Geography and ecology structure diversification in Ziziphus Mill. (Rhamnaceae)

Jess Rickenback; Caroline E. R. Lehmann; Frank Hauenschild; Mark J. Hughes; Flávia Fonseca Pezzini; R. Toby Pennington

doi:10.21425/fob.17.133817

Abstract

Understanding the processes responsible for creating and preserving biodiversity within tropical ecosystems is essential for gaining deeper insights into the history, dynamics, and resilience of these environments. We investigated the biogeography and evolution of Ziziphus Mill. (Rhamnaceae), a genus of c.50 species found in all major tropical biomes and across Africa, Asia and Australasia to evaluate how ecology and geography structures its diversification, and their associations with functional traits. We used consistency and retention indices, as well as the delta statistic to test for phylogenetic signal in ecology, geography and functional traits. We constructed a time-calibrated phylogeny using BEAST, followed by the reconstruction of biome and geographical affinities in BioGeoBEARS, and ancestral functional trait evolution using phytools and ape. At a global scale, geography structures Ziziphus’ diversification more than ecology, as evidenced by higher phylogenetic signal in retention indices and the delta statistic. Multiple ecological switches among biomes are associated with dispersal events, which structure Ziziphus’ diversification at smaller scales. Biogeographic estimation suggests that Ziziphus evolved out of closed biomes in Asia during the Oligocene. Long distance dispersal from Asia to Africa resulted in the occupation of two novel biomes, the expanding tropical grassy biome and the desertic biome. Geographic transitions across continents that retain the same biome are fewer but are demonstrated by a dispersal from the closed biome in Asia to the same ecological setting in Australia. In keeping with the younger age of the savanna biome and the Pliocene radiation of geoxylic species, the youngest species in Ziziphus are mostly found in open biomes, and the youngest habit is geoxylic. Leaf hair appears to confer advantages across biomes whilst being a synapomorphy for open and desertic clades. One of two evolutions back into closed biomes is associated with the development of the liana habit.

Highlights

We inferred the most complete phylogeny for Ziziphus to date.
Ziziphus originated in closed biomes in Asia in the Oligocene.
At a global scale, geography structures diversification more than ecology (biome preference) in Ziziphus.
Biome shifts are frequently associated with long distance dispersal events.
Shifts in growth form are frequently associated with biome shifts.

Keywords

biogeography, biomes, diversity, evolution, traits, tropical, phylogenetic structure

Introduction

Research into the mechanisms which generate and maintain tropical species diversity across biomes plays a crucial role in the deeper understanding of the history, dynamics and resilience of these systems. Diversification is driven by ecological, geographical, and evolutionary opportunity (Mitchell et al. 2018). These different processes may operate simultaneously or in succession (Kong et al. 2022). Biogeographical and evolutionary studies of genera distributed amongst a variety of biomes have successfully been used to evaluate the mechanisms structuring diversification, as in Ceiba Mill. across Latin America (Pezzini et al. 2021), Monechma (Hochst.) T. Anderson in Africa (Darbyshire et al. 2020), and Hakea Schrad. in Australasia (Cardillo et al. 2017). Ziziphus Mill. forms a good case study with which to unpick these thorny issues as its c.50 species span a wide geographic and ecological range across the subtropics and tropics of Africa, Australasia and Asia, covering all major lowland tropical biomes. Ziziphus displays a diversity of habits such as tree, shrub, liana, and geoxyle – dwarf shrubs with massive underground woody structures and limited aerial parts (Gomes et al. 2021) - and these growth forms are significantly linked to occupation of specific biomes, where lianas are restricted to closed biomes and geoxyles are only found in open biomes (Rickenback et al. 2022). Ecological associations are also seen in the development of the compact life form in Crassula L. in conjunction with diversification into arid areas (Lu et al. 2021) and across the sclerophyllous Australian flora (Crisp et al. 2004). Geographical opportunity has been linked to explosive diversification into arid Australian biomes in the mega diverse genus Solanum L. (Echeverria-Londono et al. 2020) and Ptilotus R.Br. (Hammer et al. 2021), and in CheirolophusCass. in the Canary Islands (Vitales et al. 2014). These processes can also be indistinguishable as in the rapid diversification of Oreocharis Benth. (Kong et al. 2022). Although the distribution of Ziziphus species among biomes are predominantly associated with geography - closed biome species are primarily located in Southeast Asia, and open biome species are most common in Africa - links between habit and geography, and the balance of ecological and geographical drivers of diversification in Ziziphus, have not previously been explored.

Many of the world’s significant and charismatic biomes are tropical, including rainforest, savanna, seasonally dry tropical forest (SDTF) and desert. Biomes are multiscale phenomena where ecology, climate, soils and water filter vegetational attributes producing areas of coherent physiognomies that carry a legacy of deep evolutionary processes (Higgins et al. 2016; Mucina 2019; Keith et al. 2020). Traits can be adaptive to biomes, and therefore characterize them, for example, large seeds in rainforests are associated with tolerance to shade and establishment in competitive low light environments (Poorter and Rose 2005). Research on biome-specific shared community traits is common, such as savanna grass flammability (Calitz et al. 2015), the placement of bud banks in trees in flammable systems (Charles-Dominique et al. 2015), and tree growth strategies in savannas (Dantas and Pausas 2013) but fewer data exist on evolutionary trajectories of functional traits in relation to biome switching (Donoghue and Edwards 2014; Gagnon et al. 2019; but see Onstein et al. 2014). We investigated whether functional traits in Ziziphus are adaptations to overcome barriers to evolutionary entry into biomes.

