Epiphytic leafy liverworts diversified in
angiosperm-dominated forests
Kathrin Feldberg1, Harald Schneider2, Tanja Stadler3, Alfons Scha¨fer-Verwimp4, Alexander R. Schmidt5
& Jochen Heinrichs1
1Systematische Botanik und Mykologie, Ludwig-Maximilians-Universita¨t Mu¨nchen, 80638 Mu¨nchen, Germany, 2Department of Life
Sciences, Natural History Museum, London SW75BD, United Kingdom, 3Department of Biosystems Science & Engineering, ETH
Zu¨rich, 4058 Basel, Switzerland, 4Mittlere Letten 11, 88634 Herdwangen-Scho¨nach, Germany, 5Courant Research Centre
Geobiology, Georg-August-Universita¨t Go¨ttingen, 37077 Go¨ttingen, Germany.
Recent studies have provided evidence for pulses in the diversification of angiosperms, ferns, gymnosperms,
and mosses as well as various groups of animals during the Cretaceous revolution of terrestrial ecosystems.
However, evidence for such pulses has not been reported so far for liverworts. Here we provide new insight
into liverwort evolution by integrating a comprehensive molecular dataset with a set of 20 fossil age
constraints. We found evidence for a relative constant diversification rate of generalistic liverworts
(Jungermanniales) since the Palaeozoic, whereas epiphytic liverworts (Porellales) show a sudden increase of
lineage accumulation in the Cretaceous. This difference is likely caused by the pronounced response of
Porellales to the ecological opportunities provided by humid, megathermal forests, which were increasingly
available as a result of the rise of the angiosperms.
I
n recent years, the hypothesis of the Cretaceous Terrestrial Revolution (KTR) was introduced in response to
the increasing accumulation of evidence for a substantial change in the terrestrial biodiversity around 110 to 70
million years ago (Ma)1,2. The KTR is marked by the fact that terrestrial diversity exceeded the diversity of
marine habitats for the first time3 and that it coincided with the rapid rise of the angiosperms3–10. Evidence for
coinciding Cretaceous pulses of increased diversification have been found in many lineages of terrestrial organ-
isms including various lineages of land plants11–13, insects14–17, and tetrapods1,18. In turn, the coincidence of KTR
and rise of the angiosperms implies some correlation of these events, although the causality is currently unknown.
However, this argument is consistent with the reports on the coincidences of changes in the ‘‘tree of life’’ and the
rise of the angiosperms. Examples of coinciding diversification include those observed in ants14, and ‘‘derived
ferns’’11 as well as changes in the lineage composition of dinosaurs19.
Further support of a causal link between the rise of angiosperms and the KTRwas provided by ecophysiological
studies providing evidence for a connection of several traits of angiosperms and the establishment of humid
megathermal forests2,20–24. This scenario challenged the hypothesis of the replacement of lineages considered as
primitive such as bryophytes, ferns, and gymnosperms by angiosperms as predicted by the sequential replace-
ment hypothesis4,25. In contrast, a new hypothesis emerged in which various lineages of organisms underwent
increased diversification during the KTR and early Cenozoic in response to the rapid spread of angiosperm-
dominated habitats9,10.
In contrast to other land plants, no evidence has been found for pulses in the diversification of liverworts that
coincide with the KTR or post-KTR events12,26–28. This is remarkable given the evidence for KTR and post-KTR
pulses of diversification in nearly all other groups of land plants and the ecological association of many liverwort
species and angiosperms12,29,30. Previous studies reported either constant or declining diversification rates, indi-
cating that liverworts are either in an equilibrium state or are actually in decline as expected under the replace-
ment hypothesis. However, these constant rates may be an artefact because analyses of diversification rates are
highly sensitive to sampling density31. These arguments resemble to some degree the controversy about the
impact of the Cretaceous-Palaeogene mass-extinction event on land-plant diversity32. High-resolution regional
samplings were required to detect the impact of mass extinctions on plant diversity32,33. In turn, these arguments
suggest the need for a high-resolution study that focuses on the unbalanced distribution of species diversity across
liverwort phylogeny and the high proportion of epiphytes in some lineages.
Here, we explore the hypothesis that epiphytic liverworts diversified in the shadow of angiosperms while
generalist liverworts were less affected by the KTR. The expectation of an increased speciation rate of epiphytic
OPEN
SUBJECT AREAS:
BIOGEOGRAPHY
BIODIVERSITY
PHYLOGENETICS
Received
23 April 2014
Accepted
21 July 2014
Published
7 August 2014
Correspondence and
requests for materials
should be addressed to
J.H. (jheinrichs@lmu.de)
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 1
liverworts is supported by the argument that the rise of angiosperms
increased the occurrence of humid forests20. The vast majority of
extant epiphytic liverworts are found in rainforests dominated by
angiosperms, which suggests a close link between angiosperms and
epiphytic liverworts. This hypothesis is also supported by evidence
from other studies of a causal link between angiosperm forests and
epiphytism in ferns34, and the connection between leaf-shape evolu-
tion of southern hemisphere conifers and angiosperms35. To address
this hypothesis, we selected the Palaeozoic sister orders Junger-
manniales and Porellales28 because of their high species diversity,
of about 2,600 and 2,300 species respectively, and the substantial
difference between them in the proportion of epiphytic species.
