distribution of living cupressaceae re ects the breakup of...

16
Distribution of living Cupressaceae reects the breakup of Pangea Kangshan Mao a,b,c,1 , Richard I. Milne a,b,c,1 , Libing Zhang d,e , Yanling Peng a , Jianquan Liu a,2 , Philip Thomas c , Robert R. Mill c , and Susanne S. Renner f a State Key Laboratory of Grassland Agro-Ecosystem, School of Life Sciences, Lanzhou University, Lanzhou, Gansu 730000, Peoples Republic of China; b Institute of Molecular Plant Sciences, School of Biological Sciences, University of Edinburgh, Edinburgh EH9 3JH, United Kingdom; c Royal Botanic Garden Edinburgh, Edinburgh EH3 5LR, Scotland, United Kingdom; d Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu, Sichuan 610041, Peoples Republic of China; e Missouri Botanical Garden, St. Louis, MO 63166; f Systematic Botany and Mycology, Department of Biology, University of Munich, 80638 Munich, Germany Edited by Charles C. Davis, Harvard University, Cambridge, MA, and accepted by the Editorial Board March 21, 2012 (received for review September 2, 2011) Most extant genus-level radiations in gymnosperms are of Oligocene age or younger, reecting widespread extinction during climate cooling at the Oligocene/Miocene boundary [23 million years ago (Ma)]. Recent biogeographic studies have revealed many instances of long-distance dispersal in gymnosperms as well as in angiosperms. Acting together, extinction and long-distance dispersal are likely to erase historical biogeographic signals. Notwithstanding this problem, we show that phylogenetic relationships in the gymnosperm family Cupressaceae (162 species, 32 genera) exhibit patterns expected from the Jurassic/Cretaceous breakup of Pangea. A phylogeny was gener- ated for 122 representatives covering all genera, using up to 10,000 nucleotides of plastid, mitochondrial, and nuclear sequence per species. Relying on 16 fossil calibration points and three molecular dating methods, we show that Cupressaceae originated during the Triassic, when Pangea was intact. Vicariance between the two sub- families, the Laurasian Cupressoideae and the Gondwanan Callitroi- deae, occurred around 153 Ma (124183 Ma), when Gondwana and Laurasia were separating. Three further intercontinental disjunctions involving the Northern and Southern Hemisphere are coincidental with or immediately followed the breakup of Pangea. ancestral areas reconstruction | molecular clock B etween the Early Triassic and the Middle Jurassic, virtually all continents were joined to form the supercontinent Pangea (13). Around 160138 million years ago (Ma) (1, 3), Pangea broke up into two supercontinents: Laurasia, comprising land that eventually gave rise to North America, Europe, and much of Asia, and Gondwana, made up of land that subsequently gave rise to South America, Africa, India, Antarctica, and Australia. Bio- stratigraphic data suggest that Late Triassic and Early Jurassic Pangea had a warm and equable climate without glaciation or sea ice and that it lacked signicant geographic barriers from pole to pole (4). However, because of Pangeas great latitudinal expanse, faunal provinces already had developed before its break-up, and dated molecular phylogenies of reptiles, amphibians, and mammals have made clear that subsequent lineage divergence within these groups matches the separation and fragmentation of Laurasia and Gondwana (510). Until now, there has been no equivalent evi- dence for any plant family. The fossil record shows that gymnosperms dominated the veg- etation of Pangea but declined in dominance and abundance from the Mid-Cretaceous onwards (11, 12). Perhaps because of the ex- tinction of entire clades, molecular-clock studies of gymnosperms consistently have inferred young, usually Oligocene, ages for the crown groups of living genera, e.g., Phyllocladus (13), Gnetum (14), Cedrus (15), Agathis (16, 17), Ephedra (18), Juniperus (19), Pseu- dotsuga (20), Podocarpus, Nageia, Dacrydium, Dacrycarpus (21), and Pinus subgenera Pinus and Strobus (22, 23). Radiations are especially young in the cycads (2426). Among the few spermatophyte clades that still may reect events related to the break-up of Pangea is the conifer family Cupressa- ceae (including the former Taxodiaceae) (2731). Cupressaceae occur on all continents except Antarctica and comprise 162 species in 32 genera (see Table S2 for subfamilies, genera, and species numbers). The family has a well-studied fossil record going back to the Jurassic (3236). Using ancient fossils to calibrate genetic distances in molecular phylogenies can be problematic, because the older a fossil is, the more likely it is to represent an extinct lineage that diverged somewhere along the line leading to the extant taxon with which it is being compared (37). However, probability dis- tributions on fossil calibration ages allow some manipulation of this uncertainty (38), and judicious use of multiple fossils also may help circumvent calibration pitfalls (39). Here we present a phylogeny for 122 species from the 32 genera of Cupressaceae (plus 22 species representing relevant outgroups) and use 16 fossil calibration points and three dating approaches to estimate divergence times in the Cupressaceae. We then perform ancestral area reconstructions (AARs) using maximum likelihood based on datasets with or without incorporated fossils. Possible changes in diversication rates were inferred with an approach that accounts for nonrandom taxon sampling in molecular phylogenies (40). Nonrandom sampling arises when phylogenies include at least one species per genus but not all congenerics, thereby over- representing deep nodes (diversication events) in the tree. Ex- periments have conrmed the theoretical expectation that such sampling leads to the erroneous inference of diversication rate downturns (41, 42). We aimed to test the hypothesis that, given their fossil record, the deepest Cupressaceae divergences should reect the break-up of Pangea and that evolution of the family then continued on the separating continental landmasses. Results Cupressaceae Phylogenetics. After sequence alignment and removal of ambiguous regions, we obtained two datasets, one of 56 taxa and 10,472 aligned nucleotides from plastid, mitochondrial, and nuclear DNA, the other of 144 taxa and 7,171 nucleotides from plastid DNA only. Maximum likelihood, parsimony, and Bayesian opti- mization inferred similar topologies from both datasets. Support Author contributions: J.L., K.M., and R.I.M. designed research; K.M., R.I.M., L.Z., Y.P., P.T., R.R.M., and J.L. performed research; S.S.R. contributed new reagents/analytic tools; K.M., J.L., and S.S.R. analyzed data; and K.M., J.L., S.S.R., and R.I.M. wrote the paper. The authors declare no conict of interest. This article is a PNAS Direct Submission. C.C.D. is a guest editor invited by the Editorial Board. Freely available online through the PNAS open access option. Data deposition: The sequences reported in this paper have been deposited in the Gen- Bank database (JF725702JF725991). GenBank accession numbers and provenance of se- quenced samples are provided in Table S1. 1 K.M. and R.I.M. contributed equally to this work. 2 To whom correspondence should be addressed. E-mail: [email protected]. This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1114319109/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1114319109 PNAS Early Edition | 1 of 6 EVOLUTION

Upload: others

Post on 04-Oct-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Distribution of living Cupressaceae reflects thebreakup of PangeaKangshan Maoa,b,c,1, Richard I. Milnea,b,c,1, Libing Zhangd,e, Yanling Penga, Jianquan Liua,2, Philip Thomasc,Robert R. Millc, and Susanne S. Rennerf

aState Key Laboratory of Grassland Agro-Ecosystem, School of Life Sciences, Lanzhou University, Lanzhou, Gansu 730000, People’s Republic of China;bInstitute of Molecular Plant Sciences, School of Biological Sciences, University of Edinburgh, Edinburgh EH9 3JH, United Kingdom; cRoyal Botanic GardenEdinburgh, Edinburgh EH3 5LR, Scotland, United Kingdom; dChengdu Institute of Biology, Chinese Academy of Sciences, Chengdu, Sichuan 610041, People’sRepublic of China; eMissouri Botanical Garden, St. Louis, MO 63166; fSystematic Botany and Mycology, Department of Biology, University of Munich, 80638Munich, Germany

Edited by Charles C. Davis, Harvard University, Cambridge, MA, and accepted by the Editorial Board March 21, 2012 (received for review September 2, 2011)

Most extant genus-level radiations in gymnosperms are of Oligoceneage or younger, reflecting widespread extinction during climatecooling at the Oligocene/Miocene boundary [∼23 million years ago(Ma)]. Recent biogeographic studies have revealed many instances oflong-distance dispersal in gymnosperms as well as in angiosperms.Acting together, extinction and long-distance dispersal are likely toerase historical biogeographic signals. Notwithstanding this problem,we show that phylogenetic relationships in the gymnosperm familyCupressaceae (162 species, 32 genera) exhibit patterns expected fromthe Jurassic/Cretaceous breakup of Pangea. A phylogeny was gener-ated for 122 representatives covering all genera, using up to 10,000nucleotides of plastid, mitochondrial, and nuclear sequence perspecies. Relying on 16 fossil calibration points and three moleculardating methods, we show that Cupressaceae originated during theTriassic, when Pangea was intact. Vicariance between the two sub-families, the Laurasian Cupressoideae and the Gondwanan Callitroi-deae, occurred around 153 Ma (124–183 Ma), when Gondwana andLaurasia were separating. Three further intercontinental disjunctionsinvolving the Northern and Southern Hemisphere are coincidentalwith or immediately followed the breakup of Pangea.

ancestral areas reconstruction | molecular clock

Between the Early Triassic and the Middle Jurassic, virtuallyall continents were joined to form the supercontinent Pangea

(1–3). Around 160–138 million years ago (Ma) (1, 3), Pangeabroke up into two supercontinents: Laurasia, comprising land thateventually gave rise to North America, Europe, and much of Asia,and Gondwana, made up of land that subsequently gave rise toSouth America, Africa, India, Antarctica, and Australia. Bio-stratigraphic data suggest that Late Triassic and Early JurassicPangea had a warm and equable climate without glaciation or seaice and that it lacked significant geographic barriers from pole topole (4). However, because of Pangea’s great latitudinal expanse,faunal provinces already had developed before its break-up, anddatedmolecular phylogenies of reptiles, amphibians, andmammalshave made clear that subsequent lineage divergence within thesegroups matches the separation and fragmentation of Laurasia andGondwana (5–10). Until now, there has been no equivalent evi-dence for any plant family.The fossil record shows that gymnosperms dominated the veg-

etation of Pangea but declined in dominance and abundance fromthe Mid-Cretaceous onwards (11, 12). Perhaps because of the ex-tinction of entire clades, molecular-clock studies of gymnospermsconsistently have inferred young, usually Oligocene, ages for thecrown groups of living genera, e.g., Phyllocladus (13),Gnetum (14),Cedrus (15), Agathis (16, 17), Ephedra (18), Juniperus (19), Pseu-dotsuga (20), Podocarpus, Nageia, Dacrydium, Dacrycarpus (21),and Pinus subgenera Pinus and Strobus (22, 23). Radiations areespecially young in the cycads (24–26).Among the few spermatophyte clades that still may reflect events

related to the break-up of Pangea is the conifer family Cupressa-ceae (including the former Taxodiaceae) (27–31). Cupressaceae

occur on all continents except Antarctica and comprise 162 speciesin 32 genera (see Table S2 for subfamilies, genera, and speciesnumbers). The family has a well-studied fossil record going backto the Jurassic (32–36). Using ancient fossils to calibrate geneticdistances inmolecular phylogenies can be problematic, because theolder a fossil is, the more likely it is to represent an extinct lineagethat diverged somewhere along the line leading to the extant taxonwith which it is being compared (37). However, probability dis-tributions on fossil calibration ages allow somemanipulation of thisuncertainty (38), and judicious use of multiple fossils also may helpcircumvent calibration pitfalls (39).Here we present a phylogeny for 122 species from the 32 genera

of Cupressaceae (plus 22 species representing relevant outgroups)and use 16 fossil calibration points and three dating approaches toestimate divergence times in the Cupressaceae. We then performancestral area reconstructions (AARs) using maximum likelihoodbased on datasets with or without incorporated fossils. Possiblechanges in diversification rates were inferred with an approach thataccounts for nonrandom taxon sampling in molecular phylogenies(40). Nonrandom sampling arises when phylogenies include atleast one species per genus but not all congenerics, thereby over-representing deep nodes (diversification events) in the tree. Ex-periments have confirmed the theoretical expectation that suchsampling leads to the erroneous inference of diversification ratedownturns (41, 42). We aimed to test the hypothesis that, giventheir fossil record, the deepest Cupressaceae divergences shouldreflect the break-up of Pangea and that evolution of the familythen continued on the separating continental landmasses.