Plant functional traits are the hereditary morphological and physiological characters contributing to competitive plant life history strategies in an environment. Functional traits can be chemical such as primary and secondary metabolites used for herbivory defense (Coley et al. 2019), or physical such as corky bark to enable trees to survive fires (Dantas and Pausas 2013). On a larger scale, traits can be integrative, such as habit which includes height, architecture, and annual or perennial growth. Some traits, such as the climbing habit, are key innovations for certain clades including palms (Arecaceae), enabling exploitation of novel resources and habitats (Bateman et al. 1998; Gianolo 2004; Couvreur et al. 2015). In an evolutionary context, plant traits may evolve as adaptations to new environments, or be ‘exaptive’ – pre-existing traits used for a new purpose (Armbruster 1997). Focusing on the species-level we link species-specific trait knowledge with well sampled phylogenies to infer traits reflective of the adaptative history of Ziziphusand associations with biome switching.

The major tropical and sub-tropical biomes vary widely in age, influencing how and when lineages have moved amongst them. Phylogenetic evidence suggests that closed rainforests date from the mid Cretaceous (Davis et al. 2005), although fossil evidence dates rainforests in their modern form to the early Paleocene (Carvalho et al. 2021). In contrast, fossil evidence suggests tropical dry forests date from the mid to late Eocene (Martinez et al. 2021), and open grassy biomes such as savanna did not expand widely until ~11 mya associated with a global increase in aridity (Edwards, Smith and Donoghue 2010; Couvreur et al. 2021). While these assemblages are separated in time, plant lineages demonstrate repeated evolution between closed and open biomes (Simon and Pennington 2012; Armstrong et al. 2014; Veranso-Libaleh et al. 2018), forming an integral part of the process of tropical diversification (Zizka et al. 2020). These divergent biome ages can be reflected in older lineages of species associated with closed biomes, and younger species associated with open biomes, congruent with the rise in dominance of tropical dry forests and savannas, for example in African Melastomataceae (Veranso-Libalah et al. 2018) and Entandophragma C.DC. (Meliaceae) (Monthe et al. 2019), although see Donkpegan et al. (2020). While there are uncertainties associated with ascertaining precise dates of diversification (Ho and Phillips 2009), dated phylogenies can shed light on the mechanisms underlying diversification, including long distance dispersal, continental movement, or ecotone speciation linked to a changing climate (Armstrong et al. 2014; Antonelli et al. 2015; Nge et al. 2020; Couvreur et al. 2021).

Biome boundaries can present formidable barriers to adaptive radiation (Pennington and Lavin 2015) or, as in the tropical Americas, frequent independent evolution of lineages into savanna from surrounding biomes suggest that fire adaptations such as corky bark or geoxylic growth forms have been easy to evolve (Simon et al. 2009). Cambial variants, a key trait of lianas, have evolved across the angiosperm phylogeny (Angyalossy et al. 2012) and therefore seem likely to be both easy to evolve and confer significant fitness for survival in tropical closed biomes. Phylogenetic biome conservatism suggests species tend to stay within biome boundaries, and that intra-biome diversification is more common than inter-biome (Crisp et al. 2009; Eiserhardt et al. 2017). By using a combination of phylogenetic and trait data, we can assess the magnitude of two of the key drivers of trait evolution linked to biome occupation – geography and ecology. If biome conservatism structures diversification, the same lineage may move through space between similar environments by dispersal. If geography is a barrier to dispersal then biome shifting may occur, where lineages evolve traits allowing survival in adjacent biomes in one area. However, geographical structure does not preclude ecology as a driver of diversification because biome switches can occur within continents.

Here we use a case study of Ziziphus to investigate some of the fundamental processes underlying tropical diversification by answering the following questions:

Questions

How do geography and ecology structure Ziziphus’ phylogeny?
What diversification opportunities are associated with shifts in functional traits?
How old are Ziziphus’ transitions into biomes- are open biome species younger than species in other biomes?

We answer these questions using a time-calibrated phylogeny of Ziziphus, which we use to examine phylogenetic signal in traits, ecology and geography, and to reconstruct trait evolution and biogeography.

Methods

Ziziphus is a genus of c.50 species that, in comparison to its sister genera Hovenia Thunb. and PaliurusMill., has a wider geographic and ecological distribution occupying tropical rainforest, desert, and savanna across Asia, Africa, the Arabian Peninsula and Australasia (Rickenback et al. 2022). Ziziphus is fly and occasionally wind-pollinated (Zietsman 1990). The fruits are globose, and are eaten and dispersed by birds and mammals including gibbons (Z. mauritiana Lam., Z. pubinervis Rehder) (Fan et al. 2011), foxes and other carnivores (Cancio et al. 2017). Ziziphus contains species of economic importance, foremost among them the globally cultivated Ziziphus jujuba Mill. (Gao et al. 2013).

Taxon sampling

Hauenschild et al. (2016a) recircumscribed Ziziphus as monophyletic, with Paliurus strongly resolved as its closest sister group, making it a suitable outgroup. Recent taxonomic and genetic work suggests that three species, Z. angustifolia, Z. pubescens, and Z. rivularis, should be excluded from Ziziphus (Islam and Simmons 2006; Onstein et al. 2015; Hauenschild et al. 2018; Tian et al. 2024) and these are therefore not included in this study. With the inclusion of nine species of Ziziphus and five species of Paliurus downloaded from GenBank (Clark et al. 2016), taxon sampling for Ziziphuscovered 38 of 50 species– a coverage of 76%, and all 5 species of Paliurus(Suppl. material 1: table S1). Sequences covered as broad a geographic range as possible (Fig. 1).

