RESEARCH ARTICLE

Three-dimensional preservation of cellular

and subcellular structures suggests 1.6 billion-
year-old crown-group red algae
Stefan Bengtson1,2*, Therese Sallstedt1,2, Veneta Belivanova1,2, Martin Whitehouse2,3

1 Department of Palaeobiology, Swedish Museum of Natural History, Stockholm, Sweden, 2 Nordic Center
for Earth Evolution (NordCEE), Odense, Denmark; Copenhagen, Denmark; Stockholm, Sweden,
3 Department of Geosciences, Swedish Museum of Natural History, Stockholm, Sweden

a1111111111 * [email protected]
a1111111111
a1111111111
a1111111111
a1111111111 Abstract
The ~1.6 Ga Tirohan Dolomite of the Lower Vindhyan in central India contains phosphatized
stromatolitic microbialites. We report from there uniquely well-preserved fossils interpreted as
probable crown-group rhodophytes (red algae). The filamentous form Rafatazmia chitrakoo-
OPEN ACCESS tensis n. gen, n. sp. has uniserial rows of large cells and grows through diffusely distributed
Citation: Bengtson S, Sallstedt T, Belivanova V, septation. Each cell has a centrally suspended, conspicuous rhomboidal disk interpreted as a
Whitehouse M (2017) Three-dimensional pyrenoid. The septa between the cells have central structures that may represent pit connec-
preservation of cellular and subcellular structures
tions and pit plugs. Another filamentous form, Denaricion mendax n. gen., n. sp., has coin-like
suggests 1.6 billion-year-old crown-group red
algae. PLoS Biol 15(3): e2000735. doi:10.1371/ cells reminiscent of those in large sulfur-oxidizing bacteria but much more recalcitrant than
journal.pbio.2000735 the liquid-vacuole-filled cells of the latter. There are also resemblances with oscillatoriacean
Academic Editor: David Penny, Massey University, cyanobacteria, although cell volumes in the latter are much smaller. The wider affinities of
New Zealand Denaricion are uncertain. Ramathallus lobatus n. gen., n. sp. is a lobate sessile alga with
Received: August 5, 2016 pseudoparenchymatous thallus, “cell fountains,” and apical growth, suggesting florideophy-
cean affinity. If these inferences are correct, Rafatazmia and Ramathallus represent crown-
Accepted: February 7, 2017
group multicellular rhodophytes, antedating the oldest previously accepted red alga in the fos-
Published: March 14, 2017
sil record by about 400 million years.
Copyright: © 2017 Bengtson et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original Author summary
author and source are credited.
The last common ancestor of modern eukaryotes is generally believed to have lived during
Data Availability Statement: All fossil specimens
the Mesoproterozoic era, about 1.6 to 1 billion years ago, or possibly somewhat earlier.
used for this investigation have been deposited in
the collections of the Swedish Museum of Natural
We studied exquisitely preserved fossil communities from ~1.6 billion-year-old sedimen-
History, where they are available for study. tary rocks in central India representing a shallow-water marine environment character-
Tomographic data used in the investigation are ized by photosynthetic biomats. We discovered amidst extensive cyanobacterial mats a
available from the Dryad Digital Repository: http:// biota of filamentous and lobate organisms that share significant features with modern
dx.doi.org/10.5061/dryad.gh221. eukaryotic algae, more specifically red algae. The rocks mainly consist of calcium and
Funding: Paul Scherrer Institute https://www.psi. magnesium carbonates, but the microbial mats and the fossils are preserved in calcium
ch/useroffice/useroffice (grant number 20070197, phosphate, letting us view the cellular and subcellular structures in three dimensions with
20080872, 20100167, 20110963, 20130185, the use of synchrotron-radiation X-ray tomographic microscopy. The most conspicuous
20141047). The funder had no role in study design,

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 1 / 38

 Red algae at 1.6 Ga

data collection and analysis, decision to publish, or

preparation of the manuscript. Swedish Research internal objects in the cells of the filamentous forms are rhomboidal platelets that we
Council http://www.vr.se/inenglish.4. interpret to be part of the photosynthetic machinery of red algae. The lobate forms grew
12fff4451215cbd83e4800015152.html (grant
as radiating globular or finger-like protrusions from a common centre. These fossils pre-
number 2007-4484, 2010-3929, 2013-4290). The
funder had no role in study design, data collection
date the previously earliest accepted red algae by about 400 million years, suggesting that
and analysis, decision to publish, or preparation of eukaryotes may have a longer history than commonly assumed.
the manuscript.

Competing interests: The authors have declared

that no competing interests exist.
Introduction
Abbreviations: ESEM, environmental scanning
electron microscopy; FECA, first eukaryote Multicellular eukaryotes rose to prominence around the Proterozoic–Phanerozoic transition
common ancestor; LECA, last eukaryote common coupled to evolving ecological interactions between megascopic autotrophs and heterotrophs.
ancestor; SEM, scanning electron microscopy; Animals in particular are thought to have had a pivotal role as—for example—predators, graz-
SRXTM, synchrotron-radiation X-ray tomographic
ers, and filterers in expanding the food web through interactive processes not available to
microscopy.
microbial life [1]. Fungi, the other major opisthokont multicellular clade, are important het-
erotrophic consumers, but their role in Phanerozoic ecosystem evolution is generally consid-
ered to be focused on the terrestrial biosphere [2–4]. The three major lineages of multicellular
eukaryotes—Plantae, Fungi, and Metazoa—diversified already during the Proterozoic, but the
time of eukaryote origin is under dispute (reviews by [5–9]). Estimates of the age of the last
eukaryote common ancestor, LECA (the last ancestor of the eukaryote crown group), vary
between extremes of about 0.8 Ga [10] and 2.3 Ga [11], though most proposed dates based on
modern molecular-clock work land within the Mesoproterozoic [7,12–16] or late Palaeopro-
terozoic [17]. The considerable interval of uncertainty is due to difficulties in establishing the
origination time of fossil taxa used for calibration as well as to other problems involved in
molecular-clock dating. One major issue is the lack of robust phylogenies in the base of the
eukaryotic tree [6,18–23].
The age of the first eukaryote common ancestor, FECA (the last ancestor of the eukaryote
total group), is even more uncertain, an extreme proposal being early Archaean, based on evi-
dence of methanogenesis at 3.5 Ga taken to herald the appearance of the euryarchaeotan sister
group (containing methanogenic archaea) of the clade containing the Eukaryota [9]. The pres-
ence of more widespread methanogenesis in the basal archaean tree [24], however, leaves this
inference without support. A more conservative minimum estimate suggests that FECA had
arrived some time before 1.9 Ga based on the earliest widely acceptable eukaryote fossils [7].
However, it is in the nature of stem-group organisms (see [25] for a discussion of the stem-
group and crown-group concepts in palaeontology) that initially they do not differ phenetically
from representatives of their sister group, and thus, attempts to identify FECA in the fossil
record are a hopeless exercise.
More pertinent is the question of when stem-group eukaryotes had acquired fossilizable
features that would enable us to recognize them in the fossil record. The path from FECA to
LECA may have been a fast endosymbiosis-dominated set of events [10] or a slow process
marked by successive events of horizontal gene transfer [26]. To distinguish between those
two scenarios, it is necessary to identify eukaryote characters correctly in the deep fossil
record.
The fossil morphological record of early eukaryotes is thus of crucial importance, but most
reports of pre-Cryogenian eukaryotes have at some point been challenged. The interpretation
of morphologically simple fossils as early eukaryotes has mainly been based on large size,
ornamentation of walls, and a complex, resistant wall ultrastructure [5,7,9,27,28]. The iconic
Palaeoproterozoic tubular fossil reported as Grypania (which lacks the segmented structure of
the Mesoproterozoic type species and may not be a true Grypania [29]) is generally heralded as

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 2 / 38

 Red algae at 1.6 Ga

the first plausible eukaryote in the fossil record, mainly based on its size [7,30–34], though the
eukaryotic affinity has been questioned [29], with an alternative interpretation being its identi-
fication as a giant cyanobacterium [35,36]. Other early eukaryote-like fossils [37–40] are plau-
sible as such but unconfirmed as crown-group eukaryotes.
Geochemical evidence has also been summoned in support of early eukaryotes, but even
more controversially. A much-cited report of 2.7 Ga sterane biomarkers attributed to eukary-
otes [41,42] has been shown to be in error, the biomarkers having been introduced as younger
contaminants [43]. In spite of subsequent attempts to identify Archaean eukaryote biomarkers,
all such evidence for eukaryote presence is now considered invalid, not only because of con-
tamination issues [44] but also because the putative Archaean steranes that were taken as prox-
ies for eukaryotic membrane sterols may also derive from noneukaryote life [45].
Establishing a minimum age for LECA implies identifying crown-group eukaryotes in the
fossil record. Bangiomorpha, interpreted as a bangiophycean rhodophyte [46], is dated to
about 1.1–1.25 Ga [47]. It represents the oldest generally accepted (though see [16,17,35,48])
fossil crown-group eukaryote. Earlier unicellular fossils interpreted as rhodophytes [49] have
not found acceptance [50]. Simple acritarchs at 1.8 Ga [51] have been argued to indicate a
presence of chlorophytes at that time [39], but the identification of the fossils as crown-group
eukaryotes is controversial [7].
We have studied uniquely preserved fossils from the ~1.6 Ga Tirohan Dolomite at Janki-
kund, Chitrakoot, central India, and present here cellularly and subcellularly preserved diverse
forms that show characters of rhodophytes, implying that crown-group eukaryotes, and conse-
quently LECA, are no younger than the Palaeoproterozoic.

Geological background
The Vindhyan Supergroup consists of a thick, unmetamorphosed sequence of sandstones,
shales, and carbonate rocks, as well as volcanoclastic rocks. The ~1.6 Ga [52] Chitrakoot For-
mation, which includes the Tirohan Dolomite, belongs to the Lower Vindhyan (the Semri
Group) and is exposed north of the Son River Valley in the Chitrakoot region, Uttar Pradesh
and Madhya Pradesh, central India (Fig 1). The formation rests unconformably on the Bundel-
khand Granite and is unconformably overlain by the Kaimur Sandstone [53,54].
The phosphatized stromatolitic rocks of the Tirohan Dolomite at the Paisuni River section,
Jankikund, are known for their exquisite preservation of microbial fossils [52,55,56]. The unit
consists of stromatolitic carbonates with phosphorite occurring as bands within and capping
the stromatolites. The intercolumnar matrix contains phosphatic intraclasts. The phosphatized
fossils are found both in the intercolumnar matrix and in the phosphate bands in the stromato-
lites. Parts of the rock are silicified. The Tirohan Dolomite was deposited in a marine shallow-
subtidal to supratidal environment [53,54,57]. Deposition in a photic and oxygenic environ-
ment is demonstrated by the dominance of gas-producing cyanobacteria [52].