ResultsCupressaceae Phylogenetics.After sequence alignment and removalof ambiguous regions, we obtained two datasets, one of 56 taxa and10,472 aligned nucleotides from plastid, mitochondrial, and nuclearDNA, the other of 144 taxa and 7,171 nucleotides from plastidDNA only. Maximum likelihood, parsimony, and Bayesian opti-mization inferred similar topologies from both datasets. Support

Author contributions: J.L., K.M., and R.I.M. designed research; K.M., R.I.M., L.Z., Y.P., P.T.,R.R.M., and J.L. performed research; S.S.R. contributed new reagents/analytic tools; K.M.,J.L., and S.S.R. analyzed data; and K.M., J.L., S.S.R., and R.I.M. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. C.C.D. is a guest editor invited by the EditorialBoard.

Freely available online through the PNAS open access option.

Data deposition: The sequences reported in this paper have been deposited in the Gen-Bank database (JF725702–JF725991). GenBank accession numbers and provenance of se-quenced samples are provided in Table S1.1K.M. and R.I.M. contributed equally to this work.2To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1114319109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1114319109 PNAS Early Edition | 1 of 6

EVOLU

TION

Page 2: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

values for major groups are high, with three-quarters of the genus-level nodes having >95% posterior probability (Fig. S1).

Divergence Times. Bayesian coestimation of topology and diver-gence time (BEAST, using uniform prior distributions on cali-bration ages) (43) generally gave the oldest ages, and Penalizedlikelihood (44, 45) gave the youngest (Table S3). An alternativeBayesian approach, which used a fixed topology (MULTI-DIVTIME) (46), yielded ages for short-branched nodes (mostnodes within Cupressoideae; Fig. S1) that were similar to oryounger than those obtained with BEAST; ages for long-branchednodes (most nodes within Callitroideae; Fig. S1) were similar to orolder than those obtained with BEAST. Confidence intervalsaround estimates from the two Bayesian approaches overlapped(Table S3). With all three dating approaches, the more denselysampled 144-taxon dataset produced slightly older age estimates(compare Fig. S2 A and B), a result that is consistent with the ef-fects of undersampling observed elsewhere (47). Because BEASTallows more complex nucleotide-substitution models than do theother two dating approaches, and because dates from the 56-taxon matrix might be less accurate because of undersampling,the following discussion focuses on the results obtained with the144-taxon matrix analyzed using BEAST (Fig. 1 and Fig. S2B).Cupressaceae split from their sister lineage during the late

Permian and early Triassic (209–282 Ma; node 1 in Fig. 1, Table 1,and Table S3) and began to diversify into seven major lineages(commonly ranked as subfamilies) during the Triassic (184–254Ma; node 2 in Fig. 1, Table 1, and Table S3). The genera be-longing to each subfamily are shown in Fig. S1. The stem line-ages of Cunninghamioideae, Taiwanioideae, Athrotaxidoideae,Sequoioideae, and Taxodioideae appeared around 184–254 Ma,170–238 Ma, 157–224 Ma, 150–215 Ma, and 140–201 Ma, re-spectively (nodes 2–6 in Fig. 1, Table 1, and Table S3). Theyoungest subfamilies are the Cupressoideae and Callitroideae,which diverged from each other 124–183 Ma (node 7 in Fig. 1,Table 1, and Table S3). Most cupressaceous genera with two ormore species diversified after the Cretaceous/Tertiary boundary(65.5 ± 0.3 Ma) (48) (Fig. S2B); the only exception is Chamae-cyparis, which is dated to 61–108 Ma (node 23 in Fig. S2B; notethat this is the crown age for the Chamaecyparis-Fokienia clade).BEAST analyses with different uniform distribution priors on

the calibration closest to the root (calibration point P; Figs. S2and S3) yielded largely overlapping 95% highest posterior den-sity (HPD) age ranges for all nodes of interest, indicating thatthe chosen maximum constraint (the only such constraint used inthe analysis) had no overly strong effect on the remaining dates(Fig. S3; compare run 1 with runs 2–4). BEAST analyses withdifferent subsets of calibration points, all with uniform priors,showed that calibration P plus calibrations A, B, E, F, G, J, K,and L (subset VND; SI Text and Table S4) (Fig. S3, run 5)yielded node ages similar to those obtained with calibrations Athrough P (Table S4) (Fig. S3, run 1), whereas calibration P pluscalibrations C, D, H, I, M, N, O, and P (subset NVND; SI Textand Table S4) gave much younger ages (Fig. S3, run 6).A BEAST analysis that used lognormal prior distributions on the

ages of calibration P and subset VND (and uniform priors forsubset NVND) (run 7) generated age estimates younger than butlargely overlapping those obtained with uniform priors for cali-brationA through P (run 1) (Fig. S3, Table 1, and Tables S3 and S4;see SI Text for a detailed comparison). In all nine BEAST analyses,effective sample sizes for each parameter were well above 200.

Ancestral Areas and Diversification Rate Changes. Likelihood AARswere implemented under the dispersal-extinction-cladogenesismodel in LAGRANGE (49). We defined eight continent pairs(NS, NE, SF, NA, AE, FE, SU, NF; area codes are explainedin Materials and Methods and are illustrated in Fig. 2A) and onecontinent group (NAE), which reflect continental connections

known from plate tectonics (1, 3). The most likely scenarios(Fig. 2B) required 31 dispersal events, 21 vicariance events, andfive local extinctions. Likelihood AAR for living Cupressaceae(Fig. 2B) suggested that the family originated in Asia and itsearly members expanded to North America from where Calli-troideae and Athrotaxis entered Gondwanan South America. Theintegration of fossil Cupressaceae (Fig. 2 C–E, Fig. S4, and Table

(A) 150 Mya

CenozoicMesozoic

JurassicTriassicPermian

Paleozoic

Early Cretaceous L Cretaceous P. Eocene Oli. Mio.

(Ma)275 250 225 200 175 150 125 100 75 50 25 0

eaecasserpuC

eaediosserpuC

eaed iortillaC

TaxSeqAthTaiCun

1

2

3

4

5

6

7

8

9

10

(Ma)275 250 225 75 50 25 0001002 150 125175

(B) Present

Gondwanan continents:

Central & South Africa

Gondwanan continents:

South America and Oceania

Gondwanan continents:

North Afria

Laurasian continents: Eurasia,

North & Central America

Other Cupressaceae subfamilies

Outgroups

Calibration points

Fig. 1. (Upper) Chronogram for 122 Cupressaceae species and 22 outgroupsbased on an alignment of >7,000 nucleotides of plastid DNA (144-taxondataset). A geological time scale is shown at the bottom (48). Blue lines rep-resent Cupressoideae restricted to the area of Laurasian continents. Red linesrepresent Callitroideae restricted to Gondwanan continents. Pink lines repre-sent species occurring in Africa in and north of the Sahara. Yellow lines rep-resent species occurring in Africa south of the Sahara. Gray hexagonsrepresent calibration points. Gray bars represent 95% HPD intervals for nodes1–10. Gray (run 1) and purple (run 7) normal distributions represent the pos-terior for the BEAST age estimate of node 7 when uniform or lognormal priorswere applied to calibration points. Orange shading indicates the period ofdecreasing feasibility of floristic exchange between Laurasia and Gondwana.Divergence times of nodes 5, 6, 7, 8, and 10 overlap with the fragmentationof Pangea. (Lower) Maps show (A) a paleocontinent reconstruction at 150Ma and (B) the current distribution of Callitroideae and Cupressoideae. Thestippled circle in A emphasizes island chains between North and SouthAmerica; Ath, Athrotaxidoideae; Cun, Cunninghamioideae; Seq, Sequoioi-deae; Tai, Taiwanioideae; Tax, Taxodioideae. Reprinted with permission fromRon Blakey, Colorado Plateau Geosystems.

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1114319109 Mao et al.

Page 3: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

S5) in likelihood AAR resulted in a similar scenario but with twomore incursions from Laurasian to Gondwanan continents(Austrosequoia and Austrohamia minuta). Reconstructions usingalternative placements of certain fossil taxa (two Austrohamiaspecies and Sewardiodendron laxum) affected only the ancestralareas inferred for the nodes closest to them (compare Fig. 2 Band C–E).Because the 122 species that we sequenced are not a random

sample of the family’s 162 species but instead overrepresent deepnodes, we fitted birth/death diversification models to the maxi-mum likelihood topology after statistical completion, assumingnonrandom species sampling (40). When the nonsequenced 40species were added to the tree under a constant-rate birth-deathmodel, assuming they originated during the past 10 million y, thebest fit to the 1,000 simulated completed trees was a two-ratemodel with a decrease in diversification rate at 1.37 Ma. Of thealternative models [constant-rate pure-birth (CR-PB), constant-rate birth-death, logistic density dependence, and exponentialdensity dependence], the CR-PB model provided the second-best fit to the completed trees.

DiscussionThe dense taxon sampling and large amount of sequence data usedhere yielded a solidly supported phylogeny for the Cupressaceae(Fig. S1), which are monophyletic and sister to the Taxaceae sensulato (i.e., Taxaceae sensu stricto plus Cephalotaxaceae and Amen-totaxaceae) (31), as found previously (50, 51). The divergenceof Cupressaceae from their sister lineage occurred >200 Ma (node1 in Fig. 1 and Table 1; 209–282 Ma), while Pangea was stillintact, matching fossil evidence of Cupressaceae in the Jurassic ofEurope (Hughmillerites juddii) (35), Asia (Sewardiodendron laxumand Austrohamia acanthobractea) (32, 36), and South America(Austrohamia minuta) (34). Cupressaceae diversified into sevenmajor lineages (subfamilies) during the Triassic and Jurassic (nodes2–7 inFig. 1, Fig. S2, Table 1, andTable S3), predating or coinciding

with the separation of Gondwana and Laurasia (orange column inFig. 1). Furthermore,AAR (with orwithout fossil taxa) yieldedAsiaas theancestral area for the family (Fig. 2).Cunninghamioideaemayhave originated in Asia (Fig. 2), and the divergence of Taiwanioi-deae from their sister lineage (Fig. 2B) maymatch the separation ofAsia from North America at ∼200 Ma (3); the three subfamilies,Athrotaxidoideae, Sequoioideae, and Taxodioideae (nodes 4–6 inFig. 1) probably diverged from their sister lineage inNorth America(Fig. 2 C–E); the divergence of Callitroideae from Cupressoideaewas dated to 124–183 Ma (node 7 in Fig. 1 and Table 1; mean: 153Ma), an age range almost coinciding with the separation of Gond-wana from Laurasia (Fig. 1) during the Late Jurassic (160–138Ma)(1, 3). Living members of Cupressoideae occur mainly in formerLaurasian continents (Fig. 1 Lower, B), whereas Callitroideae areendemic to fragments of Gondwana (30) (Fig. 1 Lower, B). AfricanCupressoideae apparently derived from a series of southwardexpansions during the middle and late Tertiary (Fig. 1). Un-ambiguous fossils (with reproductive organs) of Cupressoideae areknown only from former Laurasia and those of Callitroideae fromGondwana (33). Overland connections between Laurasian andGondwanan continents were severed from the Late Jurassic untilthemiddleTertiary,when India connectedwithEurasia, followedbythe subsequent reconnection of Africa to Eurasia and SouthAmerica with North America (1, 3). It is clear from our results thatthe divergence between Cupressoideae andCallitroideae correlateswith the break-up of Pangea (Fig. 1) and most likely was caused byit, as shown in the likelihood AARs (Fig. 2B).In the remaining five subfamilies, we further inferred three

intercontinental disjunctions between the Northern and SouthernHemispheres (Fig. 2 and Fig. S4). The most recent involves theextinct Austrosequoia and its extant sister lineage, comprisingSequoia and Sequoiadendron (Fig. 2 C–E and Fig. S4). Austro-sequoia dispersed from North America (via South America) toAustralia around 94–100 Ma, as judged from the mid-Cretaceous(Cenomanian) fossil remains in Australia (33, 52). The seconddisjunction involves Athrotaxis and its putative sister speciesAthrotaxites berryi (53) (Fig. S5I). Athrotaxis currently is endemicto Australia (30) but is known from fossils in Argentina (Athro-taxis ungeri) (54), and Athrotaxites berryi is known from the Aptian(ca. 111–126 Ma) (48) of North America (53). The Gondwanantaxon Athrotaxis probably originated from a southward expansionfrom North America, as suggested by likelihood AARs (Fig. 2 C–E). The third inferred intercontinental disjunction involves theextinct Austrohamia, with one species (Austrohamia minuta) fromthe Jurassic of southern Argentina (34) and the other (Austro-hamia acanthobractea) from the late Jurassic of northern China(36). The South American A. minuta might have arrived therefollowing dispersal from the Laurasian North America, as sug-gested by likelihood AAR (Fig. 2 C–E). These three instances ofintercontinental disjunctions all involve north-to-south expan-sion. We found no instance of range expansion from the Southernto the Northern Hemisphere but detected a clear signal of dis-persal or overland expansion among the Southern Gondwanancontinents themselves (Fig. S4).Previous studies of gymnosperm radiations mostly have inferred