Figure 1.

Sampling locations for Ziziphus species comprising 21 sequences for 15 species downloaded from GenBank (in pink) where location data was available (Suppl. material 1: table S1) and 45 new sequences of 29 Ziziphus species including one Central American sample of the globally cultivated Z. mauritiana (in turquoise). Yellow shading represents approximate native distributions for Ziziphus excluding Z. jujuba, from Hauenschild et al. 2016.

DNA extraction

We extracted DNA from 73 herbarium leaf samples of Ziziphus spp and two silica gel dried leaf samples, using the Plant DNeasy kit (Qiagen) and following the manufacturers protocol with the following changes; at the lyse step the lysate was incubated on ice for one hour to precipitate proteins, detergent, and polysaccharides.

PCR

A combination of nuclear ribosomal internal transcribed spacer region (i.e., the 5.8S and flanking ITS1 and ITS2 spacers) (ITS) and plastid (trnL-trnF) markers were used because they are suitable for reconstructing species-level topologies in Rhamnaceae (Hauenschild et al. 2016a, 2016b).

For ITS, primers AB101 which anneals in the 18S gene, and AB102 which anneals in the 26S gene were used (Suppl. material 1: table S2). Where sequencing was unsuccessful, samples underwent a further round of PCR for ITS2 using primers S2F and S3R. All reagents were fully defrosted, and dNTPs, 10xNH₄ buffer, MgCl₂, and DNA samples were vortexed before use. Each 20ul PCR amplification reaction contained 12.1ul denatured (Sigma) water, 2.0ul buffer, 2.0ul dNTPs, 1.0ul DMSO, 0.6ul MgCl₂, 0.5ul Bovine SA, 0.3ul forward primer, 0.3ul reverse primer, and 0.2ul Taq. 1ul of DNA (diluted 1:10) was added to each reaction tube. Twenty-five cycles of 95 °C for 30 seconds, 50 °C for 20 seconds, and 60 °C for four minutes were run before the samples were chilled to 4 °C.

PCR of plastid regions was carried out using trnLc and trnLf primers. Where unsuccessful, samples underwent a further round of PCR using trnLc and trnLd, and trnLe and trnLf. Each 20ul PCR amplification reaction contained 6.1ul denatured (Sigma) water, 4.0ul TBT-PAR, 2.0ul dNTPs, 2.0ul buffer, 0.22ul MgCl₂, 0.5ul Bovine SA, 2.0ul forward primer, 2.0ul reverse primer, and 0.3ul Taq. 1ul of DNA (diluted 1:10) was added to each reaction tube. The tubes were briefly centrifuged. Initial denaturing was at 95 °C for three minutes. 36 cycles of three steps took place. Step one was denaturation at 95 °C for one minute. Step two was annealing at 56.4 °C for one minute. Step three was extension at 72 °C for one minute. The final extension was 72 °C for ten minutes before samples were chilled to 10 °C. Gel electrophoresis was used to separate molecules of PCR product according to size.

For both ITS and trnL-F, where gel bands were bright, PCR product was diluted 50:50 with denatured (Sigma) water before purification. Where bands were faint, twice as much purified product was used for sequencing. Samples were submitted to the Edinburgh Genomics laboratory at the University of Edinburgh for Sanger sequencing.

Alignment and phylogenetic analysis

Sequences were edited with Sequencher v5.4.6 (Gene Codes Corporation 2017) and aligned using the ClustalW algorithm (Larkin et al. 2007) before additional manual editing in BioEdit v7.2.5 (Hall 1999). Multiple accessions of species were included where possible.

After trimming and aligning, the ITS data set for Ziziphus consisted of 55 accessions and 837 nucleotide characters representing 36 species, and the trnL-F data set consisted of 29 accessions and 973 nucleotide characters representing 21 species. For Paliurusthe ITS data set consisted of three accessions representing three species and the trnL-F data set consisted of four accessions representing four species. We found no strongly supported incongruencies between the topologies of the ITS and trnL-F trees and concatenated sequences from the same accessions using SequenceMatrix v1.8 (Vaidya et al. 2010). The combined cpDNA and ITS data set consisted of 66 taxa and 1810 nucleotide characters. Within these, in order to maximise similarities and avoid artificial differences we excluded 22 consecutive bases between 580–601.

Maximum likelihood analyses of individual and combined data sets ran on RAxML v8 (Stamatakis 2014) in the CIPRES Science Gateway (Miller et al. 2010). Data from ITS and trnL-trnF were partitioned because mutation rates vary between these regions and within the cpDNA region (Wolfe et al. 1987). Based on the Akaike information criterion (AIC) (Akaike 1974) and BIC (Schwarz 1978) as implemented in the program JModelTest v2.1.4 (Darriba et al. 2012) the nucelotide substitution model GTR+I+G was used for each partition.

Bayesian analyses of individual and combined data sets were performed in MrBayes v3.2.6 (Ronquist et al. 2012) (Suppl. material 1: fig. S1). The data were partitioned as before with the best fitting nucleotide substitution model for each partition selected via the Akaike information criterion (AIC) (Akaike 1974) and BIC (Schwarz 1978) as implemented in the program JModelTest v2.1.4 (Darriba et al. 2012). The substitution model GTR+I+G was chosen for ITS. GTR was selected for trnL-trnf.