Results and discussion

Tubular forms
The tubular components of the Tirohan stromatolitic biota mostly consist of aggregating mas-
ses of entangled filaments, interpreted to represent filamentous cyanobacteria [52,56]. Further-
more, scattered among the stromatolitic layers and the intercolumnar debris are isolated
tubular structures. Most of these are less than 10 μm in width and are probably sheaths of fila-
mentous bacteria or chains of bacterial cells. Some tubes, however, are considerably larger and
represent potential eukaryotic organisms. Being rare, these have not been observed in petro-
graphic sections but have been recovered by manual selection from acid residues.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 3 / 38

 Red algae at 1.6 Ga

Fig 1. Geological map of the Vindhyan basin, central India. After Azmi et al. [55], based on several sources.
doi:10.1371/journal.pbio.2000735.g001

The larger tubes vary in width from 58 to 275 μm. Similar forms from the Tirohan phos-
phorites were initially reported and interpreted as Cambrian taxa, Cambrotubulus decurvatus
and Hyolithellus vladimirovae [55]. The ensuing controversy about the age of the Lower
Vindhyan and the provenance of the fossils was resolved by the demonstration that the Tiro-
han fossil tubes were morphologically analogous to modern filamentous algae rather than to
the named Cambrian taxa and that Pb–Pb isochron analysis confirmed a late Palaeoprotero-
zoic rather than Cambrian age of the phosphorites [52]. However, the biological nature of
these enigmatic fossils has not until now been seriously investigated.
Our microtomography data clarify the internal structure of the tubular fossils, suggesting
that they represent two significantly different organisms, here named Rafatazmia chitrakooten-
sis n. gen., n. sp. and Denaricion mendax n. gen., n. sp. Rafatazmia ranges in width from 58 to
175 μm, Denaricion from 130 to 275 μm. Rafatazmia consists of concatenated cell-like com-
partments, often collapsed or distorted, resulting in buckling and folding of the exterior,
whereas Denaricion has a solid internal structure similar to a stack of coins, upholding the
integrity of the outer shape.
Taphonomy and diagenesis of Rafatazmia tubes. The fossilized structure of most Rafa-
tazmia tubes is dominated by void-filling apatite similar to that which is common in Neopro-
terozoic and Cambrian phosphatized fossils [58–60]. It takes the shape of fibronormal apatite
crystals forming a layered growth on the substrate (Fig 2C and 2D). Because of irregularities

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 4 / 38

 Red algae at 1.6 Ga

Fig 2. Rafatazmia chitrakootensis n. gen., n. sp., Scanning Electron Microscopy (SEM) images. (A–D) NRM S156424 (also figured in [52]); C and D
are backscatter images of a polished transverse section. (E) NRM X5647. (F, G) NRM X5648.
doi:10.1371/journal.pbio.2000735.g002

on the overgrown surface, the crystals frequently fan out to produce a spherulitic or botryoidal
texture (Fig 2D and 2E). Growth of the layers may be directed only inwards toward the central
cavities (Fig 2D) or in opposite directions starting out from both sides of the original walls
(Figs 3J–3L, 4B–4D, 4F and 4G). Void-filling apatite often forms irregular structures within
the compartments, occasionally even filling them completely; in the latter case, the filling can
be seen to have protected the outer walls from buckling (compare the two upper compart-
ments with the bottom one in Fig 4H–4J).
A unique specimen with one compartment open to the outside, through apparent loss of
part of the outer wall, shows internal diagenetic apatite growth on both sides of the original
septal wall (Fig 4D, sw) as well as an external growth layer that is continuous with a secondary
layer covering the original internal coating layer (Fig 4D, ec). Taken together, these observa-
tions indicate that there are at least two generations of diagenetic void-filling apatite: the inter-
nal coating formed early and was the main agent for preserving the physical integrity of the
outer and inner walls; the outer diagenetic layer, when present, is a later addition.
Rafatazmia tubes showing original intracellular features are preserved in microcrystalline
apatite (Fig 5). This suggests that the microcrystalline apatite is an early diagenetic replacement

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 5 / 38

 Red algae at 1.6 Ga

Fig 3. Rafatazmia chitrakootensis n. gen., n. sp., Synchrotron-Radiation X-ray Tomographic Microscopy (SRXTM) renderings. (A–C)
NRM X5646, surface, volume, slice. (D–F) NRM X5592, surface, volume, slice. (G, H) NRM X4230, surface, volume. (I–L) NRM X4229, surface,
volume, slice. Legend: sw, septal wall; is, incomplete septum. Scale bars 50 μm.
doi:10.1371/journal.pbio.2000735.g003

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 6 / 38

 Red algae at 1.6 Ga

Fig 4. Rafatazmia chitrakootensis n. gen., n. sp., SRXTM renderings. (A–D) NRM X4251, surface, volume, slice. (E–G) NRM X5572,
surface, volume, slice. (H–J) NRM X5562, surface, volume, slice. Legend: ec, external coating; db, diagenetic boundary; sw, septal wall; tw,
thick wall. Scale bars 50 μm except where otherwise noted.
doi:10.1371/journal.pbio.2000735.g004

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 7 / 38

 Red algae at 1.6 Ga

Fig 5. Rafatazmia chitrakootensis n. gen., n. sp., SRXTM renderings. (A–L) Holotype, NRM X4258. (A) Surface rendering. (B) Volume rendering with
rhomboidal disks coloured for visibility. (C) Virtual slice. (D) Surface. (E) Volume. (F–L) Transverse slices (positions indicated in B). (M–O) NRM X5620,
surface, volume, slice. (P–R) NRM X5574, surface, volume, slice. Scale bars 50 μm.
doi:10.1371/journal.pbio.2000735.g005

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 8 / 38

 Red algae at 1.6 Ga

Fig 6. Rafatazmia chitrakootensis n. gen., n. sp., SEM image (A) and SRXTM renderings (B, C). A: NRM S156422 (also figured in [52]). (B, C) NRM
X5544, surface, volume.
doi:10.1371/journal.pbio.2000735.g006

of biological substance, whereas the layered fibronormal apatite represents later events of dia-
genetic incrustation. Comparable conclusions have been reached with regard to similar diage-
netic fabrics in the phosphatized fossils of the Ediacaran Doushantuo Formation [59,61].
Morphology of Rafatazmia tubes. The 58–175 μm wide Rafatazmia tubes are up to ca. 2
mm in length (Fig 6A); the ends appear not to be preserved. The tubes are divided by trans-
verse septa into compartments that range from 30 μm to 165 μm in length. The positions of
the septa are usually seen on the outside as annular grooves (Figs 2, 3D, 4, 5 and 6B). The exter-
nal annulation may be visible even when the internal preservation is too poor to show any
remnants of septa (Fig 3D–3F). Although the length can vary significantly between adjacent
compartments (Figs 3J–3L, 4B and 4F), there is a general isometric trend in which the width of
the filaments is linearly correlated with the length of the cells (Fig 7).
Most compartments are empty, meaning that they are not filled with apatite but with sec-
ondary carbonate that was mostly dissolved during acid extraction of the fossils (see remaining
carbonate crystals in Figs 2C, 2D, 4F, 4G, 5Q and 5R). A few have an irregular content of void-
filling phosphate that envelopes undefined objects or fills the compartment completely, but
most of these structures do not show any recurring pattern and are not deemed to be signifi-
cant in terms of original morphology of biological features (Figs 4J and 8).
The septa between compartments are typically flat but sometimes bulge into adjacent com-
partments to form concavo-convex partitions (Figs 4B, 4C, 4F, 4G, 5Q and 5R). Some irregu-
larities in the septa can be explained as the result of taphonomic collapse of the tube structure
(Fig 8A–8D), but generally, the septa also remain undisturbed when the outer tube wall is
strongly distorted (Figs 4J, 8D and 8G). The specimen in Fig 4E–4G has a general pattern of

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 9 / 38

 Red algae at 1.6 Ga

Fig 7. Relationships between filament width and cell volume in Rafatazmia, Denaricion, and some modern filamentous organisms. Data
sources for the modern taxa: Oscillatoriacea [62–64] and Soft-Bodied Stream Algae of California (http://dbmuseblade.colorado.edu/DiatomTwo/sbsac_
site/). Beggiatoa [65–70]. Spirogyra (Soft-Bodied Stream Algae of California). Individual measurements are in Supporting Information in the file S1 Data.
doi:10.1371/journal.pbio.2000735.g007

125–145 μm distance between septa, which is broken by a single septum in the middle of one
compartment (Fig 4G, arrow). This septum has a pore in the central part and possibly repre-
sents a new dividing wall under formation. A similar phenomenon is seen in Fig 8I and 8J,
where two incomplete septa (arrows in J) partition larger (80–110 μm long) compartments,
though not into equal parts. A more irregular pattern of incomplete septa can be seen in Fig
3K and 3L, where four consecutive septa are open in or near the middle. The irregular appear-
ance of minute septum-like projections from the outer wall within the same region of the tube
(Fig 3L, is) complicates the picture, however, emphasizing the difficulties of recognizing origi-
nal structure through a taphonomic/diagenetic overprint.
The best-preserved specimens of Rafatazmia are shown in Fig 5. They are preserved in
microgranular apatite with little addition of void-filling matter. The holotype of R. chitrakoo-
tensis (NRM X4258, Fig 5A–5L) has six compartments, separated by septa that are internally
complete and externally visible as distinct furrows between the slightly convex walls of the
compartments. Two of the compartments are distorted by collapse, but all have a distinct and
consistent internal structure. This consists of a cylindrical body attached to the outer wall and
extending toward the middle of the compartment, suspending a flattened-rhomboidal object
in the centre. This object is 50–60 μm diagonally (except one 93 μm wide, Fig 5K) and 20 μm
thick and is clearly visible in five of the six compartments (Fig 5F–5L). The flattened plane is
aligned with the length axis of the filament. The supporting body is detached from all but the

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 10 / 38

 Red algae at 1.6 Ga

Fig 8. Rafatazmia chitrakootensis n. gen., n. sp., SRXTM renderings. (A–D) NRM X4246, surface, volume, slice. (E–G) NRM X4240,
surface, volume, slice. (H–J) NRM X4242, surface, volume, slice; arrows point to new septa with central pore. Scale bars 50 μm except where
otherwise noted.
doi:10.1371/journal.pbio.2000735.g008

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 11 / 38

 Red algae at 1.6 Ga

Fig 9. Rafatazmia chitrakootensis n. gen., n. sp., SRXTM virtual slices. Arrows point to diagenetic overgrowths indicating the presence of
irregularities consistently placed at the centre of septa and interpreted as possible pit plugs. (A) NRM X5573. (B) NRM X5645. (C) NRM X5626. (D) NRM
X5610. (E) NRM X5613. Scale bars 50 μm.
doi:10.1371/journal.pbio.2000735.g009

peripheral parts of the dividing septum, forming a thin planoconvex space by the septum. In
the middle, where the rhomboidal object is suspended, there is a lenticular space (Fig 5C, 5F,
5G, 5I and 5K).
The two other specimens in Fig 5 (Fig 5M–5R) are not as completely preserved as the holo-
type. There is no cylindrical internal body, but most septa have a flattened or globular object
attached to the central part, in some cases expressed on both sides of the septum (Fig 5R, top).
In at least one case (Fig 5M–5O), their size and morphology suggest that they represent
degraded specimens of the rhomboidal objects that have lost their suspending material and
attached to the septal membrane. This may, however, not explain the several cases in which
there are small objects consistently situated on or through the centre of septa, where they typi-
cally nucleate diagenetic spherules (Figs 4E–4G, 5P–5R and 9).
The original wall in most cases appears to have been very thin (black line at “sw” arrow in
Fig 3L). In the specimen in Fig 4A–4D, the boundary representing the wall (sw) is considerably
less distinct than that between the primary and secondary diagenetic overgrowth (db). A divid-
ing gap of 2–3 μm in a polished transverse section may represent the original wall between out-
ward- and inward-growing apatite (Fig 2C, arrow). Exceptionally, a thicker wall (about 7 μm)
composed of homogenous apatite surrounds a compartment (Fig 4D, tw); this feature is inter-
preted to represent a modified cell with a thickened cell wall.
Morphology, taphonomy and diagenesis of Denaricion tubes. The 130–275 μm wide
tubes of Denaricion (Fig 10) are always broken at both ends, usually at planes of weakness pro-
ducing flat surfaces of breakage. The longest fragment (Fig 10A–10C) is 755 μm in length, and
the preserved stretches of tube are straight to evenly curved. The cross section is circular to
slightly oval.
In contrast to Rafatazmia, Denaricion tubes have an almost solid interior, consisting of
microgranular apatite that weakly expresses a structure like a stack of coins (Fig 10C, 10E, 10F,
10H, 10I, 10K, 10M, 10O, 10R, 10T, 10U and 10V). Void-filling apatite is generally absent,
which suggests early replacement of biological tissue by microgranular apatite, not affected by
later void-filling phosphatization such as in Rafatazmia. The “coin-stack” structure is faintly

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 12 / 38

 Red algae at 1.6 Ga

Fig 10. Denaricion mendax n. gen., n. sp., SRXTM renderings. (A–C) Holotype, NRM X5644, surface, volume, slice. (D–F, V) NRM X5604,
surface, volume, slice, volume/slice. (G–I) NRM X5634, surface, volume, slice. (J, K) NRM X4233, surface, volume. (L, M, U) NRM X4234,
surface, volume, volume/slice. (N–P) NRM X4244, surface, volume, slice. (Q, R) NRM X4256, surface, volume. (S–T) NRM X4235, surface,
volume. Arrows in U and V point to dark lines suggesting incipient septation. Scale bars 50 μm except where otherwise noted.
doi:10.1371/journal.pbio.2000735.g010

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 13 / 38

 Red algae at 1.6 Ga

Fig 11. Denaricion mendax n. gen., n. sp., blowups of SRXTM volume renderings from Fig 10 to show 4/8/16 pattern of dividing cells within
compartments. Asterisks show divisions between 8-cell compartments; rings mark weaker divisions between the fourth and the fifth cell within some
compartments. (A) Same as in Fig 10B. (B) Same as in Fig 10M. (C) Same as in Fig 10K. (D) Same as in Fig 10E.
doi:10.1371/journal.pbio.2000735.g011

or not-at-all visible in virtual slices but can be seen in semitransparent volume renderings in
projections perpendicular to the tube axes (cf. the left [volume rendering] and right [virtual
slice] parts of Fig 10U and 10V, respectively). The “coins” appear to be in touch at the periph-
ery but have a somewhat concavo-concave shape resulting in a lens-shaped volume of lower
X-ray attenuation (dark) between the “coins.” Where the structure is distinctly expressed,
there is a thin line of low attenuation in the middle of each “coin” (Fig 10U and 10V, arrows).
In addition to the fine stacking pattern, there is a compartmentalization expressed as a
more distinct plane cutting from edge to edge (Fig 11, asterisks); this is also a plane of weak-
ness where the tubes are usually broken off. In several cases, the planes delimit sections of dif-
ferent degrees of mineralization, as apparent through differences in colour and/or X-ray
attenuation (e.g., Fig 10B and 10M). With minor exceptions, the number of “coins” within one
such compartment is 8 or 16 (Fig 11); some compartments also have a somewhat less distinct
plane between the fourth and fifth “coin” in a sequence (Fig 11C and 11D, rings), suggesting a
general geometric pattern of 4/8/16.
The outer surface of the tubes is normally smooth, though it may be obscured by a diage-
netic coating of apatite (Fig 10A, 10D, 10G, 10J, 10L and 10N). Occasionally, the planes of
compartmentalization are visible as faint annulations beneath the coating (Fig 10L).