The majority of Porellales species grow either epiphytically or epi-
phyllically, whereas Jungermanniales show a dominance of species
with a more generalistic strategy. Based on evidence provided in
ecophysiological studies on early angiosperms, we expected the
majority of epiphytic and epiphyllic liverworts to have diversified
either during the KTR or after the Cretaceous-Palaeogene (K-Pg)
boundary and, in contrast, generalists to have accumulated their
diversity under a rather constant diversification rate as inferred in
preceding studies12,27. Specifically, we explore this hypothesis by
examining the evidence supporting two pertinent arguments: 1) that
the two orders show distinct patterns of diversification during the
Mesozoic and early Cenozoic; 2) that these differences coincide with
the ecological preferences of themajority of species belonging to each
order, in particular the strong preference in Porellales for epiphytic
habitats.
Results
Lineage diversification. The orders Jungermanniales and Porellales
showdistinct patterns of biodiversity assembly through theMesozoic
and early Cenozoic (Figs 1A–1D, Supplementary 2.3–2.7). Junger-
manniales accumulated their diversity with a relatively constant
diversification rate until about 20–40 Ma, whereas Porellales show
a delay of the accumulation of lineages from the Permian-Triassic
border onwards followed by a sudden increase in the Cretaceous
(Fig. 1B, Supplementary 2.4). A decline of the diversification rate
from ca. 20–40 Ma onwards is indicated for both orders (Fig. 1B,
Supplementary 2.3). This is consistent with the negative gamma
statistics of constant diversity accumulation over time with -5.98
for Jungermanniales and -2.12 for Porellales (Supplementary 2.4).
Compared to Jungermanniales, the Porellales show not only
evidence for the shared reduction of the diversification rate around
40 Mabut also an evident increase of the diversification from themid
Cretaceous to the early Cenozoic (Fig. 1B, Supplementary 2.5).
Evidence was found for statistically significant changes in the
diversification rates of leafy liverworts (Fig. 1B, Supplementary 2.6
and 2.7). Diversification models assuming non-constant rates fitted
better with the recovered phylogeny and main clades recognised
(Suppementary 2.6), and up to seven significant shifts in the diver-
sification rates were recovered in the MEDUSA analyses (Fig. 1A,
Supplementary 2.7). Three breakpoints with rate slow-downs were
associated with nodes containing rather few extant species (Fig. 1A,
Supplementary 2.7; breakpoints 1, 2 & 6), whereas two out of the four
rate changes indicating increased rates were associated with the two
most species rich families of liverworts, the Lejeuneaceae and
Plagiochilaceae (Fig. 1A, Supplementary 2.7; breakpoints 3, 4 & 5).
The results suggested an increased rate during the KTR associated
with the onset of the diversification of species-rich families in
Porellales such as Lejeuneaceae and Frullaniaceae (Fig. 1B,
Supplementary 2.3–2.8). These results are consistent with the esti-
mates of birth-death rates obtained using DIVERSITREE (Supple-
mentary 2.9), which indicate a slightly higher speciation rate for
Porellales (l 5 0.0120) compared to Jungermanniales (l 5
0.0093) in the Bayesian mcmc estimation with the extinction rate
(m) to be close to zero.
Differences in diversification between epiphytes and generalists.
A much larger fraction of Porellales grows preferably as epiphytes,
with . 95% in Porellales versus , 30% in Jungermanniales. BiSSE
analyses with DIVERSITREE found evidence supporting the
hypothesis of interdependence of epiphytism and diversification
rates (Table 1). A diversification rate of lE50.08906 for epiphytes
and lG50.07725 for generalists was found with the six parameter
model; the other models likewise indicated higher diversification
rates of epiphytes compared to generalists (Table 1, Supplementary
2.8). The selected models suggested a significant (p , 0.005 in
Table 1) interdependence of the diversification rates and prefe-
rence scored as epiphyte or generalist combined with the asymme-
tric transition between the two preference states. The loss of
epiphytism was found to be more frequent than the gain of epiphy-
tism (model with six parameters, loss qE-G 5 0.008 vs. gain qG-E 5
0.0006 with a significance of p, 0.005; Supplementary 2.8 and 2.9).
