Genomics 113 (2021) 681–692
Available online 27 January 2021
0888-7543/© 2021 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
Genome-wide supermatrix analyses of maples (Acer, Sapindaceae) reveal
recurring inter-continental migration, mass extinction, and rapid
, Damien D. Hinsinger
, Joeri S. Strijk
Biodiversity Genomics Team, Plant Ecophysiology & Evolution Group, Guangxi Key Laboratory of Forest Ecology and Conservation, College of Forestry, Guangxi
University, DaXueDongLu 100, Nanning, Guangxi 530005, China
Alliance for Conservation Tree Genomics, Pha Tad Ke Botanical Garden, PO Box 959, 06000 Luang Prabang, Laos
etabolique, Genoscope, Institut de Biologie François Jacob, Commisariat `
Energie Atomique (CEA), CNRS, Universit´
Institute for Biodiversity and Environmental Research, Universiti Brunei Darussalam, Jalan Tungku Link, BE1410, Brunei Darussalam
Acer (Sapindaceae) is an exceptional study system for understanding the evolutionary history, divergence, and
assembly of broad-leaved deciduous forests at higher latitudes. Maples stand out due to their high diversity,
disjunct distribution pattern across the northern continents, and rich fossil record dating back to the Paleocene.
Using a genome-wide supermatrix combining plastomes and nuclear sequences (~585 kb) for 110 Acer taxa, we
built a robust time-calibrated hypothesis investigating the evolution of maples, inferring ancestral ranges,
reconstructing diversication rates over time, and exploring the impact of mass-extinction on lineage accumu-
lation. Contrary to fossil evidence, our results indicate Acer rst originated in the (north)eastern Palearctic re-
gion, which acted as a source for recurring outward migration. Warm conditions favored rapid Eocene-onward
divergence, but ranges and diversity declined extensively as a result of the Plio-Pleistocene glacial cycles. These
signals in genome-wide sequence data corroborate paleobotanical evidence for other major woody north-
temperate groups, highlighting the signicant (disparate) impact of climatic changes on the evolution, compo-
sition, and distribution of the vegetation in the northern hemisphere.
Eastern Asia is a core area of north temperate tree diversity. The
region harbors at least 170 temperate woody genera (roughly four times
more than Europe and two times more than North America), many of
which reach their maximum in terms of species diversity there [1–3].
Many of these genera are relicts of the mesophytic vegetation that once
occupied large areas throughout the northern hemisphere during the
Cenozoic. These forests fragmented as a result of global cooling and the
breakup of land connections during the second half of this era (Oligo-
cene, Neogene, and Quaternary) . The exceptional tree diversity at
generic and species level in East Asia has been attributed to a larger
refuge area with suitable climatic conditions during the Quaternary, a
higher spatial heterogeneity promoting speciation, and to lower
extinction rates than those prevalent in North America and Europe
Acer L., the maple genus, is among the largest Cenozoic tree groups
that best exemplies this pattern. The genus includes approximately 155
species , all conned to the northern hemisphere except one, Acer
laurinum Hassk., whose range includes the islands of Sulawesi and Java
in Indonesia . More than two-thirds of species occur in eastern Asia
with a major center of diversity in China (over 100 species, 61 endemic)
[9,11]. About 15 native species occur in Europe, northern Africa, and
western Asia, and only 13 are found in North and Central America
Maples are among the most prevalent trees in the broad-leaved de-
ciduous forests of the northern hemisphere [15,16]. They are often the
dominant element of the vegetation and can also act as keystone species
maintaining fundamental ecosystem processes and the biodiversity of
communities [17,18]. In addition to their ecological importance, maples
are well-known and highly appreciated as ornamental trees for their
attractive leaf shapes and autumn foliage colors . A number of
* Corresponding author at: Institute for Biodiversity and Environmental Research, Universiti Brunei Darussalam, Jalan Tungku Link, BE1410, Brunei Darussalam
E-mail address: firstname.lastname@example.org (J.S. Strijk).
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/ygeno
Received 12 October 2020; Received in revised form 5 January 2021; Accepted 22 January 2021
Genomics 113 (2021) 681–692
species are also valued as a source of timber and wood , sugar
products , and medicinal or bioactive compounds .
Historically, the genus has been included either in the Sapindaceae
[23,24] or in its own family Aceraceae together with the genus Dipter-
onia Oliv. consisting of two species endemic to China [11,25–28]. Mo-
lecular phylogenetic studies support the placement of Acer within the
Sapindaceae, in the Hippocastanoideae clade together with Dipteronia
(tribe Acereae), Aesculus L. (~13 spp.), Billia Peyritsch (two spp.), and
Handeliodendron Rehder (one spp.) of tribe Hippocastaneae [29–33].
Given the global importance and distribution of the group, Acer has
received substantial attention with regards to its biogeographic history
and that of its role in the evolution of north temperate forests [34–37].
The genus has a remarkably rich fossil record that dates back to the late
Paleocene (~60–56 Ma) and extends through most of the Cenozoic of
the northern continents, most notably of western North America
[15,34,38–40]. Because some of the earliest fossil occurrences and the
greatest diversity (at both sectional and species level) have been
recorded from this continent, it has been suggested that Acer originated
in North America, where it experienced a major diversication by mid-
Eocene (~49–47 Ma) followed by migrations to Asia and Europe
[34,39]. The American origin is further supported by the fact that all
reliable fossils of its sister genus, Dipteronia, are also from North America
Phylogenetic analyses in Acer suggest a complex biogeographic his-
tory involving multiple migrations between the northern continents
followed by waves of radiations and extinctions. For example, Renner
et al.  found that the North American species diverged from their
respective European and Asian relatives at widely different times be-
tween the late Eocene and late Miocene. These authors inferred a burst
of diversication about 40 Ma followed by a decrease in the diversi-
cation rate between 30 and 20 Ma. This study, one of the largest in the
genus in terms of species number and one of the few that have provided
a time frame for some biogeographic events, however, was based on a
few plastid loci and did not provide adequate resolution and support for
the majority of major clades.
In a recent biogeographic analysis using a nuclear dataset of over
400,000 sites,  inferred eastern Asia as the most likely ancestral area
for the genus, contradicting the American origin suggested by the fossil
record. However, this study was limited by a reduced taxon sampling
(30 taxa) and the fact that fossil data were not taken into account for the
analysis. Here we use a supermatrix combining whole plastid genomes
and nuclear data from 110 maple taxa to reconstruct the genus’ evolu-
tionary history. Our aim is to investigate the biogeographic history with
an emphasis on the impact of climatic changes on its diversication and
distribution. This is the rst study that combines plastomes with a large
nuclear dataset to generate a comprehensive phylogenomic evolu-
tionary hypothesis for this diverse and economically important tree
2. Material and methods
2.1. Taxon sampling
We obtained leaf samples belonging to 87 species of Acer and four
outgroup genera (Aesculus, Dimocarpus Lour., Koelreuteria Laxm. and
Litchi Sonn) selected based on availability and results from previous
phylogenetic studies [31,32,37]. Samples were collected from living
plants in private and state institutions in Spain, France, and China
(Supplementary Table S1) andkept frozen to generate whole plastomes
and nuclear ribosomal cistron (NRC) sequences. Voucher specimens
were deposited in the BGT herbarium (Guangxi University, China).