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 14 / 38

 Red algae at 1.6 Ga

Interpretation of the tubular fossils. Filamentous organisms are widespread among

microbial taxa. The first question to address is whether the tubular fossils described herein rep-
resent pro- or eukaryotes.
Size often takes a central position in the identification of fossil eukaryotes, but it has long
been recognized that prokaryotic and eukaryotic cell sizes overlap considerably and that size
cannot be used as an absolute criterion in fossil interpretation [71]. Bacterial cells, although
typically of micrometre size, can attain dimensions of hundreds of micrometres [63]: cells of
the sulfur bacterium Thiomargarita may be as large as 750 μm in diameter [72], and the fila-
mentous sulfur bacterium Beggiatoa reaches 190 μm in width [65].
The compartments in Rafatazmia can confidently be identified as concatenated cells of a
filamentous organism. The 4/8/16 pattern of distribution of the Denaricion “coins” shows
them to be formed by duplication through coordinated septation, providing strong support
for their interpretation as cells. Fig 7 shows a comparative diagram of filament width and cell
volume in the two Vindhyan taxa versus modern oscillatoriacean cyanobacteria (Oscillatoria
and Lyngbya), sulfur-oxidizing filamentous bacteria (Beggiatoa), and zygnematalean fila-
mentous green algae (Spirogyra). The Vindhyan taxa both cluster at the upper right end of
the diagram. Conspicuously, there is no overlap with the filamentous cyanobacteria, which
have a filament width not exceeding 80 μm [63,64] and a maximum cell volume almost four
times smaller than that of the smallest Vindhyan form. The sulfur-oxidizing Beggiatoa fol-
lows the same trajectory as the cyanobacteria but reaches a considerably wider diameter,
about 190 μm, and in its upper range it clusters together with Denaricion, which, however,
reaches a diameter of 275 μm. The zygnematalean Spirogyra in its upper size range follows
the same trajectory as Rafatazmia. Throughout its size range, and particularly in the lower
part of the range, Spirogyra exhibits a considerably larger cell volume than the prokaryotic
filaments of similar width.
These size comparisons show that Rafatazmia conforms to a eukaryotic cell size distribu-
tion, with an approximately isometric size increase. Denaricion, on the other hand, follows a
trajectory of filamentous prokaryotes, characterized by more severe constraints on cytoplasmic
volume. In terms of filament width, it exceeds Rafatazmia and known prokaryote filamentous
forms, but the tendency of the Denaricion cells to become thinner and form “coin stacks” with
increasing filament diameter is characteristic of large filamentous cyanobacteria [64].
Yet, as Denaricion clearly falls outside the size range of known cyanobacteria (Fig 7;
[63,64]), comparisons may rather be made with the large sulfur bacteria, in particular, Beggia-
toa. The constraints on prokaryote cell volume are thought to be due to the limits of diffusion
speed within a cytoplasm that lacks membrane-bound cytoskeletal transport mechanisms. The
extreme sizes of Beggiatoa and some other sulfur bacteria do not reflect cytoplasmal volume,
however, but rather the presence of intracellular vacuoles; the cytoplasm typically occupies
only a thin veneer between the cell membrane and the vacuole [65], perhaps amounting to
some 2% of the total cell volume [63]. The vacuoles of sulfur bacteria, commonly a storage site
for nitrate, are prone to rapid collapse during degradation, as has been confirmed by decay
experiments in the laboratory [73]. The apparent stiffness of the Denaricion filaments and the
solid structure of its fossilized cell contents speak strongly against the cells being occupied by a
liquid vacuole. Furthermore, there is no evidence in Denaricion of the sulfur granules that
would be expected in the cytoplasm of a sulfur-oxidizing bacterium [72]. The systematic place-
ment of Denaricion is thus problematic. Its living possible analogues among prokaryotes either
are too small (filamentous cyanobacteria) or lack the preservation potential (sulfur-oxidizing
giant filamentous bacteria) to be likely relatives. While acknowledging the possibility of extinct
prokaryote branches with larger cytoplasmic volumes than in known forms, we leave the ques-
tion of the affinities of Denaricion open at present.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 15 / 38

 Red algae at 1.6 Ga

A eukaryotic affinity of Rafatazmia is suggested primarily by its cell volume (Fig 7) and by
the presence of large organelle-like objects inside the cells (Fig 5A–5L). Subcellular structures
in microbial fossils are notoriously difficult to interpret, and much literature has been devoted
to controversies about the nature of globular or irregular objects seen within Proterozoic
microfossils [50,71,74–82]. The gist of the controversies has been whether such objects are
taphonomic/diagenetic features, such as degraded and shrunken protoplasts, or whether they
can be interpreted as more distinctly defined subcellular objects, in particular, nuclei or pyre-
noids. A middle road was recently taken by Pang et al. [82], who suggested that globular
objects in acritarchs of the late Palaeoproterozoic Ruyang biota in northern China represent
shrunken protoplasts formed by the organism during encystment.
Without morphological or structural data in support, the interpretation of intracellular
structures in fossils is bound to be controversial. Numerous identifications of fossilized nuclei
have been made, but most of them have not withstood scrutiny, although it is known from
Phanerozoic occurrences that cell nuclei, even chromosomes, can be faithfully replicated by
fossilization [83]. Nuclei have been identified in Neoproterozoic embryo-like fossils based on
features such as recurrence, position, shape, volumetric relationships, and evidence of closed
mitosis [78,79,81], but without support from such morphological evidence, the nature of
alleged fossilized nuclei is bound to be controversial.
The intracellular features of Rafatazmia cells are distinct, involving a rhomboidal disc sus-
pended in the middle of each cell by a regular framework of cytoplasmic matter. Growth seem-
ingly took place by septation of cells, and the organism occasionally produced thick-walled
cells, maybe as part of an encystment process. The central object is preserved in early-diage-
netic fine-grained apatite, and the recurring features of rhomboidal shape, central position,
and orientation with its major plane within the length axis of the filament strongly indicate
that original biological morphology is represented. Given the size of these intracellular objects,
an interpretation as eukaryotic organelles is reasonable. Bacterial organelles, such as carboxy-
somes, are enclosed within a proteinaceous shell in microcompartments that typically are only
a few hundreds of nanometres in diameter [84]; the rhomboidal bodies of Rafatazmia are
more than a hundred times wider. Within a eukaryotic frame of reference, the Rafatazmia
rhomboids may represent, e.g., nuclei, plastids, or pyrenoids. Given the polygonal (rhomboi-
dal) shape, a pyrenoid is the more likely alternative.
Pyrenoids are dense proteinaceous bodies associated with chloroplasts in most groups of
algae [85,86]. They are commonly involved in the production of starch—or the floridean
starch of red algae—and tend to take on a polygonal shape [87]. The relation between polygo-
nal shape and the surrounding starch plates is most commonly observed in green algae, chlor-
ophytes [77,88], but even though floridean starch is stored in the cytoplasm rather than in the
chloroplast, polygonal pyrenoids are also known from red algae, rhodophytes (Figure 2A, 2B
in [89]). The number of pyrenoids in a cell or in a chloroplast varies enormously between taxa,
but basal rhodophyte lineages [90–92] are generally characterized by cells having a single,
centrally located plastid with a conspicuous pyrenoid [88,93], as exemplified by filamentous
Porphyridium [94], Chroodactylon [95,96], and Bangia [97]. Later lineages tend to lose the
encasing starch shells [98] and thus their angular pyrenoids.
Pyrenoids are of different sizes, but most are much smaller than the rhomboidal bodies in
Rafatazmia. The size of a pyrenoid, however, is under comparable conditions correlated with
cell size [99], and the relative sizes of the Rafatazmia internal bodies versus cells are fully com-
parable to those seen in modern algae with singular pyrenoids (Figure 4 in [97]).
Filament endings are not preserved in the fossilized Rafatazmia, but the evidence of incom-
plete septa forming within individual cells (Fig 4G, arrows, and Fig 8I and 8J) suggests that
growth took place by diffusely distributed septation rather than from apex tips or specialized

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 16 / 38

 Red algae at 1.6 Ga

meristems. In rhodophytes, diffuse growth characterizes the early branching lineages, such as
the bangiophytes, whereas filamentous florideophyceans have apical growth meristems [88].
The presence of a central pore in Rafatazmia septa and of small bodies in corresponding posi-
tions in other septa is reminiscent of the mode of septation in rhodophytes, in which new
septa are commonly incomplete and the central pore may later be filled with a tubular mem-
brane called a primary pit plug, connecting the two daughter cells; pit plugs are unique to mul-
ticellular red algae and are particularly common in the Bangiophycea/Florideophycea clade
[92,100]. The preservation of the Rafatazmia structures in diagenetic apatite, however, does
not allow any definitive comparison with rhodophyte pit plugs.
Cells with single plastids having a large pyrenoid occur in other filamentous algae, such as
the chlorophytes Klebsormidium and Uronema, but these forms have parietal chloroplasts,
resulting in a nonaxial position of the pyrenoid. A few zygnematalean chlorophytes (the group
including Spirogyra, which was used for morphological comparisons in Fig 7), such as Zyg-
nema, have a single pyrenoid per plastid but two plastids per cell.
The presence of a single pyrenoid per cell in Rafatazmia is therefore not an unambiguous
character to distinguish rhodophyte versus chlorophyte affinity. The indications of pit connec-
tions and pit plugs, if corroborated, may place Rafatazmia within the crown-group Rhodo-
phyta. The diffuse growth would further place it within the Bangiophycea, but the possibility
that diffuse growth of filaments is a plesiomorphy for the rhodophytes makes this assignment
also uncertain.

Lobate thalli
The fabric of cyanobacteria-dominated biofilms in the stromatolitic columns contains embed-
ded lobate thalli with cellular preservation. These differ significantly in size, structure, and
preservation from the surrounding microbial communities and are analysed here as potential
eukaryotes, Ramathallus lobatus n. gen., n. sp.
Taphonomy and diagenesis of Ramathallus thalli. The millimetre-sized lobate thalli are
part of the dense apatitic stromatolitic fabric and have only been observed in sections and in
untreated stromatolite slabs, not in acid-isolated residues. They appear light grey in sections
under reflective light (Fig 12A), the cell lacunes being light grey or white and the walls between
the cells a darker grey (Fig 12B). Under transmitted light, the 1–3 μm thick cell walls are light
beige-pink with a central refractive or dark line (Fig 13B and 13C). Cell interiors are an opaque
brown to black, often with a granular appearance (Fig 13C). Under the scanning electron
microscope, the walls are seen as a 0.2–0.5 μm thin microcrystalline sheet coated on both sides
by a more coarsely crystalline layer (e.g., Fig 14). This suggests that the original wall was con-
siderably thinner than the 1–3 μm thick walls seen under the optical microscope, most of the
thickness representing diagenetic coating. The presumed original cell walls consist of equant,
approximately 0.2–0.5 μm, apatite crystals (Fig 14B, fc). The larger crystals coating the walls
are dominantly hexagonal apatite tablets and prisms, 0.5–1 μm in cross-section and up to 3 μm
in length (Fig 14B, cc).
Nondiagenetic differences in cell size are common (e.g., Fig 15B and 15C), but the exact
size of a cell is difficult to judge because of diagenetic overprint. The content of the volume
inside the cell walls and the degree of apatite encrustation vary, even between adjacent cells
(Fig 15D and 15E). Some cells are hollow, and some are more-or-less completely filled with
apatite crystals (Fig 15E, gm), but the majority contain residual carbonaceous matter with
embedded apatite crystals (Fig 15E, cm), corresponding to the dark granular cell interiors seen
under the light microscope (Fig 13C). Zones of stronger phosphatization within a thallus may
appear as compact material with only sporadically visible cell interiors (light horizontal zone

