (Degnan &
Rosenberg 2009). Fujita et al. (2012) comment that it might seem somewhat paradoxical
that these methods interpret genealogical discordance as deep coalescence, given that
gene exchange among populations and/or species is more probable among recently
diverged species.
Data suggest that coalescent models will find accurate phylogenies for both shallow
and deep divergences (Lin et al. 2011, 2012, Than & Rosenberg 2011, 2014), whereas
concatenated analyses may not and yield lower bootstrap support values (Kubatko
& Degnan 2007; Edwards, 2009, Liu et al. 2010, Kumar et al. 2013, Oliver, 2012,
Song et al. 2012). Table 1 shows a selection of studies using coalescent models with
deep divergences and the taxa whose phylogenetic discordances have been resolved
with these methods. Edwards (2009) suggests that deep coalescences can be found
in all groups and with all genes and is the major cause of phylogenetic discordances
especially in large population sizes, as one would expect to find in unicellular organisms.
Lin et al. (2011) specifically tested for the effect of population sizes in Oliver (2012)
proposes that the potential for discordance between gene trees and species histories
caused by incomplete lineage sorting will increase as taxon sampling increases, so the
problems with incomplete lineage sorting is likely to become more evident as more
diatom species/genera are added to the data set because branch lengths separating
taxa become shorter.
Thus, lineages can sort so that monophyly of lineages is violated for a [taxon] deep
in the tree (Avise 2000, Whitfield & Lockhart 2007). Short branches, possibly only
one, deep in the tree are all that are needed for incomplete lineage sorting in deep
phylogenies, such as we have in the diatom tree (Fig. 2) where we have an ancient rapid
radiation (Whitfield & Lockhart 2007, Medlin 2009a, 2015). Degnan & Rosenberg
(2009) have identified several ancient rapid radiations that are good candidates for
incomplete lineage sorting. In the latest molecular clocks produced by Medlin (2009a,
2015), there is only an 8 My gap between the divergences of the three diatom classes.
In Fig. 1 the tree to the left is made from only 2 bolidomonads outgroups and recovers
paraphyletic classes, whereas the tree to the right has been made multiple outgroups
(ciliates, haptophytes, chlorophytes heterokonts) and recovers monophyletic classes.
The branch lengths separating these clades are relatively short and without the proper
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
7
outgroups to recover their monophyly (Fig. 1), the conclusion that will be reached in
the analysis is that the classes are not monophyletic. The same argument of incomplete
lineage sorting has been made to explain the deep radiation and evolution of the land
plants (Zhong et al. 2013) and mammals (Sánchez-Gracia & Castresana 2012, Kumar et
al. 2013). According to Nordberg (2000), "no matter how many individuals we sample,
there is still only a single underlying genealogy to estimate", whether that genealogy
is reflecting shallow species level anomalies or deep divergences and radiations. This
opinion about invoking deep coalescent population models to infer molecular evolution
in the diatoms is something that can be tested mathematically as more genes are added
to the diatom molecular tool kit. Presently we do not have enough molecular data in
Fig. 3. Paths of Evolution in the Diatoms. Modified from Leaché et al. (2009) with permission from
PNAS. A. Schematic diagram showing a imaginary evolutionary time scale on the horizontal line
and the resulting phylogenetic results of the monophyletic, paraphyletic and polyphyletic predictions
of the diatoms. At the evolutionary time marked polyphyletic, likely the pennates and centrics
would be mixed in any tree recovered (see Medlin 2014) for examples of this type of analysis).
B. Schematic diagram showing a imaginary evolutionary time scale on the horizontal line and the
resulting phylogenetic results of new genera separated from Fragilaria if sampled at three different
time points. C. Schematic diagram showing a imaginary evolutionary time scale on the horizontal
line and the resulting phylogenetic results of Fragilariopsis and Pseudo-nitzschia if sampled at three
different time points. X marks the lineage that must go extinct for Fragilariopsis and Pseudo-nitzschia
to become monophyletic genera.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
8
the diatoms to test these hypotheses empirically (Table 1). Edwards (2009) and Lin et
al. (2011, 2012) can be consulted to find a range of programs that would be suitable
for testing these hypotheses on deep coalescence in the diatoms.
The time frame for coalescent theory using shallow points in the tree is from now to ca.
1,000,000 generations ago, which, for Protistan and other microbial organisms, would
be in the range of 3000 years if those organisms divide once a day as most diatoms do.
From each mitotic division, there is one diatom cell that remains the same size as the
mother cell and one that is smaller. It is the smaller cells that move into the reproductive
window. There is always a population whose size never changes. This population has
the possibility to survive indefinitely. When you isolate cells for molecular analysis,
you have no idea which population you have isolated the cell from unless thousands
of cells are measured, so the time frame above is relevant. Coalescent theory is dealing
with the entire population, not just the population that is going through the sexual
window. Many diatoms are known to undergo vegetative cell enlargement and thus
avoid sexual reproduction and others are known to be completely apomictic and how
these species could be analyzed with coalescent models is unknown.
Table 1. Summary of studies using colalescence models to resolve deep divergencies and phylogenetic
incongruencies.
Reference Groups studied Number of
genes/trees
analysed
Phylogeny Resolved by
Coalescence Models Age
potential for discordances
Oliver 2013 Angiosperms, birds,
harpaline beetles, mammals,
and nymphalid butterflies
100 Determined deep coalescence
resolve
discordances between 80–90 My
Kumar et al.
2012 Super clade Euarchontoglires
(Mammals) 19 whole
genomes
(5,875 genes)
Placement of Scandentia, few
million years rapid radiation
Sánchez-Gracia
& Castresana
2012
Mammals Up to 50 Studied in time intervals from
0.5 to 10 my, fast radiations
could not be resolved with low
numbers of genes
Zong et al. 2012 Land plants 185 Zynematales is closest relative
to land plants
Kubatko &
Degnan 2007 Simulations 10 gene trees Coalescence models
outperformed concatenated
datasets because of incomplete
lineage sorting
Liu et al. 2010 Simulations from mammal
data set Up to 2500 New program developed,
coalescence models outperformed
concatenated datasets
Lin et al. 2011 Simulations 20 gene trees
50 256-taxon
species trees
Tested for effects of different
population sizes and for the
Pareto property in tree searches
Than &
Rosenberg 2014 Simulations 46 gene trees Tested for effects of balanced vs.