Additional plastome sequences from ve maple species, two species
of Dipteronia, and Spondias mombin L. (Anacardiaceae) were obtained
from GenBank (Table S1). The plastid and NRC datasets were combined
with the dataset of  consisting of over 500 nuclear loci obtained with
hybrid enrichment for 65 species of Acer (https://doi.org/10.5061/drya
d.g9j13fm). The total number of Acer taxa included in our supermatrix
was 110 (about 71% of the genus) (Supplementary Table S1). Almost all
American and Eurasian species were included, and all taxonomic sec-
tions were represented except for section Wardiana, whose only species,
Acer wardii W.W. Sm., could not be sampled.
2.2. Library construction, sequencing, and assembly
DNA extraction, library construction, and sequencing were per-
formed by Annoroad Gene Technology (Beijing, PR China) Co., Ltd. as
described by . The assembly and annotation of the chloroplast ge-
nomes also followed . Newly generated sequences were deposited in
Genbank under the accession numbers MW067026-MW067100
The nuclear ribosomal cistron region was assembled and annotated
using Geneious v. 11.0.4 (Biomatters, Auckland, New Zealand) as
described by . For most species, we obtained partial NRC sequences
from GenBank, which were used as references for the extended assembly
of our sequences as detailed elsewhere . The paired reads were
mapped to the reference sequence with medium-low sensitivity for 100
To annotate the assemblies, we rst transferred the annotations of
the ITS1, ITS2, and 5.8S regions from a published partial NRC sequence
of Acer campestre L. (DQ238434) to our sequence of this species. To
establish the boundaries of the 18S and 26S regions we used the NRC
sequence of Spondias tuberosa Arruda (KX522674). The boundary of the
ETS region was identied searching for the transcription initiation site
(TIS) sequence, which in Acer is TCTTTAGGGGGG. The resulting an-
notated NRC sequence of A. campestre was subsequently used as a
reference to annotate the remaining species. NRC sequences were sub-
mitted to GenBank and deposited under the accession numbers
MW070114-MW070204 (Table S1).
2.3. Phylogenomic reconstruction
Phylogenomic reconstruction was performed for the combined
supermatrix consisting of the whole plastome sequences, the nuclear
ribosomal cistrons, and the nuclear dataset of . Alignments of the
plastome and the NRC datasets were performed using MAFFT  with
default settings in Geneious and then concatenated with the nuclear
dataset of  for a total-evidence analysis. The plastome dataset was
divided into three partitions, each corresponding to one of its regions
(LSC, SSC, and one IR). We did not perform a ner-scale partitioning (e.
g., into genes and introns) because in a previous study we found no effect
of partitioning at this scale on tree inference . The NRC sequences
were divided into six partitions (ETS, 18S, ITS1, 5.8S, ITS2, and 26S),
whereas the nuclear dataset was treated as a single partition. This
supermatrix was deposited in the Dryad repository with doi: 10.5061/
A partitioned maximum likelihood (ML) analysis was conducted in
RaxML-NG  using the best-t substitution model selected with
ModelTest-NG  for each partition (Supplementary Table S2). We
performed an ‘all-in-one’ analysis (ML tree search +standard non-
parametric bootstrap)  with 25 randomized parsimony starting
trees and 1000 bootstrap replicates to assess branch support.
2.4. Dating analyses and fossil selection
Divergence times among Acer lineages were estimated in BEAST
2.6.1 [49–52]. Data partitions were unlinked and priors were set for the
parameters of the corresponding evolutionary models. The Relaxed
Clock Log Normal was selected as clock model , and the Birth-Death
model  was selected as tree prior. The two clock model parameters
were assigned an exponential distribution with a mean of 10 for the
ucldMean, and a mean of 0.333 for the ucldStdev. The parameters of the
Birth-Death model (birth rate and relative death rate) were assigned an
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
exponential distribution with a mean of 1. Two independent runs of 1.2
billion generations each were performed in BEAST 2 on the CIPRES
portal . The convergence of both runs and ESS values were inspected
with Tracer v1.7.1 . Trees were sampled every 4000 generations
and combined with LogCombiner v2.6.0 (part of the BEAST 2 package,
available at https://www.beast2.org/). The maximum clade credibility
tree was constructed with TreeAnnotator v2.6.0 (also part of the BEAST
2 package) and edited in FigTree v1.4.3.
Fig. 1. Maximum likelihood tree of Acer based on the combined supermatrix (plastomes +nuclear loci) inferred with RaxML-NG. Names on the right indicate Acer
sections. Letters A and B indicate the two major clades of Acer (see text for details).
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
To calibrate our phylogenetic tree, we selected four fossil species of
Acer and the oldest fossil record of Dipteronia as an external calibration
point. Dipteronia, now consisting of two species endemic to China, is
relatively well represented in the Paleogene of western North America
but is absent from the fossil record of Asia. The age of the oldest known
fruits is 63–60 Ma old . This was used to set the minimum age of this
The oldest Acer-like leaves and Acer-like fruits are from the late
Paleocene of several high-latitude locations of the Palearctic region
[40,57]. These fossils are often reported in the literature as ‘Acer arcti-
cum Heer’, a variable taxon comprising a number of early morphospe-
cies from the Paleocene and Eocene [34,40]. The oldest unequivocal
fossil leaves of Acer from North America correspond to A. alaskense
Wolfe & Tanai from the late Paleocene of Southern Alaska . Sa-
maras, leaves, and Acer-like pollen have also been recorded from late
Paleocene locations of Canada and the US [58–60], indicating that the
genus was already well established by the end of this epoch about 60 Ma.
We used this age of 60 Ma to constrain the crown age of the genus Acer.
Despite a rich fossil record, it is very difcult to place extinct species
of Acer in modern sections. Several authors [34,40] have assigned fossil
leaves and detached samaras to currently recognized sections, but most
of these assignments are not reliable because the fossils lack the diag-
nostic characteristics of the sections, and can match the morphology of
more than one section . For this reason, we did not include a high
number of Acer fossils in our analysis but rather selected two fossil
species that can be condently assigned to extant clades and thus can
provide internal calibration points.
The clade formed by members of section Macrantha was assigned a
minimum age of 43 Ma based on fossil leaves of Acer dettermani Wolfe &
Tanai from the middle Eocene of Alaska . This species appears to be
the oldest taxon assignable to this section. Its placement was conrmed
by McClain (2000), who studied the venation pattern and the margin
characters of the leaves. Acer trifoliatum Geng, described from the mid-
dle Miocene (16–13 Ma) of China , was assigned to the clade formed
by three species of section Trifoliata: Acer triorum Kom., A griseum
(Franch.) Pax, and A. maximowiczianum Miq. Acer trifoliatum was found
to be closely related to A. triorum based on a study of the leaf
morphology that included many micromorphological features of the
epidermal cells .
2.5. Ancestral range estimation
Ancestral ranges of Acer were inferred with the package Bio-
GeoBEARS [63–65] in R . We assigned Acer species to ve biogeo-
graphic regions based on their native distribution: Western Nearctic (A),
Eastern Nearctic (B), Western Palearctic (C), Eastern Palearctic (D), and
Indomalaya (E). Geographic distribution was compiled from the litera-
ture [11,13] as well as online databases [67,68].
We performed non-stratied analyses, with and without fossil data,
to test for several possible biogeographic scenarios. In the analysis
incorporating fossil data, we constrained the ranges at three nodes based
on the occurrence of the fossil species we used for calibration. Acer
alaskense, A. dettermani, and Dipteronia brownii McClain & Manchester
are from western North America and so the respective nodes were ‘xed’
to include this information. In both types of analyses, we implemented
the DEC, DIVALIKE, and BAYAREA models along with their +J versions
and determined the one that best ts our data by using the AIC scores
and Akaike weights. The probabilities of the ancestral states estimated
with the best-tting model were plotted as pie charts at the nodes of the
chronogram inferred with BEAST.
