Fertile Prototaxites taiti: A basal ascomycete with inoperculate, polysporous asci lacking croziers

Abstract


The affinities of Prototaxites have been debated ever since its fossils, some attaining tree-trunk proportions, were discovered in Canadian Lower Devonian rocks in 1859. Putative assignations include conifers, red and brown algae, liver- worts and fungi (some lichenised). Detailed anatomical investigation led to the reconstruction of the type species, P. logani, as a giant sporophore (basidioma) of an agaricomycete (1⁄4 holobasidiomycete), but evidence for its reproduction remained elusive. Tissues associated with P. taiti in the Rhynie chert plus charcoalified fragments from southern Britain are investigated here to describe the reproductive characters and hence affinities of Prototaxites. Thin sections and peels (Pragian Rhynie chert, Aberdeenshire) were examined using light and confocal microscopy; Pridoli and Lochkovian charcoalified samples (Welsh Borderland) were liberated from the rock and examined with scanning electron microscopy. Prototaxites taiti possessed a superficial hymenium comprising an epihymenial layer, delicate septate paraphyses, inoperculate polysporic asci lacking croziers and a subhymenial layer composed predomi- nantly of thin-walled hyphae and occasional larger hyphae. Prototaxites taiti combines features of extant Taphrinomycotina (Neolectomycetes lacking cro- ziers) and Pezizomycotina (epihymenial layer secreted by paraphyses) but is not an ancestor of the latter. Brief consideration is given to its nutrition and potential position in the phylogeny of the Ascomycota.



Full text

rstb.royalsocietypublishing.org
Research
Cite this article: Honegger R, Edwards D, Axe
L, Strullu-Derrien C. 2017 Fertile Prototaxites
taiti: a basal ascomycete with inoperculate,
polysporous asci lacking croziers. Phil.
Trans. R. Soc. B 373: 20170146.
http://dx.doi.org/10.1098/rstb.2017.0146
Accepted: 15 August 2017
One contribution of 18 to a discussion meeting
issue ‘The Rhynie cherts: our earliest terrestrial
ecosystem revisited’.
Subject Areas:
palaeontology, evolution
Keywords:
dispersed cuticle, epihymenial layer,
hymenium, septate paraphyses, polyspory,
phylogeny
Author for correspondence:
Rosmarie Honegger
e-mail: rohonegg@botinst.uzh.ch
Fertile Prototaxites taiti: a basal
ascomycete with inoperculate,
polysporous asci lacking croziers
Rosmarie Honegger1, Dianne Edwards2, Lindsey Axe2
and Christine Strullu-Derrien3
1
Institute of Plant and Microbiology, University of Zu¨rich, Zollikerstrasse 107, 8008 Zu¨rich, Switzerland
2
School of Earth and Ocean Sciences, Cardiff University, Park Place, Cardiff CF10 3AT, UK
3
Department of Earth Sciences, The Natural History Museum, Cromwell Road, London SW7 5BD, UK
RH, 0000-0001-5430-2408; CS-D, 0000-0003-1131-9546
The affinities of Prototaxites have been debated ever since its fossils, some
attaining tree-trunk proportions, were discovered in Canadian Lower Devonian
rocks in 1859. Putative assignations include conifers, red and brown algae, liver-
worts and fungi (some lichenised). Detailed anatomicalinvestigationled to the
reconstruction ofthe type species, P. logani, as a giant sporophore (basidioma) of
an agaricomycete (¼holobasidiomycete), but evidence for its reproduction
remained elusive. Tissues associated with P. taiti in the Rhynie chert plus
charcoalified fragments from southern Britain are investigated here to describe
the reproductive characters and hence affinities of Prototaxites. Thin sections and
peels (Pragian Rhynie chert, Aberdeenshire) were examined using light
and confocal microscopy; Pr
ˇ
ı
´dolı
´and Lochkovian charcoalified samples
(Welsh Borderland) were liberated from the rock and examined with scanning
electron microscopy. Prototaxites taiti possessed a superficial hymenium
comprising an epihymenial layer, delicate septate paraphyses, inoperculate
polysporic asci lacking croziers and a subhymenial layer composed predomi-
nantly of thin-walled hyphae and occasional larger hyphae. Prototaxites taiti
combines features of extant Taphrinomycotina (Neolectomycetes lacking cro-
ziers) and Pezizomycotina (epihymenial layer secreted by paraphyses) but is
not an ancestor of the latter. Brief consideration is given to its nutrition and
potential position in the phylogeny of the Ascomycota.
This article is part of a discussion meeting issue ‘The Rhynie cherts: our
earliest terrestrial ecosystem revisited’.
1. Introduction
The genus Prototaxites, whose fossils extend from Silurian (Ludlow) to Upper
Devonian strata (Famennian) [1,2], remains one of the most debated conundra
in the Palaeozoic record. Some Prototaxites ‘logs’ reached more than 1 m in
diameter and several metres in height, the maximum being 8.8 m; they were
the largest terrestrial organisms of their time [1]. Some, but not all Prototaxites
axes reveal irregular growth rings. Its habitat has been much debated [3,4],
but based on sedimentological studies Prototaxites was found to be truly terres-
trial [5]. Being part of terrestrial food chains, its axes have been invaded by
boring terrestrial arthropods [6].
Approximately 14 species have so far been described. These have been col-
lected in North America [1,7– 10], northern Europe (UK [3,11,12], Germany
[1,13– 17]), North Africa (Libya [18,19]), western Asia (Saudi Arabia [20,21]) and
Australia [1]. Prototaxites spp. were locally abundant, asconcluded from allochtho-
nous charcoalified fragments in Lower Devonian fluvial rocks in the Welsh
Borderland, where Prototaxites often makes up a high percentage of fossil remains.
Eroded, fragmented silicified samples have been found as pebbles in fluvial
deposits, e.g. in gravel quarries (Middle Eocene [22] from the Eifel area to coastal
Belgium and Netherlands [23]). However, only very rarely were Prototaxites
&2017 The Author(s) Published by the Royal Society. All rights reserved.
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