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 17 / 38

 Red algae at 1.6 Ga

Fig 12. Ramathallus lobatus n. gen., n. sp. (A, B) NRM X5639, reflected light. (C, D) NRM X5643, thin section.
doi:10.1371/journal.pbio.2000735.g012

of upper part in Fig 16A); such fabric may be difficult to separate from noncellular features
such as the light apatitic crust covering many thalli (e.g., Fig 12D, right, and Fig 16A, left). The
latter is up to 40 μm thick and lacks the botryoidal fabric typical of void-filling apatite (cf. the
diagenetic coating in Rafatazmia; Fig 2C and 2D). It shows undulating extinction, however,
and thus implies a diagenetic overprint that may be coeval with the void-filling apatite, which
frequently overlies it (see Figs 17B and 18C, upper right). The morphology and distribution
suggest that it replaces an original biological feature of the fossils.
Morphology of Ramathallus thalli. The thalli are megascopic, ranging from approxi-
mately half a millimetre up to well over 3 mm across. Aggregations of thalli can attain a size of

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 18 / 38

 Red algae at 1.6 Ga

Fig 13. Ramathallus lobatus n. gen., n. sp., holotype, NRM X5638. Thin section. (A) Overview of specimen. (B) Detail
of A to show cell structure in finger-like protrusions and noncellular apatitic coating. (C) Detail of B to show dark granular
material within cells.
doi:10.1371/journal.pbio.2000735.g013

more than a centimetre (Fig 12A). The thalli are lobate in shape and typically project bulbous
extensions (Figs 12, 13, 17 and 18) radiating from a central point of the structure (Fig 13A).
Protrusions may take the form of separate globular or finger-shaped objects or be more closely
adpressed to each other, separated by deep invaginations (Figs 12C, 13B and 18C). Some pro-
trusions have a narrower stalk-like base and a more bulbous distal part (Fig 16A). The size of
single cells ranges from approximately 5 μm to 15 μm. In the projecting parts of the thallus, the

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 19 / 38

 Red algae at 1.6 Ga

Fig 14. Ramathallus lobatus n. gen., n. sp., NRM X5640. Environmental scanning electron microscopy (ESEM) images, backscatter mode. (A)
Globular bodies (gb) within cells. (B) Coarse crystals (cc) and fine crystals (fc) in diagenetically phosphatized cell walls.
doi:10.1371/journal.pbio.2000735.g014

cells are lined up in “fountains,” producing a pattern typical of pseudoparenchymatous tissue

(Figs 15B and 16–18). These cells are typically elongated in the direction of the cell row, and in
the narrower regions of such a protrusion, the cell rows take on a stretched-out, filamentous
appearance (Fig 18B).
The smallest cells are usually in the “cell fountains,” whereas larger ones may appear in the
bulbous upper parts of a protrusion (Figs 12D and 16B), in globular bodies appearing near the
base of larger protrusions (Fig 16A, lower left), or as separate smaller protrusions (Figs 15A,
15C–15E, 18C and 19). The larger-celled globular bodies often do not have the same pseudo-
parenchymatous pattern as the larger protrusions but are more irregularly packed. In parts of
the smaller bodies, the cells are repeatedly arranged as an isolated quadruplet lying in one
plane, forming a four-leaf-clover pattern (Fig 19B–19G, arrows, Fig 19H). Larger cells com-
monly have a small internal globular body, coated with apatite crystals similar to those lining
the inside of the cell walls (Fig 14A, gb)
Surrounding most of the thalli is an up to 40 μm thick homogenous apatitic coating (Figs
12C, 12D, 13A, 13B, 16A, 18A and 18C). It is continuous with, and identical in structure and
colour with, the sheaths separating adjacent thallus protrusions (Figs 12C and 13B) and shows
no evidence of a cellular organization. A similar coating also lines lacunae that are sometimes
found in the interior of the complex thalli (Fig 13A; cf. also Figs 12C, 15A and 17B). The pres-
ence of a diagenetic layered overgrowth makes it sometimes difficult to determine the presence
and extension of the homogenous outer layer (Figs 12B, 16A, 17B and 18C, upper right).
Interpretation of Ramathallus. Ramathallus was a sessile benthic organism, as indicated
by its growth mode, with lobate protrusions radiating from a centre, and its in situ preserva-
tion within the Vindhyan microbial mats. The lobes were mainly composed of pseudoparen-
chymatous tissue in the form of “cell fountains.” Cell sizes were highly variable. Apparent
tetraspores (the quadruplets in Fig 19) were present in localized places in the smaller globular
protrusions. The thallus was embedded in a noncellular matrix that also penetrated between
the lobes. No distinct cellular cortex has been observed, but the expanded cells at the outer por-
tions of the thallus underneath the noncellular coat (Figs 12D and 16B) may have been special-
ized as meristematic cells and thereby represent apical growth.
Pseudoparenchymatous tissue formation is a basic way of making a resilient thallus through
the coordinated growth of coalescing filaments. This type of structure is widespread among

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 20 / 38

 Red algae at 1.6 Ga

Fig 15. Ramathallus lobatus n. gen., n. sp., NRM X5638. ESEM images, backscatter mode. (A) Overview. (B and C) Details of A to show
pseudoparenchymatous structure with “cell fountains” and varying cell sizes. (D and E) Details of C to show preservation of cells. Legend: cm,
carbonaceous matter; gm, granular matter.
doi:10.1371/journal.pbio.2000735.g015

organisms with fundamentally filamentous growth, and as such, it is prone to convergence. It

is characteristic of fungi and algae, particularly florideophycean rhodophytes, as well as some
phaeophytes and chlorophytes [88]. Certain chroococcalean cyanobacteria, such as Chloroglea
and Entophysalis, form three-dimensional multilayered colonies in which cells may be lined up
in a pseudoparenchymatous pattern, but they tend to be irregularly arranged [101], not show-
ing the distinct “cell fountains” of eukaryotic pseudoparenchyma. Nor do they display the
complex and fleshy thallus characteristic of many eukaryotic algae as well as Ramathallus.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 21 / 38

 Red algae at 1.6 Ga

Fig 16. Ramathallus lobatus n. gen., n. sp., NRM X5638. Thin section. (A) Thallus with well-developed “cell fountains” in club-like distal
part. (B) Thallus with “cell fountains” and cortex-like structure of enlarged cells.
doi:10.1371/journal.pbio.2000735.g016

Ramathallus resembles the similarly preserved and similarly sized algal fossils in the Edia-
caran Doushantuo phosphorites of China—in particular, the pseudoparenchymatous, lobate
taxa Thallophyca Zhang (1989 [102]), Thallophycoides (Zhang and Yuan (1992 [103]), Parame-
cia Zhang and Yuan (1992 [103]), and Gremiphyca (Zhang et al. (1998 [104]). These taxa were
reinvestigated by Xiao et al. [105], who compared them mutually and with fossil and modern
algae on the basis of thallus morphology, cellular architecture, mode of growth, and putative
reproductive structures. The authors concluded tentatively that Thallophycoides and Gremi-
phyca are likely stem-group florideophyceans, whereas Paramecia and Thallophyca are stem-
group corallines and consequently belong within the crown-group florideophyceans, [105].
Thallophycoides and Gremiphyca have globular to nodular thalli with no evidence of attach-
ment. The pseudoparenchymatous structure is simple, with uniform cell size and no cortex–
medulla differentiation. The spherical thalli of Wengania Zhang (1989 [102]) may represent
early developmental stages of these simple forms [105].
Thallophyca has a foliose or lobate thallus with conspicuous pseudoparenchyma built up of
“cell fountains,” probable tetraspores, and a distinct cellular cortex with cells smaller than

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 22 / 38

 Red algae at 1.6 Ga

Fig 17. Ramathallus lobatus n. gen., n. sp. (A) NRM X5641, SRXTM volume rendering. (B) NRM X5642,
thick section.
doi:10.1371/journal.pbio.2000735.g017

those of the medulla and oriented parallel or perpendicular to the surface of the thallus. In
addition, it has clusters of larger cells—“cell islands”—embedded in the tissue, sphaeroidal cav-
ities, cylindrical invaginations, and sorus-like structures that have all been interpreted as possi-
ble reproductive features [105]. Paramecia has a nodular, compartmentalized thallus with less-

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 23 / 38

 Red algae at 1.6 Ga

Fig 18. Ramathallus lobatus n. gen., n. sp., NRM X5638. Thin section. (A) Overview of complex thallus with cell
fountains. (B) Detail of A to show elongated cells. (C) Different area to A of the same thin section, showing globular or
finger-shaped protrusions.
doi:10.1371/journal.pbio.2000735.g018

conspicuous pseudoparenchyma than Thallophyca and a distinct cortex–medulla differentia-

tion; it too has embedded “cell islands” and cavities that may be involved in reproduction
[105].
The Vindhyan Ramathallus differs from Thallophycoides and Gremiphyca in its more com-
plex, lobate thallus, nonuniform cell size, distinct “cell fountains,” and a cortex-like region of
enlarged apical cells. It is most similar to Thallophyca in its complex thallus and conspicuous

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 24 / 38

 Red algae at 1.6 Ga

Fig 19. Ramathallus lobatus n. gen., n. sp., NRM X5641. SRXTM slices. (A) Larger field of view to show the location of the imaged volume. (B–G)
Consecutive sections through volume of larger cells, where putative tetraspores are marked with arrows. (H) Stereo anaglyph showing surface-rendered
cell volumes of three putative tetraspores (yellow) surrounded by nontetradially arranged cells (grey, transparent).
doi:10.1371/journal.pbio.2000735.g019

“cell fountains” but differs from both Thallophyca and Paramecia in lacking a distinct cortex,
“cell islands,” and cavities of possible reproductive nature (the cavities seen in Figs 12C and
13A may be of diagenetic origin). The presence of tetraspore-like structures in both Ramathal-
lus and Thallophyca may possibly signify a common feature, but their occurrence in the respec-
tive taxa is not consistent. In Ramathallus, the structures occur scattered among cells of
comparable size in large-celled globules, whereas in Tetrathallus they occur as large-celled

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 25 / 38

 Red algae at 1.6 Ga

aggregates within masses of considerably smaller cells. In neither case is the biological signifi-
cance of the structures clear.
Ramathallus is thus taxonomically distinct from the billion-year younger Doushantuo
forms, but its lobate thalli with distinct pseudoparenchyma, “cell fountains,” apical growth,
and cortex-like modifications of the distal parts of the lobes all suggest affinity to the Doushan-
tuo taxa; comparisons can be made with Thallophyca in particular. As discussed by Xiao et al.
[105], the listed characters are most consistent with florideophycean rhodophytes, although
like Ramathallus the Doushantuo taxa lack evidence of pit connections, a common feature
among florideophyceans. Based on the vegetative characters of Ramathallus, in the absence of
clear reproductive features in the thallus, we tentatively propose a stem-group florideophycean
affinity of the Vindhyan taxon, which would place them within the crown-group Rhodophyta.