2.6. Diversication analyses
We used the R package Phytools  to generate a lineage-through-
time (LTT) plot for maples and to perform the Monte Carlo constant-
rates (MCCR) test . Rates of diversication through time were
estimated with the R package TESS . We considered the following
three models: a constant-rate birth-death (BD) process, an episodically
variable-rate birth-death (EBD) model, and a birth-death process with
one mass-extinction event (based on the high number of fossil species of
maples) [71,72]. For the constant-rate BD process, we used an expo-
nential prior distribution for both the speciation and extinction-rate
parameters. For the EBD model, we assumed a single speciation-rate
shift and a single extinction rate-shift. We used exponential prior dis-
tributions for the four model parameters (speciation and extinction rates
before and after the shifts). For the BD process with one mass-extinction
event, we assumed a survival probability of 30%. The speciation and
extinction-rates parameters were assigned an exponential distribution,
whereas the mass extinction time was assigned a uniform distribution.
To account for incomplete taxon sampling, the parameter rho was set to
0.71 (the sampled fraction in Acer) in all models assuming a uniform
(random) sampling. All analyses were run for 50,000 generations with a
burnin of 20%. Stationarity was visually inspected using the trace plots.
The relative t of each model to the data was assessed via Bayes Factors
Additionally, we performed a compound Poisson process on Mass-
Extinction Times (CoMET) analysis  in the package TESS to
further explore the impact of mass-extinction events on the diversica-
tion of Acer. The CoMET method uses rjMCMC to estimate the parameter
probabilities of an episodic BD model, including the number and timing
of shifts in speciation and extinction rates, the speciation and extinction
rates between the shifts, and the number, timing, and magnitude of
mass-extinction events . We assumed a survival probability of 30%
and set the sampling fraction to 0.71, similar to previous models.
3.1. Genomic data and evolutionary relationships
A total of 75 plastomes and 91 NRC sequences were generated in this
study. Twenty-four previously published plastomes were incorporated
from GenBank for a total of 99 plastid genomes, all representing
different taxa (Table S1). The plastomes of Acer ranged in size from
155,212 (A. carpinifolium Siebold & Zucc.) to 157,046 bp (A. tenuifolium
(Koidz.) Koidz.). The NRC sequences varied from 7264 (A. micranthum
Siebold & Zucc.) to 7641 bp (A. pycnanthum K. Koch). The supermatrix
combining these sequences with the nuclear dataset of Li et al. (2019)
consisted of 117 taxa and 585,780 aligned sites, of which 155,166 were
plastid, and 430,614 were nuclear (Table S2).
The Maximum Likelihood analysis recovered two major clades of
Acer with maximum support, one predominantly Asian and comprising
species of sections Negundo, Arguta, Spicata, and Palmata (clade A in
Fig. 1), and the other comprising the remaining sections which include
Asian, European, and American species (clade B). Section Negundo was
placed as sister to Arguta (BS =100%), whereas Spicata was sister to
Palmata (BS =80%) (Fig. 1). Within the second major clade (B), species
were grouped into ve mutually exclusive clades with monotypic sec-
tion Glabra plus section Parviora sister to the clade including all other
sections (BS =100%). Most sections were recovered as monophyletic
except for Trifoliata, Pentaphylla, Acer, and Lithocarpa. Sections Trifoliata
and Pentaphylla were paraphyletic with A. oblongum Wall. ex DC. and
A. pentaphyllum Diels more closely related to members of Trifoliata than
Pentaphylla. Sections Acer and Lithocarpa were both biphyletic. Acer
caesium Wall. ex Brandis (section Acer) and A. yangbiense Y.S. Chen & Q.
E. Yang (section Lithocarpa) fell outside their respective clades, the two
forming a sister pair with A. pilosum Maxim. of section Pubescentia (BS =
100%) (Fig. 1).
3.2. Divergence times
The maximum clade credibility tree resulting from the Bayesian
analysis of the supermatrix in BEAST 2 was strongly supported and
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
Fig. 2. Chronogram of Acer resulting from the analysis of the supermatrix (plastomes +nuclear loci) in BEAST 2. Numbers at the nodes are ages in million years
(shown only for major clades and sections). All branches have posterior probabilities of 1.00 except for those with an asterisk, which have PP between 0.5 and 0.97.
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
highly congruent with the ML tree (Fig. 2), with only minor differences
in the placement of a few species within sections Palmata and Plata-
noidea, and among subspecies of Acer tataricum L. (section Ginnala). The
analysis placed the split between Acer and Dipteronia in the late Creta-
ceous (~71 Ma), and the crown age of Acer in the middle Paleocene
(~61 Ma). The two major lineages within the genus began to diversify in
the late Paleocene-early Eocene, ca. 56 and 51 Ma, respectively (Fig. 2,
The four sections of clade A (Negundo, Arguta, Spicata, and Palmata)
differentiated in the middle Eocene (Fig. 2). Section Palmata separated
from section Spicata 45.5 Ma, whereas Arguta and Negundo diverged
from each other about 6 Ma later. Section Palmata, the largest and most
diverse of the genus, began to diversify in the Late Eocene (~35 Ma).
The other three much smaller sections diversied during the Oligocene
Within clade B, sections Macrantha, Glabra, and Parviora diverged
in the early Eocene (53–50 Ma), followed by section Ginnala (~48 Ma).
Sections Lithocarpa, Indivisa, Macrophylla, Platanoidea, and Rubra
differentiated around the same time in the middle Eocene (45–43 Ma).
Sections Acer and Trifoliata +Pentaphylla, the most recently divergent
groups, split in the late Eocene (~37 Ma). Diversication of these sec-
tions was estimated to have begun in the middle Eocene (43.8 Mya) and
continued throughout the Oligocene and Miocene. Several closely
related species (and their subspecies) within sections Acer, Trifoliata,
Platanoidea, Lithocarpa, Ginnala, and Macrantha appear to have origi-
nated more recently during the Plio-Pleistocene (Fig. 2).
3.3. Ancestral range estimation
Both, the unconstrained and the fossil-constrained analyses in Bio-
GeoBEARS strongly favored the DEC +J model over the six models
tested (Tables S3 and S4). However, the LnL value for this model was
considerably lower (ΔAIC =22.4) in the fossil-constrained analysis, thus
favoring the unconstrained analysis. This placed the origin of the genus
in the Eastern Palearctic region, where it underwent its initial diversi-
cation to later spread to North America, Europe, and south into the
Indomalayan region (Fig. 3).
Migrations from the Eastern Palearctic region to the Nearctic were
inferred to have occurred at least seven times between the early Eocene
(A. glabrum Torr. lineage) and early Miocene (A. rubrum L. +
A. saccharinum L.). In all probability, this movement took place overland
via the Bering Land Bridge, which allowed the oristic exchange be-
tween Eastern Asia and North America from the late Cretaceous to late
Movement from the Eastern Palearctic region westwards to West
Asia and Europe also occurred multiple times (Fig. 3). The earliest
colonization to the west took place in the late Eocene (~37 Ma) with the
separation of the lineage leading to section Acer. A second migration
around 20 Ma gave rise to a group of European and West Asian species of
section Platanoidea (A. cappadocicum Gled. subsp. divergens (Pax) A.E.
Murray, A. lobelii Ten., and A. platanoides L.). Acer campestre, another
widely distributed Eurasian species from section Platanoidea, diverged
from its East Asian sister A. miyabei Maxim. around 15 Ma, whereas Acer
tataricum subsp. tataricum, the westernmost subspecies of A. tataricum
(Section Ginnala) appears to have originated as result of a recent
migration ca. 7 Ma (Fig. 3).