This is consistent with the observation that more than 95% of the
putative transitions to epiphytism were younger than 150 million
years (Fig. 1C). More importantly, the number of cladogenesis
events per 10 million years (Fig. 1D) showed a substantially higher
increase during the KTR compared to the Late Jurassic and Early
Cretaceous in epiphytes (x5.7) than in generalists (x1.8). Notably,
53% of cladogenesis events involving epiphytes happened in the KTR
(110-70 Ma) and 70% in the period from 110 to 30 Ma. In contrast,
conspicuously smaller proportions, 25% and 53% respectively,
contributed to the total of cladogenesis events associated with
generalists.
Discussion
Our analyses point to remarkable differences in the diversification
rates of the predominantly epiphytic Porellales and the predomi-
nantly generalistic Jungermanniales. Porellales show a sudden
increase of lineage accumulation in the Cretaceous, whereas
Jungermanniales had a relatively constant diversification rate since
the Palaeozoic. The declining accumulation of both lineages around
40 Ma in the LTTPs (Fig. 1B, Supplementary 2.3) and the TreePAR
analyses (Supplementary 2.5) is very likely the result of the density of
the taxon sampling, but the differences between Porellales and
Jungermanniales in the LTTPs, gamma values, and TreePAR ana-
lyses (Supplementary 2.3–2.5) are unlikely to be the result of unbal-
anced sampling, because we sampled,5% of the extant diversity of
each order. Within each order, taxa within families were sampled
randomly under the constraint of the proportional contribution of
families to the taxon diversity of the order. The low transition rates
from a generalistic to an epiphyticmode of life estimated in the BiSSE
analysis (Table 1, Supplementary 2.8) indicate that the delayed diver-
sification of the Porellales crown group may be caused by their pref-
erence for epiphytic niches. This argument is consistent with the
results of diversification models recovering slightly lower extinction
rates and higher speciation rates for Porellales than those estimated
for Jungermanniales. The divergence time estimated for the rise of
the Porellales suggests a coincidence of increased accumulation of
epiphytic diversity with the rise of the angiosperms.
The results of the LTTPs and the BiSSE analyses raise the issue of
why epiphytic liverworts appear to be more sensitive to the trans-
formation of the environment caused by angiosperms, compared to
generalists. The latter may show a higher ecological flexibility and
thus they may be able to colonize new habitats or tolerate local
environmental changes without a significant change in the rate of
diversification, meaning that extinction and speciation events are
balanced. Some of these taxa are also able to grow epiphytically,
but they are not restricted to this habitat. In contrast, species belong-
ing to epiphytic lineages exhibit a marked preference for forests with
high humidity. A considerable portion of Porellales shows a unique
preference for the surface of the leaves of vascular plants. Although
these epiphylls are not restricted to growing on angiosperms and are
www.nature.com/scientificreports
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 2
Figure 1 | Diversification of liverworts. (A) Mean age consensus chronogam with time scale. Green branches are epiphytes, black branches are
generalists. Grey horizontal bars show confidence intervals. Black dots indicate nodes with calibrations (see Supplementary 2.1), white dots with small
letters indicate the shifts in the diversification rates (breakpoints 1 to 7 estimated by theMEDUSA analysis, see Supplementary 2.7). Vertical bars indicate
higher taxonomic units marked using the colour code shown at the lower left corner of the figure. We employed the current classification differentiating
into classes (-opsida), subclasses (-idae), and orders (-ales). (B) Lineage through time plot for the Jungermanniopsida (purple), Jungermanniidae (blue),
Jungermanniales (orange), Porellales (red). Arrows indicate rate shifts as estimated with LASER (Supplementary 2.6) with rate decline indicated by
arrows pointing downwards and rate incline by arrows pointing upwards. (C) Accumulation of epiphytic and generalist species diversity through time
(green epiphytes, black generalists). (D) Number of cladogenesis events through time for epiphytes (green), generalists (black), both (blue).
www.nature.com/scientificreports
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 3
also found on the leaves of other land plants such as ferns, epiphyllic
species are nested in clades that diversified not earlier than the Late
Cretaceous26. In general, there is no strict host preference for epi-
phytic liverworts36–38 and substrate preferences are likely shaped by
humidity of the environment with a strong tendency for epiphytic
growth in superhumid forests39.
The new results on the diversification of liverworts provide further
support for the hypothesis that various groups of land plants
exploited the opportunities created by the rise of the angiosperms11.
In turn, our study provides further evidence for the rise of the angios-
perms as the putative trigger of the KTR. Specifically, the opportun-
ities for leafy liverworts were mainly provided by the expansion of
epiphytic and epiphyllic habitats offered by humid, megathermal
forests dominated by angiosperms. This resembles the situation of
ferns in humid, megathermal forests11 but the Cretaceous fern diver-
sificationwas likelymainly due to ferns exploiting terrestrial habitats.