Geological age
Given the remarkably old age that we cite for the rhodophyte-like fossils reported herein, it is
necessary to discuss in some detail the reliability of the ~1.6 Ga date for the Tirohan Dolomite.
Vindhyan stratigraphy has generated a considerable amount of discussion and controversy
over the years, and even in the recent literature, age estimates for the Semri Group (Lower
Vindhyan) are given as either Palaeo–Mesoproterozoic, as indicated by geochronology
[106,107], or Ediacaran–Cambrian, as suggested with reference to fossil content [108,109].
Correlation between the Chitrakoot Formation and the Semri sequence in the Son Valley is
not straightforward [110], but regional isopach and lithofacies interpretations suggest that the
Chitrakoot Formation belongs within the upper part of the Semri Group; in particular, correla-
tion is suggested between the Tirohan Dolomite at Chitrakoot and the Rohtas Limestone
(Rohtasgahr Limestone of some authors) in the Son Valley [111].
Recent geochronological results from the Semri Group in the region are summarized in Fig
20. The Rohtas Limestone in Son Valley has been dated by Pb–Pb isochrons to 1,601±130 Ma
[110], 1,599±48 Ma [112], and 1,514±120 Ma [113], respectively. These dates are in agreement
with the more precise U–Pb dates of 1,602±10 Ma and 1,593±12 Ma obtained from euhedral
magmatic zircon grains in tuffs from the immediately underlying Rampur Shale [114]. Further
down in the Son Valley sequence, tuff layers in the Deonar Porcellanite have been dated with
the same zircon U–Pb method by two independent laboratories to 1,631±5 Ma, 1,631±1 Ma
[115], and 1,628±8 Ma [114], respectively. Still further down, the Kajrahat Limestone has
yielded two Pb–Pb isochron dates of 1,707±190 Ma and 1,729±110 Ma [112]. From the Kajra-
hat and Rohtas Limestones have also been published two Rb–Sr dates of 1,014±18 Ma and 939
±22 Ma, respectively [116]. These anomalous dates may be affected by late illite formation
[112] but in any case do not support an Ediacaran–Cambrian age of the Semri Group.
A kimberlite pipe at Majhgawan, about 30 km southwest of Chitrakoot, intrudes the Kai-
mur Sandstone, the lowermost unit of the Upper Vindhyan sequence, which unconformably
overlies the Semri Group. The Kimberlite has yielded a mean Ar–Ar age of 1,073.5±13.7 Ma
[117], confirming the pre-Neoproterozoic age of the Lower Vindhyans. For broader reviews of
geochronological work on the Vindhyans, see Ray [106], Azmi et al. [108], and Basu and Bick-
ford [107].
The presumed correlative equivalents of the Tirohan Dolomite in the Son Valley have thus
been dated with precision and congruence to be very close to 1.6 Ga. This is within error mar-
gins identical to the 1,650±89 Ma age obtained from a Pb–Pb isochron dating directly of the
fossiliferous phosphorite in the Tirohan Dolomite at Chitrakoot [52]. Also from the Chitra-
koot area, Kumar et al. [118] obtained six Rb–Sr dates from glauconies in the Chitrakoot For-
mation, ranging from 1,409±14 Ma to 1,531±15 Ma. Although younger by some 100–200

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 26 / 38

 Red algae at 1.6 Ga

Fig 20. Summary of recent published results from radiometric age determinations of Lower Vindhyan
rocks in the Son Valley and the Chitrakoot area. The shaded band indicates correlation of the Tirohan
Dolomite (Chitrakoot Formation) with the Rohtas Limestone. Uncertainties are 2σ.
doi:10.1371/journal.pbio.2000735.g020

million years than the above cited dates, the model Rb–Sr ages from the glauconite remain in
agreement with an old, Palaeo–Mesoproterozoic age of the Lower Vindhyan. The compara-
tively young age obtained by Kumar et al. may be due to the well-known tendency of Palaeo-
zoic or older glauconite to yield Rb–Sr ages that appear to be too young by 10%–20% [119], a
phenomenon that can persist even through leaching experiments using a variety of agents
[120], including the HCl leaching performed by Kumar et al. [118]. Ages that are apparently
too young even after leaching may also indicate later recrystallization of glauconite [120],
although Kumar et al. [118] consider this unlikely for their samples.
Biostratigraphic arguments have been forwarded in support of a considerably younger age,
Ediacaran–Cambrian, of the Lower Vindhyans [55,108,109]. With regard to the claims of
Cambrian fossils in the Tirohan Dolomite [55,56,108], these were discussed and refuted by us
[52] as being based on misidentified algae (Rafatazmia and Denaricion herein), cyanobacterial
biomat fragments, and gas bubbles. The only cited radiometric date that would support an

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 27 / 38

 Red algae at 1.6 Ga

Ediacaran–Cambrian age of the Son Valley Lower Vindhyans is a “preliminary” Ar–Ar date of
617±3.5 Ma from the Deonar Porcellanite Formation, published only in a conference abstract
[121] and never substantiated.
A recent study of drill cores through allegedly Lower Vindhyan sedimentary rocks in the
western Vindhyan sub-basin, Chambal Valley, Rajasthan, revealed a succession of diverse and
well-preserved acanthomorph acritarchs of Ediacaran aspect [109]. The authors offer two
alternative interpretations of this discovery: the Semri Group is Ediacaran in age or complex
acanthomorph acritrarchs evolved much earlier than hitherto supposed. An additional alterna-
tive may be that the correlation of the investigated sequence with the Lower Vindhyan in Son
Valley/Chitrakoot is incorrect. A major problem in Vindhyan stratigraphy is the lack of reli-
able correlation between the eastern (Son Valley and Chitrakoot) and western (Rajasthan)
sub-basins owing to differences in lithology, lack of common marker horizons, lack of congru-
ence in chemostratigraphy, lack of biostratigraphically useful fossils, and lack of continuous
outcrops [110,122]. The Chitrakoot Formation indeed contains large acritarchs with complex
wall structure [40,54], but these are not acanthomorphs and rather seem comparable to the
assemblages in the ~1.7 Ga Ruyang Group in northern China [38] and the 1.5–1.4 Ga Roper
Group in northern Australia [123].
The ~1.6 Ga geochronological date of the Tirohan Dolomite is robust. This conclusion is
based on a congruence of radiometric measurements, but particularly important are the U–Pb
analyses of magmatic zircon, the most precise and accurate absolute dating method for rocks
of this age, in the Son Valley Lower Vindhyan [114,115], as well as the direct Pb–Pb isochron
dating of the fossiliferous phosphorite [52]. Apparent conflicts with biostratigraphic consider-
ations are likely to be resolved through reexamination of the fossil evidence [52], but in general
they serve as an indication that the Proterozoic fossil record of both micro- and megafossils is
richer than hitherto recognized. The preservation window opened by the phosphorite deposit
in the Tirohan Dolomite thus underscores that we have been missing a lot of the biotic diver-
sity in the Palaeo–Mesoproterozoic because of the dearth of lithologies suitable for fossilization
of delicate tissues.

Rhodophyte phylogeny and age

Together with chlorophytes and glaucophytes, the rhodophytes belong to the Archaeplastida
[124], which seem to have derived their chloroplasts from a single endosymbiotic event
involving a cyanobacterial donor [125]. Other algal groups obtained their chloroplasts from
secondary endosymbiosis with different archaeoplastid groups [126]. The Vindhyan putative
rhodophytes, having already diversified into multicellular filamentous and lobate forms,
would have been subsequent to the primary symbiotic event, and their photosynthetic ability
is suggested by the presence of pyrenoid-like structures in Rafatazmia and an apparent pho-
totactic growth mode in Ramathallus.
Yoon et al. [127] recognized seven rhodophyte lineages, among which the florideophyceans
and bangiophyceans were united as sister groups within one clade. The basal phylogenetic
relationships were largely unresolved, although the unicellular acidophilic Cyanidiophyceae
came out as a sister group to the remaining rhodophytes.
The presented data suggest that crown-group rhodophytes—the putative florideophycean
Ramathallus and possible bangiophycean Rafatazmia—were present in the ~1.6 Ga Vindhyan
stromatolitic microbialites. Whether or not these two taxa belong to the extant clades as pro-
posed, they seem to be part of the unresolved clade that forms a sister group to the unicellular
cyanidiophyceans and so would belong to crown-group rhodophytes. This runs contrary to
recent concepts about the evolutionary timing of eukaryote lineages. Several recent molecular-

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 28 / 38

 Red algae at 1.6 Ga

clock studies [16,17,48] have had problems in using the ~1.2 Ga Bangiomorpha [46] as a cali-
bration point for bangiophycean or even rhodophyte appearance, because it had seemed too
old to make sense.
Molecular-clock estimates of lineage divergence times in the tree of life are heavily depen-
dent on calibration based on occurrences and events in the fossil record. Uncertainties in
fossil calibration have a dominant effect on the dating of nodes in molecular phylogenetic
trees [128,129]. The first representative of a lineage in the fossil record postdates the origina-
tion of that lineage by an unknown interval of time; thus, the age of a fossil only gives a
minimum age for the lineage. Depending on the fossilization potential of the particular
organisms and the availability of suitable rocks, the interval of uncertainty can be of any
length and may be more or less easy to assess. For the Phanerozoic, the richness of the rock
record and the presence of easily fossilized organisms allow the construction of reasonably
reliable calibration intervals [130], but the Proterozoic is by comparison unchartered terri-
tory in which fossiliferous deposits are scarce and fossils often controversial in nature. Calcu-
lated Proterozoic divergence dates are commonly based on extrapolations from Phanerozoic
calibration intervals.
Berney and Pawlowski [48] based a calibration of the eukaryote tree on the Phanerozoic
microfossil record of coccolithophorids, diatoms, and dinoflagellates. Using a Bayesian
relaxed-clock model, they arrived at dramatically different dates for the divergence of major
eukaryote groups depending on whether or not Proterozoic fossils were included in the cali-
bration. When Bangiomorpha was used to set a minimum age for the red algae, the resulting
age for LECA was 3,868 Ma, with a 95% confidence interval of 4,182–1,830 Ma, whereas exclu-
sion of Bangiomorpha gave a LECA age of 1,126 Ma (1,357–947 Ma). For this reason, they con-
cluded that Bangiomorpha and other proposed Proterozoic algal and fungal taxa more likely
represent basal lineages of eukaryotes convergent on extant lineages and that calibration
should be based only on Phanerozoic data.
Parfrey et al. [17] similarly compared divergence dates depending on whether or not Prote-
rozoic fossils were used for calibration. Their relaxed-clock results were more stable to the
inclusion/exclusion of Proterozoic fossils than those of Berney and Pawlowski [48], yet they
found the origination date of the rhodophytes to shift considerably, from 1,285–1,180 Ma
(95% credible range) to 959–625 Ma, when Proterozoic taxa were excluded. Parfrey et al. [17]
favoured the inclusion of the Proterozoic fossils for calibration, leading to a LECA age of
1,866–1,679 Ma, older than that of most modern molecular-clock estimates. This result, how-
ever, required Bangiomorpha to be placed at the base of the crown-group rhodophytes, not
with the bangiophyceans.
Sharpe et al. [16], using a relaxed-clock model, estimated the origin of rhodophytes to 996
Ma, with a 95% credible interval of 1,186–801 Ma. They found Bangiomorpha, treated as a
stem-group multicellular rhodophyte lineage, to be at odds with the calculations and excluded
it as a calibration point. They argued that unless Bangiomorpha was not a red alga or incor-
rectly dated, currently used Bayesian relaxed-clock models might be insufficient to deal with
changes in evolutionary rates within lineages.
Yang et al. [47] accepted Bangiomorpha as a rhodophyte, though again not as a bangiophy-
cean but as an early rhodophyte stem group, for calibration. When testing two approaches to
assigning prior distributions within their Bayesian relaxed-clock model, they found that the
date for bangiophycean/florideophycean divergence shifted from 943 Ma to 1,661 Ma, and
because the older date seemingly contradicted the fossil evidence, they preferred the approach
yielding the lower age. They also found that their results were sensitive to the inclusion of Ban-
giomorpha, but having included it, they arrived at a rhodophyte/chlorophyte divergence date
of 1,693 Ma (1,925–1,484 Ma). This is older even than the LECA age of most studies.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 29 / 38

 Red algae at 1.6 Ga

There remain a number of theoretical problems to be resolved with molecular-clock datings

of ancient divergences, but the pattern that emerges from the above examples is that current
models are not adequate to deal with a minimum-age constraint for crown-group multicellular
rhodophytes at 1.2 Ga, as defined by Bangiomorpha. In order to fit the pattern of calibrations
based mainly on Phanerozoic data, Bangiomorpha has to be reclassified as a stem-group
eukaryote, as a stem-group rhodophyte, or, at best, near the first representative of the crown-
group rhodophytes.
Rafatazmia and Ramathallus are some 400 million years older than Bangiomorpha and are
tentatively interpreted as representing multicellular crown-group rhodophytes already diversi-
fied into filamentous and fleshy forms. This would change the constraints dramatically for the
timing of divergences within early rhodophytes and within Eukaryota as a whole. Maybe this
can be accommodated by taking into account previously downplayed possibilities, such as
changes in substitution rates in multiple lineages over time [131,132], but the basic task should
be to discover and study Archaean and Proterozoic fossils and base interpretations on the
available evidence rather than on how they happen to fit into currently accepted time frames
[133–137].

Systematic palaeontology
Genus Rafatazmia Bengtson n. gen.
• Type and only species. Rafatazmia chitrakootensis Bengtson n. sp.
• Etymology. In honour of Dr. Rafat Azmi. The name is feminine in gender.
• Diagnosis. Nonbranching filamentous alga, 58–175 μm in width. Cells of different sizes, with
lengths from half to more than twice the width. Diffuse growth by septation. New septa with
central pore; some septa also with central globular object, sometimes penetrating the septum;
these structures are possible pit connections and pit plugs. Large rhomboidal disk suspended
in the middle of each cell; interpreted as a pyrenoid.

Rafatazmia chitrakootensis Bengtson n. sp. Figs 2–6, 8 and 9.