Unbalanced trees
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
9
Table 2. Summary of characters used to define the classes and subclasses of diatoms. Important review articles/books illustrating the variation across
the taxa are noted where available.
Taxon Valve
shape Location of Tube
Processes 1Location
of cribrum
in loculate
areolae1
Golgi
Arrangement2Sexual
Reproduction3Gametangia/Sexual
Cells3Auxospore
Covering3
Class
Coscinodiscophyceae radial Marginal, if cen-
tral, then moved
from margin to the
center
External, no
exceptions G-ER-M unit
reported in
many taxa,
2 exceptions
Oogamous, no
exception Sperm released from
gametangia, egg may or
may not be released
Scales
Class Mediophyceae Bipolar +
radial Central, inside
annulus, +/-
marginal ones
Internal,
a few
exceptions
Perinuclear, in
all taxa studied,
1 exception
Oogamous, no
exception Sperm released from
gametangia, egg may or
may not be released
Scales +
properizonia
Class
Bacillariophyceae
Sub class
Urneidophycidae
bipolar Sternum (com-
pressed and elon-
gated annulus)
without raphe,
processes usually
outside sternum
n/a Perinuclear, in
all taxa studied Anisogamous, no
excpetions where
studied
Male and/or female
sex cells released from
gametangia, male sex
cell with non-motile but
contractile appendage
Scales +
properizonia
+ perizonia
Class
Bacillariophyceae
Sub class
Fragilariophycidae
bipolar Sternum (com-
pressed and elon-
gated annulus)
without raphe,
processes usually
outside sternum
n/a Perinuclear in
all taxa studied Anisogamous,
no excpetions
where studied
Male and/or female
sex cells released from
gametangia, male sex
cell with non-motile but
contractile appendage
Scales +
perizonia
Class
Bacillariophyceae
Sub class
Bacillariophycidae
bipolar Sternum
(compressed and
elongated annulus)
with raphe,
processes outside
sternum
n/a Perinuclear, in
all taxa studied
some very
elaborate
Isogamous, or
physiological
anisogamous,
heterothallic or
homothallic
Male and female sex
cells not released from
gametangia, conjugation
tubes/envelopes common
Scales +
perizonia
1 Round et al. 1990, Sims & Hasle 1987, 2 Schmid 2001, 3 Kaczmarska et al. 2013
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
10
It appears that in the diatoms and likely in many other unicellular lineages, ancestral
polymorphisms and incomplete lineage sorting can be traced beyond the species level
and herein the examples shown in Fig. 2: ancestral polymorphisms can be hypothesized
to explain disparities at the class level in the diatoms because of rapid radiations deep
in their phylogenies. It is likely that the rapid turnover, the large population sizes
and generation time of microscopic organisms likely contributes to the extension
of these processes beyond the population level and may be one of the contributing
factors leading to the assumption that everything is everywhere, in which similar
morphological features are present in various clades of microbial organisms that are
molecularly distinct. Even more evidence is accumulating among Protistan and bacterial
taxa that there are real barriers to dispersal in the oceans so that everything cannot be
everywhere (Foissner 2006, Sul et al. 2013) and incomplete lineage sorting and ancestral
polymorphisms are likely contributing to this confusion. The suggestions proposed
here can be tested with multiple gene data sets and could be the object of future work
because in the diatoms, phylogenies from multiple genes are rapidly accumulating.
Paraphyly of the araphid diatoms
Returning to the evolution of the diatoms, the araphid diatoms are routinely recovered
in two groups, the basal and the core araphids (Table 3). The basal araphid lineage
possesses the properizonial bands found exclusively in the bipolar centrics plus the
perizonial bands found exclusively in the core araphid and raphid groups. The core
araphids only contain the latter arrangement of bands. Kaczmarska et al. (2013) have
recommended that both the properizonial and perizonial bands should be referred to
under the same name: perizonia. Although the structure of the bands may be the same,
their arrangement is not. Hence the two types of bands have been retained here because
their arrangement is different in the two groups and thus phylogenetically informative.
The properizonial bands are complete rings of silica strips that overlap each other like
a telescope and appear saddle shaped (Fig. 4 and fig. 65 in Round et al. 1990). Because
of this complex arrangement of the properizonial bands, one side of the auxospore is
free of bands, leading to a clear area of varying width along that side of the auxospore
free of any band protection. The perizonial bands are two systems of split ring/strips
of silica that run both longitudinally and transversely around the auxospore, with the
longitudinal bands usually lying under the transverse ones (Fig. 4).
The araphid diatoms have traditionally been defined by the absence of a feature:
the raphe (Round et al. 1990). The newly discovered feature of the appendage with
microtubules on the araphid male sex cell that draws the sex cells together in araphid
Pseudostaurosia trainorii (Sato et al. 2011) could be considered the defining feature of
an araphid diatom. This feature has also been found in other core araphid taxa: Tabularia
fasciculata, T. tabulata (Davidovich et al. 2012) and in Ulnaria ulna (Davidovich
2012). Amoeboid gametes are known also from Grammatophora and Rhabdonema
(Stosch 1958, Magne-Simon 1962). If one assumes that this character, now found in
four genera, two of which are very common araphid genera, and presumed to be present
in two other common genera with amoeboid gametes, is found in all araphids, then
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
11
Table 3. Summary of the results from major studies on the evolution of the diatoms and whether or not the classes were recovered as monophyletic (M)
and if not how many clades could be assigned to each class for the centric diatoms. For the araphid diatoms, the recovery of basal and core araphids
is indicated. Numbers in brackets refer to the highest bootstrap or posterior probability support for the clade, even if multiple types of analyses were
conducted. The number of outgroups used in each study is also indicated because this will have an effect on the monophyly of the clades (Medlin 2014).