The majority of epiphytic fern lineages, on the other hand, shows a
post K-Pg onset of their diversification34. Nonetheless, some epi-
phytic filmy ferns were found to diversify as early as the KTR40.
This difference may be explained by various factors, including pre-
adaptation to epiphytic habitats in leafy liverworts and filmy ferns
compared to other epiphytic fern groups, and by the higher tolerance
of epiphytic liverworts to disturbance, e.g. the K-Pg mass extinction.
The dissimilar accumulation of liverwort and fern epiphyte diversity
coincided with the phases of vein density increase reported for
angiosperms in the mid-Cretaceous and after the K-Pg boundary21.
In this context, it is important to note that some controversy still
exists concerning the origin of the first forests dominated by angios-
perms. Based on Cretaceous North American floras, angiosperm-
dominated forests were assumed to be not older than K-Pg.
However, increasing evidence has accumulated suggesting that
angiosperms may have formed a substantial component of the Late
Cretaceous forest vegetation in at least some areas10,41. Arguments
supporting the pre-Cenozoic existence of angiosperm-dominated
megathermal forests were also provided in studies focusing on trop-
ical angiosperm families such as the custard apple family
(Annonaceae42), palms (Arecaceae43), and rosids5.
In summary, we found evidence for distinct responses of leafy
liverwort clades to the opportunities and challenges provided by
the rise of the angiosperms. Ecological generalistsmay have exploited
these opportunities without changes in the speciation rates, whereas
epiphytic specialists show a strong increase in the diversification rate
coinciding with the rise of the angiosperms. In turn, the diversifica-
tion of ecological generalists appears not to be affected by the great
mass-extinction events, e.g. at the Permian-Triassic, Triassic-
Jurassic, and Cretaceous-Palaeogene boundaries or by the rise of
the angiosperms. For Porellales, the increase of the speciation rate
was found to be correlated with epiphytism and thus the evolutionary
success of the order is likely closely linked to the establishment of
humid forests dominated by angiosperms. We also considered the
alternative that the low diversification rate in the Triassic and
Jurassic is an artefact of our study having been based on extant taxa.
It is possible that epiphytic diversity occurring mainly on Mesozoic
gymnosperms was replaced by epiphytic diversity occurring mainly
on angiosperms. However, no evidence exists to support the assumed
lineage-specificity of epiphytic liverworts to either angiosperms or
gymnosperms. Instead, the humidity maintained in forests is the
most probable factor controlling the assembly of epiphytic liverwort
diversity.
Methods
Taxon sampling. In total, we sampled 303 species of liverworts with the two orders
Porellales and Jungermanniales sampled proportionally to represent about 5% of
their extant species diversity. We also sampled representatives of the Metzgeriidae
and Pelliidae (Supplementary Information 4: Accession numbers). The sampling was
designed based on our current understanding of the phylogeny and the systematics of
these liverworts44. Species numbers were compiled based on newly obtained estimates
and taking into account evidence for a higher number of currently unrecognized
species in Porellales versus Jungermanniales45,46. The obtained sampling is best
described as random but constrained by our current knowledge of the phylogeny and
distribution of species diversity. Several analyses were specifically applied to
determine the impact of the sampling density and species number estimates, e.g.
MEDUSA. The design of the dataset took into consideration challenges such as
limited sampling by carrying out analyses allowing to correct species numbers and by
employing a random sampling within well-defined clades.
Divergence time estimates. Bayesian divergence time estimates were generated with
the software package BEAST 1.6.247,48 using a dataset including 303 species and
sequences of the plastid gene rbcL. Twenty fossil calibrations were selected and
assigned based on recent research (Supplementary 1.2). The assignment of these
calibrations was based on our current understanding of the phylogeny and taxonomy
of the related liverwort clades whereas the assigned age intervals were based on the
recent literature on the topic (Supplementary 1.2.1 and 1.2.2). Age intervals were
integrated as age constraints with a uniform prior distribution for the minimum age
of the fossil and a maximum age of the root (475 Ma). The latter accounts for
uncertainties ofmaximum clade age due to the relative sparseness of fossils older than
35 Ma49,50. The maximum age of liverworts was set to 475 Ma based on the age of the
oldest known cryptospores51,52. Final estimates were obtained using the GTR1I1G
model as determined with jModeltest 1.053, lognormal relaxed molecular clock48, and
a birth-death tree model54. The analysis was run for 200,000,000 generations and by
sampling every 20,000th tree, so that the resulting tree file contained 10,000 trees.