• Holotype. Swedish Museum of Natural History NRM X4258 (Fig 5A–5L).
• Type stratum. Tirohan Dolomite, Jankikund section, Chitrakoot, India. 25˚090 46@N, 080˚
520 05@E.
• Etymology. From its occurrence at Chitrakoot.
• Diagnosis. As for the genus.

Genus Denaricion Bengtson n. gen.

• Type and only species. Denaricion mendax Bengtson n. sp.
• Etymology. From Latin denarius, a silver coin, and Greek kion, “pillar,” referring to the like-
ness of the internal structure to a stack of coins. The name is masculine in gender.
• Diagnosis. Nonbranching filamentous alga or prokaryote, 130–275 μm in width. Cells short,
coin-shaped. Growth by coordinated septation producing a geometric 4/8/16 pattern of cells
within compartments that are at least partly sealed from each other by a solid septum.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 30 / 38

 Red algae at 1.6 Ga

Denaricion mendax Bengtson n. sp. Figs 10 and 11.

• Holotype. Swedish Museum of Natural History NRM X5644 (Fig 10A–10C).
• Type stratum. Tirohan Dolomite, Jankikund section, Chitrakoot, India. 25˚090 32@N, 080˚
520 52@E.
• Etymology. Latin mendax, “deceptive,” for its enigmatic nature.
• Diagnosis. As for the genus.

Genus Ramathallus Sallstedt n. gen.

• Type and only species. Ramathallus lobatus Sallstedt n. sp.
• Etymology. From Rama and Latin thallus, “branch.” The name is masculine in gender.
• Diagnosis. Lobate thallus with globular and finger-like branches diverging from a central
area. Pseudoparenchymatous tissue often forming “cell fountains.” Cell sizes variable from
about 5 to 15 μm. Apparent tetraspores occur in globular bodies. Noncellular matrix up to
40 μm thick covers the thallus. No distinct cortex, but distal layer of cells in a branch are
sometimes larger than those of the adjacent pseudoparenchymatous tissue.

Ramathallus lobatus Sallstedt, n. sp. Figs 12–19.

• Holotype. Swedish Museum of Natural History NRM X5638, large specimen in Fig 13.
• Type stratum. Tirohan Dolomite, Jankikund section, Chitrakoot, India. 25˚090 46@N, 080˚
520 06@E.
• Etymology. Latin lobatus, “lobate.”
• Diagnosis. As for the genus.

Materials and methods

The studied material derives from field work at Jankikund in November 2006 and January
2011 and was augmented by microfossils kindly supplied by R. Azmi. To isolate phosphatized
fossils from the carbonate rocks, samples were dissolved in 10% buffered acetic acid, and the
residues were manually picked for microfossils. SEM/ESEM was carried out on a Hitachi
S4300 Field Emission scanning electron microscope, a Philips XL 30 ESEM-FEG, and a
Quanta ESEM-FEG 650 field emission microscope. Synchrotron-radiation X-ray tomographic
microscopy (SRXTM) was done at the X02DA TOMCAT beamline at the Swiss Light Source,
Paul Scherrer Institut, Villigen, Switzerland. Visualization was done using Avizo 9.1.1 (FEI
Company). All specimens are in the collections of the Swedish Museum of Natural History.
Each figured specimen has a unique museum number, stated in the respective figure caption
in the article. Tomographic data are deposited in the Dryad Digital Repository: http://dx.doi.
org/10.5061/dryad.gh221 [138].

Nomenclature
The electronic version of this article in Portable Document Format (PDF) in a work with an
ISSN or ISBN will represent a published work according to the International Code of

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 31 / 38

 Red algae at 1.6 Ga

Nomenclature for algae, fungi, and plants, and hence, the new names contained in the elec-
tronic publication of a PLOS article are effectively published under that Code from the elec-
tronic edition alone, so there is no longer any need to provide printed copies. The online
version of this work is archived and available from the following digital repositories: PubMed
Central, LOCKSS, and DiVA (www.diva-portal.org/).

Supporting information
S1 Data. Data for Fig 7. Measurements of filamentous taxa used to produce Fig 7. See figure
text for literature sources.
(XLSX)

Acknowledgments
We thank Rafat Azmi, Christina Franzén-Bengtson, Anindya Sarkar, Melinda Bera, Arpita
Samanta, Arvind Singh, and Partha Chakraborty for guidance and assistance with fieldwork.
Marco Stampanoni and Federica Marone gave expertise and help at the TOMCAT synchro-
tron beamline.

Author Contributions
Conceptualization: Stefan Bengtson, Therese Sallstedt.
Data curation: Stefan Bengtson, Therese Sallstedt, Veneta Belivanova.
Formal analysis: Stefan Bengtson.
Investigation: Stefan Bengtson, Therese Sallstedt, Veneta Belivanova, Martin Whitehouse.
Methodology: Stefan Bengtson.
Project administration: Stefan Bengtson.
Resources: Stefan Bengtson.
Supervision: Stefan Bengtson.
Validation: Stefan Bengtson.
Visualization: Stefan Bengtson, Therese Sallstedt.
Writing – original draft: Stefan Bengtson.
Writing – review & editing: Stefan Bengtson, Therese Sallstedt, Veneta Belivanova, Martin
Whitehouse.

References
1. Butterfield NJ. Animals and the invention of the Phanerozoic Earth system. Trends in Ecology and
Evolution. 2011; 26(2):81–7. doi: 10.1016/j.tree.2010.11.012 PMID: 21190752
2. James TY, Kauff F, Schoch CL, Matheny PB, Hofstetter V, Cox CJ, et al. Reconstructing the early evo-
lution of Fungi using a six-gene phylogeny. Nature. 2006; 443(7113):818–22. doi: 10.1038/
nature05110 PMID: 17051209
3. Lücking R, Huhndorf S, Pfister DH, Plata ER, Lumbsch HT. Fungi evolved right on track. Mycologia.
2009; 101(6):810–22. PMID: 19927746
4. Taylor TN, Krings M, Taylor EL. Fossil Fungi. Amsterdam: Elsevier; 2015.
5. Xiao S. Written in stone: the fossil record of early eukaryotes. In: Trueba G, Montúfar C, editors. Evolu-
tion from the Galapagos, Social and Ecological Interactions in the Galapagos Islands 2. New York:
Springer; 2013. p. 107–24.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 32 / 38

 Red algae at 1.6 Ga

6. Eme L, Sharpe SC, Brown MW, Roger AJ. On the age of eukaryotes: evaluating evidence from fossils
and molecular clocks. Cold Spring Harbor Perspectives in Biology. 2014; 6:a016139:14 pp.
7. Knoll AH. Paleobiological perspectives on early eukaryotic evolution. Cold Spring Harbor Perspectives
in Biology. 2014; 6:a016121:14 pp.
8. Cohen PA, Macdonald FA. The Proterozoic record of eukaryotes. Paleobiology. 2015; 41(4):610–32.
9. Butterfield NJ. Early evolution of the Eukaryota. Palaeontology. 2015; 58(1):5–17.
10. Cavalier-Smith T. Predation and eukaryote cell origins: A coevolutionary perspective. The Interna-
tional Journal of Biochemistry & Cell Biology. 2009; 41:307–22.
11. Hedges SB, Blair JE, Venturi M, Shoe JL. A molecular timescale of eukaryote evolution and the rise of
complex multicellular life. BMC Evol Biol. 2004; 4(2):1–9.
12. Yoon HS, Hackett JD, Ciniglia C, Pinto G, Bhattacharya D. A molecular timeline for the origin of photo-
synthetic eukaryotes. Mol Biol Evol. 2004; 21(5):809–18. doi: 10.1093/molbev/msh075 PMID:
14963099
13. Douzery EJP, Snell EA, Bapteste E, Delsuc F, Philippe H. The timing of eukaryotic evolution: Does a
relaxed molecular clock reconcile proteins and fossils? Proc Natl Acad Sci U S A. 2004; 101
(43):15386–91. doi: 10.1073/pnas.0403984101 PMID: 15494441
14. Bhattacharya D, Yoon HS, Hedges SB, Hackett JD. Eukaryotes (Eukaryota). In: Hedges SB, Kumar
S, editors. The Timetree of Life. Oxford: Oxford UP; 2009. p. 116–20.
15. Chernikova D, Motamedi S, Csürös M, Koonin EV, Rogozin IB. A late origin of the extant eukaryotic
diversity: divergence time estimates using rare genomic changes. Biology Direct. 2011; 6(26):18 pp.
16. Sharpe SC, Eme L, Brown MW, Roger AJ. Timing the origins of multicellular eukaryotes through
phylogenomics and relaxed molecular clock analyses. In: Ruiz-Trillo I, Nedelcu AM, editors. Evolu-
tionary Transitions to Multicellular Life. Advances in Marine Genomics. Dordrecht: Springer; 2015.
p. 3–29.
17. Parfrey LW, Lahr DJG, Knoll AH, Katz LA. Estimating the timing of early eukaryotic diversification with
multigene molecular clocks. Proc Natl Acad Sci USA. 2011; 108(33):13624–9. doi: 10.1073/pnas.
1110633108 PMID: 21810989
18. Keeling PJ, Gertraud Burger G, Durnford DG, Lang BF, Lee RW, Pearlman RE, et al. The tree of
eukaryotes. Trends in Ecology and Evolution. 2005; 20(12):670–6. doi: 10.1016/j.tree.2005.09.005
PMID: 16701456
19. Sanderson MJ. Phylogenetic signal in the eukaryotic tree of life. Science. 2008; 321:121–3. doi: 10.
1126/science.1154449 PMID: 18599787
20. Baldauf SL. An overview of the phylogeny and diversity of eukaryotes. Journal of Systematics and
Evolution. 2008; 46(3):263–73.
21. Gribaldo S, Poole AM, Daubin V, Forterre P, Brochier-Armanet C. The origin of eukaryotes and their
relationship with the Archaea: are we at a phylogenomic impasse? Nature Reviews Microbiology.
2010; 8(10):743–52. doi: 10.1038/nrmicro2426 PMID: 20844558
22. Rochette NC, Brochier-Armanet C, Gouy M. Phylogenomic test of the hypotheses for the evolution-
ary origin of eukaryotes. Mol Biol Evol. 2014; 31(4):832–45. doi: 10.1093/molbev/mst272 PMID:
24398320
23. Cavalier-Smith T, Chao EE, Snell EA, Berney C, Fiore-Donno AM, Lewis R. Multigene eukaryote phy-
logeny reveals the likely protozoan ancestors of opisthokonts (animals, fungi, choanozoans) and
Amoebozoa. Mol Phylogen Evol. 2014; 81:71–85.
24. Borrel G, Adam PS, Gribaldo S. Methanogenesis and the Wood-Ljungdahl pathway: an ancient, ver-
satile, and fragile association. Genome Biology and Evolution. 2016; 8(6):1706–11. doi: 10.1093/gbe/
evw114 PMID: 27189979
25. Budd GE, Jensen S. A critical reappraisal of the fossil record of the bilaterian phyla. Biological
Reviews. 2000; 75:253–95. PMID: 10881389
26. Lester L, Meade A, Pagel M. The slow road to the eukaryotic genome. Bioessays. 2006; 28(1):57–64.
doi: 10.1002/bies.20344 PMID: 16369937
27. Graham LE, Cook ME, Wilcox LW, Graham J, Taylor W, Wellman CH, et al. Resistance of filamentous
chlorophycean, ulvophycean, and xanthophycean algae to acetolysis: testing Proterozoic and Paleo-
zoic microfossil attributions. Int J Plant Sci. 2013; 174(6):947–57.
28. Agić H, Moczydłowska M, Yin L-M. Affinity, life cycle, and intracellular complexity of organic-walled
microfossils from the Mesoproterozoic of Shanxi, China. J Paleontol. 2015; 89(1):28–50.
29. Samuelsson J, Butterfield NJ. Neoproterozoic fossils from the Franklin Mountains, northwestern Can-
ada; stratigraphic and palaeobiological implications. Precambrian Res. 2001; 107(3–4):235–51.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 33 / 38