Source Coscinodisco-
phyceae Mediophyceae Bacillario-
phyceae Basal + Core araphids No. of outgroups
Alverson et al. 2006, fig. 3 M4Myes 2 bolidomonads
Alverson et al. 2006, fig. 4 M3Myes 2 bolidomonads
Alverson et al. 2006, fig. 5 2 plus Ellerbeckia 2MBasal plus 4 core Multiple
Alverson et al. 2006, fig. 6 3 plus Ellerbeckia 1+5 clade polytomy MBasal plus 5 core Multiple
Ashworth et al. 2012, fig. 48 3 M M 1 bolidomonad
Cavalier-Smith & Chao 2006 2 MM 2 basal plus one core Multiple
Choi et al. 2008 M3MMultiple
Ehara et al. 2000 M M M Single
Lee et al. 2013 2 2 n/a One Pennate
Kooistra & Medlin (1996) M2–3 clades MBasal and core Multiple
Kooistra et al. (2004) M3MBasal and core 4 radial centrics
Medlin & Kaczmarska 2004, fig. 1 M (63) M (98) M (100) Not shown Multiple
Medlin & Kaczmarska 2004, fig. 3 4 4 M (100) Mulitple bollidomonads
Medlin et al. 1993, fig. 3 M (68) M (95) M (100) yes (53+93) Multiple
Medlin et al. 1996a M (91) 3 M (100) yes (93+100) Single
Medlin et al. 1996b M3Myes (100+3 clades) Multiple
Medlin et al. 2000 M plus Ellerbeckia 4M (91) yes (90+ <50) Single
Medlin et al. 2008 M (100) M (100) M (99) yes (100+97)
Sato 2008, fig. 4 publication 10 M ( <50) M (51) M (91) yes (76+ <50) Multiple
Sims et al. 2006 M (100) M (100) M (100) yes (100 +100) Multiple
Sorhannus 2004, fig. 1 3 3 Myes Single
Sorhannus 2004, fig. 2 3, polytomy M M yes
Sorhannus 2004, fig. 3 M4M
Sorhannus 2007 3 plus Ellerbeckia 8Myes Single
Theriot et al. (2009, fig.1) Unweighted MP 2 3 Mone basal multiple core Multiple
Theriot et al. (2009, fig. 2) 3 M M one basal multiple core Single
Theriot et al. (2009, fig. 8) 3 M (52) + Attheya M (100) Basal and core (>95+95) Rooted midway
Coscinodiscophytes
Theriot et al. (2010, fig. 3) 2 4 M (100) basal; core Single
Li et al. unpubl. M ( 100) M (71) M (100) yes (76+<50) Multiple
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
12
Fig. 4. Summary of the phylogeny of the diatoms using the auxospore and the sperm cells as the character defining the lower branches in the tree.
Redrawn from Medlin & Kaczmarska (2004) and Medlin & Sato (2009).
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
13
araphids could be defined by the presence of this feature, viz., amoeboid spermatia
with an appendage that is not a flagellum. Most araphid diatoms also release their
gametes from the gametangia. Known exceptions to this are Grammatophora and
Rhabdonema of the family Tabellariaceae/Rhabdonemataceae; they retain the egg
within their gametangia, albeit in Grammatophora, the egg is in an open theca and
within an entire gametangium in Rhabdonema. Amoeboid sperm are released in close
proximity to the egg (Stosch 1958, Magne-Simon 1962).
If the appendage were found on the spermatia of most araphid genera then the release
of one or both gametes from the gametangia, would satisfy Williams and Kociolek’s
recommendation not to use the absence of a feature (viz., the raphe) as the defining
feature of this group. Araphid diatoms could thus be defined as those diatoms possessing
only a sternum, having an auxospore with a combination of properizonial and perizonial
bands (basal araphid) or only perizonial bands (core araphids), releasing one or both of
their sex cells from their gametagania and male sex cells consisting motile amoeboid
gametes with hair like appendages that facilitate the amoeboid movement of the
male spermatia to the female sex cell if the female sex cell is also released from a
gametangium. The cells change in shape from globular to angular when the appendage
is produced (Davidovich 2012). Sexual reproduction in the basal araphid taxa is needed
to verify if they also possess this feature on male sex cells.
The two groups (basal and core) are each strongly supported monophyletic clades
in nearly all analyses with multiple genes (monophyletic groups in Sato 2008,
Ashworth et al. 2012, Medlin et al. 2012, grades in Theriot et al. 2010 with the same
three genes as Ashworth et al. 2012 but with fewer taxa, Li et al. submitted, Table 3)
and will be formally described below at the subclass level so that within the Class
Bacillariophyceae, there will now be three subclasses: basal araphid, core araphid,
and raphid diatoms.
Thus, the terms centric and araphid diatoms can be used in a descriptive context to
convey an image of a diatom and the araphids (further defined as core or basal) can be
used taxonomically, whereas centrics cannot but radial or bipolar centrics can. Centric
and pennates have different types of gametogenesis (see diagrams in Kaczmarska et
al. (2013) and Fig. 4).
Paraphyly in the diatoms at lower taxonomic levels
It would also seem at both a higher clade level and at the species level, there are
excellent examples of paraphyly among the diatoms. Incomplete lineage sorting is
associated with the persistence of ancestral alleles in recently diverging lineages, gene
trees will inevitably go through phases where individuals of a species are polyphyletic
and paraphyletic before becoming monophyletic as alleles become fixed through time
(Leliaert et al. 2014). The fact that taxa have to go through polyphyly to paraphyly to
reach a monophyletic state has been clearly discussed and illustrated in Avise (2000)
and is an accepted trajectory of evolution by molecular evolutionary biologists. Thus,
depending on when in the course of evolution the taxa and traits are sampled, one will
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
14
recover a polyphyletic, a paraphyletic or a monophyletic state. In Fig. 3A, three time
points are drawn across the diatom tree to illustrate when each of these states would be
sampled at the class level. A possible scenario of having polyphyletic diatoms where
centric and pennates would be highly mixed would be a sampling very early in their
evolution (Fig. 3A). Extensive re-descriptions of taxa are not needed if paraphyletic
taxa are accepted, some paraphyletic taxa can be well defined using both unique and
shared traits, e.g. Coscinodiscophyceae and Mediophyceae (as define above and see
Table 3 and keys below), whereas others are truly cryptic and can only be distinguished
by molecular methods, e.g. Skeletonema tropicum and S. menzelii (Kooistra et al. 2008).