Oligocene-age crown groups (14–26), and a recent meta-analysisfound a median crown age for gymnosperm genera of 32 Ma,younger than that found for angiosperm genera (25). Our dating ofthose genera with more than one species in the Cupressaceaesimilarly suggests relatively recent diversifications (Fig. S2B). Theyoung ages of most living gymnosperm clades probably reflectrediversification following extinction. In Cupressaceae, the evi-dence for widespread extinction and range shrinkage is particularlystrong (as visualized in Fig. S6). For example, Cunninghamioideae(Fig. S6A) and Taiwanioideae (55) were widely distributed in theNorthern Hemisphere during the Cretaceous but now are re-stricted to Asia. Sequoioideae and Taxodioideae were widespreadin the Northern Hemisphere in the Cretaceous and Early Tertiary,

Table 1. Divergence times for the Cupressaceae obtained undera Bayesian relaxed clock as implemented in the program BEAST

Node Node description

Ages (Ma)*

Uniform priors(P, subset VND)†

Lognormal priors(P, subset VND)‡

Uniform priors (subset NVND)

1 Stem lineage of Cupressaceae 245 (209–282) 242 (194–293)2 Crown lineage of Cupressaceae 219 (184–254) 211 (168–259)

(Stem of Cunninghamioideae)3 Stem of Taiwanioideae 204 (170–238) 195 (157–240)4 Stem of Athrotaxoideae 190 (157–224) 182 (145–222)5 Stem of Sequoioideae 183 (150–215) 174 (139–213)6 Stem of Taxodioideae 170 (140–201) 159 (128–194)7 Divergence between

Cupressoideae andCallitroideae

153 (124–183) 143 (114–175)

8 Crown lineage of Callitroideae 128 (98–159) 121 (92–152)9 Stem lineage of Widdringtonia

(endemic in southern Africa)65 (42–92) 62 (40–86)

10 Crown lineage of Cupressoideae 134 (104–164) 123 (93–154)

The nodes are numbered as in Fig. 1 (144-taxon dataset: 144 taxa and7,171 nucleotides).*Million year ranges in parentheses denote the 95% HPD.†BEAST run 1 as described in SI Text and Fig. S3 in which uniform priors wereapplied to calibration P, subset VND (C, D, H, I, M, N, and O) and subsetNVND (A, B, E, F, G, J, K, and L).‡BEAST run 7 as described in SI Text and Fig. S3 in which lognormal priorswere assigned to calibration P and subset VND, whereas uniform priors wereretained for subset NVND; see SI Text for a full explanation of all runs.

Mao et al. PNAS Early Edition | 3 of 6

EVOLU

TION

Page 4: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

with Sequoioideae also found in Australia, but now are reducedto two species each in southern North America and one and twospecies, respectively, in East Asia (Fig. 2 C–E and Fig. S6 B and

C). Athrotaxidoideae were present in both North and SouthAmerica during the Cretaceous (53, 54) but today consist ofthree species in Australia (Fig. 2 C–E). Finally, the genera

N

S

F

E A

U

Cladogenesis vicarianceby

Local extinction

Dispersal resulting in range expansion

Metasequoia glyptostroboides

Cupressoideae

Callitroideae

Taxodium mucronatum

Taxodium distichum

Glyptostrobus pensilis

Cryptomeria japonica

Sequoiasempervirens

Sequoiadendrongiganteum

Athrotaxis selaginoides

Athrotaxis laxifolia

Athrotaxis cupressoides

Taiwania cryptomerioides

Cunninghamia lanceolata

Amentotaxaceae

N

A

E

F

F

S

U

N

U

N

A

A

N

N

A

U

U

A

A

N A

Taxaceae sensu strictoN A E

CephalotaxaceaeA

SciadopityaceaeA

eaecasserpuC

Tax

Seq

Ath

TaiCun

A

A

A

A

N

S

N

N

N

N

S

N

S

AN

A0.61

S

0.35

U

U

0.52

0.59

0.52

0.28

0.34

0.48

0.33

0.62

A

A 0.42

U

U

0.91

N

0.96

A

N

0.51

0.37

AN

A

A

N N

0.99

0.60

0.69

050100150200250 (Ma)

E

A

A

E

Cupressoideae

Callitroideae

Taxodium mucronatum

Taxodium distichum

Taxodium European fossils

Glyptostrobus pensilis

Glyptostrobus European fossils

Taxodium N American fossils

Cryptomeria japonica

Cryptomeria European fossils

Sequoiasempervirens

Sequoiadendron giganteum

Austrosequoia

Sequoia Sequoiadendron& related fossils

Metasequoia glyptostroboides

Metasequoia N Hemisphere fossils

Athrotaxis S Amercian fossils

Athrotaxites berryi

Taiwania cryptomerioides

Taiwania European fossils

Taiwania Asian fossils

N

A

E

F

F

S

U

N

N

E

A

N

E

A

N

N

U

N A E

A

N A E

SS

N

A

E

A

Cunninghamia lanceolata

Cunninghamia NAmerican fossils

Sewardiodendron laxum

Austrohamia minuta

A

N

SS

Athrotaxis selaginoides

Athrotaxis laxifolia

Athrotaxis cupressoidesU

U

U

AmentotaxaceaeN A

Taxaceae sensu strictoN A E

CephalotaxaceaeA

SciadopityaceaeA

Glyptostrobus N American fossilsN

Taiwania N American fossilsN

Cunninghamia related fossil fromAsiaA

Cunninghamia related fossil from Europe

Austrohamia acanthobractea

S

N

N

0.32

S

0.47

0.63

N

0.99

E

N

0.81

E

EN

0.53

N

0.55

N

0.66

E

N

0.17

E 0.68

N

E

A0.86

0.52

N

0.66

N

0.62

N

0.31

A

A

A

A

A

A

N

0.33

0.21

0.30

0.36

0.38

0.25

N A

A0.34

A 0.81

A

0.52A 0.79

A 0.64

N

A0.39A

0.40

A

E0.64

SS

N

0.65SS 0.51

U

U

0.90

N A

0.18 N

N

0.50

SS

N0.87 N 0.92

A 0.37

eae casserpuC

Cun

Tai

Tax

Seq

Ath

050100150200250 (Ma)

050100150200250 (Ma)050100150200250 (Ma)

Cupressoideae

Callitroideae

Taiwanioideae and related fossils

N

A

E

F

F

S

U

Athrotaxites berryi

Sequoiasempervirens

Sequoiadendron giganteum

Austrosequoia

Sequoia Sequoiadendron& related fossils

Metasequoia glyptostroboides

Metasequoia N Hemisphere fossils

Athrotaxis S Amercian fossils

N

N

U

N A E

A

N A E

SS

N

Athrotaxis selaginoides

Athrotaxis laxifolia

Athrotaxis cupressoidesU

U

U

AmentotaxaceaeN A

Taxaceae sensu strictoN A E

CephalotaxaceaeA

SciadopityaceaeA

A

Austrohamia minutaSS

Austrohamia acanthobractea

A Sewardiodendron laxum

E

Cunninghamia lanceolata

Cunninghamia NAmerican fossils

A

N

Cunninghamia related fossil fromAsiaA

Cunninghamia related fossil from Europe

N A E

Taxodioideae and related fossilsN A E

N

Cupressoideae

Callitroideae

Taiwanioideae and related fossils

N

A

E

F

F

S

U

Athrotaxites berryi

Sequoiasempervirens

Sequoiadendron giganteum

Austrosequoia

Sequoia Sequoiadendron& related fossils

Metasequoia glyptostroboides

Metasequoia N Hemisphere fossils

Athrotaxis S Amercian fossils

N

N

U

N A E

A

N A E

SS

N

Athrotaxis selaginoides

Athrotaxis laxifolia

Athrotaxis cupressoidesU

U

U

Taxaceae sensu strictoN A E

CephalotaxaceaeA

SciadopityaceaeA

E

Cunninghamia lanceolata

Cunninghamia NAmerican fossils

A

N

Cunninghamia related fossil fromAsiaA

Cunninghamia related fossil from Europe

N A E

Taxodioideae and related fossilsN A E

A

Austrohamia minutaSS

Austrohamia acanthobractea

A Sewardiodendron laxum

AmentotaxaceaeA

eaecasse rp uC

Cun

Tai

Tax

Seq

Ath

eaecasserpuC

Cun

Tai

Tax

Seq

Ath

N

0.64

N

0.62

N

0.27

A

A

A 0.38

0.41

0.29

N

0.51

S

N

0.63

N

0.32

S

0.47

E

N

0.17

N A

0.18 N

N

0.50

SS

NN 0.92

A 0.37

0.87

SS

N

0.650.51

U

U

0.90

SS

N

A0.38

A

A 0.22

0.30

A

0.32

A

0.32

A

0.64 A 0.53

A

A 0.40

0.56

N

0.63

N

0.51

S

N

0.65

N

0.32

S

0.48

E

N

0.17

N A

0.19 N

N

0.50

SS

NN 0.93

A 0.38

0.88

N

0.61

N

0.42

SS

N

0.670.50

U

U

0.90

SS

N

A0.43

A

A 0.40

0.43

A

A 0.22

0.32

A 0.33

A

0.31

E

N A

0.41

N A

A

0.42N

A

0.41

N A

A0.37

A

N

0.38

A

B

C

DE

Fig. 2. AARs for Cupressaceae. (A) The six areas (“N,” “S,” “E,” “F,” “A,” and “U”) used in the analyses (Left) and the modeled biogeographic processes(Right). (B) Likelihood reconstruction without fossil lineages. (C–E) Likelihood reconstructions that include fossil taxa in the tree and assume alternativeplacements of three early Cupressaceae fossils (see SI Text for details). The AARs with the highest likelihood are shown as colored boxes at each node. Single-area boxes indicate an ancestor confined to a single geographic area; combined boxes indicate an ancestor with a distribution encompassing two or moreareas; two boxes separated by a space indicate the ancestral ranges inherited by each of the daughter lineages arising from the node. For each node withalternative reconstructions (within log2 likelihood units of the maximum), the relative probability of the global likelihood for the optimal reconstruction isgiven. Fossil lineages are shown by a dashed line, indicating their extinct status. Area codes are explained in SI Text.

4 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1114319109 Mao et al.

Page 5: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Austrocedrus (56), Calocedrus (57), Chamaecyparis (58), Fitzroya(59), Papuacedrus (60), and Tetraclinis (61) all had wider dis-tributions in the past. The ice ages of the past 2 million yearsfurther contributed to population extinction and reductions inspecies range, as inferred from our diversification modeling,which indicated a downturn in Cupressaceae diversification ratesat 1.37 Ma. Relatively few Cupressaceae lineages have adaptedto the strongly seasonal and semiarid habitats that became morewidespread with the global cooling during the Oligocene/Miocene(30). Those that did, such as Juniperus, experienced a diversifi-cation burst during the Miocene (19).Besides throwing light onto the diversification of a Triassic/

Jurassic Pangean spermatophyte lineage, our findings confirmand illustrate the power of incorporating fossils directly intoAARs, rather than using them only for molecular-clock calibra-tion (62, 63). Specifically, it was the incorporation in the AARs ofup to 29 fossil taxa (groups) (Fig. S4 and Table S5), most fromareas where the respective lineage no longer occurs, that providedinsights about range changes, but with the direction of range ex-pansion (predominantly north-to-south) being inferred less fromthe fossil record than the molecular topology.