Dispersal from Europe to eastern North America was inferred to have
occurred only once, at the Oligocene/Miocene boundary (~23 Ma). This
event led to the origin of the American series Saccharodendron of section
Acer (represented in our study by A. saccharum Marshall,
A. grandidentatum Nutt., and A. skutchii Rehder) (Fig. 3).
Southward spread from the Eastern Palearctic into the Indomalayan
region appears to have occurred relatively recently, beginning in the
middle Miocene (~15 Ma) and onward, except for the A. laurinum
lineage, the southernmost species, which separated from the other
species of section Rubra in the late Oligocene, ca. 25 Ma.
3.4. Diversication analyses
The LTT plot of Acer shows an initial period of constant accumulation
of lineages followed by an increase between ~50 and ~ 40 Ma after
which the lineage accumulation appears to slow slightly and remain
nearly constant to the present (Fig. 4). The results of the MCCR test
indicate that this pattern observed in Acer is not signicantly different
from the one expected under the null model of constant speciation and
extinction rates (γ = − 0.8609, p =0.92).
Of the three branching-process models compared in TESS, we found
positive but low support for the birth-death process with one mass-
extinction event over the episodic BD model (BF =3.12), but no sup-
port over the constant-rate BD process (Table S5). The CoMET analysis
identied one signicant extinction event (2 ln BF =6) at ~3 Ma (Fig. 5
H, I). No shifts in the speciation or extinction rates were detected (Fig. 5
B, C, and E, F). The net-diversication rate was highest at 50 Ma in the
early Eocene and then gradually decreased to remain relatively constant
from the middle Eocene (~40 Ma) to the late Miocene (~10 Ma). It was
lowest in the Plio-Pleistocene (last 4 Ma) due to increased extinction in
this period (Fig. 5 A, D, G).
Age estimates for the Acer genus and its sections inferred from the analysis in
Age in My
Age in My
Acer genus 70.9
Clade B (all
Section Glabra 50.4
Section Ginnala 48.2
Section Palmata 45.5
Section Spicata 45.5
Section Rubra 45.3
Section Indivisa 44
Section Negundo 39.6
Section Arguta 39.6
Section Acer 37.1
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
Fig. 3. Ancestral geographic ranges of Acer inferred with BioGeoBEARS using the DEC +J model in the unconstrained analysis (ancstates: global optim, 2 areas max.
d =0.0022; e =0; j =0.0174; LnL = − 152.92). The inset map shows the present distribution of the genus.
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
4.1. Phylogenomic relationships
Our ML and Bayesian trees obtained with the combined plastid and
nuclear datasets largely agree with the infrageneric classication of
maples. Of the 18 currently accepted sections [9,75], 13 are well-
supported in our study (BS =100%, PP =1.00) and only four, Trifo-
liata, Pentaphylla, Acer, and Lithocarpa, are non-monophyletic due to the
placement of four species: A. oblongum, A. pentaphyllum, A. caesium, and
A. yangbiense (Fig. 1). Only the monotypic section Wardiana was not
included in this study.
Our results conrm previous analyses showing the close relationship
between sections Trifoliata and Pentaphylla [37,42,76,77] and provide
further evidence for merging of the two groups (Fig. 1). Shared
morphological features, including the presence of compound leaves,
corymbiform inorescences, extrastaminal oral disk, and convex or
inated seed locules [11,14] also point to the strong association between
the two sections.
Sections Acer and Lithocarpa are both biphyletic due to the exclusion
of A. caesium and A. yangbiense, respectively (Fig. 1). Acer caesium, the
only extant East Asian member of section Acer, appears to be a distinct,
genetically isolated taxon that has been previously recovered outside the
Acer core clade both in the ITS [76,78] and plastid trees [37,79]. This
species was placed in our trees sister to the Chinese endemic A. yang-
biense and forming a strongly supported (BS =100%, PP =1.00) clade
together with A. pilosum of section Pubescentia (Fig. 1). However, in the
nuclear tree of Li et al. (2019) A. caesium and A. yangbiense were grouped
together within section Acer, suggesting that the dissimilar placement in
our trees is given by the signal contained in the plastid genome. A
plausible explanation might be chloroplast capture via ancient
hybridization between A.caesium-A. yangbiense and A. pilosum lineages.
The latter species was not included in the study of Li et al. (2019) and is
represented in our analyses only by the plastid genome and the NRC
cistron. It was recovered in the plastid tree of Renner (2008) forming a
sister pair with A. caesium as in our study, but it was placed sister to the
Pentaphylla-Trifoliata clade in the ITS tree of Grimm et al. (2006).
Aside from the different placement of the A. caesium -A. yangbiense
lineage, relationships among sections are consistent with the nuclear
tree of . The only exception concerns section Spicata, which was
recovered in our analyses as sister to section Palmata (Fig. 1). By
contrast, it was placed as sister to a clade formed by sections Negundo,
Arguta, and Palmata .
4.2. Historical biogeography
Our analyses placed the origin of Acer in the Eastern Palearctic re-
gion, and a crown age of 61 Ma, similar to previous estimates , but
15 Ma older than the age inferred by . Estimates for the divergence
times between the New World and Old World lineages are, in general,
older than the ones inferred by these authors (Table 1, Figs. 2 and 3).
Wolfe and Tanai  proposed an American origin for Acer based on
a detailed revision of North American megafossils. Around 100 fossil
species have been described from this continent [34,80] compared to
approximately 45 and 50 from Eurasia and East Asia, respectively
[40,81–83] This greatest diversity led to suggest an initial radiation in
North America with subsequent dispersals to East Asia and Europe
[34,39,40]. However, no migrations from North America to Asia or
Europe were inferred in the most likely biogeographic scenario. When
the deepest nodes of our tree were constrained to include Western North
America within the ancestral distribution range, the resulting LnL values
were signicantly lower (ΔAIC >10) than the ones obtained in the
Fig. 4. Lineage-through-time (LTT) plot for Acer (black line). The dashed red line is the expected LTT plot under a pure-birth (no extinction) process. The pink area
represents the 95% condence interval based on 1000 simulated trees assuming a pure-birth process. The tree in the background is the maximum clade credibility
tree. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
Fig. 5. Results of the CoMET analysis. A, D, G: Plots of speciation, extinction, and diversication rates over time. Shaded areas indicate the 95% credible interval. B,
E: Posterior probabilities for speciation and extinction-rate shifts. No signicant shifts were detected (all bars are below the signicance thresholds). H: Posterior
probabilities for the mass-extinction events. A signicant mass extinction event (2 ln BF ~ 6) that occurred ca. 3 Ma was identied. C, F: Bayes Factor values for the
speciation and extinction-rate shifts. I: Bayes Factor values for the mass extinction events.
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
unconstrained analysis (Tables S3 and S4).
Acer rst emerged in the fossil record of the Russian Far East and
Alaska almost simultaneously [34,40,57]. The genus most likely origi-
nated in the north-eastern Palearctic from where it dispersed to North
America soon after its appearance. Based on fossil evidence, it appears to
have diversied rapidly in western North America to later undergo
signicant extinction in this continent. More than 50 fossil species from
at least 15 sections are recorded just from the Eocene of North America
compared to six in Asia [34,40], but only 12–13 American species exist
today. Most of these, e.g., Acer macrophyllum Pursh, A. glabrum, and
A. negundo L. are placed in our time-calibrated tree on long branches that
go back to the Eocene and Oligocene (Figs. 2 and 3), reecting the past
extinctions along these old lineages.