In instances where the molecular data shows a monophyletic genus arising from within
another genus, such as Fragilariopsis from within Pseudo-nitzschia (Lundholm et al.
2002), Asterionella from within Diatoma (Medlin et al. 2008b), Didymosphenia from
within Cymbella (Nakov & Theriot 2010), Geissleria from within Placoneis (Nakov
et al. 2014), the parent genus becomes paraphyletic. It is easy to place these genera
onto the tree to understand how these paraphyletic taxa have arisen and which branch
would have to go extinct over time for each paraphyletic genus to become eventually
monophyletic (Fig. 3B). Each of these genera making the parent genus paraphyletic
represents a good morphological genus in terms of having unique characters to define
it, which would have to be rejected for the sake of insisting on monophyletic taxa. It
is noteworthy that in none of these phylogenetic analyses above have the researchers
recommended the demise of any of these genera and only Nakov et al. (2014)
acknowledge the problem but only for some and not all of the genera in their study.
In fact to the contrary, new taxa have been described when they render older genera
paraphyletic/polyphyletic by the same workers who insist that paraphyletic/polyphyletic
taxa should not be recognized (see Shinodiscus Alverson, Kang & Theriot in Alverson
et al. 2007). Polyphyletic taxa are more problematic to accept because there is no
consistent morphology that can be used to define any of the clades. The best example
of this stage of evolution in the diatoms is the new genera separated from Fragilaria,
which are polyphyletic (Fig. 3C, Medlin et al. 2008a, b, 2011). Yet, despite it being
polyphyletic, diatomists still use and retain the genus Thalassiosira.
Major evolutionary steps in the diatoms
A summary of the major evolutionary steps in the evolution of the diatoms using
features of sexual reproduction is shown in Fig. 4. Beginning with the UR diatom/
bolidomonad, a cell, with scales as a protective coating, capitalizes on the ability to
metabolize silica, which is exported from the cell as a waste product in the form of
precipitated silica scales, see Medlin (2002) for a fuller explanation of how the diatoms
could possibly have evolved the use of silica. This scenario was published before the
placement of the Parmales in the Bolidophyceae (Ichinomiya et al. 2011). The scales
evolve into the valves but are retained at the zygote stage to protect this cell. Sexual
reproduction is by oogamy. The first major divergence is the addition of bands (B+C),
which separates the bipolar centric and pennate lineages (B+C) from the radial centric
lineage (A). The B + C lineage diverge into two lineages with the B lineage retaining
oogamy and the properizonial bands (B) and the B+C lineage with both perizonial (C)
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
15
and properpizonial bands (B), loosing oogamy but gaining anisogamy with the sperm
changing into spermatia with an appendage that is not a flagellum (this assumes that
all araphids have this new appendage). The B+C lineage next diverges into the B+C
lineage, which retains both B+C types of bands (properizonial and perizonial bands,
releasing sex cells from the gametangia and the spermatia with an appendage to pull
the spermatia to the egg and the C lineage, which retains only perizonial bands and
have spermatia with an appendage or have gametes that are either physiologically
anisogametic or isogametic. The final divergence in the C lineage is into a lineage (C)
with a slit in the valve for movement, all sex cells retained within gametangia and true
anisogamy of the gametes and into a lineage (C') without this slit with one or both
types of sex cells being released from the gametangia and with spermatia that retain
appendages. The accepted systematics of the cells is illustrated at the top of the figure
for each of the terminal groups (A, B, B+C, C' and C). The evolution of sperm and
spermatia are indicated with arrows on the figure.
As a final point as to whether or not the terms centric and araphid can be used, they can
be used as a collective noun. Algae, as a collective noun, can be defined as photosynthetic
organisms whose tissues are not differentiated into roots, stems and leaves (loss of
photosynthesis is a secondary feature), although they are molecularly polyphyletic. This
definition precludes the inclusion of rabbits, ducks, echinoderms, fishes, bacteria… and
any other taxa related to these organisms whose ancestral state was not photosynthetic
(see discussion in Williams & Kociolek 2009). Groups of convenience, even bacteria,
are used throughout science and centric and pennate, raphid and araphid diatoms should
not be an exception. They all convey an image of a diatom and there is neither rhyme nor
reason for abandoning them or any evolutionary reason for not recognizing paraphyletic
taxa, they are only a stage in the evolution of any group and the relationships recovered
by any phylogenetic analysis are only a snapshot in time.
Taxonomic keys
For the teaching of diatom classification, the following keys are offered to separate
higher groups in the diatom classification of Medlin & Kaczmarska (2004), using in
key 1: features of the auxospore and the valve and in key 2: features of the whole cell
or cleaned valves. Using the latter key, the major groups of diatoms can be identified
and separated and using the former key (based on the less commonly observed trait,
the auxospores), an overview of the evolution of the classes can be taught and better
understood, and basal and core araphids can be separated.
Key 1: Key to higher level groups in the diatoms based on auxospore
and some morphological valve features
1 a. Auxospores contain only scales ..................................................................................................4
b. Auxospores contain scales and bands .........................................................................................2
2 a. Auxospores contain only properizonial bands .......................Bacillariophytina (Mediophyceae)
b. Auxospores otherwise .................................................................................................................3
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
16
3 a. Auxospores contain all perizonial bands ....................................................................................5
b. Auxospores contain both perizonial and properizonial bands ......................................................