Materials and MethodsPlant Material, DNA Isolation and Sequencing, and Sequence Alignment. TableS1 lists all plant materials used in this study, with species name and author,geographic provenance, herbarium voucher and deposition, and GenBankaccession numbers. A total of 290 sequences were newly produced. Phylo-genetic and dating analyses were conducted on two datasets. The 56-taxondataset comprised 35 ingroup species, 21 outgroups, and 10,472 alignednucleotides from 10 DNA regions (see below). The 144-taxon dataset com-prised 122 ingroup species, 22 outgroups, and 7,171 aligned nucleotides fromsix plastid DNA regions. The ingroup species represented all 32 Cupressaceaegenera; outgroups represented the other conifer families, Cycas, Ginkgo, anda basal angiosperm for rooting purposes. We sequenced the mitochondrialregions atpA and cox1, the nuclear regions 18S and 26S, and the plastidregions matK, rbcL, psbB, petB-D, rps4, and trnL-F (for primer sequences, seerefs. 19 and 64). For DNA extraction, PCR, and sequencing procedures, seeMao et al. (19). The sequences produced were aligned using ClustalX version1.83 (65), followed by manual adjustments in Mega4 (66).

Phylogenetic Analyses. Phylogenetic relationships were reconstructed usingparsimony, Bayesian, and maximum likelihood inference. Parsimony analysesrelied on PAUP version 4.10b (67) and the University of Oslo Bioportal (http://www.bioportal.uio.no) (68) using heuristic searching, starting trees obtained viastepwise addition, tree-bisection-reconnection branch swapping, steepest de-scent, and the MulTrees and Collapse options in effect, with no upper limit forthe number of trees held in memory; support values for all nodes (on a 50%majority rule bootstrap tree) were calculated with the same settings as abovefor 1,000 replicates; 10 searches with random taxon additions were conductedfor each replicate, and the strict consensus tree of all shortest trees were saved.Bayesian analysis relied onMrBayes version 3.1.2 (69) and the GTR+I+Gmodel assuggested byMrModeltest version 2.3 (70).We used the default of one cold andthree heated Markov chain Monte Carlo chains, starting from random initialtrees, and chains were run for 6,000,000 generations, sampling every 200th. Thedefault options in MrBayes were used for chain heating and mixing. We dis-carded a burn-in of the first 2,000,000 generations and used 20,000 trees fromthe posterior distribution to obtain a majority rule consensus tree. Maximumlikelihood analyses relied onGarli version 1.0 (71) with the GTR+I+G substitution

model, starting from random trees and using 5,000,000 generations per search;30 independent searches were performed, and the best tree was saved.

Separate phylogenetic analyses of the nuclear, plastid, and mitochondrialdatasets did not yield statistically supported (>75% likelihood bootstrapsupport) topological contradictions (data available upon request). Thereforewe combined the three data partitions in the 56-taxon dataset.

Molecular-Clock Models and Calibration. A likelihood ratio test in PAUP 4.10b(67, 72) suggested that the 56-taxon and 144-taxon datasets reject a strictmolecular clock (P < 0.01), and we therefore used relaxed molecular-clockapproaches: Bayesian coestimation of branch lengths and topology with un-correlated lognormally distributed rates in BEAST 1.5.3 (43), Bayesian esti-mation with an input phylogeny in MULTIDIVTIME (46), and penalizedlikelihood rate smoothing in R8S (44, 45). In each case, genetic distances weretransformed into absolute time (in million years) by using 16 fossil calibrationpoints, of which 12 were within Cupressaceae (Fig. S2 and Table S4). Fossilswere assigned to the stem of their most closely related lineages; Table S4 liststhe morphological features used for each fossil taxonomic assignment. ForBEAST analyses, we used uniform prior distributions for minimum constraint(calibration points A–O), with the younger bound set by the youngest date ofthe respective fossil and the older bound set to 366.8 Ma (maximum constraintfor calibration point P place near the root) (Table S4). Calibration point P wasrestricted to fall between 306.2 and 366.8 Ma (Table S4). We tested the effectsof other time intervals at calibration point P, calibrations with different sub-sets of fossils, and different distribution prior for calibrations by carrying outeight additional BEAST runs (SI Text and Fig. S3). BEAST analyses were run onthe Cyberinfrastructure for Phylogenetic Research (CIPRES) Science Gateway(http://www.phylo.org/portal2) (73). For MULTIDIVTIME and penalized likeli-hood, constraints were as in BEAST, except that these programs do not allowspecific prior distributions on fossil constraints.

Ancestral Area Reconstructions (AAR). AAR relied on the likelihood dispersal-extinction-cladogenesis approach implemented in LAGRANGE (49). Thematrix of migration probabilities among continents in LAGRANGE (SI Textand Table S6) allowed dispersal between six operational geographic areas: E,Europe, north Africa, and northern Arabia; A, Asia; N, North America, Ca-ribbean, and Central America; S, South America; F, south to middle Africa andsouthern Arabia; and U, Australia, New Guinea, New Caledonia, and NewZealand (see Fig. 2A). Boundaries between A, E, and F were defined to min-imize the number of species that fell in two areas. The northern boundarybetween A and E was defined by the Ural Mountains, which is the conven-tional boundary between European Russia and Asian Russia. The boundarybetween E and F is the Tropic of Cancer, which runs along the middle ofa broad belt of very low precipitation (<100 mm y−1) stretching across all ofNorth Africa and most of Arabia (74); this belt of low precipitation is a sig-nificant biogeographic barrier for Cupressaceae (30).

For details on the selection of fossils, the inference of calibration fossils’phylogenetic position, cross validation of calibration fossils (Fig. S7), Ances-tral Area Reconstructions, and diversification modeling, see SI Text.

ACKNOWLEDGMENTS. We thank Damon Little, Elena Conti, Richard Abbott,the editor, and three anonymous reviewers for their constructive sugges-tions. The University of Oslo Bioportal, CIPRES Science Gateway, and theWilliHennig Society provided computation and software resources. This researchwas supported by Ministry of Science and Technology of China Grant2012CB114504 and 2010DFB63500, National Natural Science Foundation ofChina Grants 30725004 and 31100488, and International Joint Program(“111” project) of China. K.M. was supported by the China Scholarship Coun-cil for 1 y study at the University of Edinburgh. The Royal Botanic GardenEdinburgh is supported by the Scottish Government’s Rural and EnvironmentScience and Analytical Services Division.

1. Scotese CR (2001) Atlas of Earth History (Department of Geology, University of Texas,

Arlington, TX).2. Rogers JJW, Santosh M (2003) Supercontinents in earth history. Gondwana Res 6(3):

357–368.3. Smith AG, Smith DG, Funnell BM (2004) Atlas of Cenozoic and Mesozoic Coastlines

(Cambridge Univ Press, Cambridge, UK).4. Whiteside JH, Grogan DS, Olsen PE, Kent DV (2011) Climatically driven biogeographic

provinces of Late Triassic tropical Pangea. Proc Natl Acad Sci USA 108:8972–8977.5. Springer MS, Murphy WJ, Eizirik E, O’Brien SJ (2003) Placental mammal diversification

and the Cretaceous-Tertiary boundary. Proc Natl Acad Sci USA 100:1056–1061.6. Roelants K, Bossuyt F (2005) Archaeobatrachian paraphyly and Pangaean

diversification of crown-group frogs. Syst Biol 54:111–126.7. San Mauro D, Vences M, Alcobendas M, Zardoya R, Meyer A (2005) Initial diversification

of living amphibians predated the breakup of Pangaea. Am Nat 165:590–599.

8. Wildman DE, et al. (2007) Genomics, biogeography, and the diversification of

placental mammals. Proc Natl Acad Sci USA 104:14395–14400.9. Gamble T, Bauer AM, Greenbaum E, Jackman TR (2008) Evidence for Gondwanan

vicariance in an ancient clade of gecko lizards. J Biogeogr 35(1):88–104.10. Roelants K, Haas A, Bossuyt F (2011) Anuran radiations and the evolution of tadpole

morphospace. Proc Natl Acad Sci USA 108:8731–8736.11. Crane PR (1987) Vegetational consequences of the angiosperm diversification. The

Origin of Angiosperms and Their Biological Consequences, eds Friis EM, Chaloner WC,

Crane PR (Cambridge Univ Press, Cambridge, UK), pp 107–144.12. Lupia R, Lidgard S, Crane PR (1999) Comparing palynological abundance and di-

versity: Implications for biotic replacement during the Cretaceous angiosperm radi-

ation. Paleobiology 25(3):305–340.13. Wagstaff SJ (2004) Evolution and biogeography of the austral genus Phyllocladus

(Podocarpaceae). J Biogeogr 31:1569–1577.

Mao et al. PNAS Early Edition | 5 of 6

EVOLU

TION

Page 6: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

14. Won H, Renner SS (2006) Dating dispersal and radiation in the gymnosperm Gnetum(Gnetales)—clock calibration when outgroup relationships are uncertain. Syst Biol 55:610–622.

15. Qiao CY, Ran JH, Li Y, Wang XQ (2007) Phylogeny and biogeography of Cedrus(Pinaceae) inferred from sequences of seven paternal chloroplast and maternalmitochondrial DNA regions. Ann Bot (Lond) 100:573–580.

16. Knapp M, Mudaliar R, Havell D, Wagstaff SJ, Lockhart PJ (2007) The drowning of NewZealand and the problem of Agathis. Syst Biol 56:862–870.

17. Biffin E, Hill RS, Lowe AJ (2010) Did Kauri (Agathis: Araucariaceae) really survive theOligocene drowning of New Zealand? Syst Biol 59:594–602.

18. Ickert-Bond SM, Rydin C, Renner SS (2009) A fossil-calibrated relaxed clock forEphedra indicates an Oligocene age for the divergence of Asian and NewWorld clades and Miocene dispersal into South America. J Syst Evol 47(5):444–456.

19. Mao K, Hao G, Liu J, Adams RP, Milne RI (2010) Diversification and biogeography ofJuniperus (Cupressaceae): Variable diversification rates and multiple intercontinentaldispersals. New Phytol 188:254–272.

20. Wei XX, Yang ZY, Li Y, Wang XQ (2010) Molecular phylogeny and biogeography ofPseudotsuga (Pinaceae): Insights into the floristic relationship between Taiwan and itsadjacent areas. Mol Phylogenet Evol 55:776–785.

21. Biffin E, Brodribb TJ, Hill RS, Thomas P, Lowe AJ (2012) Leaf evolution in SouthernHemisphere conifers tracks the angiosperm ecological radiation. Proc Biol Sci 279:341–348.

22. Willyard A, Syring J, Gernandt DS, Liston A, Cronn R (2007) Fossil calibration ofmolecular divergence infers a moderate mutation rate and recent radiations forpinus. Mol Biol Evol 24:90–101.

23. Gernandt DS, et al. (2008) Use of simultaneous analyses to guide fossil-basedcalibrations of pinaceae phylogeny. Int J Plant Sci 169:1086–1099.

24. Treutlein J, Wink M (2002) Molecular phylogeny of cycads inferred from rbcLsequences. Naturwissenschaften 89(5):221–225.

25. Crisp MD, Cook LG (2011) Cenozoic extinctions account for the low diversity of extantgymnosperms compared with angiosperms. New Phytol 192:997–1009.

26. Nagalingum NS, et al. (2011) Recent synchronous radiation of a living fossil. Science334:796–799.

27. Eckenwalder JE (1976) Re-evaluation of Cupressaceae and Taxodiaceae: A proposedmerger. Madrono 23(5):237–256.

28. Brunsfeld SJ, et al. (1994) Phylogenetic relationships among the genera of Tax-odiaceae and Cupressaceae: Evidence from rbcL sequences. Syst Bot 19(2):253–262.

29. Gadek PA, Alpers DL, HeslewoodMM, Quinn CJ (2000) Relationships within Cupressaceaesensu lato: A combined morphological and molecular approach. Am J Bot 87:1044–1057.