samples found with an intact surface layer, exceptions being a
few permineralized fossils in the Rhynie chert palaeoecosystem,
and fertile parts have been largely missing [10].
As the etymology of the name suggests, its originator
Dawson [7] initially identified the large ‘trunks’ of Prototaxites
as a conifer with anatomy superficially similar to that in the
wood of the extant genus Taxus (yew) with its tertiary spirals.
Such assignation was subsequently challenged by the phycol-
ogist Carruthers [24] who, illegitimately based on precedence,
renamed the Lower Devonian Gaspe
´fossils as Nematophycus,
here with connotations to brown algal affinity and resemblance
to the laminarialean genus Lessonia. Subsequent authors
(e.g. Penhallow) [25] supported an algal affinity, while
Dawson [26] changed the name to Nematophyton, a name,
although again illegitimate, which was used by a number of
authors, including Kidston & Lang [3] (N. taiti). Later a red
algal affinity was postulated [27], and recently Prototaxites
axes have beeninterpreted as rolled mats of liverworts (march-
antiophytes), which had rolled down a slope and stabilized
themselves with their own rhizinae [28,29]; this interpretation
has been rejected [30,31].
An early dissenting voice had been that of Church [32]
who saw resemblances with fungi, a relationship reinforced
by Hueber [1]. In a very detailed reinvestigation of Dawson’s
type, P. logani, and more recently acquired material, Hueber
described the organism as possessing three forms of interac-
tive hyphae (plectenchyma as opposed to embryophyte
parenchyma) with similarities to other Prototaxites species.
Definitive evidence for the fungal status of Prototaxites and
then its position in the fungal clades requires information on
its reproduction. Hueber [1] described structures on the
margin of a growth increment in P. logani, which he interpreted
as remnants of a hymenium on a previous outer surface in which
dendritically branching hyphae (dendrophyses) were associated
with ‘apparent remnants of primitive basidia in theform of a clo-
sely linear or clustered individually inflated sterigmata, each
with a prominent speculum’ (p. 145; Plate VII). Neither spores
nor basal metabasidia were preserved. On such somewhat
equivocal evidence, Hueber concluded that a hymenium orig-
inally present on the outermost surface of the presumed
sporophore (trunk) supported basidiomycete affinity.
Here we describe a different type of hymenium associated
with Prototaxites taiti in the Rhynie chert [3], which offers an
alternative solution to relationships within the fungi. Although
the hymenial layer of the type specimen is not in organic con-
tinuity with the axis, we are convinced, as were Kidston &
Lang [3], that it was neither a parasite nor a saprotorophic
degrader, but part of P. taiti. The new material studied here
shows a similar juxtapositioning with P. taiti and contains
medullary spots. Peels of P. taiti comprise detached fragments
of the hymenial layer and a larger fragment whose entire sur-
face is covered by a hymenium. Similar organization detected
in charcoalified fragments from the Welsh Borderland leads
to the inference that some members of the Nematothallus com-
plex, which were common elements of terrestrial ground
covers in the Late Silurian– Early Devonian [33], also belong
to Prototaxites.
2. Material and methods
Anatomy was studied using petrographic thin sections and
cellulose acetate peels of the silicified matrix from Rhynie chert:
(1) Figured specimen, i.e. thin section no. 2525 from the Hunter-
ian Museum, is part of the original Kidston’s collection (Plate
X, figs 113, 115, 116, 118 in [3]); this section was made by
W Hemingway. Courtesy Dr Neil Clark.
(2) One section and numerous peels from Block 149 in the Lyon
Collection at Aberdeen University; thin section 149/CT/B
had been prepared by AG Lyon in Cardiff and is now in
the Aberdeen collection. Courtesy Prof. Nigel Trewin.
Sections had been prepared by standard petrographic tech-
niques. Peels had been prepared by AG Lyon by etching
smooth surfaces of chert with HF before washing and applying
a superficial sheet of cellulose acetate moistened with acetone.
Peels and thin sections were examined in transmitted light
using a LEICA DMR microscope fitted with LEICA LAS software,
and a Nikon Eclipse LV100ND compound microscope. Confocal
images from the thin sections were acquired with a Nikon A1-Si
laser-scanning confocal microscope. Autofluorescence of the
sample was excited with four laser lines. Autofluorescence signal
was collected with four photomultiplier detectors with the follow-
ing wavelength emission windows: 425 –475 nm for the 405 nm
laser, 500–550 nm for the 488 nm laser, 570– 620 nm for the
561 nm laser and 675–725 nm for the 640 nm laser. Samples
were visualized using a 29.9 mm (1.2 airy units) confocal pinhole
and a number of z-stacks (typically between 100 and 400) with
optical thickness between 200 and 300 nm each were acquired.
The fluorescence signal from each z-stack was then projected
onto a maximum projection image.
For SEM investigations coalified material collected at Ludford
Corner (Pr
ˇ
ı
´dolı
´) and North Brown Clee Hill (Lochkovian) in the
Welsh borderland was isolated from the matrix using HCl and
HF (protocol in [33,34]). Selected specimens were mounted on
stubs and examined using an environmental scanning electron
microscope (ESEM-FEG, Philips XL 30, with field emission gun;
FEI).
3. Results
(a) Light microscopy preparations from Rhynie chert
fossils
(i) Figured specimens in same block as the type specimens of
Prototaxites taiti
Slide no. 2525 of the Kidston collection, the petrographic thin
section with the type specimen of P. taiti Kidston & Lang 1921
[3], contains an isolated fragment interpreted by Kidston and
Lang as its peripheral layer (figure 1a). This was based on the
presence of medullary spots (figure 1b–d) noted in P. taiti and
common in other Prototaxites species; the presence of ‘tubes’
within this region thought equivalent to those in the tissues of
Prototaxites, although with smaller diameter in the peripheral
tissue (figures 1aand 2e); and the close proximity to P. taiti in
the same block. Accordingto Kidston & Lang [3] the peripheral
layer, which ‘has a thickness of about 1 mm’ ( p. 888), is built up
by ‘tubes’ arranged in parallel at right angles to the surface with
homogeneous brown contents and fine hyphae in between
these (p. 886). They describe a narrow outer zone above the
summits of the vertical tubes as a ‘structureless, and perhaps
mucilaginous, layer during life. It has no sharply defined
outer limit, though it is marked as a dark line owing to a gran-
ular deposit’.
We interpret this peripheral zone of P. taiti as the hyme-
nial layer of an apothecium-bearing ascomycete, the section
plane being slightly tangential, the tubes with granular
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
2
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

contents being asci with large numbers of small ascospores
and the fine hyphae between them as paraphyses. The thin,
amorphous and pigmented outermost layer is presumably
an amorphous epihymenial layer; the granular deposits
therein might have been secondary metabolites, as secreted
by the tip cells of the paraphyses prior to crystallization.
Such gelatinous, often brightly pigmented epihymenial
layers are common and widespread in the apothecia of
extant non-lichenized and lichenized Pezizomycotina.
(ii) Slide no. 149/CT/B, petrographic thin section
In this section (figures 2aand 3 –5), several detached fragments
of the presumed hymenial layer of P. taiti are preserved. Most
of them are tangentially sectioned, but a comparatively large
fragment of a median section can be explored (marked in
figure 3, white arrow). The hymenial layer is approximately
1100 mm thick. Moreover, there are two dark areas contained
in this slide with hundreds of sporangia and resting sporangia
of a zoosporic fungus [35]. This large assembly probably devel-
oped on detritus, as accumulated in the chert palaeoecosystem.
In the thinner part of the median section of the hymeniumof
P. taiti, the asci, paraphyses and epihymenial layers were
resolved (figures 4 and 5a,b). At maturity the ascus tip reached
the surface of the hymenium. The delicate paraphyses
(approx. 5– 6 mm thick) are septate (figure 4) and end with a
globose to pyriform tip cell (figure 4a,d); anastomoses occur
(figure 4e). The asci contained more than 100, at presumed
maturity, well-defined spores with thin, translucent walls
measuring approximately 6 mm in diameter (figure 4a–c);
these are best visible in confocal micrographs (figure 5a,b), the
chemical composition of the autofluorescent material being
unknown, fluorescence resulting either from mineral or organic
preservation. Some of the asci had already released their spores.
The amorphous epihymenial layer of P. taiti, with wall
fragments of tip cells of the paraphyses and granular deposits
embedded therein, was locally peeling off the hymenial
surface and detached fragments were found in the surround-
ings of the hymenia (figure 5d,e). The apical region of
asci after spore release appears as holes in these detached
epihymenial layers.
Various hyphae of fungal and bacterial invaders are seen
below and partly within the hymenial layer of P. taiti
(figure 5c). They might have grown into the detached hymenial
fragments post-mortem. Moreover, there are manysilicate crys-
tals, often rosette-like, around and within the fossils (figure 1a).
(iii) Slides 149/B 2, 3, 5, 8 & 9 with peels of P. taiti
Serial peels of a block containing P. taiti (figure 2a–d) con-
tained several detached fragments of hymenial layers in
various section planes (marked with pen in figure 2a), and a
medullary spots
epihymenium
hymenium
7 µm 15 µm
(a)
(a)
(c)(d)
(b)
1 mm1 mm
1 mm1 mm
medullary spots
Figure 1. Petrographic thin section of Prototaxites taiti in bright field (a) and confocal laser microscopy (b,c), and peel in bright field microscopy (d). (a–c) The type
specimen of P.taiti from the Rhynie cherts, new photograph of slide no. 2525 of the Kidston Collection. (a) This fragment is depicted in black and white in
Kidston & Lang [3] in fig. 113, with the following legend: ‘View of one of the sections of the second fragment which shows the peripheral tissues’. Here
these peripheral tissues are interpreted as the hymenial layer of the ascoma of an ascomycete, the section plane being slightly tangential. Note the innumerable
silicate crystals, best visible outside the tissues. (b–d) Medullary spots: in confocal laser microscopy (b,c), (slide no. 2525), intense wefts of hyphae in close contact
with a population of minute cells visible in bright field microscopy (d), (slide 149 B 3).
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
3
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

near-median section of the hymenial layer covering the surface
of an axis (figure 2b,d). A cross-section of presumed cords are
seen (figure 2b,c), but not in conjunction with the axes.
In these serial peels of P. taiti two large, dark fragments and
a distinctly brighter one are seen (figure 2a,b). The dark frag-
ments lack a peripheral layer, whereas the surface of the
bright fragment is covered by an extensive hymenium, the
latter being tangentially sectioned. Although this fertile frag-
ment reveals the anatomy characteristic of Prototaxites it
seems to be more ‘fleshy’. With high probability the detached
hymenial fragments had peeled off such fertile parts when
these were either senescent or otherwise damaged. The large
number of detached hymenial fragments, as found in these
Rhynie chert specimens, infers an abundance of fertile parts.
C
D
B-D
60 µm
(a)
(c)(d)
(b)
(e)
Figure 2. Slides of a petrographic thin section and a series of peels of Prototaxites taiti, from Rhynie chert as prepared by Prof. AG Lyon. (a) Overview of petrographic
thin section 149 CT B (see details in figures 3–5) and serial peels in slides 149/B 2, 3, 8 & 9. (b) Detail of slide 149/B 8, the position of figures (c)and(d)being
marked. (c) Cross-section of a cord (or a very young axis?) with a somewhat denser peripheral zone. (d) Slightly tangential section of the peripheral layer of a ‘fleshy’
axis which, in the present study, is interpreted as the hymenial layer of an apothecium-bearing ascomycete. (e) Confocal laser micrograph of the hymenial layer of the
type specimen of P. ta it i (slide no. 2525 of the Kidston and Lang collection; see figure 1a), showing the tubes, i.e. asci with ascospores, in the hymenial layer.
0.2 cm
30 µm
Figure 3. Petrographic thin section of Prototaxites taiti from Rhynie chert specimen no. 149 CT B shown in figure 2a. The white arrow points to a median-long-
itudinally sectioned fragment of the presumed hymenial layer, black arrows to additional hymenial fragments; framed is a thin area of which photographs in figures
4 and 5 had been taken. The asterisk refers to part of a thin, cross-sectioned axis of Prototaxites with a simple peripheral zone. White circles refer to areas with
hundreds of sporangia of a zoosporic fungus (insert: resting sporangium; see [35]). Photo: Thomas Honegger.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
4
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