Bacillariophytina (Urneidophycidae = basal araphid diatoms)
4 a. Radial valves with a central process ...................................... Bacillariophytina (Mediophyceae)
b. Radial valves without a central process ................Coscinodiscophytina (Coscinodiscophyceae)
5 a. Male gametes with hair like appendage........................................................................................
Bacillariophytina, Fragilariophycidae = core araphids diatoms
b. Male gametes motile without any appendage ...............................................................................
Bacillariophytina Bacillariophycidae = raphid diatoms)
Key 2: Key to higher level groups in the diatoms based on morphological features of whole
cells or cleaned valves (Fig. 4)
1 a. Pervalvar axis 5 or more times greater in length than the apical axis ..........................................
Coscinodiscophytina Aulacosirales, Rhizosoleniales, Leptocylindrales, Corethrales)
b. Valves otherwise .........................................................................................................................2
2 a. Valves radial................................................................................................................................3
b. Valves .........................................................................................................................................4
3 a. Areolation radial, no central process, external cribra in loculate areolae, internal cribra in pseudo-
loculate areolae with eccentric areolation ....................................................................................
Coscinodiscophytina (Coscinodiscophyceae, all remaining orders with a narrow pervalvar axis)
b. Areolation eccentric, central process present, internal cribra in loculate areolae ........................
Bacillariophytina (Mediophyceae, Thalassiosirales)
4 a. Valves bipolar with radial areolation .....................................Bacillariophytina (Mediophyceae)
b. Valves bipolar with bilateral areolation ......................................................................................5
5 a. Valves with a sternum and raphe on at least one of the two valves ..............................................
Bacillariophytina Bacillariophyceae, raphid diatoms)
b. Valves with a sternum and without a raphe on either valve ..........................................................
Bacillariophytina Bacillariophyceae, basal and core araphid diatoms)
Nomenclatural changes
Three subclasses of diatoms under the Class Bacillariophycideae are defined below.
Urneidophycidae Medlin subclass nov.
Valves with a medial sternum, lacking a raphe, labiate process(es) usually present,
sexual reproduction anisogamous with one or both sex cells released from the
gametagania, with a non-motile female cell and an amoeboid male cell with a hairlike
appendage extruded to attach the female to the male and pull it towards the male cell
for fertilization, auxospore bands composed of properizonial and perizonial bands.
Areolar covering normally rotae. Apical pore fields usually present not delimited from
the valve areolation.
Type genus: Rhaphoneis
Derivation of the Name: a combination of Greek and Latin, meaning original boat,
suggested by Prof. D.G.Mann.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
17
Other genera in this new subclass includes Talaroneis, Delphineis, Asteroplanus,
Asterionellopsis (found by Sato 2008, Medlin and Kaczmarska 2004, Sorhannus 2004)
and Plagiogramma, Neofragillaria, Dimmeregramma, Psammoneis, Pseudostriatella
and Striatella (found by Sato 2008).
Fragilariophycidae (Round) Medlin subclass rev.
Valves with a medial sternum, lacking a raphe, labiate process(es) may be present,
sexual reproduction anisogamous with one or both sex cells released from the
gametagania, when both sex cells are released from the gametangium, there is a non-
motile female cell and an amoeboid male cell with a hairlike appendage extruded to
attach the female to the male and pulling it towards the male cell for fertilization,
auxospore bands composed of perizonial bands. Areolae covering variable. Apical
pore fields may or may not be from the valve areolation by a rim of silica.
Bacillariophycidae (Mann) Medlin subclass rev.
Valves with a sternum and a raphe. A labiate process is present only in one group, the
Eunotiales, who have a reduced raphe system, sexual reproduction isogamous with
non-motile female and male cells, fertilization by conjugation/juxtapositioning/pairing
of the gametangania next to one another within various types of copulation envelopes
and not released from the gametangia, auxospore bands composed of perizonial bands.
Acknowledgements
I would like to thank Dr. Shinya Sato for his many valuable observations on the araphid diatoms and
Dr. Frederik Leliaert for providing me with illustrations for their adaptation into Fig. 1.
References
ALVERSON, A.J., R.K. JANSEN & E. THERIOT 2007: Bridging the Rubicon, Phylogenetic analysis
reveals repeated colonization of marine and fresh waters by thalassiosiroid diatoms. – Mol. Phylo.
Evol. 45: 193–210.
ALVERSON, A.-J., J.-J. CANNONE, R.R. GUTELL & E.C. THERIOT 2006: The evolution of
elongate shape in diatoms. – J. Phycol. 42: 655–68.
ANDERSEN, R.A., G.W. SAUNDERS & M.P. PASKIND 1993: Ultrastructure and 18S rRNA gene
sequence for Pelagomonas calceolata gen. and sp. nov. and the description of a new algal class, the
Pelagophyceae classis nov. – J. Phycol. 29: 701–715.
AnOnyMOus 1975: Proposal for a standardization of diatom terminology and diagnoses. – Nova
Hedw. Beih. 53: 323–354.
ASHWORTH, A., E. RUCK, C. LOBBAN, R. ROMANOVICZ & E. THERIOT 2012: A revision
of the genus Cyclophora and description of Astrosyne gen. nov. Bacillariophyta, two genera with
the pyrenoids contained within pseudosepta. – Phycologia 51: 684–699.
AVISE, J.C. (2000: Phylogeography, The History and Formation of Species. – Harvard University
Press, Cambridge, Massachusetts, viii + 447 pp.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
18
BOWLER, C., A.E. ALLEN, J.H. BADGER, J. GRIMWOOD, K. JABBARI et al. 2008: The
Phaeodactylum genome reveals the dynamic nature and multi–lineage evolutionary history of diatom
genomes. – Nature 456: 239–244.
CAVALIER-SMITH, T. & E.-Y. CHAO 2006: Phylogeny and megasystematics of phagatrophic
heterokonts Kingdom Chromista. – J. Mol. Evol. 62: 388–420.
CHOI, H.G., H.M. JOO, W. JUNG, S.S. HONG, J.S. KANG & S.H. KANG 2008: Morphology and
phylogenetic relationships of some psychrophilic polar diatoms (Bacillariophyta). – Nova Hedwigia
Beih. 133: 7–30.