30. Farjon A (2005) A Monograph of Cupressaceae and Sciadopitys (Royal Botanic Gar-dens Kew, London), pp 228–400.

31. Christenhusz MJM, et al. (2011) A new classification and linear sequence of extantgymnosperms. Phytotaxa 19:55–70.

32. Yao XL, Zhou ZY, Zhang BL (1998) Reconstruction of the Jurassic coniferSewardiodendron laxum (Taxodiaceae). Am J Bot 85:1289–1300.

33. Stockey RA, Kva�cek J, Hill RS, Rothwell GW, Kva�cek Z (2005) Fossil Record ofCupressaceae s. lat. A Monograph of Cupressaceae and Sciadopitys, ed Farjon A(Royal Botanic Gardens Kew, London), pp 54–68.

34. Escapa I, Cúneo R, Axsmith B (2008) A new genus of the Cupressaceae (sensu lato)from the Jurassic of Patagonia: Implications for conifer megasporangiate conehomologies. Rev Palaeobot Palynol 151:110–122.

35. Rothwell GW, Stockey RA, Mapes G, Hilton J (2011) Structure and relationships of theJurassic conifer seed cone Hughmillerites juddii gen. et comb. nov.: Implications forthe origin and evolution of Cupressaceae. Rev Palaeobot Palynol 164:45–59.

36. Zhang JW, D’Rozario A, Wang LJ, Li Y, Yao JX (2012) A new species of the extinctgenus Austrohamia (Cupressaceae s.l) in the Daohugou Jurassic flora of China and itsphytogeographical implications. J Syst Evol 50(1):72–82.

37. Doyle JA, Donoghue MJ (1993) Phylogenies and angiosperm diversification. Paleo-biology 19(2):141–167.

38. Ho SY, Phillips MJ (2009) Accounting for calibration uncertainty in phylogeneticestimation of evolutionary divergence times. Syst Biol 58:367–380.

39. Clarke JT, Warnock RC, Donoghue PC (2011) Establishing a time-scale for plantevolution. New Phytol 192:266–301.

40. Cusimano N, Stadler T, Renner SS (2012) A new method for handling missing speciesin diversification analysis applicable to randomly or non-randomly sampled phylog-enies. Syst Biol, 10.1093/sysbio/sys031.

41. Pybus OG, Harvey PH (2000) Testing macro-evolutionary models using incompletemolecular phylogenies. Proc Biol Sci 267:2267–2272.

42. Cusimano N, Renner SS (2010) Slowdowns in diversification rates from real phylog-enies may not be real. Syst Biol 59:458–464.

43. Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by samplingtrees. BMC Evol Biol 7:214.

44. Sanderson MJ (2002) Estimating absolute rates of molecular evolution and divergencetimes: A penalized likelihood approach. Mol Biol Evol 19(1):101–109.

45. Sanderson MJ (2003) R8S: Analysis of rates (“r8s”) of evolution (and other stuff),version1.60. Available at: http://ginger.ucdavis.edu/r8s.

46. Thorne JL, Kishino H (2002) Divergence time and evolutionary rate estimation withmultilocus data. Syst Biol 51:689–702.

47. Linder HP, Hardy CR, Rutschmann F (2005) Taxon sampling effects in molecular clockdating: An example from the African Restionaceae. Mol Phylogenet Evol 35:569–582.

48. Ogg JG (2010) International Stratigraphic Chart (International Commission on Stra-tigraphy, West Lafayette, IN). Available at: https://www.seegrid.csiro.au/wiki/pub/CGIModel/GeologicTime/StratChart2010.jpg.

49. Ree RH, Smith SA (2008) Maximum likelihood inference of geographic range evolu-tion by dispersal, local extinction, and cladogenesis. Syst Biol 57:4–14.

50. Cheng YC, Nicolson RG, Tripp K, Chaw SM (2000) Phylogeny of Taxaceae and Ceph-alotaxaceae genera inferred from chloroplast matK gene and nuclear rDNA ITSregion. Mol Phylogenet Evol 14:353–365.

51. Rai HS, Reeves PA, Peakall R, Olmstead RG, Graham SW (2008) Inference of higher-order conifer relationships from a multi-locus plastid data set. Botany 86:658–669.

52. Peters MD, Christophel DC (1978) Austrosequoia wintonensis, a new taxodiaceouscone from Queensland, Australia. Can J Bot 56:3119–3128.

53. Miller CN, Jr., LaPasha CA (1983) Structure and affinities of Athrotaxites berryi Bell, anEarly Cretaceous conifer. Am J Bot 70:772–779.

54. Del Fueyo GM, Archangelsky S, LlorensM, Cuneo R (2008) Coniferous ovulate cones fromthe Lower Cretaceous of Santa Cruz Province, Argentina. Int J Plant Sci 169:799–813.

55. Lepage BA (2009) Earliest occurrence of Taiwania (Cupressaceae) from the EarlyCretaceous of Alaska: Evolution, biogeography, and paleoecology. Proc Acad Nat SciPhiladelphia 158(1):129–158.

56. Paull R, Hill RS (2008) Oligocene Austrocedrus from Tasmania (Australia): Comparisonswith Austrocedrus chilensis. Int J Plant Sci 169(2):315–330.

57. Shi G, Zhou Z, Xie Z (2012) A new Oligocene Calocedrus from South China and itsimplications for transpacific floristic exchanges. Am J Bot 99:108–120.

58. Liu YS, Mohr BAR, Basinger JF (2009) Historical biogeography of the genusChamaecyparis (Cupressaceae, Coniferales) based on its fossil record. PalaeobioPalaeoenv 89:203–209.

59. Paull R, Hill RS (2010) Early Oligocene Callitris and Fitzroya (Cupressaceae) fromTasmania. Am J Bot 97:809–820.

60. Wilf P, et al. (2009) Papuacedrus (Cupressaceae) in Eocene Patagonia: A new fossil linkto Australasian rainforests. Am J Bot 96:2031–2047.

61. Kva�cek Z, Manchester SR, Schorn HE (2000) Cones, seeds, and foliage of Tetraclinis Sal-icornioides (Cupressaceae) from the Oligocene and Miocene of Western North America:A geographic extension of the European Tertiary species. Int J Plant Sci 161:331–344.

62. Clayton JW, Soltis PS, Soltis DE (2009) Recent long-distance dispersal overshadowsancient biogeographical patterns in a pantropical angiosperm family (Simar-oubaceae, Sapindales). Syst Biol 58:395–410.

63. Crisp MD, Trewick SA, Cook LG (2011) Hypothesis testing in biogeography. TrendsEcol Evol 26:66–72.

64. Liu N, et al. (2009) Phylogenetic relationships and divergence times of the familyAraucariaceae based on the DNA sequences of eight genes. Chin Sci Bull 54:2648–2655.

65. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The CLUS-TAL_X windows interface: Flexible strategies for multiple sequence alignment aidedby quality analysis tools. Nucleic Acids Res 25:4876–4882.

66. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary GeneticsAnalysis (MEGA) software version 4.0. Mol Biol Evol 24:1596–1599.

67. Swofford DL (2002) PAUP*: Phylogenetic Analyses Using Parsimony (*and OtherMethods), Version 4 (Sinauer Associates, Sunderland, MA).

68. Kumar S, et al. (2009) AIR: A batch-oriented web program package for construction ofsupermatrices ready for phylogenomic analyses. BMC Bioinformatics 10:357.

69. Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetictrees. Bioinformatics 17:754–755.

70. Nylander JA (2004) MrModeltest v2 (Program distributed by the author, EvolutionaryBiology Centre, Uppsala University, Uppsala, Sweden). Available at: http://www.abc.se/znylander/mrmodeltest2/mrmodeltest2.html.

71. Zwickl DJ (2006) Genetic algorithm approaches for the phylogenetic analysis of largebiological sequence datasets under the maximum likelihood criterion. Ph.D. disser-tation (University of Texas, Austin, TX).

72. Felsenstein J (1981) Evolutionary trees from DNA sequences: A maximum likelihoodapproach. J Mol Evol 17:368–376.

73. Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science Gateway forinference of large phylogenetic trees. Proceedings of the Gateway Computing Envi-ronments Workshop (GCE) New Orleans [Institute of Electrical and ElectronicsEngineers (IEEE), New York], pp 1–8.

74. Geelan P, Lewis H (1992) Times Atlas of the World (Comprehensive Edition) (TimesBooks, London), 9th Ed.

6 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1114319109 Mao et al.

Page 7: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Supporting InformationMao et al. 10.1073/pnas.1114319109SI TextSelection of Fossil Taxa and Their Phylogenetic Positions. The in-tegration of fossil calibrations is themost critical step inmoleculardating (1, 2). We only used the fossil taxa with ovulate cones thatcould be assigned unambiguously to the extant groups (Table S4).The exact phylogenetic position of fossils used to calibrate themolecular clocks was determined using the total-evidence analy-ses (following refs. 3−5). Cordaixylon iowensis was not included inthe analyses because its assignment to the crown Acrogymno-spermae already is supported by previous cladistic analyses (alsousing the total-evidence approach) (6). Two data matrices werecompiled. Matrix A comprised Ginkgo biloba, 12 living repre-sentatives from each conifer family, and three fossils taxa relatedto Pinaceae and Araucariaceae (16 taxa in total; Fig. S5A). In thismatrix, the 105 morphological characters and their states followGernandt et al. (ref. 7 and references therein). Matrix B com-prised Pinus sylvestris, Sciadopitys verticillata, 39 living taxa, and 17fossil taxa that are closely related to Cupressaceae (58 taxa intotal). In this matrix, the 53 morphological characters and theirstates follow Farjon (8) and two updates (9-10). Both data ma-trices include 5,476 molecular characters for all living taxa ex-tracted from the plastid DNA sequence matrix (144-taxondataset), with the fossil taxa coded as having “missing data.”Molecular and morphological characters were concatenated inboth matrices, and parsimony analyses were performed using theprogram TNT 1.1 (11), with heuristic searching based on 5,000random addition sequences and Tree-Bisection-Reconnectionswapping (saving 10 trees per replication). All shortest trees weresaved and summarized into a strict consensus tree. All characterswere weighted equally, and the resulting phylogenetic trees wererooted on Ginkgo biloba (matrix A) (Fig. S5A) or on Pinus syl-vestris (matrix B) (Fig. S5 B–I).As shown in Fig. S5A, Matrix A yielded a tree in which the

phylogenetic positions of the fossil taxa were well resolved.Matrix B, however, yielded a tree in which phylogenetic positionsof more than half of the fossil taxa were unresolved. We then rana series of analyses, each of which included a different subset offossils, and determined that seven fossils (Hughmillerites juddii,Athrotaxis ungeri, Austrosequoia wintonensis, Glyptostrobus sp.,Papuacedrus prechilensis, Thuja polaris, and Fokienia raven-scragensis) are responsible for the collapsing of the tree, likelybecause that fewer morphological characters are available forthem than for the others included in Matrix B. We thereforedetermined the position of the other 10 fossils by reanalyzingMatrix B with only these 10 fossils included (Fig. S5I). We fur-ther studied the placement of each of the seven problematicfossils by analyzing a version of Matrix B containing only onetarget fossil alongside all the living taxa (Fig. S5 B–H). All datamatrices and resulting trees have been submitted to TreeBASE(study accession no. S12554).Cross-validation tests of the different fossil calibrations (Fig. S7

and ref. 12) were performed with the program R8S under Pe-nalized Likelihood rate smoothing using the 56-taxon dataset. Amaximum constraint was used only for calibration P while theremaining 16 calibrations used the minimum constraints (Fig. S7and Table S4). Only the calibration Wa (Widdringtonia ameri-cana) (13) resulted in the node ages significantly older than thoseof the other fossil calibrations (Fig. S7). This fossil appears to bewrongly placed in the living genus Widdringtonia, as suggestedalso by Crisp and Cook (14). We therefore excluded the cali-bration Wa from further analyses.