Effective sample size was investigated with dedicated tools such as TRACER 1.5 that
are part of the BEAST package. After a burn-in of 25% amaximum credibility tree was
compiled with TreeAnnotator 1.6.2. Phylogenetic uncertainty, e.g. node robustness
and branch length, were taken into account by employing methods using at least 100
randomly selected chronograms generated in the BEAST analyses instead of the
single maximum credibility tree (Supplementary 2.1–2.4). We carefully explored the
distribution of substitution rate changes to detect biases such as incorrectly applied
time constraints and correlation between diversification rates and substitution rates.
No significant correlation was found.
Diversification rate analyses. These analyses were carried out with 100 randomly
selected chronograms or with the maximum credibility tree from the BEAST results
for Jungermanniopsida, Jungermanniidae, Jungermanniales, and Porellales (for
Table 1 | Results of the BiSSE analyses exploring the interdependence of diversification rate on ecological preference (epiphytic or gen-
eralist). Seven models were tested and compared using likelihood values (ln) and Akaike Information Criterion (AIC). The significance of the
differences between models was explored using a X2 test. The seven models considered the following parameters: speciation rate (l),
extinction rate (m), and character state transition rate (q). These models parameters were either treated as dependent on the ecological
preferences (?) or as constant (5). The upper threemodels were found to be the best fit using aX2 test and significance value of p,0.01. The
three selected models shared the prediction of the interdependence of character transition (q) and the two parameters (l and m) contributing
to the diversification rate
parameters model ln AIC X2 values p value
6 l0?l1,m0?m1,qG-E?qE-G 21729.0 3469.9
5 l05l1,m0?m1,qG-E?qE-G 21729.2 3468.4 0.469 0.493505
l0?l1,m05m1,qG-E?qE-G 21729.0 3468.0 0.051 0.821696
l0?l1,m0?m1,qG-E5qE-G 21732.7 3475.4 7.521 0.006097
4 l05l1,m05m1,qG-E?qE-G 21750.2 3508.4 42.501 5.901e-10
l0?l1,m05m1,qG-E5qE-G 21733.7 3475.4 9.462 0.008816
l05l1,m0?m1,qG-E5qE-G 21735.4 3478.8 12.852 0.001619
www.nature.com/scientificreports
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 4
details see Supplementary Information 1: Material and Methods). Lineage through
time plots (LTTPs) were generated with TreeSIM55, and the gamma statistic values56
were calculated in APE57 while the null distribution of the gamma statistic was
sampled with GEIGER58. Evidence for non-constant rates was further investigated in
LASER59 by testing the fit of several time-dependent models (assuming constant rates
or variable rates). The results of these analyses were investigated using the MEDUSA
approach as implemented in GEIGER58. TreePAR analyses60 were carried out to
detect changes in the diversification rates through time. A penalized likelihood
analysis was performed with the chronoplot function in APE57 to visualize changes in
the substitution rates across the chronograms. The analysis was performed for nine
values of the smoothing parameter lambda, and with the cross-validation option on
(Supplementary 2.2).
Diversification rate and ecological preference. The ecological preference of the
species included was inferred based on reports in the literature, our own observations,
and herbarium specimens. Two distinct conditions were found: 1) Generalists show a
wide range of preferences including terrestrial, saxicolous, and sometimes epiphytic
habitats. However, they rarely show a strong preference for a certain substrate. The
few epiphyte lineages of Jungermanniales do not show special morphological
adaptations such as complicate bilobed leaves with the ventral lobes modified into
pockets or watersacs; 2) Epiphytic taxa, in contrast, grow exclusively or nearly
exclusively on the branches, stems or leaves of vascular plants and have adapted
morphologically to this habitat by developing the above-mentioned types of leaves,
bundled rhizoids, and exclusively lateral branching that allows a flat growth of the
shoots pressed to their substrate61.
These data were used in BiSSE analyses62 calculated in DIVERSITREE63 (for details
see Supplementary 1.1.8). The distribution of cladogenesis events associated with
both classes was inferred by counting the number of events per time unit of 10 Ma for
the whole chronogram or for three age classes of 40 Ma each, pre KTA (150-110),
KTA (110-70), post-KTA (70-30). All analyses were designed and carried out con-
sidering the known challenges of BiSSE analyses62,64–66. Several analyses were
employed to explore the impact of the taxon sampling and phylogenetic uncertainty
to the obtained results (gamma statistics, Supplementary 2.4; mcmc-cross-validation
of the birth-death model, Supplementary 2.9).
1. Benton, M. The origins of modern biodiversity on land. Philos. Trans. Roy. Soc.
London B 365, 3667–3679 (2010).
2. Vermeij, G. J. The energetics of modernization: the last one hundredmillion years
of biotic evolution. Paleontol. Res. 15, 54–61 (2011).
3. Vermeij, G. J. & Grosberg, R. K. The great divergence: when did diversity on land
exceed that in the sea? Integ. Comp. Biol. 50, 675–682 (2010).