DAUGJBERG, N., G. HANSEN, J. LARSEN & Ø. MOESTRUP 2000: Phylogeny of some of
the major genera of dinoflagellates based on ultrastructure and partial LSU rDNA sequence data,
including the erection of three new genera of unarmoured dinoflagellates. – Phycologia 39: 302–317.
DAVIDOVICH, N.A. 2012: Pseudopodial activity of the gamete surface in araphid pennate diatoms.
– Poster Abstracts, XXII International Diatom Symposium page 150.
DAVIDOVICH, N.A., I. KACZMARsKA, s.A. KARpOV, O.I. DAVIDOVICH, L. MICHAeL et
al. 2012: Mechanism of male gamete motility in araphid pennate diatoms from the genus Tabularia
(Bacillariophyta). – Protist 163: 480–494.
DEGNAN, J.H. & N.A. ROSENBERG 2009: Gene tree discordance, phylogenetic inference and
the multispecies coalescent. – Trends Ecol. Evol. 24: 332–340.
EDVARDSEN, B., W. eIKReM, J.C. gReen, R.A. AnDeRsen, s.y. MOOn- VA n DeR sTAAy
& L.K. MeDLIn 2011: Ribosomal DNA phylogenies and a morphological revision set the basis for
a new taxonomy of the Prymnesiales (Haptophyta). – Eur. J. Phycol. 46: 202–228.
EDVARDSEN, B., W. eIKReM, J. THROnDsen, A. sAeZ, I. pROBeRT & L.K. MeDLIn 2000:
Phylogenetic reconstructions of the Haptophyta inferred from 18S ribosomal DNA sequences and
available morphological data. – Phycologia 39: 19–35.
EDWARDS, S.V. 2009: Is a new and general theory of molecular systematics emerging? – Evol.
63: 1–19.
EHARA, M., Y. INAGAKI, K.I. WATANABE & T. OHAMA 2000: Phylogenetic analysis of diatom
coxI genes and implications of a fluctuating GC content on mitochondrial genetic code evolution.
– Cur. Gen. 37: 29–33.
FOISSNER, W. 2006: Biogeography and dispersal of micro–organisms, a review emphasizing –
Protists. – Acta Protozool. 45: 111–136.
FUJITA, M.K., A.D. LEACHÉ, F.T. BURBRINK, J.A. MCGUIRE, & C. MORITZ 2012: Coalescent–
based species delimitation in an integrative taxonomy. – TREE 27: 480–488.
GUILLOU, L., M-J. CHRETIENNOT-DINET, L.K. MEDLIN, H. CLAUSTRE, S. LOISEAUX–DE
GOER & D. VAULOT 1999: Bolidomonas, a new genus with two species belonging to a new algal
class, the Bolidophyceae (Heterokonta). – J. Phycol. 35: 368–381.
ICHINOMIYA, M., S. YOSHIKAWA, M. KAMIYA, K. OHKI, S. TAKAICHI & A. KUWATA 2011:
Isolation and characterization of Parmales (Heterokonta/Heterokontophyta/Stramenopiles) from the
Oyashio region, Western North Pacific. – J. Phycol. 47: 144–151.
KACZMARSKA, I., A. pOuLÍČKOVÁ, s. sATO, M.B. eDLunD, M. IDeI, T. WATAnABe &
D.g. MAnn 2013: Proposals for a terminology for diatom sexual reproduction, auxospores and
resting stages. – Diat. Res. 28: 1–32.
KAWACHI, M., I. InOuye, D. HOnDA, C.J. O’KeLLy, J.C. BAILey, R.R. BIDegARe & R.A.
AnDeRsOn 2002: The Pinguiophyceae, classis nova, a new class of chromophyte algae whose
members produce large amounts of omega–3 fatty acids. – Phycol. Res. 50: 31–47.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
19
KOOISTRA, W.H.C.F., M. DE STEFANO, D.G. MANN, N. SALMA & L.K. MEDLIN 2003: The
phylogenetic position of Toxarium within the diatoms (Bacillariophyceae). – J. Phycol. 39: 185–197.
KOOISTRA, W.H.C.F., D. SARNO, S. BALZANO, H. GU, R.A. ANDERSEN & A. ZINGONE 2008:
Global diversity and biogeography of Skeletonema species (Bacillariophyta). – Protist 159: 177–193.
KUBATKO, L.M. & J.H. DEGNAN 2007: Inconsistency of phylogenetic estimates from concatenated
data under coalescence. – Sys. Biol. 56: 17–24. doi, 10. 1080/10635150601146041.
KUMAR, V., B.M. HALLSTRÖM & A. JANKE 2013: Coalescent–Based genome analyses
resolve the early branches of the Euarchontoglires. – PLoS ONE 8: e60019: doi:10.1371/ journal.
pone.0060019.
LEACHÉ, A.D., M.s. KOO, C.L. spenCeR, T.J. pApenFuss, R.n. FIsHeR & J.A. MCguIRe
2009: Quantifying ecological, morphological & genetic variation to delimit species in the coast
horned lizard species complex. – PNAS USA 106: 12418–12423.
LELIAERT, F., H. VERBRUGGEN, P. VANORMELINGEN, F. STEEN, J.M. LÓPEZ–BAUTISTA,
G.C. ZUCCARELLO & O. DE CLERCK 2014: DNA–based species delimitation in algae. – Eur.
J. Phycol 49:179–196, DOI: 10.1080/09670262.2014.904524.
LEE, M.-A., D.G. FARIA, M.-S. HAN & J. LEE 2013: Evaluation of nuclear ribosomal RNA and
chloroplast gene markers for the DNA taxonomy of centric diatoms. – Biochem. Sys. Ecol. 50:
163–174.
LI, C., M. ASHWORTH, A. WITKOWSKI, H. DĄBEK, J. ZGŁOBICKA et al. 2015: New insights
into Plagiogrammaceae (Bacillariophyta) based on morphological characteristics and a multigene
phylogeny with the description of new families and species, and a new genus. – Plos ONE submitted.