The several migrations to the Nearctic occurred across widely
different time periods (Fig. 3), meaning that Acer spread (and diversi-
ed) under a broad range of climatic conditions. The Beringia corridor
would have allowed the eastward movement, rst during the warm and
wet climate of the late Paleocene-early Eocene (e.g., A. glabrum lineage),
and later during the cooler and drier regimes of the Oligocene and early
Miocene (Fig. 3) [74,84]. According to the fossil record, Acer became a
common component of the broad-leaved deciduous forests of North
America by the late Eocene .
The migration from East Asia into Europe was inferred to have begun
in the late Eocene with the separation of section Acer from the Asian
Pentaphylla-Trifoliata lineage (Fig. 3). This estimate is consistent with
the fossil evidence. The oldest records of Acer in Europe are from the late
Eocene (Priabonian) of Spitsbergen , and the oldest records
assignable to section Acer (A. haselbachense Walther and A. engelhardtii
Walther), are from the early Oligocene of Central Europe [86,87]. The
presence of Acer fossils in Svalbard points to early movement from the
north through the arctic islands and across Greenland to Western
Europe. Subsequent migrations from the east would have been possible
after the closure of the Turgai Seaway in the early Oligocene (~30 Ma),
a time during which temperate broad-leaved deciduous forests were
expanding in response to a cooler, dryer, and more seasonal climate
The ancestor of the sugar maple group (series Saccharodendron of
Section Acer) would have reached Eastern North America from Europe
by the beginning of the Miocene (Fig. 3) through long-distance dispersal,
likely via Iceland-Greenland. The earliest fossils assignable to this series
are from several early Miocene localities of western North America
(British Columbia, Oregon, Nevada) , indicating rapid colonization
and broader distribution in North America than the present. Fossils of
Saccharodendron are absent from Alaska and high latitude locations of
Eastern Asia , but they have been reported from the middle-late
Miocene (12 Ma) of Iceland , lending support to the transatlantic
migration of the Saccharodendron lineage inferred in our study.
The southward movement of Acer into the Indomalayan tropics
(Fig. 3) was likely driven by the climatic cooling during the Oligocene
and Miocene. Given the strong oristic afnities with several Eocene
plant assemblages from the northern hemisphere, the Indomalayan re-
gion has been viewed as a refugium for the once widespread Eocene
‘boreotropical’ ora of which Acer was a part [89,90].
4.3. Diversication of Acer
Acer diversied most rapidly in the early to middle Eocene (55–45
Ma). Diversication remained relatively constant afterward with no
appreciable changes in the speciation and extinction rates until the late
Miocene. It declined in the last ve million years as extinction increased,
presumably with the Plio-Pleistocene glaciations (Fig. 5).
Dating analysis shows that Acer sections differentiated over a period
of ~15 Ma, between the early and mid-Eocene, except for Acer and
Pentaphylla +Trifoliata, which split at the end of the Eocene (Table 1,
Figs. 2 and 3). A similar inference was drawn by Wolfe and Tanai (1987)
based on the fossil record. These authors noted that the “simultaneous
occurrences in the early middle Eocene of species that represent di-
vergences leading to major groups indicate that evolution was pro-
ceeding at a very rapid pace” . The rapid differentiation of major
lineages may have been favored by the warm climate of the early
Eocene. Acer fossils from this period belong to megathermal and mes-
othermal plant assemblages despite the genus being now primarily
found in mesothermal to microthermal vegetation . The cool
tolerance likely evolved progressively from the late Eocene as micro-
thermal climates extended through the northern hemisphere.
Our ndings show that Acer diversied at a constant rate during the
Oligocene and Miocene and that diversication does not appear to have
been affected by the climatic cooling of the Eocene-Oligocene transition.
However, the CoMET analysis detected one extinction event at ~3.5–3
Ma that coincides with the start of the Plio-Pleistocene glacial cycles in
the northern hemisphere (Fig. 5). As in many north temperate groups,
the evolutionary history and distribution of Acer were shaped by the
climatic uctuations of the last 3.5 Ma . In particular, the climatic
changes associated with the advance (and retreat) of ice sheets over
extensive areas of North America and Europe would lead to the
extinction of a large fraction of the Acer diversity in these continents
. In East Asia, extinction was less severe due to much larger ice-free
refugia that harbored the ora during the glacial peaks [93,94].
Several of the North American and Eurasian lineages of Acer appear
to have experienced some but limited diversication during the last 5–4
Ma; for example, series Saccharodendron, and various groups of Eurasian
species within sections Acer and Platanoidea (Fig. 3.). This is likely to
have been generated through allopatric speciation driven by isolation in
glacial refugia .
Using a genome-wide supermatrix combining both plastomes and
nuclear loci we reconstructed the most complete phylogenomic evolu-
tionary hypothesis of maples to date. The results of our analyses do not
support the American origin suggested by earlier paleobotanists ,
but instead indicate an East Asian origin for the group. We argue that the
greater abundance of maple fossils in North America is due to a higher
sampling intensity in this continent, given by a long history of geological
exploration and paleobotanical research .
The biogeographic patterns inferred from our genomic data are very
similar to those of other woody north-temperate genera (e.g., Aesculus
, Prunus , Fagus [98,99]), and agree with the paleobotanical
evidence for the origin and migration of the north-temperate ora
. Our results highlight the signicance of the (north-)eastern
Palearctic region as the center of origin and early development for many
north-temperate groups, as well as the impact of climatic changes, in
particular of the climatic deterioration of the last ve million years, on
the evolution and distribution of the vegetation in the northern
This work was supported by the Bagui Scholarship (Team Funding
C33600992001) to JSS. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the
Declaration of Competing Interest
We declare that there is no conict of interest.
We would like to kindly acknowledge Jean-Louis Helardot (Arbo-
retum du Passadoux, Brive-la-Gaillarde, France), Beatrice Chasse (Ar-
boretum de Pouyouleix, France) and the Botanical Garden of Iturraran
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
(Pagoeta Natural Park, Basque Country, Spain) for sharing their
knowledge on the genus and providing essential ground support and
data during our study. We would also like to acknowledge the horti-
cultural staff of JingDong Subtropical Botanical Garden (Yunnan,
China), Xishuangbanna Tropical Botanical Garden (Yunnan, China),
Hangzhou Botanical Garden (Zhejiang, China) and Qing Xiu Shan Park
(Guangxi, China) providing additional samples. Special thanks go to
Mark Miller (CIPRES portal) for helping with our long BEAST2 runs and
providing additional CPU hours.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
 R.E. Latham, R.E. Ricklefs, Continental comparisons of temperate-zone tree
species diversity, in: R.E. Ricklefs, D. Schluter (Eds.), Species Diversity in
Ecological Communities: Historical and Geographical Perspectives, University of
Chicago Press, Chicago, 1993, pp. 294–314.
 Q. Guo, R.E. Ricklefs, Species richness in plant genera disjunct between
temperate eastern Asia and North America, Bot. J. Linn. Soc. 134 (2000)
 E.L. Little, North American trees with relationships in Eastern Asia, Ann. Missouri
Bot. Gard. 70 (1983) 605–615, https://doi.org/10.2307/2398980.
 B.H. Tiffney, The Eocene North Atlantic land bridge: its importance in tertiary
and modern phytogeography of the northern hemisphere, J. Arnold Arbor. 66
 Q.Y. Xiang, W.H. Zhang, R.E. Ricklefs, H. Qian, Z.D. Chen, J. Wen, J.H. Li,
Regional differences in rates of plant speciation and molecular evolution : a
comparison between Eastern Asia and Eastern North America, Evolution. 58
 R. Milne, Northern hemisphere plant disjunctions: a window on tertiary land
bridges and climate change? Ann. Bot. 98 (2006) 465–472, https://doi.org/
 R. Milne, R. Abbott, The origin and evolution of tertiary relict oras, Adv. Bot.
Res. 38 (2002) 281–314, https://doi.org/10.1016/s0065-2296(02)38033-9.