Divergence time estimates

We pruned our tree to one accession per species (following Hauenschild et al. 2017) to avoid violating the Yule model of diversification which is used to infer relationships between different species (Sarver et al. 2019). Where more than one sequence of a taxon matched in region coverage we chose the cleanest sequence, with the least ambiguity in base calling. The extent of missing trnL-F data in the combined nuclear and plastid dataset resulted in long basal branches and unstable age inferences in our dated phylogeny. To maximise stability, we limited our data to the ITS region and further edited sequences to result in a smaller data set of 39 taxa and 711 nucleotide characters. We used BEAST v1.10.4 (Suchard et al. 2018) which employs a Bayesian Markov chain Monte Carlo (MCMC) approach to co-estimate topology, substitution rates, and node ages (Holstein and Renner 2011) to analyse divergence times in the pruned nuclear dataset. Paliurus and Ziziphus taxon sets were defined in BEAUTi v1.10.4 and set as monophyletic based on outputs from the Bayesian inference (BI) and Maximum Likelihood (ML) trees, and previous work (Hauenschild et al. 2016a). Based on JModelTest v2.1.4 (Darriba et al. 2012), we selected the GTR + I + G substitution model for the ITS region. We used an uncorrelated log normal model (relaxed clock model and rate heterogeneity) to ensure the consistency of the relative rate of evolution on branches on the tree. We used a randomly generated starting tree with a Yule speciation prior which is appropriate for inter-specific level data (Ritchie et al. 2017; Monthe et al. 2019). We calibrated the phylogeny by setting the divergence date between Paliurus and Ziziphus as a lognormally distributed random variable (mean = 44 Ma, StDev = 11) following Burge and Manchester (2008). Two different fossils were used to calibrate independent runs to provide date estimates for the stem node (Table 1). Both Archeopaliurus boyacensis Correa, Manchester, Jaramillo & Gutierrez, and Paliurus favonii Unger are suitable for placement at the Paliurus stem – Archaeopaliurus boyacensis has previously been used as a calibration for Paliurus, and P. favonii more closely resembles extant Paliurus. We used the younger of our two calibrations (Paliurus favonii) in the biogeographic analysis, because P. favonii possesses many of the diagnostic characteristics of extant Paliurus, is based on multiple fossil records, and is found across the area of current Paliurus distribution (Burge and Manchester 2008). Paliurus favonii is morphologically similar to extant Paliurus species especially P. spina-christi Mill. Paliurus favonii is based on 37 fossil fruits and has a fleshy indehiscent drupe with a mean diameter of 13 mm and the characteristic Paliurus wing (Burge and Manchester 2008). A similar fossil (P. clarnensis) has been used in previous studies (Onstein et al. 2015). In order to provide the most robust results analyses were also run using the older fossil A. boyacensis, which has been used previously to date the Ziziphoid clade (Hauenschild et al. 2018) and is dated to the same period (66 my) as an Indian fossil Paliurussp used by Chen et al. 2017. Using the older Archaeopaliurus boyacensis calibration did not affect our results (Suppl. material 1: fig. S2). Other priors were left to the default settings. Each tree was run twice for 20000000 generations with 2000 trees sampled. Effective sample sizes were visually checked in Tracer v1.7.2 (Rambaut et al. 2018) and confirmed to be >200 and stable, before being combined in LogCombiner v1.10.4. A maximum clade credibility (MCC) tree was created in Tree Annotator v1.10.4 after a run of 20000000 and a 10% burn in which was sufficient to achieve stationarity (Suppl. material 1: fig. S3).

Table 1.

Download as

CSV

XLSX

Details of the calibration points (node age, standard deviation, and prior distribution) and their sources used to generate the dated phylogeny of Ziziphus in BEAST.

Calibration point	Age (Ma)	Prior distribution	St. dev.	Fossil
*Paliurus* stem	66	Lognormal	11	Archaeopaliurus boyacensis (Correa et al. 2010)
*Paliurus* stem	44	Lognormal	11	Paliurus favonii (Burge and Manchester 2008)

Attribution of biome, geography, and traits

Attribution of biome and traits followed Rickenback et al. (2022) with the inclusion of data for the five Paliurusspecies (Suppl. material 1: table S3). Species were assigned to biomes based on data from floras, peer-reviewed articles, grey literature such as doctoral dissertations, United Nation Development Programme (UNDP) reports, herbarium labels, and expert knowledge. Biomes were categorised as ‘open’, ‘closed’, and ‘desertic’. Following the typology of Earth’s ecosystems by Keith et al. (2020) that groups ecosystems into biomes for the purpose of enabling comparative work, in the analyses here, closed biomes are analogous to T1 ‘tropical and subtropical forests’ that include lowland rainforests, tropical montane forests, tropical dry forests, and tropical heath forests. Desertic biomes are analogous to T5 ‘deserts and semi-deserts’, that includes semi-desert steppes, thorny deserts and semi-deserts, sclerophyll hot deserts and semi-deserts, cool deserts and semi-deserts, and hyper-arid deserts. Open biomes, following Bond (2019), are analogous to T3 ‘shrublands and shrubby woodlands’ and T4 ‘savannas and grasslands’ because both ecosystems are mediated by similar processes (Keith et al. 2020). These include seasonally dry tropical shrublands, seasonally dry temperate heaths and shrublands, cool temperate heathlands, rocky pavements, screes and lava flows, trophic savannas, pyric tussock savannas, hummock savannas, temperate woodlands, and temperate subhumid grasslands. For each species, geographical regions were assigned based on flora accounts (Suppl. material 1: table S4) and confirmed with geolocation checks based on Rickenback et al. (2022). Geographic regions followed Cox’s (2001) phytogeographic classification but split the Indo-Pacific and Australasian realms along Lydekker’s Line which represents the barrier which divides the distribution on the highest number of plant species (van Welzen et al. 2011), and better reflected patterns of endemism in Ziziphus. The locations were (i) Africa and Arabian Peninsula, (ii) Indo-Pacific, (iii) Australasia, (iv) Holarctic. Where species distribution spanned multiple areas, these were coded for analyses as multistate. Distributional ranges were based on descriptions from flora such as Flora of India, Flora of Thailand, and taxonomically verified web resources such as Plants of the World Online (Suppl. material 1: table S4). We used trait data from Rickenback et al. (2022) where data were primarily acquired from floras, varied grey literature sources such as worldagroforestry.org (Orwa et al. 2009) and assessment of type specimens. Traits assessed comprised habit, spinescence, bark hair, leaf hair, leaf size, fruit hair, fruit colour and fruit size. A full trait dataset is available at doi:10.5061/dryad.34tmpg4p0.