(b) Scanning electron micrographs of charcoalified
fragments from the Lochkovian Welsh Borderland
(i) Prototaxites sp. (HD(L)171/03), figures 6 and 7
This small Lochkovian fragment reveals an intact, hummocky
surface of the hymenial layer (figure 6a,b) comprising well-
spaced structures interpreted as asci. Various developmental
stages of asci embedded in the amorphous epihymenial layer
indicate that they were ripening at different times (figure 6b).
Still closed ascus apices appear near others, which slightly
bulge above the surface prior to opening with a slit. After ascos-
pore release the wall of the empty ascus sunk back into the
hymenial layer (figure 6b), leaving a roundish slit before get-
ting covered by mucilage. Groups of ovoid, somewhat
shrivelled presumed ascospores were repeatedly seen lying
on the hymenial surface (figure 6d); their size (approx.
5.5 mm long) corresponds to those seen in the silicified speci-
mens (figure 4). Some of these presumed ascospores seemed
to germinate (figure 6e), but cannot be distinguished from con-
taminant fungal spores. Both landing of ascospores and their
germination on the hymenial surface occur frequently in
extant ascomycetes. However, crystals of the same size and
shape as the presumed ascospores are frequently found on
this type of material.
On the lower surface of this fragment are medullary spots
embedded within hyphae (figure 6c). The epihymenial layer
has locally peeled off (figure 7a), revealing a reticulum charac-
teristic of cuticles of the Nematothallus complex. The probably
largely immature asci are well preserved (figure 7a–c), but
most of the delicate paraphyses were lost, probably during
charcoalification, leaving some impressions on the surface of
the asci (figure 7c). Presumed paraphyses are seen in
figure 7c. The amorphous epihymenial layer is well preserved,
revealing impressions of tip cells of paraphyses on its inner sur-
face; these are visible at sites where the epihymenial layer
detaches (figure 7a). A subhymenial layer with crozier for-
mation was not seen here nor in the silicified material
examined by light microscopy. As described by Kidston &
Lang [3] for the latter ‘the tubes bend outwards and branch,
and in the superficial layer ... stand parallel at right angles to
the surface’ (p. 888). This specimen was invaded by either
fungal or bacterial contaminants (figure 7b).
(c) A charcoalified apothecial fragment of a presumed
pezizomycete from the Late Silurian Shropshire
The unnamed specimen LL03/02 from Ludford Lane, Shrop-
shire (figure 8), is interpreted as a fragment of at least two
apothecia, which had been growing side by side
100 µm
tip cells of paraphyses
paraphyses
polysporous
asci
ascospores
tip cells of
paraphyses
(a)
(a)(c)
(d)
(b)
(e)
Figure 4. (a–e)Prototaxites aff. taiti from Rhynie chert, details from the longitudinal section in slide no. 149 CT B (figure 3). Polysporous asci and thin-walled, delicate
paraphyses. The darkly pigmented epihymenial layer is best visible in (d). The arrow in (e) points to anastomosing paraphyses. Same magnification in (a–e).
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
5
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

(figure 8a,b,e). In contrast to the hymenium of P. taiti, a dense
subhymenial layer is seen with presumed crozier formation
(figure 8c). The apothecia have a thin margin built up by
only one cell layer (figure 8b,e). The apex of the polysporous
ascus reached the hymenial surface at maturity (figure 8e,f).
Some of the tip cells of the septate paraphyses bend and
branch (figure 8f,g), a feature which is seen also among some
extant ascomycetes. This specimen, which was unfortunately
lost while being turned over for additional fracturing and
investigations, represents the first apothecium-bearing
member of the Pezizomycotina so far found.
4. Discussion
(a) Prototaxites taiti: an ascomycete with extensive
hymenial layer
A striking structural similarity of the ascomata of Prototaxites
taiti with the sori (meiosporangia-bearing zones) of the
large sporophytes of marine brown algae (kelps) of the order
Laminariales (Phaeophyceae, Heterokonta) is evident. Sori of
Laminariales are formed as patches on the surface of the
phylloids (‘blades’); they are built up by sterile filaments
(paraphyses) interspersed with elongate, ascus-shaped meio-
sporangia comprising large numbers of biflagellate zoospores.
The morphological similarity of Prototaxites stems with cauloids
and of their basal part with holdfasts of Laminariales was
reported by earlier investigators [13,17,18,24,25]. However,
the meristem-derived tissues of laminarialean sporophytes
differ anatomically from the plectenchyma of Prototoaxites
spp. Fossilrecords are largely missing. Based on multigene phy-
logenetic analyses the branching time of Laminariales from
Ectocarpales (filamentous brown algae) is estimated around
100– 90 Ma [36]. Thus a laminarialean affiliation of the genus
Prototaxites can be excluded.
Based on the present findings P. taiti is interpreted as a
basal ascomycete; in the fertile state a hymenial layer com-
prising polysporous asci and delicate, septate paraphyses
covered the surface of either part of its axis or of fertile, pre-
sumably lateral and/or terminal outgrowths. It is not clear
how the hymenial fragment of the type specimen of P. taiti
(figure 1a) was connected to the axis; the ‘fleshy’ fertile frag-
ment in the peels (figure 2a,b) might represent a fruit body as
a lateral outgrowth of the axis. Further material is needed to
answer these questions.
The oldest ascoma of an ascomycete so far described are
the beautifully preserved perithecia of Palaeopyrenomycites
devonicus [37,38], a presumed sordariomycete, parasite or epi-
biont of Asteroxylon mackiei. However, the most basal taxa
100 µm
fungal hyphae
actinobacterial
filaments
200 µm
100 µm
tip cells of
paraphyses
ascus tips
30 µm 4 µm
(a)
(c)(d)
(b)
(e)
Figure 5. Prototaxites aff. taiti from Rhynie chert, details from slide no. 149 CT B; (a,b) confocal laser microscopy, (c–e) bright field light microscopy. (a) Detail of
the hymenial layer with asci, paraphyses and invasive, contaminating fungal filaments. (b) Spores inside an ascus. (c) Contaminating fungal and bacterial filaments
growing below the hymenial layer. (d) Laminal view, and (e) tangential (left) and cross-section (right) of detached fragments of the epihymenial layer, with tip cells
of paraphyses and open ascus tips, the interspace being filled with dark pigments.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
6
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

500 µm
50 µm
500 µm
20 µm 10 µm
*
medullary spots
epihymeniumepihymenium
hymeniumhymenium
*
*
(a)
(c)(d)
(b)
(e)
Figure 6. SEM micrographs of a charcoalified fragment of the hymenial layer of Prototaxites sp. (HD(L)171/03) from Welsh Borderland. (a,b,d,e) Laminal view of the
surface. (a) Overview; (b) detail of the hymenial surface. The asterisks mark an ascus apex prior to dehiscence. Thick arrows point to asci shortly after ascospore
release, the inoperculate ascus tips having slightly expanded and opened with a slit. Thin arrows point to empty asci whose wall sank back into the hymenial layer.
(c) Lower surface of the specimen. (d) Arrows point to shrivelled, presumed ascospores after landing on the hymenial surface. (e) Presumed ascospores having
germinated next to the apex of the empty ascus.
20 µm20 µm
200 µm
200 µm 20 µm
tip cells of
paraphyses
epihymeniumepihymenium
hymeniumhymenium
epihymeniumepihymenium
*
(a)(c)
(b)
Figure 7. SEM micrographs of the charcoalified fragment of the hymenial layer of Prototaxites sp. (HD(L)171/03) from Welsh Borderland, lateral view. (a) The dense
epihymenial layer is partly peeling off (figure 5a,b). (b). Detail of the hymenial layer with immature asci. Arrows point to presumed paraphyses and/or fungal or
bacterial invaders. (c) Arrows point to remains of a few thin-walled paraphyses. The asterisk points to an ascus with superficial impressions of lost paraphyses.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
7
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