 M.J. Donoghue, S.A. Smith, Patterns in the assembly of temperature forests
around the Northern Hemisphere, Philos. Trans. R. Soc. B Biol. Sci. 359 (2004)
 P. de Jong, Worldwide maple diversity, in: S.J. Wiegrefe, H. Angus, D. Otis,
P. Gregorey (Eds.), Proceedings of the International Maple Symposium, 2002,
Westonbirt Arboretum and the Royal Agricultural College, Gloucestershire, 2002,
 R.R. Irwanto, A.S.D. Irsyam, Acer laurinum Hassk. Sapindaceae, in: F.M. Franco
(Ed.), Ethnobotany of the Mountain Regions of Southeast Asia, Ethnobotany of
Mountain Regions, Springer, Cham, 2020, pp. 1–5, https://doi.org/10.1007/978-
 T.-Z. Xu, Y. Chen, P. de Jong, H. Oterdoom, C.-S. Chang, Aceraceae, in: Z. Wu,
P. Raven, D. Hong (Eds.), Flora of China, Science Press, Beijing, 2008,
 Y. Vargas-Rodriguez, L. Urbatsch, V. Karaman-Castro, Taxonomy and
phylogenetic insights for Mexican and Central American species of Acer
(Sapindaceae), J. Torrey Bot. Soc. 147 (2020) 49–86, https://doi.org/10.3159/
 T. Tutin, V. Heywood, N. Burges, D. Moore, D. Valentine, S. Walters, D. Webb,
Flora Europaea, Vol. 2: Rosaceae to Umbelliferae, 2nd edition, Cambridge
University Press, Cambridge, 2010.
 P. de Jong, in: H. Veenman, B.V. Zonen (Eds.), Flowering and Sex Expression in
Acer L.: A Biosystematic Study, Wageningen, 1976.
 A. Graham, Late Cretaceous and Cenozoic History of North American Vegetation
North of Mexico, Oxford University Press, New York, 1999.
 L.E. Frelich, R.A. Montgomery, J. Oleksyn, Northern temperate forests, in: K.S.-
H. Peh, R.T. Corlett, Y. Bergeron (Eds.), Routledge Handbook of Forest Ecology,
Routledge, London, 2015, pp. 30–45, https://doi.org/10.4324/9781315818290.
 P.V. Minorsky, The decline of sugar maples (Acer saccharum), Plant Physiol. 133
 D.A. Bishop, C.M. Beier, N. Pederson, G.B. Lawrence, J.C. Stella, T.J. Sullivan,
Regional growth decline of sugar maple (Acer saccharum) and its potential causes,
Ecosphere. 6 (2015) a179, https://doi.org/10.1890/es15-00260.1.
 J. Harris, Tree genera — 3. Acer — of the maple, Arboric, J. Int. J. Urban For. 2
(1975) 361–369, https://doi.org/10.1080/03071375.1975.10590443.
 H. Betts, Maple (Acer species). American Wood Series, US Department of
Agriculture. US Government Printing Ofce, Washington DC, 1959.
 D.W. Ball, The chemical composition of maple syrup, J. Chem. Educ. 84 (2007)
 W. Bi, Y. Gao, J. Shen, C. He, H. Liu, Y. Peng, C. Zhang, P. Xiao, Traditional uses,
phytochemistry, and pharmacology of the genus Acer (maple): a review,
J. Ethnopharmacol. 189 (2016) 31–60, https://doi.org/10.1016/j.
 J. Lindley, The Vegetable Kingdom, London, 1846.
 G. Bentham, J.D. Hooker, Genera Plantarum 1, Lovell Reeve & Co., London, 1862.
 F. Pax, Aceraceae, in: A. Engler (Ed.), Das Panzenreich, H. 8, Wilhelm
Engelmann, Leipzig, 1902, pp. 1–89.
 A. Cronquist, An Integrated System of Classication of Flowering Plants,
Columbia University Press, New York, 1981.
 R.F. Thorne, An updated phylogenetic classication of owering plants, Aliso. 13
 A. Takhtajan, Diversity and Classication of Flowering Plants, Columbia
University Press, New York, 1997.
 V. Savolainen, M.F. Fay, D.C. Albach, A. Backlund, M. van der Bank, K.
M. Cameron, S.A. Johnson, M.D. Lledo, J.-C. Pintaud, M. Powell, M.C. Sheahan,
D.E. Soltis, P.S. Soltis, P. Weston, W.M. Whitten, K.J. Wurdack, M.W. Chase,
Phylogeny of the eudicots: a nearly complete familial analysis based on rbcL gene
sequences, Kew Bull. 55 (2000) 257–309, https://doi.org/10.2307/4115644.
 D.E. Soltis, P.S. Soltis, M.W. Chase, M.E. Mort, D.C. Albach, M. Zanis,
V. Savolainen, W.H. Hahn, S.B. Hoot, M.F. Fay, M. Axtell, S.M. Swensen, L.
M. Prince, W.J. Kress, K.C. Nixon, J.S. Farris, Angiosperm phylogeny inferred
from 18S rDNA, rbcL, and atpB sequences, Bot. J. Linn. Soc. 133 (2000) 381–461,
 M.G. Harrington, K.J. Edwards, S.A. Johnson, M.W. Chase, P.A. Gadek,
Phylogenetic inference in Sapindaceae sensu lato using plastid matK and rbcL
DNA sequences, Syst. Bot. 30 (2005) 366–382, https://doi.org/10.1600/
 S. Buerki, F. Forest, P. Acevedo-Rodríguez, M.W. Callmander, J.A.A. Nylander,
M. Harrington, I. Sanmartín, P. Küpfer, N. Alvarez, Plastid and nuclear DNA
markers reveal intricate relationships at subfamilial and tribal levels in the
soapberry family (Sapindaceae), Mol. Phylogenet. Evol. 51 (2009) 238–258,
 S. Buerki, P.P. Lowry II, N. Alvarez, S.G. Razamandimbison, P. Küpfer, M.
W. Callmander, Phylogeny and circumscription of Sapindaceae revisited:
molecular sequence data, morphology and biogeography support recognition of a
new family, Xanthoceraceae, Plant Ecol. Evol. 143 (2010) 148–159, https://doi.
 J.A. Wolfe, T. Tanai, Systematics, phylogeny, and distribution of Acer (maples) in
the Cenozoic of Western North America, J. Fac. Sci. Hokkaido Univ. Ser. 4 (22)
(1987) 1–246. Geol. Mineral.
 M.C. Boulter, J.N. Beneld, H.C. Fisher, D.A. Gee, M. Lhotak, The evolution and
global migration of the Aceraceae, Philos. Trans. R. Soc. B Biol. Sci. 351 (1996)
 S.-F. Huang, R.E. Ricklefs, P.E. Raven, Phylogeny and historical biogeography of
Acer I - study history of the infrageneric classication, Taiwania 47 (2002)
 S.S. Renner, G.W. Grimm, G.M. Schneeweiss, T.F. Stuessy, R.E. Ricklefs, Rooting
and dating maples (Acer) with an uncorrelated-rates molecular clock:
implications for north American/Asian disjunctions, Syst. Biol. 57 (2008)
 P.R. Crane, D.L. Dilcher, S.R. Manchester, A Preliminary Survey of Fossil Leaves
and Well-Preserved Reproductive Structures From the Sentinel Butte Formation
(Paleocene) near Almont, North Dakota, Fieldiana (Geology), New Ser, 1990,
pp. 1–63, https://doi.org/10.5962/bhl.title.100826.