BEASTAnalyses. In addition to a BEAST analysis that used uniformprior distributions for all calibrations (run 1; 144-taxon dataset,calibrations as in Table S4), we performed eight additionalanalyses to explore factors affecting estimates of divergencetime (Fig. S3).First, to test the effect of calibration point P, which is close to

the root node and is the only functional hard maximum constraintin BEAST runs using uniform priors, we carried out three runswith calibrations A through O (Table S4), and calibration P set to[306.2, 351.7] (run 2), [306.2, 336.5] (run 3), and [306.2, 321.4](run 4). The age estimates obtained in runs 2, 3, and 4 largelyoverlapped with those from run 1 (Fig. S3).Second, we carried out two runs with different subsets of

calibrations using uniform priors. When parsing the log.txt file ofrun 1 with Tracer 1.4 (15), we noted that the posterior distributionof nodes calibrated with the minimum constraints (calibrations Athrough O) fell into two groups. One included calibrations C, D,H, I, M, N, and O. For each node calibrated with these fossils(Table S4), the posterior distributions of their age estimatessignificantly violated a normal distribution (hence our name forthis subset: “VND”). The other group included calibrations A, B,E, F, G, J, K, and L. For each node calibrated with these fossils(Table S4), the posterior distributions of their age estimates didnot significantly violate a normal distribution (hence, subset“NVND”). When comparing between-lineage age estimates de-rived from BEAST runs based on calibration subset VND (pluscalibration P) (run 5) and subset NVND (plus calibration P) (run6), we found that the calibration subset NVND (run 6) signifi-cantly underestimated lineage ages when compared with thecalibration subset VND (run5) or the calibrations A through P(run 1) (Fig. S3).Third, we carried out another BEAST run that incorporated

lognormal priors. Calibrations that may underestimate lineageage should not be given lognormal priors, because a lognormalprior places a rapidly declining probability on older ages (16).Nevertheless, fossils that underestimate lineage ages still may beuseful as hard minimum constraints, as suggested by previousstudies (12, 17, 18). For this reason, uniform priors were retainedfor the subset NVND, whereas lognormal priors were applied forthe calibration P and the subset VND (run 7) (see Table S4). Theage estimates obtained in run 7 largely overlapped with thosefrom runs 1 and 5 (Fig. S3). However, one run (with lognormalpriors; run 7) bias to estimate the node ages younger and anotherrun (with uniform priors; run 1) bias to estimate the node agesolder, so “the truth is likely to be somewhere in between” (19).Fourth, we also carried out a run that included Widdringtonia

americana (calibration Wa) as a hard minimum calibration inaddition to calibrations A through P (run 8). Integration of thiscalibration resulted in significantly older age estimates for nodes1–10 (Fig. S3), demonstrating that the calibration Wa has a po-tential to overestimate lineage ages. As suggested by the cross-validation test and a recent study (14), it is better to exclude thecalibration Wa.Finally, we carried out a BEAST run on the 56-taxon matrix

that assumed uniform prior for all calibrations (run 9) (Table S4).The ages obtained are slightly younger than these obtained with144-taxon matrix (Fig. S3). This result is consistent with the ef-fects of undersampling observed elsewhere (20).

Ancestral Area Reconstruction. Ancestral area reconstruction(AAR) under the dispersal-extinction-cladogenesis model, asimplemented in LAGRANGE (21), requires a matrix that defines

Mao et al. www.pnas.org/cgi/content/short/1114319109 1 of 10

Page 8: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

migration probabilities among the six operational geographicareas (see Fig. 2A for their boundaries). Migration probabilitiesamong areas were based on geological history, climate history,and the presence and dissolution of land bridges and islandchains. Geological events considered were the presence of a landconnection between all operational area before the break-up ofPangea, the separation between Laurasia and Gondwana, thegradual fragmentation of Gondwana and Laurasia, island chainsbetween North and South America, the collision of Australianand Asian plates, and the collision of African and Asian plates(22-26). Migration probabilities range from 0.1 for well-sepa-rated areas to 1.0 for contiguous landmasses (Table S6), and theLAGRANGE online configuration tool (21) was used to devisea matrix that reflected different migration probabilities betweenareas at different periods (i.e., time slices) in the past. Based onplate tectonics, we defined eight area pairs and one area com-bination as ancestral area candidates in AAR analyses: NS, NE,SF, NA, AE, FE, SU, NF, and NAE (in which “E” stands forEurope, north Africa, and northern Arabia; “A” represents Asia;“N” represents North America, Caribbean, and Central Amer-ica; “S” represents South America; “F” represents south tomiddle Africa and southern Arabia; and “U” represents Aus-tralia, New Guinea, New Caledonia, and New Zealand). Becausethe connectivity between our six operational geographic unitschanged during the past 275 million years, we decided to developa time-slice model that would reflect changing continental con-nectivity. To find the best-fitting time-slice model, we comparedmodels with five time slices (275–160 Ma, 160–125 Ma, 125–70Ma, 70–30 Ma, and 30 Ma to the present), six time slices (275–160 Ma, 160–125 Ma, 125–105 Ma, 105–70 Ma, 70–30 Ma, and30 Ma to the present), seven time slices (275–160 Ma, 160–125Ma, 125–105 Ma, 105–70 Ma, 70–45 Ma, 45–30 Ma, and 30 Mato the present), and eight time slices (275–160 Ma, 160–125 Ma,125–105 Ma, 105–70 Ma, 70–45 Ma, 45–30 Ma, 30–5 Ma, and 5Ma to the present). For each of these four time-slice schemes, wecompiled a separate migration probability matrix and calculatedits global maximum likelihood in LAGRANGE. Comparison ofthe resulting global likelihoods suggested that the eight-time-slicematrix (Table S6) fit our data best, and we therefore adopted thismigration probability matrix for all subsequent AAR analyses.We performed additional AAR analyses on trees that com-

prised 29 fossils or groups of fossils representing extinct taxa (Fig.2 C–E, Fig. S4, and Table S5) (26, 27), the 122 sequenced livingtaxa of Cupressaceae and four outgroups. For these analyses,fossils were placed as extinct sister lineages to those living line-ages with which they showed the closest morphological affinitiesas assessed in their original publications and related updates (fordetailed placement justifications, see Table S5). The divergencebetween extinct lineages and their sister lineages (living taxa) wasdetermined based on the earliest fossil of each extinct lineage(always determined as the youngest possible age of the formationor stratum in which a fossil occurred). For the absolute age ofeach geological stratum, we relied on the latest geologic timescale (28).As shown in Fig. 2 and Fig. S4, our AARs inferred four types of

events affecting geographic ranges of either living or extinctlineages. First, instances of dispersal result in range expansion(indicated by black arrows on a lineage); such events are commonthroughout the Cupressaceae tree. Second, local extinction events(indicated by a red “X” on a lineage; 11 are hypothesized in Fig.S4) are inferred when a daughter lineage (i) inherits a rangedifferent from that of its parent (a range expansion before local

extinction is inferred; 10 such events are hypothesized in Fig. S4)or (ii) inherits a reduced range relative to its parent (one in-stance related to Metasequoia is hypothesized in Fig. 2 C–E andin Fig. S4). Third, cladogenesis events caused by vicariance (in-dicated by blue arrows on a lineage) are inferred where the an-cestral range encompassing two or more areas subdivides betweendaughter lineages. Fourth, a combination of range expansion andsubsequent cladogenesis caused by vicariance is inferred whenone daughter lineage inherits the range of its parent, and theother inherits a different range (e.g., the separation betweenAthrotaxites berryi and Athrotaxis in Fig. 2 C–E). Note that each ofthe fossil taxa shown by dashed lines in Fig. 2 and Fig. S4 expe-rienced one extra total extinction event compared with living taxashown by solid lines.Our total-evidence analyses fully resolved the phylogenetic

position of Athrotaxites berryi and partly resolved the phyloge-netic positions of Austrohamia minuta, Austrohamia acantho-bractea, and Sewardiodendron laxum (Fig. S5I) but failed forHughmillerites juddii (Fig. S5B). We therefore excluded Hugh-millerites juddii from AAR analyses (its inclusion resulted in thecollapse of the Cupressaceae phylogenetic tree: Fig. S5B). Toinclude Austrohamia minuta, Austrohamia acanthobractea, andSewardiodendron laxum in the AARs, we assigned them to threepossible placement scenarios and performed likelihood AARsfor each scenario. As shown in Fig. 2 C–E, A. minuta and A.acanthobractea always were assumed to be sister to each other.Scenario 1 (Fig. 2C) assumes that Sewardiodendron divergedfrom Cunninghamioideae (both living and extinct members)(Fig. 2 and Fig. S4) at 157.2 Ma, and both of them share a mostrecent common ancestor (MRCA) with Austrohamia at 164.2 Ma.Scenario 2 (Fig. 2D) assumed that Sewardiodendron and Austro-hamia shared a MRCA at 157.2 Ma and diverged from Cun-ninghamioideae 164.2 Ma. Scenario 3 (Fig. 2E) assumed thatSewardiodendron and Austrohamia shared a MRCA at 164.2 Maand diverged from the MRCA of all Cupressaceae subfamiliesexcept Cunninghamioideae at 211.5 Ma (the intermediate agebetween nodes 2 and 3 in Table 1).

Diversification Modeling. Using the TreeSim R package (29) andthe BEAST highest posterior probability chronogram obtainedfor the Cupressaceae (from the 144-taxon dataset) as input, wesimulated 1,000 trees with the number of tips corresponding to thetotal number of extant species of Cupressaceae (162 species) andspeciation and extinction rates obtained by fitting the constantrate birth-death model to the chronogram. To add the 40 extantspecies that were not sequenced, we used the sim.missing functionin the CorSiM R package (30) and simulated 1,000 trees undera constant-rate birth-death model, assuming that the missingspeciation events are not distributed randomly over the tree butprobably happened during the past 10 million years. We thenapplied birth/death likelihood (BDL) analysis (using TreePar) tothe 1,000 completed trees to obtain means and SDs for the γstatistic, the Akaike information criterion values, and the in-ferred rate parameters from the BDL analyses. TreePar alsocalculates the percentage of trees to which a particular model fitsbest. Among the five models under comparison (CR-PB, con-stant-rate birth-death, logistic density dependence, exponentialdensity dependence, and a two-rate variant of the pure-birthmodel with a rate shift at a certain time point), the two-ratemodel provided the best fit for all 1,000 trees. The CR-PBprovided the second-best fit.

1. Clarke JT, Warnock RC, Donoghue PC (2011) Establishing a time-scale for plantevolution. New Phytol 192:266–301.

2. Hedges SB, Kumar S (2004) Precision of molecular time estimates. Trends Genet 20:242–247.

3. Sun G, et al. (2002) Archaefructaceae, a new basal angiosperm family. Science 296:899–904.

4. Peterson KJ, Butterfield NJ (2005) Origin of the Eumetazoa: Testing ecologicalpredictions of molecular clocks against the Proterozoic fossil record. Proc Natl AcadSci USA 102:9547–9552.

5. Manos PS, et al. (2007) Phylogeny of extant and fossil Juglandaceae inferred from theintegration of molecular and morphological data sets. Syst Biol 56:412–430.

Mao et al. www.pnas.org/cgi/content/short/1114319109 2 of 10

Page 9: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

6. Magallon S (2010) Using fossils to break long branches in molecular dating: A comparisonof relaxed clocks applied to the origin of angiosperms. Syst Biol 59:384–399.

7. Gernandt DS, et al. (2008) Use of simultaneous analyses to guide fossil-basedcalibrations of pinaceae phylogeny. Int J Plant Sci 169:1086–1099.

8. Farjon A (2005) A Monograph of Cupressaceae and Sciadopitys (Royal BotanicGardens, Kew, UK), pp 47–53, 601−602.

9. Escapa I, Cúneo R, Axsmith B (2008) A new genus of the Cupressaceae (sensu lato)from the Jurassic of Patagonia: Implications for conifer megasporangiate conehomologies. Rev Palaeobot Palyno 151:110–122.

10. Rothwell GW, Stockey RA, Mapes G, Hilton J (2011) Structure and relationships of theJurassic conifer seed cone Hughmillerites juddii gen. et comb. nov.: Implications forthe origin and evolution of Cupressaceae. Rev Palaeobot Palyno 164:45–59.