among the extant Pezizomycotina (i.e. the Peziomycetes and
Orbiliomycetes; [39,40]) produce neither ostiolate peri- or
pseudothecia, nor closed cleisto- or gymnothecia, but apothe-
cia, the former ascomal types having evolved independently
in diverse taxa [41–44].
The dimensions of the hymenial layer of P. taiti differ con-
siderably from extant ascomycetes, both in circumference and
height. Very large apothecial surfaces are found, e.g. among
the earth tongues (Geoglossaceae, Geoglossomycetes;
figure 9b), the elfin saddles (Helvellaceae, Pezizomycetes) and
the morels (Morchellaceae, Pezizomycetes). Their comparatively
soft, large ascomata are relatively short-lived (few weeks) and
die off after ascospore release, all asci reaching maturity at
almost the same time. In contrast, the tough, club-shaped dead
moll’s finger (Xylaria longipes) and related taxa among the Xylar-
iaceae as well as caterpillar fungi (Cordiceps spp.,
Sordariomycetes) have no superficial hymenial layer, but large
numbers of minute perithecia peripherally inserted in their
robust sclerotia. Long-living hymenial layers are found in the
apothecia of the majority of lichen-forming ascomycetes,
especially in the Lecanoromycetes (e.g. Cladonia spp., Cladonia-
ceae, or Baeomyces spp., Baeomycetaceae; figure 9c,d), their fruit
bodies producing asci over many months to a few years. How-
ever, their apothecia are minute compared with the
aforementioned taxa.
The hymenial layer of P. taiti is approximately 1100 mm
high and thus more than twice the size as reached by extant
apothecia-bearing ascomycetes. The length of the mature
ascus, as recorded in the taxonomic literature, does not necess-
arily reflect the full height of the hymenium; the apex of the
mature ascus may be below the hymenial surface and the epi-
hymenial layer often accounts for additional micrometres. In
extant pezizomycetes, ascus lengths around 300 –380 mmare
found in morels (Morchella spp.), up to 390 mm in cedar cup
200 µm 200 µm 2 apothecia
subhymenial
layer
hymenial surface
hymenial surface
hymenial surface
100 µm
polysporous polysporous
ascusascus
hymenial hymenial
layerlayer
subhymenial
layer
50 µm
polysporous asci
apothecial margins
ascus apex
presumed
ascus apices
tip cells of
paraphyses
10 µm
septa in
paraphyses
septa in
paraphyses
20 µm
2 µm
(a)(b)
(b)
(c)
(c)(d)(e)
(f)(f)(g)
subhymenial
layer
Figure 8. SEM micrographs of a charcoalified apothecial fragment from the Late Silurian (LL03/02) from Ludford Lane, Shropshire. (a) Upside-up, and (b,e) upside-
down positional views of the specimen. (c) Presumed crozier in the basal part of the hymenial layer. (d) The longitudinally fractured hymenial layer is built up by
slim, unbranched paraphyses. The numerous ascospores within the fractured, polysporous ascus reveal irregular outlines, i.e. are deformed. Arrows point to bacterial
contaminants which had grown post-mortem over the fragment. (e) Detail of surface in (b): the thin apothecial margin is built up by parallel hyphae. ( f) The
rounded tip cells of paraphyses are grouped around presumed ascus apices. (g) Septa are visible in fractured paraphyses.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
8
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

(Sepultaria sumneriana), and up to 450 mm in midnight disco
(Pachyella violaceonigra), all representatives of the Pezizales/
Pezizomycetes (data from [45]).
(b) The oldest fossil apothecia
This is the first report on fossil apothecia, the characteristic
ascomata as formed by numerous taxa within the subphylum
Pezizomycotina, which are considerably older than those
found in Cenozoic amber. So far a striking discrepancy
between the presumed age of the earliest apothecia-bearing
ascomycetes (Ordovician), as concluded from molecular data-
sets [38,39], and the fossil record was evident. So-called
calicioid ascomata (a paraphyletic assemblage of taxa with
stalked apothecia and superficial mazaedium, i.e. a powdery
spore mass, as released by deliquescent asci [46,47]), were
found in Cenozoic amber [48–51].
Extant representatives of the Pezizomycotina have a hap-
loid vegetative mycelium. Ascoma formation is initiated by
dikaryon formation, either via spermatisation of ascogonia
(the contact sites being thin-walled trichogynes, which grow
out of the ascogonia) or via fusionof non-differentiated vegeta-
tive hyphae (somatogamy). Upon successful dikaryotisation a
system of dikaryotic hyphae grows towards the future hyme-
nial layer of the ascoma. The haploid hyphal system in close
contact with the ascomal primordium builds up the ascoma
with characteristic margin and the hymenial layer with para-
physes, which, in many groups, secrete the hydrophilic
hymenial gelatin and the epihymenial layer plus the pigments
located therein. In the majority of Pezizomycotina, crozier for-
mation with nuclear migration into the penultimate cell along
the dikaryotic ascogenous hyphae, precedes karyogamy; this
allows the maintenance of the dikaryotic state with nuclei of
opposite mating types.
(c) Neolectomycetes (Taphrinomycotina): basal
Ascomycota with club-shaped ascomata
The Taphrinomycotina, formerly termed Archaeascomycetes,
is the most basal out of three subphyla of the Ascomycota;
it comprises the classes Taphrinomycetes (plant pathogens),
Schizosaccharomycetes (fission yeasts), and the monotypic
class Neolectomycetes (with one order, one family and one
genus, Neolecta Speg., comprising four presumably sapro-
trophic species) [52]; these are the only extant representatives
of the Taphrinomycotina which form ascomata.
The stipitate, clavate ascomata of Neolecta spp. are covered
by a coloured hymenial layer (figure 9a), no ascomal margin
being differentiated. The vegetative hyphae are binucleate,
the ascogenous hyphae multinucleate, the nuclei being haploid
[53]. When asci are to be formed in Neolecta spp. pairs of nuclei,
presumably of opposite mating types, are formed and segre-
gated via septum formation from the rest of the hypha [53]; a
dense subhymenial layer is missing and no croziers are
formed. The mating system of Neolecta spp. has not yet been
genetically analysed, and it is not known how the pairing of
nuclei is achieved prior to karyogamy and meiosis. In the
more advanced, plant pathogenic Taphrinomycetes (Taphrina
spp.) the dikaryon is formed soon after ascospore germination
via fusion of budding cells, only the dikaryotic (bi-nucleate)
mycelium being infective. Based on molecular datasets
[54,55] Neolecta (i.e. the Neolectomycetes) was referred to as
‘a fungal dinosaur’ [56].
(d) Prototaxites taiti: a basal ascomycete
According to the present findings P. taiti represents a basal
ascomycete which combines characteristic features of Neolec-
tomycetes/Taphrinomycotina, such as asci without crozier
formation (in addition to the SEM specimen several millimetres
of hymenial length were studied in petrographic thin sections),
their inoperculate apex opening with a slit, and of Pezizomyco-
tina, i.e. a system of paraphyses which secrete mucilage into the
hymenial and epihymenial layer. However, Upper Silurian
and Lower Devonian Prototaxites spp. were unlikely ancestors
of the Pezizomycotina since they were contemporaries of
apothecia-bearing ascomycetes with more advanced character
traits. The unnamed Upper Silurian apothecial fragment
(LL03/02) with a thin apothecial margin (figure 8b,e), distinct
paraphyses and polysporous asci, reveals a dense subhymenial
layer (figure 8d) and presumed croziers (figure 8c), both being
characteristic features of Pezizomycotina. No croziers could be
resolved in Palaeopyrenomycites devonicus [37,38].
How does our interpretation of P. taiti as a basal ascomy-
cete correlate with Francis Hueber’s [1] emendation of the
genus Prototaxites as a representative of the class
Taphrinomycotina
Neolectomycetes
Pezizomycotina
Geoglossomycetes Lecanoromycetes (lichenized)
(a)(c)(d)
(b)
Figure 9. Extant ascomycetes with clavate to roundish sporophores (ascomata). (a) The presumably saprotrophic Neolecta vitellina, one out of four species of
Neolectomycetes, the only Taphrinomycotina with ascomata. The extensive hymenial layer is bright yellow. Photo courtesy of Raymond McNeil. (b) The saprotrophic
dark-purple earth tongue (Geoglossum atropurpureum, syn. Thuemenidium a.). Photo courtesy of Knud Knudsen. (c) The stalked, non-lichenized ascomata of Dibaeis
baeomyces (syn. Baeomyces roseus) growing out of the crustose lichen thallus. (d) Convex ascomata of Cladonia bellidiflora with red coloration (due to crystals of the
anthraquinone bellidiflorin in the epihymenial layer) on an erect podetium whose surface is covered by dorsiventrally organized, lichenized squamules.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
9
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