4. Dilcher, D. Toward a new synthesis: major evolutionary trends in angiosperm
fossil record. Proc. Natl. Acad. Sci. USA 97, 7030–7036 (2000).
5. Wang, H. et al. Rosid radiation and the rapid rise of angiosperm-dominated
forests. Proc. Natl. Acad. Sci. USA 106, 3853–3858 (2009).
6. Davies, T. J. et al. Darwin’s abominable mystery: Insights from a supertree of the
angiosperms. Proc. Natl. Acad. Sci. USA 101, 1904–1909 (2009).
7. Magallo´n, S. & Castillo, A. Angiosperm diversification through time. Am. J. Bot.
96, 349–365 (2009).
8. Clarke, J. T., Warnock, R. C. M. & Donoghue, P. C. J. Establishing a time-scale for
plant evolution. New Phytol. 192, 266–301 (2011).
9. Morley, R. J. [Cretaceous and Tertiary climate change and the past distribution of
megathermal rainforests]. Tropical Rainforest Responses to Climatic Change
[Bush, M. B., Flenley, J. R. & Gosling,W. D. (eds)] [1–34] (Springer, Berlin, 2011).
10. Coiffard, C., Gomez, B., Daviero-Gomez, V. &Dilcher, D. L. Rise to dominance of
angiosperm pioneers in European Cretaceous environments. Proc. Natl. Acad. Sci.
USA 109, 20955–20959 (2012).
11. Schneider, H. et al. Ferns diversified in the shadow of angiosperms. Nature 428,
553–557 (2004).
12. Fiz-Palacios, O., Schneider, H., Heinrichs, J. & Savolainen, V. Diversification of
land plants: insights from a family-level phylogenetic analysis. BMCEvol. Biol. 11,
341 (2011).
13. Villarreal, J. C. & Renner, S. S. Hornwort pyrenoids, carbon-concentrating
structures, evolved and were lost at least five times during the last 100 million
years. Proc. Natl. Acad. Sci. USA 109, 18873–18878 (2012).
14. Moreau, C. S., Bell, C. D., Vita, R., Archibald, S. B. & Pierce, N. E. Phylogeny of the
ants: diversification in the age of angiosperms. Science 312, 101–104 (2006).
15. Moreau, C. S. & Bell, C. D. Testing the museum versus cradle tropical biological
diversity hypothesis: phylogeny, diversification, and ancestral biogeographic
range evolution of the ants. Evolution 67, 2240–2257 (2013).
16. Wahlberg, N., Wheat, C. W. & Pena, C. Timing and patterns in the taxonomic
diversification of Lepidoptera (butterflies and moths). PLoS ONE 8, e80875
(2013).
17. Wilson, J. S., von Dohlen, C. D., Forister, M. L. & Pitts, J. P. Family-level
divergence in the stinging wasps (Hymenoptera: Aculeata), with correlations to
angiosperm diversification. Evol. Biol. 40, 101–107 (2013).
18. Meredith, R. W. et al. Impacts of the Cretaceous Terrestrial Revolution and KPg
extinction on mammal diversification. Science 334, 521–524 (2011).
19. Butler, R. J., Barrett, P. M., Kenrick, P. & Penn, M. G. Diversity patterns amongst
herbivorous dinosaurs and plants during the Cretaceous: implications for
hypotheses of dinosaur/angiosperm co-evolution. J. Evol. Biol. 22, 446–459
(2009).
20. Boyce, C. K. & Lee, J.-E. An exceptional role for flowering plant physiology in the
expansion of tropical rainforests and biodiversity. Proc. R. Soc. B 277, 3437–3443
(2010).
21. Feild, T. S. et al. Fossil evidence for Cretaceous escalation in angiosperm leaf vein
evolution. Proc. Natl. Acad. Sci. USA 108, 8363–8366 (2011).
22. de Boer, H. J., Eppinga, M. B.,Wassen, M. J. & Dekker, S. C. A critical transition in
leaf evolution facilitated the Cretaceous angiosperm revolution.Nature Commun.
3, 1221 (2012).
23. Boyce, C. K. & Leslie, A. B. The paleontological context of angiosperm vegetative
evolution. Int. J. Plant. Sci. 173, 561–568 (2012).
24. Peralta-Medina, E. & Falcon-Lang, H. J. Cretaceous forest composition and
productivity inferred from a global fossil wood database. Geology 40, 219–222
(2012).
25. Kenrick, P. & Crane, P. R. The Origin and Early Diversification of Land Plants: A
Cladistic Study (Smithsonian Institution Press, Washington DC, 1997).
26. Wilson, R., Heinrichs, J., Hentschel, J., Gradstein, S. R. & Schneider, H. Steady
diversification of derived liverworts under Tertiary climatic fluctuations. Biol.
Lett. 3, 566–569 (2007).