11. Goloboff PA, Farris JS, Nixon KC (2008) TNT, a free program for phylogenetic analysis.Cladistics 24(5):774–786.

12. Near TJ, Meylan PA, Shaffer HB (2005) Assessing concordance of fossil calibration pointsin molecular clock studies: An example using turtles. American Naturalist 165:137–146.

13. McIver EE (2001) CretaceousWiddringtonia Endl. (Cupressaceae) from North America.Int J Plant Sci 162:937–961.

14. Crisp MD, Cook LG (2011) Cenozoic extinctions account for the low diversity of extantgymnosperms compared with angiosperms. New Phytol 192:997–1009.

15. Rambaut A, Drummond AJ (2007) Tracer v1.4. Available at http://beast.bio.ed.ac.uk/Tracer.

16. Ho SY, Phillips MJ (2009) Accounting for calibration uncertainty in phylogeneticestimation of evolutionary divergence times. Syst Biol 58:367–380.

17. Near TJ, Sanderson MJ (2004) Assessing the quality of molecular divergence timeestimates by fossil calibrations and fossil-based model selection. Philos Trans R SocLond B Biol Sci 359:1477–1483.

18. Near TJ, Bolnick DI, Wainwright PC (2005) Fossil calibrations and molecular divergencetime estimates in centrarchid fishes (Teleostei: Centrarchidae). Evolution 59:1768–1782.

19. Davis CC, Schaefer H (2011) Plant evolution: Pulses of extinction and speciation ingymnosperm diversity. Curr Biol 21(24):R995–R998.

20. Linder HP, Hardy CR, Rutschmann F (2005) Taxon sampling effects in molecular clockdating: An example from the African Restionaceae. Mol Phylogenet Evol 35:569–582.

21. Ree RH, Smith SA (2008) Maximum likelihood inference of geographic rangeevolution by dispersal, local extinction, and cladogenesis. Syst Biol 57:4–14.

22. Tiffney BH (1985) The Eocene North-Atlantic Land-Bridge: Its importance in tertiaryand modern phytogeography of the Northern Hemisphere.2. J Arnold Arboretum 66:243–273.

23. Scotese CR (2001) Atlas of Earth History (Department of Geology, University of Texas,Arlington, TX).

24. Morley RJ (2003) Interplate dispersal paths for megathermal angiosperms. PerspectPlant Ecol Evol Syst 6:5–20.

25. Smith AG, Smith DG, Funnell BM (2004) Atlas of Cenozoic and Mesozoic Coastlines(Cambridge University Press, Cambridge, UK).

26. Clayton JW, Soltis PS, Soltis DE (2009) Recent long-distance dispersal overshadowsancient biogeographical patterns in a pantropical angiosperm family (Simaroubaceae,Sapindales). Syst Biol 58:395–410.

27. Crisp MD, Trewick SA, Cook LG (2011) Hypothesis testing in biogeography. TrendsEcol Evol 26:66–72.

28. OggJG(2010) InternationalStratigraphicChart (InternationalCommissiononStratigraphy,West Lafayette, IN) Available at https://www.seegrid.csiro.au/wiki/pub/CGIModel/GeologicTime/StratChart2010.jpg.

29. Stadler T (2011) Simulating trees on a fixed number of extant species. Syst Biol 60:676–684.

30. Cusimano N, Stadler T, Renner SS (2012) A new method for handling missing speciesin diversification analysis applicable to randomly or non-randomly sampledphylogenies. Syst Biol, 10.1093/sysbio/sys031.

Mao et al. www.pnas.org/cgi/content/short/1114319109 3 of 10

Page 10: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Fig. S1. Maximum likelihood trees based on (A) the 56-taxon dataset (nuclear ribosomal DNA regions: 18S, 26S; mitochondrial DNA regions: coxI, atpA; plastidDNA regions: rbcL, matK, psbB, petB-D, rps4, and trnL-F) and (B) the 144-taxon dataset (plastid DNA regions only: rbcL, matK, psbB, petB-D, rps4, and trnL-F).Strongly supported nodes are marked by asterisks. A black asterisk indicates parsimony bootstrap support (PBS) ≥85% and Bayesian posterior probability (BPP)≥0.98; a blue asterisk indicates PBS ≥85% but BPP <0.98; an orange asterisk indicates BPP ≥0.98 but PBS <85%.

Mao et al. www.pnas.org/cgi/content/short/1114319109 4 of 10

Page 11: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

050100150200250300350

Dacrycarpus

Ginkgo

Cedrus

Araucaria

Sciadopitys verticillata

Pinus

Picea

Podocarpus

Cathaya

Cycas

Larix

Keteleeria

Abies

Agathis

Phyllocladus

Wollemia

Acmopyle

eaediosserpuC

eaediortillaC

eaecasserp uC

spuorg tuO

Tax

Seq

AthTai

Cun

surepinuJ

Tax

SeqAth

TaiCun

e aecass erpuC

(b)1

3

5

6

10

7

41/P

2/A

4/B

22

24

28

32

38/N

34

3735

13/D

11/C

14

12

8/E15

16

18

20/G

919/F

17

21/H

23/I

25/J26/K 27

30

31

29/L

(Ma)

Fokienia hodginsii

Tetraclinis articulata

Taiwania cryptomerioides

Callitris drummondii

Widdringtonia schwarzii

Callitris macleayana

Widdringtonia whytei

Widdringtonia cedarbergensis

Nageia

Callitris endlicheri

Sequoia sempervirens

Platycladus orientalis

Widdringtonia nodiflora

Athrotaxis laxifolia

Fitzroya cupressoides

Athrotaxis selaginoides

Thuja koraiensis

Chamaecyparis lawsoniana

Thujopsis dolabrata

Papuacedrus papuana

Actinostrobus acuminatus

Libocedrus plumosa

Metasequoia glyptostroboides

Chamaecyparis pisifera

Sequoiadendron giganteum

Actinostrobus pyramidalis

Diselma archeri

Athrotaxis cupressoides

Taxus wallichiana

Callitris preissii

Calocedrus decurrens

Thuja plicata

Amentotaxus argotaenia

Cryptomeria japonica

Taxodium distichum

Callitris rhomboidea

Austrocedrus chilensis

Cephalotaxus sinensis

Glyptostrobus pensilis

Chamaecyparis formosensis

Thuja standishii

Calocedrus macrolepis

Neocallitropsis pancheri

Callitris canescens

Callitris verrucosa

Libocedrus bidwillii

Callitris sulcata

Callitris muelleri

Chamaecyparis obtusa

Taxodium mucronatum

Cunninghamia lanceolata

Pilgerodendron uviferum

Chamaecyparis thyoides

Microbiota decussata

Thuja occidentalis

Cupressus sensu stricto

Sect. Juniperus

Sect. Caryocedrus

Sect. Sabina

Hesperocyparis

Xanthocyparis vietnamensis

Callitropsis nootkatensis

39/O

40

33

36/M

0 (Ma)100200300400

Phyllocladus

Wollemia

Acmopyle

Juniperus formosana

Juniperus drupacea

Platycladus

Tetraclinis

Austrocedrus

Ginkgo

Callitris muelleri

Nageia

Cephalotaxus

Papuacedrus

Glyptostrobus

Diselma

Sequoiadendron

Araucaria

Thujopsis

Athrotaxis

Taxodium

Cedrus

Abies

Keteleeria

Actinostrobus

Thuja

Widdringtonia

Cathaya

Metasequoia

Agathis

Neocallitropsis

Cycas

Xanthocyparis

Chamaecyparis

Fokienia

Pinus

Libocedrus

Cryptomeria

Calocedrus

Dacrycarpus

Podocarpus

Juniperus virginiana

Amentotaxus

Taxus

Callitropsis

Callitris preissii

Cunninghamia

Microbiota

Cupressus

Sequoia

Fitzroya

Picea

Hesperocyparis

Taiwania

Sciadopitys

Pilgerodendron

( )a

41/P

34

38/N

37

35

321

3

5

2/A

4/B

13/D

11/C

14

12

6

7

8/E15

16

18

20/G

919/F

17

24

2830

31

29/L

10

22

21/H

23/I

25/J26/K

27

40 36/M

39/O

Fig. S2. Divergence times for the Cupressaceae based on (A) the 56-taxon dataset and (B) the 144-taxon dataset, with fossil calibration points and key nodesindicated. Light blue bars represent 95% highest posterior density intervals for age estimates. The red line indicates the Cretaceous/Tertiary boundary. Lightorange shading represents the breaking up of Pangea into Laurasia and Gondwana.

Mao et al. www.pnas.org/cgi/content/short/1114319109 5 of 10

Page 12: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Legend continued on following page

1

2

3

4

5

6

7

8

9

10

34

38

40

41

400 350 300 250 200 150 100 50 0

MOEocenePL CretaceousE CretaceousJurassicTriassicPermianCarboniferousDevonians

ed

oN

Million years ago (Ma)

Silurian

450

Run1: Uniform priors, A to P (P:[306.2, 366.8])

]5.633,2.603[:P:3nuR Uniform priors, A to O,

]7.153,2.603[:P:2nuR Uniform priors, A to O,

]4.123,2.603[:P,4nuR Uniform priors, A to O,

Run5, subset VNDUniform priors, P,

Run6, subset NVNDUniform priors, P,

Run8, Uniform priors, A to P + Wa

Run9, Uniform priors, A to P, 56 taxa

Run7, Lognormal priors, P, subset VND;uniform priors, subset NVND

Fig. S3. Divergence time estimates for nodes 1–10 (see node descriptions in Table 1), the crown of Conifer clade II (node 34), Pinaceae (node 38), Conifers(Node 40), and Acrogymnospermae (Node 41) obtained with different calibration schemes and datasets. Gray bars represent age estimates from the 56-taxondataset; bars of other colors represent age estimates from the 144-taxon dataset. Pink bars represent age estimates with lognormal priors for a subset ofcalibrations; bars of other colors represent age estimates with uniform priors. Gray and dark-blue bars represent age estimates with calibrations A through P(Table S4); light-blue, purple, and green bars represent age estimates with calibrations A through O, with the age ranges for calibration P constrained to [306.2,351.7], [306.2, 366.5], or [306.2, 321.4] Ma. Yellow bars represent age estimates with calibration P plus calibrations C, D, H, I, M, N, and O (subset VND). Orange

Mao et al. www.pnas.org/cgi/content/short/1114319109 6 of 10

Page 13: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

bars represent age estimates with calibration P and calibrations A, B, E, F, G, J, K, and L (subset NVND). Red bars represent age estimation with calibrations A–Pand Wa. Pink bars represent age estimate with calibration P and subset VND set as lognormal priors and calibration subset NVND set as uniform priors. Lightorange shading represents the breaking-up of Pangea into Laurasia and Gondwana.