Agaricomycetes (¼Holobasidiomycetes)? Based on detailed
investigations of P. loganii samples, Hueber interpreted the
tangentially sectioned peripheral layer as a hymenial layer
and some of the structures therein as sterigmata of presumed
basidia. Neither basidia proper nor basidiospores were seen
in this sample. The presumed sterigmata could be remains
of ascus walls in the tip region after spore discharge.
Median sections are required to solve this problem.
In the neotype of P. loganii Hueber [1] described a cluster
of very thin, finely branched presumed dendrophyses (¼
dendrohyphidia), a type of sterile hyphae as found in the
hymenia of some basidiomycetes. No comparable structures
were found in P. taiti. Since the hymenium, as the peripheral
layer, is also easily accessible to fungal and bacterial coloni-
zers during lifetime and post-mortem (figure 5c) and
crystallization products of minerals may have developed
during the fossilization process on or near the hymenial sur-
face, a more detailed analysis of the cellular structure of the
dendrophyses would be of interest.
The septal pore apparatus, a phylogenetically informative
character trait, has been investigated in Prototaxites southworthi
with TEM techniques [57]. It shows a quite complex central
pore with surrounding material, but not a dolipore as typically
found in extant Agaricomycetes. A surprisingly similar situ-
ation was observed in Neolecta vitellina [58], whose septal
pore apparatus differs structurally and biochemically from
the one as typically found in Pezizomycotina. The HEX-1
protein, as contained in the membrane-bound Woronin
bodies which are plugging the septal pore of Pezizomycotina
[59], is missing in Neolecta [60], but a unique type of material
is present. Healy et al. [58] conclude that the septal pore-occlud-
ing structure in Neolecta might have evolved independently.
(e) Polyspory
With regard to the current discussion about the evolution of
polyspory among the Pezizomycotina [61] it is particula-
rly interesting to see polysporous asci in the perithecia of
Palaeopyrenomycites devonicus (up to approx. 16 ascospores
per ascus; [37,38]), in the earliest yet found apothecium
from the Upper Silurian (present study) and in P. taiti (pre-
sent study).
The majority of extant representatives of the Pezizomyco-
tina contain eight ascospores per ascus (octospory), these
being the result of one mitosis following meiosis. However,
polyspory is widespread, some taxa containing more than
100 spores per ascus. Examples are more than 500, as in
Melanophloea (lichenized ascomycete incertae sedis; [62]), or
more than 1000 as in Brigantiaea (Lecanorales) and Gyalidea
(Ostropales; both Lecanoromycetes). Among extant Pezizomy-
cotina polyspory occurs in Eurotiomycetes, Dothideomycetes,
Lecanoromycetes, Leotiomycetes, Lichinomycetes, Orbiliomy-
cetes, Sordariomycetes, often punctually (one or few
polysporous among many octosporous species per genus),
but sometimes prominent, with the majority of species per
genus or even family being polysporous. Polyspory is also
widespread within the Taphrinomycotina, the most basal of
all subphyla of Ascomycota.
Some authors distinguish between true polyspory, when
asci contain more than 100 spores, and non-true polyspory
for taxa with more than eight and less than 100 spores per
ascus [61]. Polyspory results from either several mitoses follow-
ing meiosis (probably the most common case), from multiple
budding (e.g. in various Taphrina sp., sometimes also in
Neolecta spp; [53,63]), from conidium formation by ascospores
within the ascus (e.g. in Rhamphoria [Sordariomycetes],
Brigantiaea and Gyalidea [Lecanoromycetes]), or from break-
ing apart of multicellular ascospores within the ascus (e.g.
various species in the genera Claussenomyces, Tympanis [Leo-
tiomycetes] or Cordyceps [Sordariomycetes]).
Among the lichen-forming Pezizomycotina polyspory is
widespread [61,62]. Few families such as the Ascosporaceae,
Biatorellaceae (Lecanoromycetes) and Thelocarpaceae (incer-
tae sedis) and at least 16 genera therein are largely
polysporous, while 38 (unrelated) genera comprise only few
polysporous species (representatives of Lecanoromycetes,
Lichinomycetes, Eurotiomycetes). Polyspory was assumed
to have evolved independently at least 57 times among
extant lichen-forming ascomycetes [62]. However, the possi-
bility of polyspory being an ancestral trait and repeatedly
lost during evolution cannot be excluded. Larger sets of mol-
ecular data and distinctly more fossil records are required.
(f) The detached epihymenial layer of apothecia: a type
of ‘dispersed cuticle’
In the palaeobotanical literature the term cuticle is used
for (usually fragmented) surface layers of various origins:
arthropods, plants or thalli of presumably lichenized fungi.
Dispersed cuticles of nematophytes had so far been interpreted
as the top layer of vegetative thalli [64,65]. Here we present the
first examples of dispersed cuticles as a detached epihymenial
layer of apothecia in permineralized Prototaxites aff. taiti from
Rhynie chert material and from charcoalified Prototaxites sp.
from the Lochkovian of the Welsh Borderland. Apothecia
might be more common in the fossil record than previously
assumed, e.g. among nematophytes with distinct palisade
zone below the peripheral layer. Asci differ in their cell-wall
structure, composition and function from paraphyses and
50 µm
(a)(c)
(b)
Figure 10. Light micrographs of iodine-stained free-hand sections of the
ascoma (apothecium) of extant Peltigera spp. (lichenized ascomycetes) whose
ascus wall differs structurally and chemically from ascospores, paraphyses and
vegetative hyphae. (a) Detail of the hymenial layer of P. canina and (b)
mature ascus with eight spindle-shaped, three-septate ascospores, both stained
with Lugol’s solution (IKI). (c) Detail of the hymenial layer and adjacent margin
of P. venosa subjected to the van Wisselingh test for chitin (red coloration of
chitosan after hydrolytic transformation of chitin). Red coloration, indicative of
chitin, is discernible in the walls of vegetative hyphae of the apothecial
margin, of paraphyses and of the ascospores. In both preparations, the ascus
wall stains blue with iodine due to amyloid compounds (presumably liche-
nin-like glucans with b-1,3 and b-1,4 linkage bonds).
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
10
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