 Red algae at 1.6 Ga

30. Walter MR, Oehler JH, Oehler DZ. Megascopic algae 1300 million years old from the Belt Supergroup,
Montana: A reinterpretation of Walcott’s Helminthoidichnites. J Paleontol. 1976; 50(5):872–81.
31. Han T-M, Runnegar B. Megascopic eukaryotic algae from the 2.1 billion-year-old Negaunee Iron-For-
mation, Michigan. Science. 1992; 257:232–5. PMID: 1631544
32. Fedonkin M. The origin of the Metazoa in the light of the Proterozoic fossil record. Paleontol Res.
2003; 7(1):9–41.
33. Porter SM. The fossil record of early eukaryotic diversification. In: Lipps JH, Waggoner BM, editors.
Neoproterozoic–Cambrian Biological Revolutions. The Paleontological Society Papers 10: The Pale-
ontological Society; 2004. p. 35–50.
34. Xiao S, Muscente AD, Chen L, Zhou C, Schiffbauer JD, Wood AD, et al. The Weng’an biota and the
Ediacaran radiation of multicellular eukaryotes. National Science Review. 2014; 1(4):498–520.
35. Cavalier-Smith T. Cell evolution and Earth history: stasis and revolution. Philosophical Transactions of
the Royal Society of London B Biological Sciences. 2006; 361(1470):969–1006. doi: 10.1098/rstb.
2006.1842 PMID: 16754610
36. Sharma M, Shukla Y. Taxonomy and affinity of Early Mesoproterozoic megascopic helically coiled and
related fossils from the Rohtas Formation, the Vindhyan Supergroup, India. Precambrian Res. 2009;
173(1–4):105–22.
37. Yan Y-z, Liu Z-l. Tuanshanzian macroscopic algae of 1700 Ma B.P. from Changcheng System of Jix-
ian, China. Acta Palaeontol Sin. 1997; 36(1):18–41.
38. Peng Y, Bao H, Yuan X. New morphological observations for Paleoproterozoic acritarchs from the
Chuanlinggou Formation, North China. Precambrian Res. 2009; 168:223–32.
39. Moczydłowska M, Landing E, Zang W, Palacios T. Proterozoic phytoplankton and timing of chloro-
phyte algae origins. Palaeontology. 2011;online.
40. Singh VK, Sharma M. Morphologically complex organic-walled microfossils (OWM) from the
late Palaeoproterozoic- early Mesoproterozoic Chitrakut Formation, Vindhyan Supergroup, central
India and their implications on the antiquity of eukaryotes. J Palaeontol Soc India. 2014; 59(1):89–
102.
41. Brocks JJ, Logan GA, Buick R, Summons RE. Archean molecular fossils and the early rise of eukary-
otes. Science. 1999; 285:1033–6. PMID: 10446042
42. Brocks JJ, Buick R, Logan GA, Summons RE. Composition and syngeneity of molecular fossils from
the 2.78 to 2.45 billion-year-old Mount Bruce Supergroup, Pilbara Craton, Western Australia. Geochim
Cosmochim Acta. 2003; 67(22):4289–319.
43. Rasmussen B, Fletcher IR, Brocks JJ, Kilburn MR. Reassessing the first appearance of eukaryotes
and cyanobacteria. Nature. 2008; 455:1101–4. doi: 10.1038/nature07381 PMID: 18948954
44. French KL, Hallmann C, Hope JM, Schoon PL, Zumberge JA, Hoshino Y, et al. Reappraisal of hydro-
carbon biomarkers in Archean rocks. Proc Natl Acad Sci USA. 2015; 112(19):5915–20. doi: 10.1073/
pnas.1419563112 PMID: 25918387
45. Kirschvink JL, Kopp RE. Palaeoproterozoic ice houses and the evolution of oxygen-mediating
enzymes: the case for a late origin of photosystem II. Philosophical Transactions of the Royal Society
B Biological Sciences. 2008; 208:2755–65.
46. Butterfield NJ. Bangiomorpha pubescens n.gen., n.sp.: implications for the evolution of sex, multicellu-
larity, and the Mesoproterozoic/Neoproterozoic radiation of eukaryotes. Paleobiology. 2000; 26
(3):386–404.
47. Yang EC, Boo SM, Bhattacharya D, Saunders GW, Knoll AH, Fredericq S, et al. Divergence time esti-
mates and the evolution of major lineages in the florideophyte red algae. Scientific Reports. 2016; 6
(21361):11 pp.
48. Berney C, Pawlowski J. A molecular time-scale for eukaryote evolution recalibrated with the continu-
ous microfossil record. Proc R Soc Biol Sci Ser B. 2006; 273:1867–72.
49. Tappan H. Possible eucaryotic algae (Bangiophycidae) among early Proterozoic microfossils. Geol
Soc Am Bull. 1976; 87(4):673–9.
50. Awramik SM, Barghoorn ES. The Gunflint microbiota. Precambrian Res. 1977; 5:121–42.
51. Lamb DM, Awramik SM, Chapman DJ, Zhu S. Evidence for eukaryotic diversification in the ~1800 mil-
lion-year-old Changzhougou Formation, North China. Precambrian Res. 2009; 173(1–4):93–104.
52. Bengtson S, Belivanova V, Rasmussen B, Whitehouse M. The controversial ‘Cambrian’ fossils of the
Vindhyan are real but more than a billion years older. Proc Natl Acad Sci USA. 2009; 106:7729–34.
doi: 10.1073/pnas.0812460106 PMID: 19416859
53. Singh IB, Kumar S. On the stratigraphy and sedimentation of the Vindhyan sediments in the Chitrakut
area, Banda district (U.P.)–Satna district, (M.P.). J Geol Soc India. 1978; 19:359–67.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 34 / 38

 Red algae at 1.6 Ga

54. Anbarasu K. Acritarchs from Mesoproterozoic Chitrakoot Formation, Semri Group, Chitrakoot area,
Central India. J Geol Soc India. 2001; 57:179–83.
55. Azmi RJ, Joshi D, Tiwari BN, Joshi MN, Mohan K, Srivastava SS. Age of the Vindhyan Supergroup of
Central India: An exposition of biochronology vs radiochronology. In: Sinha D, editor. Micropaleontol-
ogy: Application in Stratigraphy and Paleoceanography. New Delhi: Narosa Publishing House; 2006.
p. 29–62.
56. Joshi D, Azmi RJ, Srivastava SS. Earliest Cambrian calcareous skeletal algae from Tirohan Dolomite,
Chitrakoot, Central India: a new age constraint for the Lower Vindhyan. Gondwana Geol Mag. 2006;
21(2):73–82.
57. Kumar S. Mineralogy, geochemistry and genesis of middle Riphean phosphatic carbonates, Tirohan
Limestone (Lower Vindhyan Supergroup), Chitrakut area, central India. J Geol Soc India. 1993; 41
(2):133–43.
58. Bengtson S, Budd G. Comment on “Small bilaterian fossils from 40 to 55 million years before the Cam-
brian”. Science. 2004; 306:1291a.
59. Xiao S, Schiffbauer JD. Microfossil phosphatization and its astrobiological implications. In: Seckbach
J, Walsh M, editors. From Fossils to Astrobiology: Springer; 2008. p. 89–117.
60. Cunningham JA, Thomas C-W, Bengtson S, Kearns SL, Xiao S, Marone F, et al. Distinguishing geol-
ogy from biology in the Ediacaran Doushantuo biota relaxes constraints on the timing of the origin of
bilaterians. Proc R Soc Biol Sci Ser B. 2012; 279(1737):2369–76.
61. Cunningham JA, Donoghue PCJ, Bengtson S. Distinguishing biology from geology in soft-tissue pres-
ervation. In: Laflamme M, Schiffbauer JD, Darroch SAF, editors. Reading and Writing of the Fossil
Record: Preservational Pathways to Exceptional Fossilization The Paleontological Society Papers
202014. p. 275–87.
62. Graham LE, Graham JM, Wilcox LW. Algae. San Francisco: Benjamin Cummings; 2009. 720 pp.
63. Schulz HN, Jørgensen BB. Big bacteria. Annu Rev Microbiol. 2001; 55(1):105–37.
64. Demoulin V, Janssen MP. Relationship between diameter of the filament and cell shape in blue-green
algae. Br Phycol J. 1981; 16(1):55–8.
65. Larkin JM, Henk MC. Filamentous sulfide-oxidizing bacteria at hydrocarbon seeps of the Gulf of
Mexico. Microsc Res Tech. 1996; 33(1):23–31. doi: 10.1002/(SICI)1097-0029(199601)33:1<23::AID-
JEMT4>3.0.CO;2-1 PMID: 8820662
66. Strohl WR, Larkin JM. Cell division and trichome breakage in Beggiatoa. Curr Microbiol. 1978; 1
(3):151–5. doi: 10.1007/BF02601668 PMID: 23338140
67. Larkin J, Aharon P, Henk MC. Beggiatoa in microbial mals at hydrocarbon vents in the Gulf of Mexico
and Warm Mineral Springs, Florida. Geo-Mar Lett. 1994; 14(2–3):97–103.
68. Nelson DC, Wirsen CO, Jannasch HW. Characterization of large, autotrophic Beggiatoa spp. abun-
dant at hydrothermal vents of the Guaymas Basin. Appl Environ Microbiol. 1989; 55(11):2909–17.
PMID: 16348053
69. Jean MRN, Gonzalez-Rizzo S, Gauffre-Autelin P, Lengger SK, Schouten S, Gros O. Two new Beggia-
toa species inhabiting marine mangrove sediments in the Caribbean. PLoS ONE. 2015; 10(2):
e0117832. doi: 10.1371/journal.pone.0117832 PMID: 25689402
70. Mußmann M, Schulz HN, Strotmann B, Kjær T, Nielsen LP, Rosselló-Mora RA, et al. Phylogeny and
distribution of nitrate-storing Beggiatoa spp. in coastal marine sediments. Environ Microbiol. 2003; 5
(6):523–33. PMID: 12755720
71. Schopf JW, Oehler DZ. How old are the eukaryotes? Science. 1976; 193:47–9. doi: 10.1126/science.
193.4247.47 PMID: 17794005
72. Schulz HN, Brinkhoff T, Ferdelman TG, Hernández Mariné M, Teske A, Jørgensen BB. Dense popula-
tions of a giant sulfur bacterium in Namibian shelf sediments. Science. 1999; 284:493–5. PMID:
10205058
73. Cunningham JA, Thomas C-W, Bengtson S, Marone F, Stampanoni M, Turner FR, et al. Experimental
taphonomy of giant sulphur bacteria: implications for the interpretation of the embryo-like Ediacaran
Doushantuo fossils. Proc R Soc Biol Sci Ser B. 2012; 279(1734):1857–64.
74. Schopf JW. Microflora of the Bitter Springs Formation, Late Precambrian, Central Australia. J Paleon-
tol. 1968; 42:651–88.
75. Schopf JW, Blacic JM. New microorganisms from the Bitter Springs Formation (Late Precambrian) of
the north-central Amadeus Basin, Australia. J Paleontol. 1971; 45:925–59.
76. Knoll AH, Barghoorn ES. Precambrian eukaryotic organisms: A reassessment of the evidence. Sci-
ence. 1975; 190(4209):52–4.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 35 / 38