LIN, H.T., G. BURLEIGH & O. EULENSTEIN 2011: The deep coalescence consensus tree problem
is pareto on clusters. – In: CHEN, J., J. WANG & A. ZELIKOVSKY (Eds.): ISBRA, Springer-Verlag,
Berlin, Heidelberg, pp. 172–183
LIN, H.T., G. BURLEIGH & O. EULENSTEIN 2012: Consensus properties for the deep coalescence
problem and their application for scalable tree search. – BMC Bioinformatics 13 (Suppl. 10): S12.
LIU, L., Y. LILI & S.V. EDWARDS 2010: A maximum pseudo–likelihood approach for estimating
species trees under the coalescent model. – BMC Evol. Biol. 10: 302: http://www. biomedcentral.
com/1471–2148/10/302.
LUNDHOLM, N., G.R. HASLE, G.A. FRYXELL & P.E. HARGRAVES 2002: Morphology, phylogeny
and taxonomy of species within the Pseudo-nitzschia americana complex (Bacillariophyceae)
with descriptions of two new species, Pseudo–nitzschia brasiliana and Pseudo-nitzschia linea. –
Phycologia 41: 480–497.
MAGNE-SIMON, M.-F. 1962: L’auxosporulation chez une Tabellariacée marine, Grammatophora
marina (Lyngb.) Kütz. Diatomée. – Cah. Biol. Mar. 3: 79–89.
MARIN, B. & M. MELKONIAN 1999: Mesostigmatophyceae, a new class of streptophyte green
algae revealed by SSU rRNA sequence comparisons. – Protist 150: 399–417.
MARIN, B. & M. MELKONIAN 2010: Molecular phylogeny and classification of the Mamiello-
phyceae class. nov. (Chlorophyta) based on sequence comparisons of the nuclear– and plastid–encoded
rRNA operons. – Protist 161: 304–334.
MEDLIN, L.K. 2002: Why silica or better yet why not silica? Speculations as to why the diatoms
utilise silica as their cell wall material. – Diat. Res. 17: 453–459.
MEDLIN, L.K. 2009a: Diatoms (Bacillariophyta). – In: HEDGES, S.B. & S. KUMAR (eds.): The
Tree of Life. Oxford University Press, pp. 127–130.
MEDLIN, L.K. 2009b: The use of the terms centrics and pennates. – Diat. Res. 24: 499–501.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
20
MEDLIN, L.K. 2010: Pursuit of a natural classification of diatoms, an incorrect comparison of
published data. – Eur. J. Phycol. 45: 155–166.
MEDLIN, L.K. 2014: Evolution of the diatoms, VIII. Re–examination of the SSU-rRNA gene
using multiple outgroups and a cladistic analysis of valve features. – J. Biodivers. Biopros. Dev. 1:
doi:10.4172/2376–0214.1000129.
MEDLIN, L.K. 2015: Timescale for Diatom Evolution Based on Four Molecular Markers and
Reassessment of Ghost Lineages. – Vie et Millieu in press.
MEDLIN, L.K., R. GERSONDE, W.H.C.F. KOOISTRA & U. WELLBROCK 1996a: Evolution
of the diatoms (Bacillariophyta). II. Nuclear–encoded small–subunit rRNA sequence comparisons
confirm a paraphyletic origin for the centric diatoms. – Mol. Biol. Evol. 13: 67–75.
MEDLIN, L.K., I. JUNG, R. BAHULIKAR, K. MENDGEN, P. KROTH & W.H.C.F. KOOISTRA
2008: Evolution of the Diatoms. VI. Assessment of the new genera in the araphids using molecular
data. – Nova Hedwigia, Beih. 133: 81–100.
MEDLIN, L.K. & I. KACZMARSKA 2004: Evolution of the diatoms. V. Morphological and
cytological support for the major clades and a taxonomic revision. – Phycologia 43: 245–70.
MEDLIN, L.K., W.H.C.F. KOOISTRA, R. GERSONDE & U. WELLBROCK 1996b: Evolution of
the diatoms (Bacillariophyta). III. Molecular evidence for the origin of the Thalassiosirales. – Nova
Hedwigia 11: 221–234.
MEDLIN, L.K., W.C.H.F. KOOISTRA & A.-M. SCHMID 2000: A review of the evolution of the
diatoms – a total approach using molecules, morphology and geology. – In: WITKOWSKI, A. &
J. SIEMINSKA (eds.): The origin and early evolution of the diatoms. Frag. Flor. Geobot. Special
Volume, pp. 13–35.
MEDLIN, L.K., K. METFIES, U. JOHN & J. OLSEN 2007: Algal molecular systematics, a review
of the past and prospects for the future. – In: BROADIE, J. & J. LEWIS (eds.): Unravelling the
algae, the past, present and future of algal systematicsitors. Systematics Association, Special Volume
75: 234–253.
MEDLIN, L.K. & S. SATO 2009: The biological reality of the core and basal groups of araphid
diatoms. – Diat. Res. 24: 503–508.
MEDLIN, L.K., S. SATO, D.G. MANN & W.C.H.F. KOOISTRA 2008: Molecular evidence confirms
sister relationship of Ardissonea, Climacosphenia, and Toxarium within the bipolar centric diatoms
(Bacillariophyta, Mediophyceae), and cladistic analyses confirm that extremely elongated shape has
arisen twice in the diatoms. – J. Phycol. 44: 1340–1348.
MEDLIN, L.K., D.M. WILLIAMS & P.A. SIMS 1993: The evolution of the diatoms (Bacillariophyta):
I. Origin of the group and assessment of the monophyly of its major divisions. – Eur. J. Phycol. 28:
261–275.
MEDLIN, L.K., I. YANG & S. SATO 2011: Evolution of the Diatoms. VII. Four gene phylogeny
assesses the validity of selected araphid genera. Festschrift for Prof. H. Lange–Bertalot. – Nova
Hedwigia Beih. 141: 505–514
MOCK, T. & L.K. MEDLIN 2012: Genomics and Genetics of Diatoms. – In: PIGANEAU, G. (ed.):
Genomic Insights into the Biology of Algae; pp. 245–284.