vegetative hyphae (figure 10a–c) and thus might often have
been lost during fossilization.
Looking at extant apothecia-bearing ascomycetes one
should keep in mind that their vegetative mycelium is either
hidden in the substrate (saprotrophs) or in a host tissue (para-
sites) or ensheathing plant root systems (mycorrhizae), only
their ascomata being visible above ground. On suitable sub-
strates, apothecia are locally abundant and appear in large
and often colourful clusters (e.g. in the genera Sepultaria,
Peziza,Helotium,Hymenoscyphus, etc., examples in [45]).
Upon suitable fossilization, fragments of apothecial discs
should be detectable in the fossil record. Exceptions are the
lichenized taxa; these have to expose their photobiont cell
population to adequate illumination and therefore differentiate
their vegetative thallus at or above ground (rock, soil, bark,
etc.), fossilized remains of their vegetative thallus having
been found from the mid-Palaeozoic onwards [33,34], but fer-
tile fragments remain to be found.
(g) Nutritional aspects
Assignation to the heterotrophic fungi of organisms with such
enormous body mass (‘trunks’ of Saudi Arabian Prototaxites
attained a length of 8.8 m and a diameter of 1.2 m, fig. 14 in
[1]) raises questions on sources of nutrients which Hueber
suggested were acquired via a hypothetical extensive soil
mycelium exploiting decaying organic material. Our present
data do not contribute substantially to our understanding of
its nutritional strategies. There is no evidence for underground
structures in Prototaxites, although the basal part of P. hefteri
resembles the basal part of Laminaria sp. whence rhizoids
derive [17]. Edwards and coworkers described putative cords
or rhizomorphs of Nematosketum, closely allied to Prototaxites
[66], and presumed rooting structures of Prototaxites sp. in
rocks of the Early Devonian Anglo-Welsh Basin [67]; similar
structures likely occur alongside sections of P. taiti (this inves-
tigation, figure 2c). The authors speculated that such organs
might have translocated soluble carbohydrate produced by
leakage from biofilms or their decay. A similar source has
been suggested from d
13
CdatafromPrototaxites and coeval
vascular plants [68]. The former showed a considerable range
in values (215.6% to 226.6%) allowing the inference of hetero-
trophic nutrition involving a number of isotopically distinct
substrates. The values determined by Graham et al. [29] were
more consistent with those obtained experimentally from
Marchantia, but no allowance was made for differences in
atmospheric d
13
C in Devonian times. They used such data to
support their hypothesis that Prototaxites represented accumu-
lations of concentric layers of thalloid hepatic mats, bound
together by rhizoids, produced on rolling down a slope. Both
isotopic data and marchantialean affinity have subsequently
been questioned and concluded unlikely on sedimentological,
anatomical and geochemical grounds (e.g. [30,31,68,69]).
(h) Saprotrophic, lichenised or mixed nutritional
lifestyle?
Based solely on energy relationships, Selosse [70,71] had
suggested a lichen-type association in Prototaxites,anaffinity
reinforced on very tenuous anatomical grounds by Retallack &
Landing [10], but a sufficiently large and productive photobiont
cell population has not yet been convincingly demonstrated.
However, as in the other genera of nematophytes [33], the
photobiont cell population might not have been preserved
during fossilization. Almost all green algal and cyanobacterial
cells in the Lower Devonian lichens Chlorolichenomycites salopen-
sis and Cyanolichenomycites devonicus were missing, but
mucilaginous cyanobacterial sheaths were well preserved [34].
When part of the Prototaxites axes were covered by a hyme-
nial layer, as concluded from the present findings, lichenized
zones might have been either interspersed with fertile areas,
or restricted to either the surface proper or lateral outgrowths
of a non-fertile basal part, as in podetia of cup lichens
(Cladonia spp., figure 9d). Alternatively the basal cords of
Prototaxites might have connected the fertile stems with the
common and widespread, possibly lichenized nematophytes
with dorsiventrally organized thalli, common and widespread
organisms in the groundcover of terrestrial ecosystems of the
Mid-Palaeozoic [33,66]. This situation, although in miniature
format, is found in extant Dibaeis and Baeomyces spp.
(figure 9c), but also in lichenized basidiomycetes (Agaricomy-
cotina) with non-lichenized basidiomata (e.g. Lichenomphalia
spp. [72]). However, a superficial hymenial layer and an under-
lying, photosynthetically active photobiont cell population are
not mutually exclusive; this situation occurs in innumerable
representatives of Lecanoromycetes with a lichenized, so-
called thallus margin around the apothecial disc (e.g. Parmelia-
ceae, Xanthoriaceae), photobiont cells living not only in the
lateral margin, but also underneath the hymenium, the latter
allowing sufficient light transmission in the hydrated state,
especially when ascospores are hyaline. If originally present
below the hymenial layer, traces of green algae or cyanobac-
teria should have been fossilized in the Rhynie chert material
of P. taiti.
It is also possible that Prototaxites had a mixed nutritional
lifestyle with simultaneous saprotrophic and symbiotic acqui-
sition of fixed carbon. The majority of extant lichen-forming
ascomycetes are physiologically facultatively biotrophic and
thus can be axenically cultured in the aposymbiotic state. In
nature many lichenized taxa likely derive nutrients not only
as soluble, mobile photosynthates from their photoautotrophic
partner, but also saprotrophically, e.g. by enzymatic degra-
dation of organic matter, such as the cellulosic mother cell
walls of green algal photobionts after autospore formation
or decaying plant material in the substratum. Hyphae of
Icmadophila ericetorum, a lichen-forming lecanoromycete grow-
ing on rotten wood (lignin having largely been removed by
white rot fungi), were shown to grow through the cell walls
of woody tissues which had been partially degraded [73];
quantitative data are missing. The large axes of Prototaxites
spp. certainly carried a diverse and probably at least partly
beneficial microbiome which remains to be circumscribed,
epi- and endolichenic bacteria and fungi having been found
in the Lower Devonian Chlorolichenomycites salopensis [74].
(i) Fossil records and the calibration of the molecular
clock
Prior to the advent of molecular genetic tools, the shape and
ontogeny of the fruiting bodies, together with ascus types,
were prime characters in ascomycete classification. However,
in molecular phylogenies both character sets turned out to be
homoplaseous; thus, groupings such as discomycetes (com-
prising apothecia-bearing taxa), pyrenomycetes (comprising
perithecia-bearing taxa), etc., as found in the older literature,
became obsolete [41– 44]. Phylogenetic datasets indicate
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
11
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

that ascomycetes derived from other fungal lineages in the
Proterozoic, estimates ranging between 1490 and 600 Ma,
depending on calibration [40,46,75–81]. Unfortunately the
fruiting bodies of the majority of ascomycetes lack robust struc-
tures capable of withstanding microbial degradation and thus
are very rarely found in fossil records. Well-preserved, fertile
fungal fossils are urgently needed for calibrating the molecular
clock. To conclude: the beautifully preserved Rhynie chert
material and the charcoalified specimens from siltstone of the
Welsh Borderland, as investigated in this study, give new
insights into the taxonomic affiliation of the genus Prototaxites.
Data accessibility. This article has no additional data.
Authors’ contributions. R.H. did the light and most of the scanning elec-
tron microscopy and mycological interpretation, and wrote most of
the manuscript; D.E. was initially investigating specimens of Prototax-
ites with intact surface layer from various slide collections, wrote part
of the manuscript, provided the material and infrastructure,
organized financial support and helped draft the manuscript. L.A.
prepared the specimens for SEM analysis, helped with LM and
SEM investigations and contributed to the manuscript. C.S.-D. did
the confocal microscopy and contributed to the manuscript. All
authors gave final approval for publication.
Competing interests. We declare we have no competing interests.
Funding. D.E. is grateful for financial support from the Leverthulme
Trust and the Gatsby Charitable Foundation. C.S.-D. thanks the Euro-
pean Commission, Programme FP7-People-2011(SYMBIONTS 298735)
and the Paleontological Association, UK (grant no. PA-RG201602) for
financial support.
Acknowledgements. Our sincere thanks are due to Dr Neil Clark (Hunter-
ian Museum, Glasgow) and to Prof. Nigel Trewin (Aberdeen
University) for providing specimens for investigation. We gratefully
acknowledge the technical support by Dr Tomasz Goral (confocal
microscopy; NHM London) and by Peter Fisher (SEM; Cardiff
University). Raymond McNeil and Knud Knudsen provided
images of extant ascomycetes. We thank Marc-Andre
´Selosse and
an anonymous reviewer for their comments on the manuscript.
References
1. Hueber FM. 2001 Rotted wood-alga-fungus: the
history and life of Prototaxites Dawson 1859. Rev.
Palaeobot. Palynol. 116, 123– 158. (doi:10.1016/
S0034-6667(01)00058-6)
2. Taylor TN, Krings M, Taylor E. 2015 Fossil fungi.
Amsterdam, The Netherlands: Elsevier.
3. Kidston R, Lang WH. 1921 On Old Red Sandstone
plants showing structure, from the Rhynie chert
bed, Aberdeenshire. Part V. The Thallophyta
occurring in the peat-bed; the succession of plants
throughout a vertical section of the bed, & the
conditions of accumulation and preservation of the
deposit. Trans. R. Soc. Edinb. 52, 855– 902. (doi:10.
1017/S0080456800016045)
4. Niklas KJ. 1976 Chemotaxonomy of Prototaxites and
evidence for possible terrestrial adaptation. Rev.
Palaeobot. Palynol. 22, 1– 17. (doi:10.1016/0034-
6667(76)90008-7)
5. Hotton CL, Hueber FM, Griffing DH, Bridge JS. 2001
Early terrestrial plant environments: an example
from the Emsian of Gaspe
´, Canada. In Plants invade
the land: evolutionary and environmental
perspectives (eds PG Gensel, D Edwards), pp. 179–
212. New York, NY: Columbia University Press.
6. Labandeira CC. 2007 The origin of herbivory on
land: initial patterns of plant tissue consumption by
arthropods. Insect Sci. 14, 259– 275. (doi:10.1111/j.
1744-7917.2007.00152.x)
7. Dawson JW. 1859 On the fossil plants from the
Devonian rocks of Canada. Q. J. Geol. Soc. Lond. 15,
477–488. (doi:10.1144/GSL.JGS.1859.015.01-02.57)
8. Nevius JN. 1900 A fossil plant from Orange County.
Ann. Rep. Regents New York State Museum 52,79–
81.
9. Arnold CA. 1952 A specimen of Prototaxites from
the Kettle Point Black Shale of Ontario.
Palaeontographica B 93, 45– 56, Plates 22 –23.
10. Retallack GJ, Landing E. 2014 Affinities and
architecture of Devonian trunks of Prototaxites
loganii.Mycologia 106, 1143–1158. (doi:10.3852/
13-390)
11. Croft WN, Lang WH. 1942 The Lower Devonian
flora of the Senni Beds of Monmouthshire and
Breconshire. Phil. Trans. R. Soc. Lond. B 231,
131–163. (doi:10.1098/rstb.1942.0001)
12. Glasspool IJ, Edwards D, Axe L. 2006 Charcoal in the
Early Devonian: a wildfire-derived Konservat–
Lagersta
¨tte. Rev. Palaeobot. Palynol. 142, 131– 136.
(doi:10.1016/j.revpalbo.2006.03.021)
13. Kra
¨usel R. 1936 Landbewohnende Algenba
¨ume zur
Devonzeit? Eine kritische Betrachtung. Ber. Deutsch.
Bot. Ges. 54, 379–385.
14. Altmeyer H. 1978 Prototaxiten im Taunusquarzit.
Grondboor en Hamer 32, 122 –124.
15. Altmeyer H. 1979 Prototaxiten im Emsquarzit.
Grondboor & Hamer 33, 134.
16. Schaarschmidt F. 1974 Mosellophyton hefteri n. g. n.
sp. (?Psilophyta), ein sukkulenter Halophyt aus dem
Unterdevon von Alken an der Mosel. Pala
¨ontol. Z.
48, 188–204. (doi:10.1007/BF02985950)
17. Schweitzer H-J. 2000 Neue Pflanzenfunde im
Unterdevon der Eifel (Deutschland).
Senckenbergiana lethaea 80, 371– 395. (doi:10.
1007/BF03043355)
18. Chiarugi A. 1934 Tronchi silicizzati di un’ alga
arborea Silurico-Devoniana ‘Nematophyton
saharianum’ n. sp. nel Deserto Libico presso le Oasi
di Cufra. Nuovo Giorn. Bot. Ital. 40, 590– 595.
19. Arbey F, Koeniguer J-C. 1979 Les Ne
´matophytes et
les algueraies de l’ Ordovicien et Devonien saharien.
Bull. Centr. Rech. Explor.– Prod. Elf-Aquitaine 3,
409–418.
20. Bahafzallah A, Jux U, Omara S. 1981 Occurrence of
Prototaxites in Saudi Arabia. Bull. Fac. Earth Sci.,
King Abdul Aziz University, Jeddah, Kingdom of Saudi
Arabia 4, 43– 55.
21. Meissner CRJ, Dini SM, Farasani AM, Riddler GP, van
Eck M, Aspinall NC. 1989 Preliminary Geologic Map
of the Al Jawf Quadrangle, sheet 29D, Kingdom of
Saudi Arabia. Report prepared by the U.S.
Geological Survey in cooperation with the Deputy
Ministry for Mineral Resources, Saudi Arabia, Open-
File Report 89– 342. http://pubs.usgs.gov/of/1989/
0342/report.pdf.
22. Lo
¨hnertz W. 2003 Eocene paleovalleys in the Eifel:
mapping, geology, dating and implications for the
reconstruction of the paleosurfaces and vertical
movements of the lithosphere at the edges of the
Rhenish shield. Ge
´ol. France 1, 57– 62.
23. Altmeyer H. 1978 Die Prototaxiten von Arenrath.
Grondboor & Hamer 32, 9–12.
24. Carruthers W. 1872 On the history, histological
structure, and affinities of Nematophycus logani
Carr. (Prototaxites logani Dawson), an alga of
Devonian age. Monthly Microsc. J. 8, 160– 172.
(doi:10.1111/j.1365-2818.1872.tb02193.x)
25. Penhallow DP. 1888 On Nematophyton and allied
forms from the Devonian (Erian) of Gaspe and Bay
des Chaleurs. Royal Society of Canada’s (RSC)
Transactions and Proceedings 1st ser., 6, sect. iv,
27–47.
26. Dawson JW. 1888 The geological history of plants.
New York,NY: D. Appleton.
27. Jonker FP. 1979 Prototaxites in the Lower Devonian.
Palaeontographica 171B, 39– 56.
28. Graham LE, Cook ME, Hanson DT, Pigg KB, Graham
JM. 2010 Rolled liverwort mats explain major
Prototaxites features. Am. J. Bot. 97, 1079– 1108.
(doi:10.3732/ajb.1000172)
29. Graham LE, Cook ME, Hanson DT, Pigg KB, Graham
JM. 2010 Structural, physiological, and stable
carbon isotopic evidence that the enigmatic
Paleozoic fossil Prototaxites formed from rolled
liverwort mats. Am. J. Bot. 97, 268– 275. (doi:10.
3732/ajb.0900322)
30. Boyce CK, Hotton CL. 2010 Prototaxites was not a
taphonomic artefact. Am. J. Bot. 97, 1073. (doi:10.
3732/ajb.1000104)
31. Taylor TN, Taylor EL, Decombeix A-L,
Schwendemann A, Serbet R, Escapa I, Krings M.
2010 The enigmatic Devonian fossil Prototaxites is
not a rolled-up liverwort mat: Comment on the
paper by Graham et al. (AJB 97: 268–275).
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
12
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