 S.R. Manchester, Biogeographical relationships of north American tertiary oras,
Ann. Missouri Bot. Gard. 86 (1999) 472–522, https://doi.org/10.2307/2666183.
 T. Tanai, Revisions of tertiary acer from East Asia, J. Fac. Sci. Hokkaido Univ. Ser.
4 (20) (1983) 291–390. Geol. Mineral.
 A.M. Mcclain, S.R. Manchester, Dipteronia (Sapindaceae) from the Tertiary of
North America and implications for the phytogeographic history of the
Aceroideae, Am. J. Bot. 88 (2001) 1316–1325.
 F. Areces-Berazain, Y. Wang, D.D. Hinsinger, J.S. Strijk, Plastome comparative
genomics in maples resolves the infrageneric backbone relationships, PeerJ. 8
(2020), e9483, https://doi.org/10.7717/peerj.9483.
 L.A. Ripma, M.G. Simpson, K. Hasenstab-Lehman, Geneious! Simplied genome
skimming methods for phylogenetic systematic studies: a case study in Oreocarya
(Boraginaceae), Appl. Plant Sci. 2 (2014) 1400062, https://doi.org/10.3732/
 T. Cvetkovic, D.D. Hinsinger, J.S. Strijk, Exploring evolution and diversity of
Chinese Dipterocarpaceae using next-generation sequencing, Sci. Rep. 9 (2019)
 K. Katoh, D.M. Standley, MAFFT multiple sequence alignment software version 7:
improvements in performance and usability, Mol. Biol. Evol. 30 (2013) 772–780,
 A.M. Kozlov, D. Darriba, T. Flouri, B. Morel, A. Stamatakis, RAxML-NG: a fast,
scalable and user-friendly tool for maximum likelihood phylogenetic inference,
Bioinformatics (2019) 1–3, https://doi.org/10.1093/bioinformatics/btz305.
 D. Darriba, D. Posada, A.M. Kozlov, A. Stamatakis, T. Flouri, ModelTest-NG: a
new and scalable tool for the selection of DNA and protein evolutionary models,
BioRxiv 612903 (2019), https://doi.org/10.1101/612903.
 A.M. Kozlov, A. Stamatakis, Using RAxML-NG in practice, Preprints (2019) 1–25,
 R. Bouckaert, T.G. Vaughan, J. Barido-Sottani, S. Duchˆ
ene, M. Fourment,
A. Gavryushkina, J. Heled, G. Jones, D. Kühnert, N. De Maio, M. Matschiner, F.
K. Mendes, N.F. Müller, H.A. Ogilvie, L. du Plessis, A. Popinga, A. Rambaut,
D. Rasmussen, I. Siveroni, M.A. Suchard, C.H. Wu, D. Xie, C. Zhang, T. Stadler, A.
J. Drummond, BEAST 2.5: An advanced software platform for Bayesian
F. Areces-Berazain et al.
Genomics 113 (2021) 681–692
evolutionary analysis, PLoS Comput. Biol. 15 (2019) e1006650, https://doi.org/
 R. Bouckaert, J. Heled, D. Kühnert, T. Vaughan, C.H. Wu, D. Xie, M.A. Suchard,
A. Rambaut, A.J. Drummond, BEAST 2: a software platform for Bayesian
evolutionary analysis, PLoS Comput. Biol. 10 (2014), e1003537, https://doi.org/
 A.J. Drummond, M.A. Suchard, D. Xie, A. Rambaut, Bayesian phylogenetics with
BEAUti and the BEAST 1.7, Mol. Biol. Evol. 29 (2012) 1969–1973, https://doi.
 A.J. Drummond, A. Rambaut, BEAST: Bayesian evolutionary analysis by sampling
trees, BMC Evol. Biol. 7 (2007) 214, https://doi.org/10.1186/1471-2148-7-214.
 A.J. Drummond, S.Y.W. Ho, M.J. Phillips, A. Rambaut, Relaxed phylogenetics and
dating with condence, PLoS Biol. 4 (2006) 699–710, https://doi.org/10.1371/
 S. Nee, R.M. May, P.H. Harvey, The reconstructed evolutionary process, Philos.
Trans. R. Soc. B Biol. Sci. 344 (1994) 305–311.
 M.A. Miller, W. Pfeiffer, T. Schwartz, Creating the CIPRES Science Gateway for
inference of large phylogenetic trees, in: Proceedings of the Gateway Computing
Environments Workshop (GCE), New Orleans, LA, 2010, pp. 1–8, https://doi.org/
 A. Rambaut, A.J. Drummond, D. Xie, G. Baele, M.A. Suchard, Posterior
summarization in Bayesian phylogenetics using Tracer 1.7, Syst. Biol. 67 (2018)
 O.D. Lavrenko, L.I. Fot’janova, Some early Paleogene species from Western
Kamchatka, in: M.C. Boulter, H.C. Fisher (Eds.), Cenozoic Plants and Climates of
the Arctic, Springer⋅Verlag, Berlin Heidelberg, 1994, pp. 315–321.
 A. Kittle, M.L. DeVore, B. Wall, K. Pigg, Acer fruits from the Paleocene of North
Dakota, Georg. J. Sci. 63 (2005) 32.
 R. Zetter, M.J. Farabee, K.B. Pigg, S.R. Manchester, M.L. DeVore, M.D. Nowak,
Palynoora of the late Paleocene silicied shale at Almont, North Dakota, USA,
Palynology 35 (2011) 179–211, https://doi.org/10.1080/
 R.A. Stockey, G.L. Hoffman, G.W. Rothwell, Paleobotany and paleoecology of Gao
mine, a late Paleocene fossil locality near Red Deer, Alberta, Canada, Can. J.
Earth Sci. 50 (2013) 235–248, https://doi.org/10.1139/cjes-2012-0073.
 A.M. Mcclain, Systematics and Biogeography of Selected Modern and Fossil
Dipteronia and Acer (Sapindaceae), Master Thesis, University of Florida, 2000.
 J. Kovar-Eder, G. Sun, The Neogene ora from Badaogou of Changbai, NE China -
Most similar living relatives of selected taxa and relations to the European record,
Rev. Palaeobot. Palynol. 158 (2009) 1–13, https://doi.org/10.1016/j.
 N.J. Matzke, BioGeoBEARS: BioGeography With Bayesian (and Likelihood)
Evolutionary Analysis in R Scripts. R Package, v. 0.2.1. https://rdrr.io/cran/Bio
 N.J. Matzke, Probabilistic historical biogeography: new models for founder-event
speciation, imperfect detection, and fossils allow improved accuracy and model-
testing, Front. Biogeogr. 5 (2013) 242–248, https://doi.org/10.21425/
 N.J. Matzke, Model selection in historical biogeography reveals that founder-
event speciation is a crucial process in island clades, Syst. Biol. 63 (2014)
 R Development CoreTeam, R: A Language and Environment for Statistical
Computing, R Foundation for Statistical Computing, Vienna, 2019.
 GBIF.org, GBIF (Global Biodiversity Information Facility) Home Page.
 POWO, Plants of the World Online. Facilitated by the Royal Botanic Gardens,
Kew. http://www.plantsoftheworldonline.org/, 2019.