Analyses

Phylogenetic signal in traits, geography and ecology

Consistency and retention index calculations are used to measure homoplasy and implied synapomorphies respectively in phylogenetic characters. Consistency index (CI) measures homoplasy (Kluge and Farris 1969). When CI nears 1, the trait is consistent with the phylogeny and displays no homoplasy. The retention index (RI) measures the amount of implied synapomorphy in characters retained on a phylogeny, and therefore is a measure of how much synapomorphy (grouping information) a character provides (Kitching 1998). When applied to biomes and geography, CI and RI can be used as measures of fit with a phylogenetic tree. In the context of biomes and geography, high CI means that lineages have entered or left a given biome or geographic area few times (or once if CI=1). In the same context, a high RI means that multiple species in the same monophyletic group(s) are confined to a given biome or geographic area. RI is therefore a key measure of how geography or biome structure diversification because it provides a better measure of diversification confined to one biome or geographic area.

We performed CI and RI calculations using the maximum likelihood (ML) and Bayesian inference (BI) inferred phylogenies in Mesquite v.3.70 (Mesquite Project Team 2021) on the physiological traits described above as well as biome and geography. Because Mesquite allows for multistate characters, where we were unable to ascertain unambiguously a single distributional area, we coded these as multistate (ie P. spina-christi, Z. nummularia (Burm.f.) Wight & Arn. and Z. mauritiana were coded as both Indo-Pacific and African, and Z. oenopolia (L.) Mill. was coded as both Indo-Pacific and Australasian).

We also ascertained the strength of phylogenetic signal for the categorical non-binary traits geography and ecology using the delta statistic (Borges et al. 2019) which tests for phylogenetic signal between a tree and a categorical trait. Delta decreases when a trait has evolved independently: the higher the delta value, the greater the phylogenetic signal between tree and trait. We tested biome, geography and habit – the trait which scored the highest RI for which we had complete data - using phytools v1.0.3 (Revell 2012) and ape v5.6.2 (Paradis and Schliep 2019) in R v3.6.2 (R core team 2021).

Trait evolution and ancestral range estimation

Ancestral character evolution analyses were performed on the combined nuclear and plastid data for habit, the trait for which we had complete data, using function ‘ace’ of the package ape (Paradis and Schliep 2019). The best fitting model was symmetrical (SYM), selected using AIC. We generated 1000 stochastic character maps and summarized them (Fig. 2). Where trait data were incomplete, we used the AncThresh function of phytools (Revell 2012) to code multistate characters to include uncertainty.

Figure 2.

Ancestral character estimation of habit using function ace in ape (Paradis and Schliep 2019) based on a combined ITS and trnL-F dataset of 66 taxa, generated with the SYM best fitting model based on a summary of 1000 simulated stochastic character maps using empirical Bayes method.

We further inferred the biogeographic history of both biome and geography for the dated (ITS only) MCC tree using BioGeoBEARS v1.1.2 (Matzke 2013). We ran the model using Dispersal-Extinction-Cladogenesis (DEC) and a likelihood variant of the Dispersal - Vicariance - Analysis (DIVALIKE). We did not perform BayArea-like analyses as the model does not allow vicariance-driven events (Matzke 2014). The best fitting model was selected using AIC. For biome the best fitting model was DIVALIKE+J and for geography the best fitting model was DEC+J, that is, with and without founder events (+J).

Results

The combined ML and BI tree shows Ziziphus as monophyletic with strong support (Bootstrap support (BS) 97, Posterior probability (PP) 1.0) supporting previous work by Hauenschild et al. (2016) and Islam and Simmons (2006), but the relationships along the backbone are unresolved (Suppl. material 1: fig. S1). Clades with support BS >70 / PP >0.9 are (Z. xylopyrus (Retz.) Willd., Z. horrida Roth, Z. attopensis [57] Pierre, Z. cambodianus Pierre, and Z. horsfieldii Miq.), (Z. funiculosa Buch.-Ham., Z. calophylla Wall., Z. havilandii Ridl., Z. suluensis Merr., Z. elegans Wall., Z. affinis Hemsl., Z. kunstleri King, Z. attopensis [58] and Z. attopensis), (Z. hajarensis Duling, Ghaz. & Prend., Z. mairei Dode, Z. montana W.W.Sm., Z. xiangchengensis Y.L.Chen & P.K.Chou), (Z. abyssinica Hochst. Ex A.Rich, Z. zeyheriana Sond., and Z. mucronata Willd.), and (Z. brunoniana C.B.Clarke ex Brandis, Z. fungi Merr., Z. oenopolia, Z. laui Merr., Z. pubinervis, and Z. apetala Hook.f.).