Am. J. Bot. 97, 1074 –1078. (doi:10.3732/ajb.
1000047)
32. Church AH. 1919 Thalassiophyta and the subaerial
transmigration. Oxford Bot. Mem. 3, 1– 95.
33. Edwards D, Axe L, Honegger R. 2013 Contributions
to the diversity in cryptogamic covers in the mid-
Palaeozoic: Nematothallus revisited. Bot. J. Linn. Soc.
173, 505–534. (doi:10.1111/boj.12119)
34. Honegger R, Edwards D, Axe L. 2013 The earliest
records of internally stratified cyanobacterial and
algal lichens from the Lower Devonian of the Welsh
Borderland. New Phytol. 197, 264–275. (doi:10.
1111/nph.12009)
35. Strullu-Derrien C, Spencer ART, Goral T, Dee J,
Honegger R, Kenrick P, Longcore JE, Berbee ML.
2017 New insights into the evolutionary history of
Fungi from a 407 Ma Blastocladiomycota fossil
showing a complex hyphal thallus. Phil.
Trans. R. Soc. B 373, 20160502. (doi:10.1098/rstb.
2016.0502)
36. Kawai H, Hanyuda T, Uwai S. 2016 Evolution and
biogeography of laminarialean kelps. In Seaweed
phylogeography: adaptation and evolution of seaweeds
under environmental change (eds Z-M Hu, C Fraser),
pp. 227–249. Dordrecht, The Netherlands: Springer.
37. Taylor T, Hass H, Kerp H. 1999 The oldest fossil
ascomycetes. Nature 399, 648. (doi:10.1038/21349)
38. Taylor TN, Hass H, Kerp H, Krings M, Hanlin RT.
2005 Perithecial ascomycetes from the 400 million
year old Rhynie chert: an example of ancestral
polymorphism. Mycologia 97, 269–285. (doi:10.
1080/15572536.2006.11832862)
39. Prieto M, Wedin M. 2013 Dating the diversification
of the major lineages of ascomycota (Fungi).
PLoS ONE 8, e65576. (doi:10.1371/journal.pone.
0065576)
40. Beimforde C et al. 2014 Estimating the Phanerozoic
history of the Ascomycota lineages: combining fossil
and molecular data. Mol. Phylogen. Evol. 77,
307–319. (doi:10.1016/j.ympev.2014.04.024)
41. Schmitt I, del Prado R, Grube M, Lumbsch HT. 2009
Repeated evolution of closed fruiting bodies is
linked to ascoma development in the largest group
of lichenized fungi (Lecanoromycetes, Ascomycota).
Mol. Phylogen. Evol. 52, 34–44. (doi:10.1016/j.
ympev.2009.03.017)
42. Schmitt I, Mueller G, Lumbsch HT. 2005 Ascoma
morphology is homoplaseous and phylogenetically
misleading in some pyrenocarpous lichens.
Mycologia 97, 362–374. (doi:10.1080/15572536.
2006.11832813)
43. Lumbsch HT, Huhndorf SM. 2007 Whatever
happened to the pyrenomycetes and
loculoascomycetes? Mycol. Res. 111, 1064–1074.
(doi:10.1016/j.mycres.2007.04.004)
44. Lumbsch HT, Schmitt I, Mangold A, Wedin M. 2007
Ascus types are phylogenetically misleading in
Trapeliaceae and Agyriaceae (Ostropomycetidae,
Ascomycota). Mycol. Res. 111, 1133–1141. (doi:10.
1016/j.mycres.2007.06.016)
45. Breitenbach J, Kra
¨nzlin F. 1981 Pilze der Schweiz.
Band 1 Ascomyceten. Luzern, Switzerland:
Mykologia Verlag.
46. Prieto M, Baloch E, Tehler A, Wedin M. 2013
Mazaedium evolution in the Ascomycota (Fungi)
and the classification of mazaediate groups of
formerly unclear relationship. Cladistics 29,
296–308. (doi:10.1111/j.1096-0031.2012.00429.x)
47. Tuovila H, Cobbinah JR, Rikkinen J. 2011
Chaenothecopsis khayensis, a new resinicolous
calicioid fungus on African mahogany. Mycologia
103, 610–615. (doi:10.3852/10-194)
48. Go
¨ppert HR, Berend GC. 1845 1. Bd. I. Abth. Der
Bernstein und die in ihm befindlichen Pflanzenreste
der Vorwelt. In Die im Bernstein befindlichen
organischen Reste der Vorwelt, gesammelt in
Verbindung mit Mehreren bearbeitet und
herausgegeben (ed. GC Berendt). Berlin, Germany:
Nicolaische Buchhandlung.
49. Rikkinen J. 2003 Calicioid lichens from European
Tertiary amber. Mycologia 95, 1032– 1036. (doi:10.
1080/15572536.2004.11833019)
50. Rikkinen J, Poinar G. 2000 A new species of
resinicolous Chaenothecopsis (Mycocaliciaceae,
Ascomycota) from 20 million year old Bitterfeld
amber, with remarks on the biology of resinicolous
fungi. Mycol. Res. 104, 7–15. (doi:10.1017/
S0953756299001884)
51. Tuovila H, Schmidt AR, Beimforde C, Do
¨rfelt H,
Grabenhorst H, Rikkinen J. 2013 Stuck in time—a
new Chaenothecopsis species with proliferating
ascomata from Cunninghamia resin and its fossil
ancestors in European amber. Fungal Divers. 58,
199–213. (doi:10.1007/s13225-012-0210-9)
52. Eriksson OE, Winka K. 1997 Supraordinal taxa of
Ascomycota. Myconet 1, 1–16.
53. Landvik S, Schumacher TK, Eriksson OE, Moss ST.
2003 Morphology and ultrastructure of Neolecta
species. Mycol. Res. 107, 1021–1031. (doi:10.1017/
S0953756203008219)
54. Landvik S. 1996 Neolecta, a fruit-body-producing
genus of the basal ascomycetes, as shown by SSU
and LSU rDNA sequences. Mycol. Res. 100, 199–
202. (doi:10.1016/S0953-7562(96)80122-5)
55. Landvik S, Eriksson OE, Gargas A, Gustafsson P.
1993 Relationships of the genus Neolecta (Neolectales
ordo nov., Ascomycotina) inferred from 18S rDNA
sequences. Systema Ascomycetum 11, 107 –115.
56. Landvik S, Eriksson E, Berbee ML. 2001 Neolecta—
a fungal dinosaur? Evidence from b-tubulin amino
acid sequences. Mycologia 93, 1151–1163. (doi:10.
2307/3761675)
57. Schmid R. 1976 Septal pores in Prototaxites,an
enigmatic Devonian plant. Science 191,287– 288.
(doi:10.1126/science.191.4224.287)
58. Healy RA, Kumar TKA, Hewitt DA, McLaughlin DJ.
2013 Functional and phylogenetic implications of
septal pore ultrastructure in the ascoma of Neolecta
vitellina.Mycologia 105, 802–813. (doi:10.3852/
12-347)
59. Tenney K, Hunt I, Sweigard J, Pounder JI, McClain C,
Bowman EJ, Bowman BJ. 2000 Hex-1, a gene
unique to filamentous fungi, encodes the major
protein of the Woronin body and functions as a
plug for septal pores. Fungal Genet. Biol. 31,
205–217. (doi:10.1006/fgbi.2000.1230)
60. Dhavale T, Jedd G. 2007 The fungal Woronin body.
In Biology of the fungal cell, 2nd edition. The Mycota
vol. VIII (eds RJ Howard, NAR Gow), pp. 87– 96.
Berlin, Germany: Springer.
61. Reeb V, Lutzoni F, Roux C. 2004 Contribution of
RPB2 to multilocus phylogenetic studies of the
euascomycetes (Pezizomycotina, Fungi) with special
emphasis on the lichen-forming Acarosporaceae and
evolution of polyspory. Mol. Phylogen. Evol. 32,
1036–1060. (doi:10.1016/j.ympev.2004.04.012)
62. Aptroot A, Schumm F. 2012 The genus
Melanophloea, an example of convergent evolution
towards polyspory. Lichenologist 44, 501–509.
(doi:10.1017/S0024282912000035)
63. Redhead SA. 1977 The genus Neolecta (Neolectaceae
fam. nov., Lecanorales, Ascomycetes) in Canada.
Can. J. Bot. 55, 301– 306. (doi:10.1139/b77-041)
64. Edwards D. 1982 Fragmentary non-vascular plant
microfossils from the late Silurian of Wales.
Bot. J. Linn. Soc. 84, 223 –256. (doi:10.1111/j.1095-
8339.1982.tb00536.x)
65. Edwards D. 1986 Dispersed cuticles of putative non-
vascular plants from the Lower Devonian of Britain.
Bot. J. Linn. Soc. 93, 259 –275. (doi:10.1111/j.1095-
8339.1982.tb01025.x)
66. Edwards D, Axe L. 2012 Evidence for a fungal
affinity for Nematasketum, a close ally of
Prototaxites.Bot. J. Linn. Soc. 168, 1–18. (doi:10.
1111/j.1095-8339.2011.01195.x)
67. Hillier RD, Edwards D, Morrissey LB. 2008
Sedimentological evidence for rooting structures in
the Early Devonian Anglo-Welsh Basin (UK), with
speculation on their producers. Palaeogeogr.
Palaeoclim. Palaeoecol. 270, 366– 380. (doi:10.
1016/j.palaeo.2008.01.038)
68. Hobbie EA, Boyce CK. 2010 Carbon sources
for the Palaeozoic giant fungus Prototaxites inferred
from modern analogues. Proc. R. Soc. B 277,
2149–2156. (doi:10.1098/rspb.2010.0201)
69. Boyce CK, Hotton CL, Fogel ML, Cody GD, Hazen
RM, Knoll AH, Hueber FM. 2007 Devonian landscape
heterogeneity recorded by a giant fungus. Geology
35, 399–402. (doi:10.1130/G23384A.1)
70. Selosse MA. 2002 Prototaxites: a 400 Myr old giant
fossil, a saprophytic basiodiomycete, or a lichen?
Mycol. Res. 106, 641–644. (doi:10.1017/
S0953756202216317)
71. Selosse M-A, Strullu-Derrien C. 2015 Origins of the
terrestrial flora: a symbiosis with fungi? BIO Web of
Conf. 4, 00009. (doi:10.1051/bioconf/20150400009)
72. Oberwinkler F. 2012 Basidiolichens. In Fungal
associations, 2nd edn, The Mycota 9 (ed. B Hock),
pp. 341–335. Berlin, Germany: Springer.
73. Honegger R, Brunner U. 1981 Sporopollenin in the
cell wall of Coccomyxa and Myrmecia phycobionts of
various lichens: an ultrastructural and chemical
investigation. Can. J. Bot. 59, 2713– 2734. (doi:10.
1139/b81-322)
74. Honegger R, Axe L, Edwards D. 2013 Bacterial
epibionts and endolichenic actinobacteria and fungi
in the Lower Devonian lichen Chlorolichenomycites
salopensis.Fungal Biol. 117, 512– 518. (doi:10.
1016/j.funbio.2013.05.003)
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
13
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from