27. Feldberg, K., Heinrichs, J., Schmidt, A. R., Va´nˇa, J. & Schneider, H. Exploring the
impact of fossil constraints on the divergence time estimates of derived liverworts.
Plant. Syst. Evol. 299, 585–601 (2013).
28. Cooper, E. D., Henwood, M. J. & Brown, E. A. Are the liverworts really that old?
Cretaceous origins and Cenozoic diversifications in Lepidoziaceae reflect a
recurrent theme in liverwort evolution. Biol. J. Linn. Soc. 107, 425–441 (2012).
29. Nagallingum, N. S. et al. Recent synchronous radiation of a living fossil. Science
334, 796–799 (2011).
30. Davis, C. C. & Scha¨fer, H. Plant evolution: pulses of extinction and speciation in
gymnosperm diversity. Curr. Biol. 21, R995–R998 (2013).
31. Cusimano, N. & Renner, S. S. Slowdowns in diversification rates from real
phylogenies may not be real. Syst. Biol. 59, 458–464 (2010).
32. McElwain, J. C. & Punyasena, S. W. Mass extinction events and the plant fossil
record. Trends Ecol. Evol. 22, 548–557 (2007).
33. Nichols, D. J. & Johnson, K. R. Plants and the K-T Boundary (Cambridge
University Press, New York, 2008).
34. Schuettpelz, E. & Pryer, K. M. Evidence for a Cenozoic radiation of ferns in an
angiosperm-dominated canopy. Proc. Natl. Acad. Sci. USA 106, 11200–11205
(2009).
35. Biffin, E., Brodribb, T. J., Hill, R. S., Thomas, P. & Lowe, A. J. Leaf evolution in
Southern Hemisphere conifers tracks the angiosperm ecological radiation. Proc.
R. Soc. B 279, 341–348 (2012).
36. Cornelissen, J. H. C. & Ter Steege, H. Distribution and ecology of epiphytic
bryophytes and lichens in dry evergreen forest of Guyana. J. Trop. Ecol. 5, 131–150
(1989).
37. Gonza´les-Manzebo, J. M., Losada-Lima, A. & McAllister, S. Host specificity of
epiphytic bryophyte communities of a laurel forest of Tenerife (Canary Islands,
Spain). Bryologist 106, 383–394 (2003).
38. Sonnleitner, M., Dullinger, S., Wanek, W. & Zechmeister, H. Microclimatic
patterns correlate with the distribution of epiphyllous bryophytes in a tropical
lowland rain forest in Costa Rica. J. Trop. Ecol. 25, 321–330 (2009).
39. Po´cs, T. (1996) Epiphyllous liverwort diversity at worldwide level and its threat
and conservation. Anales Inst. Biol. Univ. Nac. Auto´n. Me´xico, Ser. Bot. 67,
109–127 (1996).
40. Hennequin, S., Schuettpelz, E., Pryer, K. M., Ebihara, A. & Dubuisson, J. Y.
Divergence times and the evolution of epiphytism in filmy ferns
(Hymenophyllaceae) revisited. Int. J. Plant Sci. 169, 1278–1287 (2008).
41. Friis, E.M., Pedersen, K. R. & Crane, P. R. Diversity in obscurity: fossil flowers and
the early history of angiosperms. Phil. Trans. R. Soc. B 365, 369–382 (2010).
42. Couvreur, T. L. P. et al. Early evolutionary history of the flowering plant family
Annonaceae: steady diversification and boreotropical geodispersal. J. Biogeogr.
38, 664–680 (2011).
43. Couvreur, T. L. P., Forest, F. & Baker, W. J. Origin and global diversification
patterns of tropical rain forests: inferences from a complete genus-level phylogeny
of palms. BMC Biology 9, 44 (2011).
44. Crandall-Stotler, B., Stotler, R. E. & Long, D. G. Phylogeny and classification of the
Marchantiophyta. Edinburgh J. Bot. 66, 155–198 (2009).
45. Heinrichs, J. et al. One species or at least eight? Delimitation and distribution of
Frullania tamarisci (L.) Dumort. s.l. (Jungermanniopsida, Porellales) inferred
fromnuclear and chloroplast DNAmarkers.Mol. Phylogenet. Evol. 56, 1105–1114
(2010).
46. Renner, M. A. M. et al. Integrative taxonomy resolves the cryptic and pseudo-
cryptic Radula buccinifera complex (Porellales, Jungermanniopsida), including
two reinstated and five new species. PhytoKeys 27, 1–113 (2013).
47. Drummond, A. J., Ho, S. Y. W., Phillips, M. J. & Rambaut, A. Relaxed
phylogenetics and dating with confidence. PLoS Biol. 4, e88 (2006).