 Red algae at 1.6 Ga

77. Oehler DZ. Pyrenoid-like structures in Late Precambrian algae from the Bitter Springs Formation of
Australia. J Paleontol. 1977; 51:885–901.
78. Huldtgren T, Cunningham J, Yin C, Stampanoni M, Marone F, Donoghue PCJ, et al. Fossilized nuclei
and germination structures identify Ediacaran ‘animal embryos’ as encysting protists. Science. 2011;
334(6063):1696–9. doi: 10.1126/science.1209537 PMID: 22194575
79. Huldtgren T, Cunningham J, Yin C, Stampanoni M, Marone F, Donoghue PCJ, et al. Response to
comment on “Fossilized nuclei and germination structures identify Ediacaran ‘animal embryos’ as
encysting protists”. Science. 2012; 335(6073):1169-d (2 pp.).
80. Schiffbauer JD, Xiao S, Sharma KS, Wang G. The origin of intracellular structures in Ediacaran meta-
zoan embryos. Geology. 2012; 40(3):223–6.
81. Xiao S, Knoll AH, Schiffbauer JD, Zhou C, Yuan X. Comment on “Fossilized nuclei and germination struc-
tures identify Ediacaran ‘animal embryos’ as encysting protists”. Science. 2012; 335:1169-c (3 pp.).
82. Pang K, Tang Q, Schiffbauer JD, Yao J, Yuan X, Wan B, et al. The nature and origin of nucleus-like
intracellular inclusions in Paleoproterozoic eukaryote microfossils. Geobiology. 2013; 11:499–510.
doi: 10.1111/gbi.12053 PMID: 24033870
83. Bomfleur B, McLoughlin S, Vajda V. Fossilized nuclei and chromosomes reveal 180 million years of
genomic stasis in royal ferns. Science. 2014; 343:1376–7. doi: 10.1126/science.1249884 PMID:
24653037
84. Yeates TO, Kerfeld CA, Heinhorst S, Cannon GC, Shively JM. Protein-based organelles in bacteria:
carboxysomes and related microcompartments. Nature Reviews Microbiology. 2008; 6(9):681–91.
doi: 10.1038/nrmicro1913 PMID: 18679172
85. Griffiths DJ. The pyrenoid. The Botanical Review. 1970; 36(1):29–58.
86. Badger MR, Andrews TJ, Whitney SM, Ludwig M, Yellowlees DC, Leggat W, et al. The diversity and
coevolution of Rubisco, plastids, pyrenoids, and chloroplast-based CO2-concentrating mechanisms in
algae. Can J Bot. 1998; 76(6):1052–71.
87. Timberlake HG. Starch-formation in Hydrodictyon utriculatum. Ann Bot. 1901; 15(60):619–35.
88. Graham LE, Graham JM, Wilcox LW, Cook ME. Algae. 3rd Ed. ed. Madison WI: LJLM Press; 2016.
595 pp.
89. Yang EC, Scott J, West JA, Yoon HS, Yokoyama A, Karsten U, et al. Erythrolobus australicus sp. nov.
(Porphyridiophyceae, Rhodophyta): a description based on several approaches. Algae. 2011; 26
(2):167–80.
90. Freshwater DW, Fredericq S, Butler BS, Hommersand MH, Chase MW. A gene phylogeny of the red
algae (Rhodophyta) based on plastid rbcL. Proc Natl Acad Sci USA. 1994; 91:7281–5. PMID:
8041781
91. Chapman RL, Bailey JC, Waters DA. Macroalgal phylogeny. In: Cooksey KE, editor. Molecular
Approaches to the Study of the Ocean. Dordrecht: Springer; 1998. p. 389–407.
92. Saunders GW, Hommersand MH. Assessing red algal supraordinal diversity and taxonomy in the con-
text of contemporary systematic data. Am J Bot. 2004; 91(10):1494–507. doi: 10.3732/ajb.91.10.1494
PMID: 21652305
93. Müller KM, Lynch MDJ, Sheath RG. Bangiophytes: from one class to six; where do we go from here?
Moving the bangiophytes into the genomic age. In: Seckbach J, Chapman DJ, editors. Red Algae in
the Genomic Age. Dordrecht: Springer; 2010. p. 243–59.
94. Schornstein KL, Scott J. Ultrastructure of cell division in the unicellular red alga Porphyridium purpur-
eum. Can J Bot. 1982; 60(1):85–97.
95. Vis ML, Sheath RG. Distribution and systematics of Chroodactylon and Kyliniella (Porphyridiales, Rho-
dophyta) from North American streams. Japanese Journal of Phycology. 1993; 41:237–41.
96. Wolowski K, Kowalska J, Hindák F. Chroodactylon ornatum (Rhodophyta, Porphyridiales) occurring in
Poland and Slovakia. Biologia, Bratislava. 2007; 62(6):646–9.
97. Sommerfeld MR, Nichols HW. Developmental and cytological studies of Bangia fuscopurpurea in cul-
ture. Am J Bot. 1970; 57(6):640–8.
98. Lee RE. Chloroplast structure and starch grain production as phylogenetic indicators in the lower Rho-
dophyceae. Br Phycol J. 1974; 9(3):291–5.
99. Smith GM. Cytological studies in the Protococcales. II. Cell structure and zoospore formation in Pedia-
strum Boryanum (Turp.), Menegh. Ann Bot. 1916; 30(3):467–79.
100. Pueschel CM, Cole KM. Rhodophycean pit plugs: an ultrastructural survey with taxonomic implica-
tions. Am J Bot. 1982; 69(5):703–20.
101. Komárek J. Coccoid and colonial cyanobacteria. In: Wehr JD, Sheath RG, editors. Freshwater Algae
of North America. New York: Elsevier; 2002. p. 59–116.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 36 / 38

 Red algae at 1.6 Ga

102. Zhang Y. Multicellular thallophytes with differentiated tissues from Late Proterozoic phosphate rocks
of South China. Lethaia. 1989; 22(2):113–32.
103. Zhang Y, Yuan X. New data on multicellular thallophytes and fragments of cellular tissues from Late
Proterozoic phosphate rocks, South China. Lethaia. 1992; 25(1):1–18.
104. Zhang Y, Yin L, Xiao S, Knoll AH. Permineralized fossils from the terminal Proterozoic Doushantuo
Formation, south China. J Paleontol. 1998; 72(4, Suppl.):1–52.
105. Xiao S, Knoll AH, Yuan X, Pueschel CM. Phosphatized multicellular algae in the Neoproterozoic
Doushantuo Formation, China, and the early evolution of florideophyte red algae. Am J Bot. 2004; 91
(2):214–27. doi: 10.3732/ajb.91.2.214 PMID: 21653378
106. Ray JS. Age of the Vindhyan Supergroup: A review of recent findings. Journal of Earth System Sci-
ence. 2006; 115(1):149–60.
107. Basu A, Bickford ME. Contributions of zircon U–Pb geochronology to understanding the volcanic and
sedimentary history of some Purana basins, India. J Asian Earth Sci. 2014; 91:252–62.
108. Azmi RJ, Joshi D, Tiwari BN, Joshi MN, Srivastava SS. A synoptic view on the current discordant geo-
and biochronological ages of the Vindhyan Supergroup, central India. Himal Geol. 2008; 29(2):177–
91.
109. Prasad B, Asher R. Record of Ediacaran complex acanthomorphic acritarchs from the Lower Vindh-
yan succession of the Chambal Valley (east Rajasthan), India and their biostratigraphic significance. J
Palaeontol Soc India. 2016; 61(1):29–62.
110. Ray JS, Veizer J, Davis WJ. C, O, Sr and Pb isotope systematics of carbonate sequences of the
Vindhyan Supergroup, India: age, diagenesis, correlations and implications for global events. Precam-
brian Res. 2003; 121(1–2):103–40.
111. Gupta S, Jain KC, Srivastava VC, Mehrotra RD. Depositional environment and tectonism during the
sedimentation of the Semri and Kaimur Groups of rocks, Vindhyan Basin. J Palaeontol Soc India.
2003; 48:181–90.
112. Sarangi S, Gopalan K, Kumar S. Pb–Pb age of earliest megascopic, eukaryotic alga bearing Rohtas
Formation, Vindhyan Supergroup, India: implications for Precambrian atmospheric oxygen evolution.
Precambrian Res. 2004; 132(1–2):107–21.
113. Chakrabarti R, Basu AR, Chakrabarti A. Trace element and Nd-isotopic evidence for sediment
sources in the mid-Proterozoic Vindhyan Basin, central India. Precambrian Res. 2007; 159(3–4):260–
74.
114. Rasmussen B, Bose PK, Sarkar S, Banerjee S, Fletcher IR, McNaughton NJ. 1.6 Ga U-Pb zircon age
for the Chorhat Sandstone, lower Vindhyan, India: Possible implications for early evolution of animals.
Geology. 2002; 30(2):103–6.
115. Ray JS, Martin MW, Veizer J, Bowring SA. U-Pb zircon dating and Sr isotope systematics of the
Vindhyan Supergroup, India. Geology. 2002; 30(2):131–4.
116. Bansal M, Vijan AR, Mahar YS, Ghosh N, Krishna PS, Prabhy BN. Rb–Sr dating of Lower Vindhyan
sediments from Son Valley, M. P., India. Indian Mineralogist. 1999; 33:1–9.
117. Gregory LC, Meert JG, Pradhan V, Pandit MK, Tamrat E, Malone SJ. A paleomagnetic and geochro-
nologic study of the Majhgawan kimberlite, India: Implications for the age of the Upper Vindhyan
Supergroup. Precambrian Res. 2006; 149(1–2):65–75.
118. Kumar A, Gopalan K, Rajagopalan G. Age of the Lower Vindhyan sediments, Central India. Curr Sci.
2001; 81(7):806–9.
119. Hurley PM, Cormier RF, Hower J, Fairbairn HW, Pinson WH Jr. Reliability of glauconite for age mea-
surement by K–Ar and Rb–Sr methods. American Association of Petroleum Geologists Bulletin. 1960;
44:1793–808.
120. Morton JP, Long LE. Rb–Sr dating of Paleozoic glauconite from the Llano region, central Texas. Geo-
chim Cosmochim Acta. 1980; 44:663–72.
121. Banerjee DM, Frank W. Preliminary 40Ar/39Ar dates of porcellanite and detrital mica from the Vindh-
yan of central India. Workshop on Vindhyan Stratigraphy and Palaeobiology—Volume of Abstracts.
Lucknow: University of Lucknow; 1999. p. 5.
122. Kumar S. Stratigraphy and correlation of the Neoproterozoic deposits of central and western India: an
overview. Geological Society, London, Special Publications. 2012; 366:75–90.
123. Javaux EJ, Knoll AH, Walter MR. Morphological and ecological complexity in early eukaryotic ecosys-
tems. Nature. 2001; 412:66–9. doi: 10.1038/35083562 PMID: 11452306
124. Adl SM, others. The new higher level classification of eukaryotes with emphasis on the taxonomy of
protists. J Eukaryot Microbiol. 2005; 52(5):399–451. doi: 10.1111/j.1550-7408.2005.00053.x PMID:
16248873

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 37 / 38

 Red algae at 1.6 Ga

125. Rodrı́guez-Ezpeleta N, Brinkmann H, Roure SC, Roure B, Burger G, Löffelhardt W, et al. Monophyly
of primary photosynthetic eukaryotes: Green plants, red algae, and glaucophytes. Curr Biol. 2005; 15
(14):1325–30. doi: 10.1016/j.cub.2005.06.040 PMID: 16051178
126. Yoon HS, Hackett JD, Bhattacharya D. A genomic and phylogenetic perspective on endosymbiosis
and algal origin. J Appl Phycol. 2006; 18(3–5):475–81.
127. Yoon HS, Müller KM, Sheath RG, Ott FD, Bhattacharya D. Defining the major lineages of red algae
(Rhodophyta). J Phycol. 2006; 42(2):482–92.
128. Inoue J, Donoghue PCJ, Yang Z. The impact of the representation of fossil calibrations on Bayesian
estimation of species divergence times. Syst Biol. 2010; 59(1):74–89. doi: 10.1093/sysbio/syp078
PMID: 20525621
129. dos Reis M, Donoghue PCJ, Yang Z. Bayesian molecular clock dating of species divergences in the
genomics era. Nat Rev Genet. 2016; 17(2):71–80. doi: 10.1038/nrg.2015.8 PMID: 26688196
130. Benton MJ, Donoghue PCJ, Asher RJ, Friedman M, Near TJ, Vinther J. Constraints on the timescale
of animal evolutionary history. Palaeontol Electron. 2015; 18(1):107 pp.
131. Levinton J, Dubb L, Wray GA. Simulations of evolutionary radiations and their application to under-
standing the probability of a Cambrian explosion. J Paleontol. 2004; 78(1):31–8.
132. Lee MSY, Soubrier J, Edgecombe GD. Rates of phenotypic and genomic evolution during the Cam-
brian explosion. Curr Biol. 2013; 23:1889–95. doi: 10.1016/j.cub.2013.07.055 PMID: 24035543
133. Rasmussen B, Bengtson S, Fletcher IR, McNaughton N. Discoidal impressions and trace-like fossils
more than 1200 million years old. Science. 2002; 296:1112–5. doi: 10.1126/science.1070166 PMID:
12004128
134. Bengtson S, Rasmussen B, Krapež B. The Paleoproterozoic megascopic Stirling Biota. Paleobiology.
2007; 33(3):351–81.
135. El Albani A, Bengtson S, Canfield DE, Bekker A, Macchiarelli R, Mazurier A, et al. Large colonial
organisms with coordinated growth behaviour in oxygenated environments 2.1 billion years ago.
Nature. 2010; 466:100–4. doi: 10.1038/nature09166 PMID: 20596019
136. El Albani A, Bengtson S, Canfield DE, Riboulleau A, Rollion Bard C, Macchiarelli R, et al. The 2.1 Ga
old Francevillian biota: biogenicity, taphonomy and biodiversity. PLoS ONE. 2014; 9(6):e99438. doi:
10.1371/journal.pone.0099438 PMID: 24963687
137. Javaux EJ, Marshall CP, Bekker A. Organic-walled microfossils in 3.2-billion-year-old shallow-marine
siliciclastic deposits. Nature. 2010; 463:934–8. doi: 10.1038/nature08793 PMID: 20139963
138. Bengtson S (2017) Data from: Three-dimensional preservation of cellular and subcellular structures
suggests 1.6 billion-year-old crown-group red algae. Dryad Digital Repository. Openly available via
http://dx.doi.org/10.5061/dryad.gh221.

PLOS Biology | DOI:10.1371/journal.pbio.2000735 March 14, 2017 38 / 38