MOESTRUP, Ø. 1991: Further studies of presumably primitive green algae, including the description
of Pedinophyceae class. nov. and Resultor gen. nov. – J. Phycol. 27: 119–133.
NAKOV, T. & E. THERIOT 2010: Phylogenetic position of Didymosphenia geminata (Lyngbye)
M. Schmidt and evaluating molecular markers for distinguishing populations. – Abstract ASLO/
NABS meeting June 2010.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
21
NAKOV, T., E.C. RUCK, Y. GALACHYANTS, S.A. SPAULDING & E.C. THERIOT 2014: Molecular
phylogeny of the Cymbellales (Bacillariophyceae, Heterokontophyta) with a comparison of models
for accommodating rate variation across sites. – Phycologia 53: 359–373.
NORDBERG, M. 2000: Coalescent Theory. – University of California; 1–37.
OLIVER, J.C. 2013: Microevolutionary processes generate phylogenomic discordance at ancient
divergences. – Evol. 67: 1823–1830.
ROUND, F.E., R.M. CRAWFORD & D.G. MANN 1990: The Diatoms, Biology of the Genera. –
Cambridge University Press, Cambridge, 758 pp.
SÁNCHEZ-GRACIA, A. & J. CASTRESANA 2012: Impact of deep coalescence on the reliability
of species tree inference from different types of DNA markers in mammals. – PLoS ONE 7: e30239.
doi:10. 1371/journal. pone. 0030239.
SATO, S. (2008: Phylogeny of araphid diatoms, inferred from morphological and molecular data. –
PhD Dissertation, University of Bremen. http,//elib. suub. uni-bremen. de /diss/docs/00011057. pdf,
SATO, S., G. BEAKES, M. IDEI, T. NAGUMO & D.G. MANN 2011: Novel sex cells and evidence
for sex pheromones in diatoms. – PLoS One 6: e26923.
SCHMID, A.-M. 2001: Value of pyrenoids in the systematics of the diatoms: their morphology and
ultrastructure. – In: ECONOMOU-AMILLI, A. (ed.): Proceedings of the 16th International Diatom
Symposium. Amvrosiou Press, Athens, pp. 1–31.
SIMS, P.A. & G.R. HASLE 1981: Two Cretaceous Stellarima species: S. steinyi and S. distincta;
their morphology, palaeogreography and phylogeny. – Diat. Res. 2: 229–240.
SIMS, P.A., D.G. MANN & L.K. MEDLIN 2006: Evolution of the Diatoms: Insights from fossil,
biological and molecular data. – Phycologia 45: 361–402.
SONGA, S., L. LIUB, S.V. EDWARD & S. WUB 2012: Resolving conflict in eutherian mammal
phylogeny using phylogenomics and the multi–species coalescent model. – PNAS USA 109:
14942–14947.
SORHANNUS, U. 2004: Diatom phylogenetics inferred based on direct optimization of nuclear–
encoded SSU rRNA sequences. – Clad. 20: 487–497.
SORHANNUS, U. 2007: A nuclear–encoded small–subunit ribosomal RNA timescale for diatom
evolution. – Mar. Micro. 65: 1–12.
STOSCH, H.A. VON 1958: Kann die oogame Araphidee Rhabdonema adriaticum als Bindeglied
zwischen den beiden grossen Diatomeengruppen angesehen werden? – Ber. Deut. Bot. Gell. 71:
241–249.
SUL, W.J., T.A. OLIVER, H.W. DUCKLOW, L.A. AMARAL-ZETTLER & M.L. SOGIN 2013:
Marine bacteria exhibit a bipolar distribution. – PNAS doi 10. 1073/pnas. 1212424110.
THAN, C.V. & N.A. ROSENBERG 2011: Consistency properties of species tree inference by
minimizing deep coalescences. – J. Comp. Biol. 18: 1–15.
THAN, C.V. & N.A. ROSENBERG 2014: Mean deep coalescence cost under exchangeable probability
distributions. – Disc. App. Math. 174: 11–26.
THERIOT, E.C., M. ASHWORTH, E. RUCK, T. NAKOV & R.K. JANSEN 2010: A preliminary
multigene phylogeny of the diatoms (Bacillariophyta), challenges for future research. – Pl. Ecol.
Evol. 143: 278–296.
THERIOT E.C., J.J. CANNONE, R.R. GUTELL & A.J. ALVERSON 2009: The limits of nuclear–
encoded SSU rDNA for resolving the diatom phylogeny. – Eur. J. Phycol. 44: 277–290.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
22
VAN DE MEENE A.M.L & J.D. PICKETT-HEAPS 2004: Valve morphogenesis in the centric
diatom Rhizosolenia setigera (Bacillariophyceae, Centrales) and its taxonomic implications. – Eur.
J. Phycol. 39: 93–104, DOI: 10.1080/09670260310001646522.
WHITFIELD, J.B. & P.J. LOCKHART 2007: Deciphering ancient rapid radiations. – Trends Ecol.
Evol. 22: 258–265.
WILLIAMS, D.M. & J.P. KOCIOLEK 2010: Classifications of convenience, The meaning of names.
– Diat. Res. 25: 213–216.
WILLIAMS, D.M. & J.P. KOCIOLEK 2007: The rejection of paraphyletic taxa. – Eur. J. Phycol.
42: 313–319
YOON, H.S., K.M. MULLER, R.G. SNEATH, F.D. OTT & D. BHATTACHARYA 2006: Defining
the major lineages of the red algae (Rhodophyta). – J. Phycol., 42: 482–492.
ZHONG, B., L. LIU, Z. YAN & D. PENNY 2013: Origin of land plants using the multispecies
coalescent model. – Trends Pl. Sci. 18: 492–495.
Manuscript submitted December 16, 2014; accepted June 12, 2015.
eschweizerbart_xxx
PrePub Article
eschweizerbart_xxx