Phylogenetic signal in traits, geography and ecology

CI calculations run on both ML and BI trees suggested geography as the least homoplasious and therefore most consistent trait (Table 2). Geography also showed the highest RI (0.72). Biome has an RI of 0.58, lower than geography, but still higher than any of the functional traits. Leaf hair is homoplasious (CI= 0.08, RI = 0.61) suggesting repeated independent evolutions but that it supports the structure of the phylogeny – i.e., clades where all members have hairy leaves. Fruit hair shows the lowest CI and a low RI (CI= 0.1, RI = ~0.3) suggesting both that fruits have switched between hairy and hairless states multiple times but also that the trait fruit hair does not provide much grouping information. Habit has a low CI (0.16) but one of the highest RIs (0.5) suggesting that habit is a trait which has changed multiple times independently, but that provides substantial grouping information. For spines, CI=0.5 and RI =0 indicating that spinescence is a consistent trait but provides no grouping information.

Delta statistic outputs supported CI and RI outputs in finding that geography showed the highest phylogenetic signal, with a d-score an order of magnitude larger than biome for both Bayesian and Likelihood trees (Table 3).

Ancestral character evolution analyses (Fig. 2) shows that the ancestral habit is tree, with shrub, liana and geoxyle more recently evolved. Ancestral trait reconstructions suggested that the ancestral leaf hair state was likely hairless with a low probability for hairs, with multiple evolutions of hairiness (Suppl. material 1: fig. S4). Ziziphus’ fruit was ancestrally hairless with multiple evolution of hair (Suppl. material 1: fig. S5).

Table 2.

Download as

CSV

XLSX

Consistency index (CI) and retention index (RI) values calculated for a maximum likelihood (ML) and Bayesian inference (BI) phylogram using Mesquite. Data for traits and biome allocations are from Rickenback et al. (2022). See Suppl. material 1: table S4 for geographic allocations.

Trait	Consistency Index (Maximum Likelihood)	Retention Index (Maximum Likelihood)	Consistency Index (Bayesian Inference)	Retention Index (Bayesian Inference)
Habit	0.17	0.55	0.16	0.52
Spines	0.5	0	0.5	0
Bark hair	0.09	0.38	0.09	0.38
Leaf area	0.16	0.45	0.15	0.4
Leaf hair	0.08	0.61	0.08	0.61
Fruit size	0.14	0.45	0.13	0.41
Fruit colour	0.68	0.33	0.17	0.41
Fruit hair	0.11	0.33	0.13	0.36
Biome	0.17	0.61	0.16	0.56
Geography	0.62	0.72	0.67	0.72

Table 3.

Download as

CSV

XLSX

Delta statistic values (Borges et al. 2019) calculated in phytools (Revell 2012) and ape (Paradis and Schliep 2019) for the traits with the highest RI where data were complete.

Trait	Delta statistic (Maximum Likelihood)	Delta statistic (Bayesian Inference)
Habit	5.23	5.48
Biome	11.42	7.75
Geography	61.77	78.75

Ancestral range estimation

The MCC tree shows the diversification of Ziziphus starting around 28 mya (Suppl. material 1: fig. S3), in line with previous studies (Table 4). Most diversification of extant species occurred around the Late Miocene and Pliocene (11.6–2.6 mya). It suggests that the ancestral area for Ziziphus is the Indo-Pacific (P= 54%), with dispersals into Africa and Arabia at 23, 17 and 14 mya, and two transitions into Australia (Fig. 3a). Ziziphus originated in closed biomes in Asia during the Oligocene (Fig. 3b). There are multiple switches in both directions between closed and open biomes; biome switches happened at 10 and 8 mya into closed biomes, at 23, 12, 8 and 3 mya into open biomes and at 23 mya into the desertic biome. Using A. boyacensis to calibrate the phylogeny shifts the biome switches slightly earlier; into closed biomes at 30 and 15 mya, into open biomes at 38, 19, 17, and 3 mya, and into the desertic biome at 39 mya (Suppl. material 1: fig. S6b).

Table 4.

Download as

CSV

XLSX

Comparison of stem and crown ages of Ziziphus as inferred by previous studies with different fossil calibrations. Ages marked with an asterisk are estimations based on published figures.

Source	Fossil	Stem age (my)	Crown age (my)
Onstein et al. (2015)	Paliurus clarnensis (Burge & Manchester)	45.76	38.56
Hauenschild et al. (2018)	Archaeopaliurus boyacensis	65*	25*
Chen et al. (2017)	Indian fossil	67.04	57.56
This paper	Paliurus favonii	34.4	27.51
This paper	Archaeopaliurus boyacensis	58.68	46.73

Figure 3.