48. Drummond, A. J. & Rambaut, A. BEAST: Bayesian evolutionary analysis by
sampling trees. BMC Evol. Biol. 7, 214 (2007).
49. Donoghue, P. C. J. & Benton, M. J. Rocks and clocks: calibrating the Tree of Life
using fossils and molecules. Trends Ecol. Evol. 22, 424–431 (2007).
www.nature.com/scientificreports
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 5
50. Dornburg, A., Beaulieu, J. M., Oliver, J. C. & Near, T. J. Integrating fossil
preservation biases in the selection of calibrations for molecular divergence time
estimation. Syst. Biol. 60, 519–527 (2011).
51. Wellman, C. H., Osterloff, P. L. & Mohiuddin, U. Fragments of the earliest land
plants. Nature 425, 282–284 (2003).
52. Kenrick, P., Wellman, C. H., Schneider, H. & Edgecombe, G. D. A timeline for
terrestrialization: consequences for the carbon cycle in the Palaeozoic. Phil. Trans.
R. Soc. B 367, 519–536 (2012).
53. Posada, D. jModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25,
1253–1256 (2008).
54. Gernhard, T. The conditioned reconstructed process. J. Theor. Biol. 253, 769–778
(2008).
55. Stadler, T. Simulating trees with a fixed number of extant species. Syst. Biol. 60,
676–684 (2011).
56. Pybus, O. G. & Harvey, P. H. Testing macro-evolutionary models using
incomplete molecular phylogenies. Proc. R. Soc. Lond. B 267, 2267–2272 (2000).
57. Paradis, E., Claude, J. & Strimmer, K. APE: analyses of phylogenetics and
evolution in R language. Bioinformatics 20, 289–290 (2004).
58. Harmon, L. J., Weir, J. T., Brock, C. D., Glor, R. E. & Challenger, W. GEIGER:
investigating evolutionary radiations. Bioinformatics 24, 129–131 (2008).
59. Rabosky, D. L. LASER: amaximum likelihood toolkit for detecting temporal shifts
in diversification rates from molecular phylogenies. Evol. Bioinform. Online 2,
247–250 (2006).
60. Stadler, T. Mammalian phylogeny reveals recent diversification rate shifts. Proc.
Natl. Acad. Sci. USA 108, 6187–6192 (2011).
61. Heinrichs, J., Gradstein, S. R., Wilson, R. & Schneider, H. Towards a natural
classification of liverworts (Marchantiophyta) based on the chloroplast gene rbcL.
Cryptog. Bryol. 26, 131–150 (2005).
62.Maddison,W. P., Midford, P. E. &Otto, S. P. Estimating a binary character’s effect
on speciation and extinction. Syst. Biol. 56, 701–710 (2007).
63. FitzJohn, R. G. Diversitree: comparative phylogenetic analyses of diversification in
R. Methods Ecol. Evol. 3, 1084–1092 (2012).
64. FitzJohn, R. G., Maddison, W. P. & Otto, S. P. Estimating trait-dependent
speciation and extinction rates from incompletely resolved phylogenies. Syst. Biol.
58, 595–611 (2009).
65. Davis, M. P., Midford, P. E. & Maddison, W. Exploring power and parameter
estimation of the BiSSE method for analysing species diversification. BMC Evol.
Biol. 13, 38 (2013).
66. Bloom, D. D., Weir, J. T., Piller, K. R. & Lovejoy, N. R. Do freshwater fishes
diversify faster than marine fishes? A test using state-dependent diversification
analyses and molecular phylogenetics of new world silversides (Atherinopsidae).
Evolution 67, 2040–2057 (2013).
Acknowledgments
Financial support of the German Research Foundation (grant HE 3584/4) is gratefully
acknowledged. This is publication number 129 from the Courant Research Centre
Geobiology, funded by the German Excellence Initiative.
Author contributions
K.F., H.S. and J.H. designed the research and wrote the paper; K.F., A.S.-V., A.R.S. and J.H.
assembled the data sets; K.F. analyzed the data; T.S. assisted in the TreePAR-analyses.
Additional information
Supplementary information accompanies this paper at http://www.nature.com/
scientificreports
Competing financial interests: The authors declare no competing financial interests.
How to cite this article: Feldberg, K. et al. Epiphytic leafy liverworts diversified in
angiosperm-dominated forests. Sci. Rep. 4, 5974; DOI:10.1038/srep05974 (2014).
This work is licensed under a Creative Commons Attribution 4.0 International
License. The images or other third party material in this article are included in the
article’s Creative Commons license, unless indicated otherwise in the credit line; if
the material is not included under the Creative Commons license, users will need
to obtain permission from the license holder in order to reproduce thematerial. To
view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
www.nature.com/scientificreports
SCIENTIFIC REPORTS | 4 : 5974 | DOI: 10.1038/srep05974 6