75. Heckman DS, Geiser DM, Eidell BR, Stauffer RL, Kardos
NL, Hedges SB. 2001 Molecular evidence for the early
colonization of land by fungi and plants. Science 293,
1129– 1133. (doi:10.1126/science.1061457)
76. Hedges S, Blair J, Venturi M, Shoe JL. 2004 A
molecular timescale of eukaryote evolution and the
rise of complex multicellular life. BMC Evol. Biol. 4,
2. (doi:10.1186/1471-2148-4-2)
77. Padovan ACB, Sanson GFO, Brunstein A, Briones MRS.
2005 Fungi evolution revisited: application of the
penalized likelihood method to a Bayesian fungal
phylogeny provides a new perspective on
phylogenetic relationships and divergence dates of
Ascomycota groups. J. Mol. Evol. 60, 726– 735.
(doi:10.1007/s00239-004-0164-y)
78. Sugiyama J, Hosaka K, Suh S-O. 2006 Early
diverging Ascomycota: phylogenetic divergence
and related evolutionary enigmas. Mycologia
98, 996–1005. (doi:10.1080/15572536.2006.
11832628)
79. Taylor J, Berbee M. 2006 Dating divergences in the
Fungal Tree of Life: review and new analyses.
Mycologia 98, 838–849. (doi:10.1080/15572536.
2006.11832614)
80. Lu¨cking R, Huhndorf S, Pfister DH, Plata ER,
Lumbsch HT. 2009 Fungi evolved right on track.
Mycologia 101, 810–822. (doi:10.3852/09-016)
81. Berbee ML, Taylor JW. 2010 Dating the molecular
clock in fungi—how close are we? Fungal Biol. Rev.
24, 1–16. (doi:10.1016/j.fbr.2010.03.001)
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20170146
14
on December 18, 2017http://rstb.royalsocietypublishing.org/Downloaded from