Ancestral range estimations based on the MCC tree calibrated using Paliurus favonii in BioGeoBEARS. Geography (a) model DEC+J on Ziziphus unconstrained ancstates: global optim, four areas max. anagenetic dispersal rate, d =0.0029, extinction rate, e = 0; cladogenetic dispersal rate, j = 0.0275; likelihood ratio test, LnL = -44.24. Af- Africa and Arabian Peninsula, I – Indo-Pacific, Au- Australasia, H- Holarctic; and biome (b) BioGeoBEARS DIVALIKE+J on Ziziphus unconstrained ancstates: global optim, four areas max. anagenetic dispersal rate, d =0.0019, extinction rate, e = 0; cladogenetic dispersal rate, j = 0.0785; likelihood ratio test, LnL = -50.30. O- Open biome, C- Closed biome, D- Desertic biome, Cu- Cultivated. Species which grow as lianas are highlighted in pink. The geoxylic species is highlighted in blue. Photos: Jess Rickenback (open and closed), Gail Stott (desertic), Chiring Chandan (cultivated).

Discussion

We investigated the drivers of diversification in the widely distributed genus Ziziphus (Rhamnaceae). We found that at a global scale geography structures Ziziphus’ diversification more than ecology, as evidenced by higher phylogenetic signal in retention indices, delta statistic and biogeographic analyses in BioGeoBEARS. Multiple biome switches are associated with dispersal events, which structure Ziziphus’ diversification at smaller scales. We identified a strong association of habit and leaf hair with phylogenetic structure. We suggest that Ziziphus originated in closed biomes in Asia during the Oligocene (Fig. 3). In keeping with the younger geological age of the savanna biome and the Pliocene radiation of geoxylic species (Maurin et al. 2014), most of the youngest species in Ziziphus are found in open biomes, and the youngest habit is geoxylic.

How does geography and ecology structure Ziziphus’ phylogeny?

In situ diversification within different geographic regions predominates over extensive dispersal between continents in Ziziphus. The higher RI for geography relative to biome (Table 2) suggests diversification events tend to have been confined to continental scale geographical areas more than to tropical closed, open and desertic biomes. This finding is corroborated by the delta statistic (Borges 2019) which measures phylogenetic signal (Table 3). Whilst some authors have suggested that lineages retain their ancestral niche over time (Wiens and Donoghue 2014), that the Ziziphusphylogeny is structured by geography at the global scale is perhaps not surprising given that global biogeographic regionalizations tend to emphasise differences amongst continents (Wallace 1876; Holt et al. 2013; Morrone 2018). Geographical structure does not preclude ecology as a driver of diversification because biome switches can occur within continents.

The signal of geography and ecology as structuring diversification in Ziziphus is often conflated, as shown in African Ziziphus. Geographical structure is evident both from our analyses, and the phylogeny itself (Suppl. material 1: fig. S1) where the African species Z. zeyheriana, Z. abyssinica and Z. mucronata form a clade. However, these species are also united by biome. All African Ziziphus species occupy open or arid biomes with the clade comprising Z. zeyheriana, Z. abyssinica, and Z. mucronata restricted to savanna ecosystems, and the clade comprising Z. leucodermis (Baker) O.Schwartz, Z. spina-christi (L.) Desf., and Z. hamur Engl. found in desert and dry shrublands (Fig. 3b). These close relationships are supported by previous taxonomic work (Johnston 1972). The placement of Z. lotus could represent a long-persisting arid-adapted lineage. However, this placement is poorly supported (Suppl. material 1: fig. S1) and without other extant species from this lineage, exact dating of Z. lotus’ transition into African and Arabian arid ecosystems cannot be established. Similarly, Z. hajarensis is an extant representative of a lineage which transitioned into Arabian open biomes and at 8 mya is younger than the Z. zeyheriana clade’s biome shift into African open biomes. While incomplete sampling, reliance on one fossil, and the possibility that extinctions have erased evidence of past speciation (Meseguer et al. 2015) prohibits a definitive interpretation, of the 15 unsampled Ziziphus species, none occur in Africa (Suppl. material 1: table S5) suggesting that our inferences are relatively robust.

The reason for the conflation of geography and ecology in the diversification of African Ziziphusmay be because the establishment of African Ziziphus coincides with a period of major savanna expansion in Africa and the end of the geographic separation of the African continent from Asia (Fig. 3a and Suppl. material 1: fig. S6a) – presenting both geographical and ecological opportunities. The closure of the Tethys Sea 20–14 mya ended the 80 million year isolation of Africa and created new possibilities for dispersal (Torfstein and Steinberg 2020; Couvreur et al. 2021). Both the timing and the location of the land connection are similar to the timing and location of the majority of Ziziphus’ dispersals into Africa 14–23 mya (Fig. 3a). The Miocene was a period of increased cooling and drying as a result of global tectonic uplift, and the expansion of the Antarctic Sea ice (Kissling et al. 2012; Monthe et al. 2019; Couvreur et al. 2021). This led to the expansion of C4 grasses forming extensive savannas, which in Africa took place 10–15 mya (Davies et al. 2020). Changing biomes created opportunities for diversification, reflected in genera such as Manilkara Adans. (Armstrong et al. 2014), GuibourtiaBenn. (Tosso et al. 2018) and Acridocarpus Guill., Perr. & A.Rich. (Davis et al. 2002).