 L.J. Revell, Phytools: an R package for phylogenetic comparative biology (and
other things), Methods Ecol. Evol. 3 (2012) 217–223, https://doi.org/10.1111/
 O.G. Pybus, P.H. Harvey, Testing macro-evolutionary models using incomplete
molecular phylogenies, Proc. R. Soc. Lond. B 267 (2000) 2267–2272, https://doi.
 S. H¨
ohna, M.R. May, B.R. Moore, TESS: an R package for efciently simulating
phylogenetic trees and performing Bayesian inference of lineage diversication
rates, Bioinformatics. 32 (2015) 789–791.
 S. H¨
ohna, The time-dependent reconstructed evolutionary process with a key-role
for mass-extinction events, J. Theor. Biol. 380 (2015) 321–331, https://doi.org/
 M.R. May, S. H¨
ohna, B.R. Moore, A Bayesian approach for detecting the impact of
mass-extinction events on molecular phylogenies when rates of lineage
diversication may vary, Methods Ecol. Evol. 7 (2016) 947–959, https://doi.org/
 J. Wen, Z.L. Nie, S.M. Ickert-Bond, Intercontinental disjunctions between eastern
Asia and western North America in vascular plants highlight the biogeographic
importance of the Bering land bridge from late Cretaceous to Neogene, J. Syst.
Evol. 54 (2016) 469–490, https://doi.org/10.1111/jse.12222.
 A.J. Harris, Y. Chen, R.T. Olsen, S. Lutz, J. Wen, On merging Acer sections Rubra
and Hyptiocarpa: molecular and morphological evidence, PhytoKeys. 86 (2017)
 G.W. Grimm, S.S. Renner, A. Stamatakis, V. Hemleben, A nuclear ribosomal DNA
phylogeny of Acer inferred with maximum likelihood, splits graphs, and motif
analysis of 606 sequences, Evol. Bioinforma. 2 (2006) 7–22, https://doi.org/
 J. Li, M. Stukel, P. Bussies, K. Skinner, A.R. Lemmon, E.M. Lemmon, K. Brown,
A. Bekmetjev, N.G. Swenson, Maple phylogeny and biogeography inferred from
phylogenomic data, J. Syst. Evol. (2019), https://doi.org/10.1111/jse.12535.
 G.W. Grimm, Tracing the Mode and Speed of Intrageneric Evolution - A
Phylogenetic Case Study on Genus Acer L. (Aceraceae) and Genus Fagus L.
(Fagaceae) Using Fossil, Morphological, and Molecular Data, PhD thesis,
Eberhard-Karls University Tübingen, 2003.
 S.S. Renner, L. Beenken, G.W. Grimm, A. Kocyan, R.E. Ricklefs, The evolution of
dioecy, heterodichogamy, and labile sex expression in Acer, Evolution. 61 (2007)
 H. Zhu, S.R. Manchester, Red and silver maples in the neogene of western North
America: fossil leaves and samaras of Acer section Rubra, Int. J. Plant Sci. 181
(2020) 542–556, https://doi.org/10.1086/707106.
 Y. Chen, W.O. Wong, Q. Hu, Y. Liufu, Z. Xie, A new fossil-species of Acer
(Sapindaceae) from the Ningming Basin in Guangxi, South China, Phytotaxa 298
(2017) 158–164, https://doi.org/10.11646/phytotaxa.298.2.5.
 J.H. Kim, A new species of Acer samaras from the Miocene Yeonil Group in the
Pohang Basin, Korea, Geosci. J. 12 (2008) 331–336, https://doi.org/10.1007/
 H. Walther, Studien über terti¨
are Acer Mitteleuropas, in: Abhandlungen Des
Staatl. Museums Für Mineral. Und Geol. Zu Dresden 19, 1972, pp. 1–309.
 I. Sanmartín, H. Enghoff, F. Ronquist, Patterns of animal dispersal, vicariance and
diversication in the Holarctic, Biol. J. Linn. Soc. 73 (2001) 345–390, https://doi.
 L.Y. Budantsev, L.B. Golovneva, Fossil Flora of Arctic II- Paleogene Flora of
Spitsbergen, Russian Academy of Sciences, Komarov Botanical Institute, Saint
 H. Walther, Invasion of arcto-tertiary elements in the paleogene of central
Europe, in: M. Boulter, H. Fisher (Eds.), Cenozoic Plants and Climates of the
Arctic, Springer-Verlag, Berlin Heidelberg, 1994, pp. 239–250.
 H. Walther, Z. Kvacek, Early Oligocene ora of of Seifhennersdorf (Saxony), Acta
Musei Natl. Pragae 63 (2007) 85–174.
 F. Grímsson, T. Denk, Floristic turnover in Iceland from 15 to 6 Ma - extracting
biogeographical signals from fossil oral assemblages, J. Biogeogr. 34 (2007)
 J.A. Wolfe, Some aspects of plant geography of the northern hemisphere during
the late cretaceous and tertiary, Ann. Missouri Bot. Gard. 62 (1975) 264, https://
 J.M. Baskin, C.C. Baskin, Origins and relationships of the mixed mesophytic forest
of Oregon-Idaho, China, and Kentucky: review and synthesis, Ann. Missouri Bot.
Gard. 101 (2016) 525–552, https://doi.org/10.3417/2014017.
 B. Huntley, T. Webb III, Vegetation History, Kluwer Academics, Dordrecht, 1988.
 J.C. Svenning, Deterministic Plio-Pleistocene extinctions in the European cool-
temperate tree ora, Ecol. Lett. 6 (2003) 646–653.
 H. Qian, R.E. Ricklefs, A comparison of the taxonomic richness of vascular plants
in China and the United States, Am. Nat. 154 (1999) 160–181, https://doi.org/
 H. Qian, R.E. Ricklefs, Large-scale processes and the Asian bias in species
diversity of temperate plants, Nature. 407 (2000) 180–182.
 Y. Xing, M.A. Gandolfo, R.E. Onstein, D.J. Cantrill, B.F. Jacobs, G.J. Jordan, D.
E. Lee, S. Popova, R. Srivastava, T. Su, S.V. Vikulin, A. Yabe, H.P. Linder, Testing
the biases in the rich Cenozoic angiosperm macrofossil record, Int. J. Plant Sci.
177 (2016) 371–388, https://doi.org/10.1086/685388.
 A. Harris, J. Xiang, D. Thomas, Phylogeny, origin, and biogeographic history of
Aesculus L. (Sapindales) - an update from combined analysis of DNA sequences,
morphology, and fossils, Taxon 58 (2009) 108–126.
 S. Chin, J. Shaw, R. Haberle, J. Wen, D. Potter, Diversication of almonds,
peaches, plums and cherries – molecular systematics and biogeographic history of
Prunus (Rosaceae), Mol. Phylogenet. Evol. 76 (2014) 34–48, https://doi.org/
 S. Renner, G. Grimm, P. Kapli, T. Denk, Species relationships and divergence
times in beeches: new insights from the inclusion of 53 young and old fossils in a
birth-death clock model, Phylosophical Trans. R. Soc. B. 371 (2016) 20150135.
 L. Jiang, Q. Bao, W. He, D.-M. Fan, S.-M. Cheng, J. Lopez-Pujol, M. Chung,
S. Sakaguchi, A. Sanchez Gonzalez, A. Gedik, D.-Z. Li, Y.-X. Kou, Z.-Y. Zhang,
Phylogeny and biogeography of Fagus (Fagaceae) based on 28 nuclear single/low-
copy loci, J. Syst. Evol. (2020), https://doi.org/10.1111/jse.12695.
 L.Y. Budantsev, Early stages of formation and dispersal of the temperate ora in
the boreal region, Bot. Rev. 58 (1992) 1–48.
F. Areces-Berazain et al.