N

SF

E A

U

Cladogenesis vicarianceby

Local extinction

Dispersal resulting in range expansion

E

A

A

E

Taxodium mucronatum

Taxodium distichum

Taxodium European fossils

Glyptostrobus pensilis

Glyptostrobus European fossils

Taxodium N American fossils

Cryptomeria japonica

Cryptomeria European fossils

Sequoiasempervirens

Sequoiadendron giganteum

Austrosequoia

Sequoia Sequoiadendron& related fossils

Metasequoia glyptostroboides

Metasequoia N Hemisphere fossils

Athrotaxis S Amercian fossils

Athrotaxites berryi

Taiwania cryptomerioides

Taiwania European fossils

Taiwania Asian fossils

N

N

E

A

N

E

A

N

N

U

N A E

A

N A E

SS

N

A

E

A

Cunninghamia lanceolata

Cunninghamia NAmerican fossils

Sewardiodendron laxum

Austrohamia minuta

A

N

SS

AthrotaxisU

AmentotaxaceaeN A

Taxaceae sensu strictoN A E

CephalotaxaceaeA

SciadopityaceaeA

Glyptostrobus N American fossilsN

Taiwania N American fossilsN

Cunninghamia related fossil fromAsiaA

Cunninghamia related fossil from Europe

Austrohamia acanthobractea

0.32

0.47

0.63

0.99

0.53

0.55

E 0.68

0.86

0.66

0.62

0.31

0.36

0.34

0.52 0.79

0.39

0.40

0.65

0.51

0.18

0.87 N 0.92

eaecasse rpuC

0 (Ma)50100150200250

Papuacedrus S Amercian fossilsSS

Papuacedrus papuanaU

Austrocedrus Australasian fossilsU

Austrocedrus chilensisSS

Juniperus N American serrate leaf clade ( )15 species

Juniperus QTP Clade ( )9 species

Juniperus N American smooth leaf clade ( )8 species

J sabinauniperus

J semiglobosauniperus

J microspermauniperus

J chinensisuniperus

J procumbensuniperus

J thuriferauniperus

J excelsauniperus

J procerauniperus

J phoeniceauniperus

J drupaceauniperus

J taxifoliauniperus

J communisuniperus

J rigidauniperus

J formosanauniperus

J oxycedrusuniperus

J deltoidesuniperus

C Clade (4 species)upressus QTP

C atlanticaupressus

C sempervirensupressus

C chengianaupressus

C jiangeensisupressus

C funebrisupressus

Hesperocyparis ( )12 species

Callitropsis nootkatensis

Xanthocyparis vietnamensis

Platycladus orientalis

Microbiota decussata

Tetraclinis articulata

Tetraclinis N American & European fossils

Calocedrus decurrens

Calocedrus macrolepis

Calocedrus European fossils

Chamaecyparis obtusa

Chamaecyparis European fossils

Chamaecyparis lawsoniana

Fokienia hodginsii

Fokienia N American fossils

Chamaecyparis pisifera

Chamaecyparis formosensis

Chamaecyparis thyoides

Chamaecyparis N American fossils

Thuja standishii

Thuja plicata

Thuja occidentalis

Thuja koraiensis

Thuja N American fossils

Thuja European fossils

Thujopsis dolabrata

Core 6 species( )Callitris

Actinostrobus acuminatus

Actinostrobus pyramidalis

Neocallitropsis pancheri

Callitris sulcata

Callitris macleayana

Callitris drummondii

Widdringtonia ( )4 species

Diselma archeri

Fitzroya cupressoides

Fitzroya Australian fossils

Libocedrus plumosa

Libocedrus bidwillii

PilgerodendronuviferumSS

U

U

U

SS

U

F

U

U

U

U

U

U

U

N

A

E

A

N

N

A

N

N

A

A

N

A

N

E

A

E

N

A

N E

E

A

A

A

N

N

A

A

A

E

E

A

E

E

A

A

E

A

E

E

F

E

E

A

A

A

A

A E

N

A

N

N A

Cun

Tai

Ath

Seq

Tax

eaediortil laC

eaedi osse rpuC

A 0.38

A

A 0.25

A 0.30A 0.21

A 0.64

A

A

A

N0.33

N A

AA 0.81

A

EA 0.64A

N

NSS

N

SSU

0.90

N

N

SS

N

N0.50

A 0.37

N

E

AE

N0.17

N 0.66

N

E

EN

N

E

N

0.81

N 0.52

S

N

S

N

S0.46

S 0.30

U

S0.42

U

U

S0.76

U

S0.88

U 0.99

S

U0.89

S

0.49

F

S U

U

U

S0.87

0.76

0.95

U 0.99

UU

U

U

U

N 0.34

A

A 0.29N E

A0.38

N0.65

N

A0.72

A

0.31

N AN

N A

E

0.46

EN A

N

N A N

A0.92

0.26

0.56

N

A0.94

N 0.65

A 0.97

N0.50 A

N A

0.53

N

EA0.59E

A0.86

0.55

A 0.92

N E

E0.84

N

EA

0.65

E

A

A

0.39A

A

A

E E

0.89

0.89

0.78 A

0.94

A 0.54

A

A

N0.66

NN

0.92

0.99

E

EA

0.22

E

A0.33A

AA

EN A

E 0.97

0.39

0.37

0.47

E0.39

N A

E

0.34

A

N N0.93

0.98

A0.93

E

NAE

0.20

E

N A

0.40

A

N A A

NN

0.41

0.45

0.99

A E

A0.50

E0.77 E

0.79E

A A0.93

0.98

N A

S U

Fig. S4. The likelihood-based AAR for living and extinct members of Cupressaceae, which is the basis for Fig. 2C (scenario 1 in SI Text). (Right) The AARs with thehighest likelihood are shown as colored boxes at each node. (Left) The six areas used in the analyses and the modeled biogeographic processes. Single-area boxesindicate an ancestor confined to a single geographic area; combined boxes indicate an ancestor with a distribution encompassing two or more areas; two boxesseparated by a space indicate the ancestral ranges inherited by each of the daughter lineages arising from the respective ancestor. For nodes with alternativereconstructions (within log2 likelihoodunits of themaximum), the relative probability of the global likelihood for the optimal reconstruction is given. Extinct lineagesknown from fossils are indicated by dashed lines.

Mao et al. www.pnas.org/cgi/content/short/1114319109 7 of 10

Page 14: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

(c)Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis ungeri

Athrotaxis cupressoides

Metasequoia glyptostroboides

Sequoiasempervirens

Sequoiadendrongiganteum

Cryptomeria japonica

Glyptostrobus pensilis

Taxodium distichum

Callitroideae

Cupressoideae

(b)Pinus sylvestris

Sciadopitys verticillata

Hughmillerites juddii

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis cupressoides

Sequoioideae

Taxodioideae

Callitroideae

Cupressoideae

Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis cupressoides

Metasequoia glyptostroboides

Sequoiasempervirens

Sequoiadendrongiganteum

Cryptomeria japonica

Glyptostrobus pensilis

Glyptostrobus sp

Taxodium distichum

Callitroideae

Cupressoideae

(e)

Ginkgo biloba

Compsostrobus neotericus

Cedrusdeodara

Abies firma

Keteleeriadavidiana

Larix occidentalis

Cathaya argyrophylla

Picea sitchensis

Pinus strobus

Pityostrobus bernissartensis

Podocarpus macrophyllus

Araucaria araucana

Araucaria mirabis

Scidopitys verticillata

Cryptomeria japonica

Amentotaxus argotaenia

(a)

Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis cupressoides

Sequoioideae

Taxodioideae

Papuacedrus papuana

Papuacedrus prechilensis

Austrocedrus chilensis

Pilgerodendronuviferum

Libocedrus plumosa

Widdringtonia nodiflora

Fitzroya cupressoides

Diselma archeri

Callitris macleayana

Neocallitropsis pancheri

Actinostrobus pyramidalis

Callitris drummondii

Thujopsis dolabrata

Thuja plicata

Fokienia hodginsii

Chamaecyparis lawsoniana

Calocedrus decurrens

Tetraclinis articulata

Platycladus orientalis

Microbiota decussata

Cupressus funebris

Xanthocyparisvietnamensis

Callitropsis nootkatensis

Hesperocyparis bakeri

Juniperus

(g)

Pinus sylvestris

Sciadopitys verticillata

Austrohamia minuta

Austrohamia acanthobractea

Sewardiodendron laxum

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxites berryi

Athrotaxis cupressoides

Metasequoia glyptostroboides

Metasequoia sp

Sequoiasempervirens

Sequoiadendrongiganteum

Cryptomeria japonica

Glyptostrobus pensilis

Taxodium distichum

Papuacedrus papuana

Austrocedrus chilensis

Libocedrus plumosa

Pilgerodendronuviferum

Widdringtonia nodiflora

Diselma archeri

Fitzroya cupressoides

Fitzroya acutifolia

Callitris macleayana

Callitris leaensis

Neocallitropsis pancheri

Actinostrobus pyramidalis

Callitris drummondii

Thujopsis dolabrata

Thuja plicata

Fokienia hodginsii

Chamaecyparis lawsoniana

Calocedrus decurrens

Calocedrus suleticensis

Tetraclinis articulata

Tetraclinis salicornioides

Platycladus orientalis

Microbiota decussata

Cupressus funebris

Hesperocyparis bakeri

Xanthocyparis vietnamensis

Callitropsis nootkatensis

Juniperusvirginiana

Juniperusrecurva

Juniperusphoenicea

Juniperus indica

Juniperus californica

Juniperus pauli

Juniperus oxycedrus

Juniperusdrupacea

(i)

Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis cupressoides

Sequoioideae

Taxodioideae

Callitroideae

Thujopsis dolabrata

Thuja plicata

Fokienia ravenscragensis

Fokienia hodginsii

Chamaecyparis lawsoniana

Calocedrus decurrens

Tetraclinis articulata

Microbiota decussata

Platycladus orientalis

Hesperocyparis bakeri

Cupressus funebris

Xanthocyparisvietnamensis

Callitropsis nootkatensis

Juniperus

(h)Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Athrotaxis cupressoides

Sequoioideae

Taxodioideae

Callitroideae

Thujopsis dolabrata

Thuja plicata

Thuja polaris

Fokienia hodginsii

Chamaecyparis lawsoniana

Calocedrus decurrens

Tetraclinis articulata

Microbiota decussata

Platycladus orientalis

Hesperocyparis bakeri

Cupressus funebris

Xanthocyparisvietnamensis

Callitropsis nootkatensis

Juniperus

(f)

Pinus sylvestris

Sciadopitys verticillata

Cunninghamia lanceolata

Taiwania cryptomerioides

Sequoia semprevirens

Austrosequoia wintonensis

Taxodium distichum

Glyptostrobus pensilis

Cryptomeria japonica

Callitroideae

Cupressoideae

Athrotaxis cupressoides

Metasequoia glyptostroboides

Sequoiadendron giganteum

(d)

Fig. S5. Strict consensus trees reconstructed using a total-evidence approach (SI Text). (A) Plot shows phylogenetic positions of Compsostrobus neotericus,Pityostrobus bernissartensis, and Araucaria mirabilis. (B–H) Plots illustrate phylogenetic positions of Hughmillerites juddii, Athrotaxis ungeri, Austrosequoiawintonensis, Glyptostrobus sp., Thuja polaris, Papuacedrus prechilensis, and Fokienia ravenscragensis. (I) Plot shows the phylogenetic positions of theremaining 10 Cupressaceae fossils. Extinct lineages known from fossils are indicated by dashed lines, and their names are highlighted on a gray background.

Mao et al. www.pnas.org/cgi/content/short/1114319109 8 of 10

Page 15: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Fig. S6. Distributions of fossil (yellow solid circles) and living (green shading or green solid circles) Cupressaceae. (A) Cunninghamioideae. (B) Sequoioideae. (C)Taxodioideae. All three subfamilies underwent range contraction over time. Fossil distribution maps were compiled from the Paleobiology Database (http://paleodb.org).

Mao et al. www.pnas.org/cgi/content/short/1114319109 9 of 10

Page 16: Distribution of living Cupressaceae re ects the breakup of ...renners/Mao_Cupressaceae_PNAS2012_withOSM.pdfPangea had a warm and equable climate without glaciation or sea ice and that

Fossil calibration node Fossil calibration node

Fossil calibration node removed

S

SS

DχSS

(a) (b)

(d)c)(All Calibration Nodes

Fig. S7. Cross-validation of fossil calibrations A–O and Wa (Table S4) based on the 56-taxon dataset. (A) Histogram of the mean deviation (�D) (1) betweenmolecular and fossil age estimates for all nodes, using a single fossil-dated node as a calibration point. (B) Histogram of the SS values (1) for a given fossilcalibration node when it was used as the sole calibration point. (C) 2D plot for a given fossil calibration node with SS values (the sum of the squared differencesbetween the molecular and fossil age estimates at all other fossil-dated nodes) (1) on the x axis and the mean deviation on the y axis. (D) Plot illustrating theeffect on s (an average squared deviation for the deviation between molecular and fossil age estimates for all fossil calibrations in the analysis) of removingfossil calibration points (1). Open points indicate that the removal of the respective fossil calibration resulted in a significant reduction in the variance ofs, based on a one-tailed Fisher’s test (1). A maximum constraint of 366.8 Ma was placed on the older bound of calibration point P based on the argumentsprovided in Table S4.

Other Supporting Information Files

Table S1 (DOC)Table S2 (DOC)Table S3 (DOC)Table S4 (DOC)Table S5 (DOC)Table S6 (DOC)

1. Near TJ, Meylan PA, Shaffer HB (2005) Assessing concordance of fossil calibration points in molecular clock studies: an example using turtles. Am Nat 165:137–146.

Mao et al. www.pnas.org/cgi/content/short/1114319109 10 of 10