In contrast to the geographical and ecological unity of the African Ziziphusclade, we found no close relationship between Australian species Z. quadrilocularis F.Muell. and its geographically closest sampled species Z. suluensis (Borneo, the Philippines) and Z. oenopolia (Australia, New Guinea, Asia) (Suppl. material 1: fig. S1). This lack of close relationship of Australian species could indicate multiple dispersal events from Asia (Fig. 3a), a pattern also seen elsewhere in Themeda triandra Forssk. (Dunning et al. 2017; Dunning et al. 2022). Alternatively, patterns of disjunction could represent extinction events (Sanmartín and Meseguer 2016). The Australasian species also differ in ages – Z. suluensis and Z. oenopolia both evolved less than 5 mya, whereas Z. quadrilocularis evolved 17 mya – which suggests long distance dispersal (LDD) (Crisp and Cook 2007) although it is worth noting that Z. quadrilocularis’ position in the MCC tree is uncertain whereas Z. suluensis and Z. oenopolia are both well supported (Suppl. material 1: fig. S3). The recent dispersal of Z. oenopolia (4.5 my) to Australia is in line with Bridelia Willd. which dispersed 1–2 mya when islands between Asia and Australia were fully exposed (Li et al. 2009). Within Rhamnaceae LDD is not unusual; the island distribution in Phylica L. may be a result of LDD (Richardson et al. 2003) as is the Australian - New Zealand disjunction in Pomaderris Labill. (Nge et al. 2021), the Pacific coastal distribution of Colubrina Rich. ex Brongn. (Richardson et al. 2004), and the Andes-sub-Antarctic distribution of Ochetophila trinervis (Gillies ex Hook. & Arn.) Poepp. ex Endl. (Kalwij et al. 2019).

A west to east dispersal route from Asia to Australia is the most common direction of plant dispersal (Richardson et al. 2012; Crisp and Cook 2013; Richardson et al. 2014; Crayn et al. 2015). For most of the Miocene the Sunda shelf was exposed and biotic interchange with mainland Asia was frequent (Boer et al. 2015). The Sahul shelf (Australia and New Guinea) has a geological history independent of mainland Asia (Boer et al. 2015). New Guinea arose in the late Miocene but the majority of its landmass remained underwater until 5 mya (Hall 2002, Hall et al. 2005; Hall 2009; Li et al. 2009; Toussaint et al. 2014). Including the two endemic New Guinean species, Z. djamuensis Lauterb. and Z. papuana Lauterb., would help clarify not only the route into and age of the Australian species – whether dispersal was via New Guinea or whether the colonisation of New Guinea was via Australia – but also the relative structural forces of ecology and geography. Unfortunately, the preserved New Guinean specimens available to us are old, and yielded small amounts of low quality DNA. Collection of fresh material is therefore a priority.

While in many genera phylogenetic biome conservatism structures phylogenetic patterns (Yin et al. 2021), Ziziphus’ prevalence of biome switching is more in line with the within-continent diversification amongst biomes demonstrated by Hakea in Australia (Cardillo et al. 2017) and Coccinia Wight & Arn. in Africa (Holstein and Renner 2011). The origin of Ziziphus is a transition from the open biome characteristic of Paliurus to a closed biome (Fig. 3b). There are multiple switches in both directions between closed and open biomes; with two reversals into closed biomes (10 and 8 mya) and four major transitions into open biomes (at 23, 12, 8 and 3 mya) (Fig. 3b). The transition into the desertic biome is the least common, occurring only once at 23 mya. Biome conservatism amongst continents occurs less frequently than biome switching in Ziziphus. Biome conservatism suggests that if species or clades of species are found in the same biome but in disjunct geographical areas, that LDD events are responsible, and predominate in comparison to evolutionary switches amongst different biomes. This is seen in Z. oenopolia, a closed biome species which likely dispersed from Asia to Australasia (Fig. 3). Australia was dominated by rainforest during the Oligocene (Macphail 2007; Crisp and Cook 2013; Boer et al. 2015) with a warm humid climate continuing into the Miocene (Herold et al. 2011). This similarity of biomes may have made establishment relatively simple. However such long-distance dispersals within the same biome are comparatively rare in Ziziphus, although are seen in clades such as Guibourtia (Tosso et al. 2018) and the mimosoids (Bouchenak-Khelladi et al. 2010). The dispersals from Asia to Africa both involved biome switching; into the desertic biome (Z. hamur, Z. leucodermis, Z. spina-christi and Z. lotus (L.) Lam.) and into the open biome (Z. zeyheriana, Z. abyssinica, Z. mucronata, and Z. hajarensis). Similar patterns have been seen in Bombacoideae (Zizka et al. 2020) as well as across the tropical Americas (Antonelli et al. 2018).

Of the eight biome shifts detected in the data, half occurred from closed biomes into open biomes, with the closed biome acting as a source, despite multiple reversals. Ziziphus evolved repeatedly into the open biome and once into the desert, but always from the closed biome (Fig. 3b). There is only one instance of evolution from desertic to open biomes, with open biome species Z. oxyphlla Edgew. nested in the desertic Z. leucodermis clade (Fig. 3b). This biome shift is associated with a geographical transition from Africa and the Arabian Peninsula to Pakistan and Northwest India (Fig. 3).

How old are Ziziphus’ transitions into biomes- are open biome species younger than species in other biomes?

Open biome species in Ziziphus are generally younger than closed biome species. It seems probable that there is a link between Ziziphus’ Asian origin, and its ancestral biome being closed (Fig. 3). The Indochinese region is notable for its climate and biome stability dating from the Oligocene (Huang et al. 2022). The area was covered in lowland tropical evergreen forest, a floristic composition present today and which supports the largest number of Ziziphus species. In general, global climates in the Eocene - when Ziziphus evolved - were wetter and warmer (Armstrong et al. 2014), although open biomes in parts of China date to 18 mya (Ratnam et al. 2016). While ancestral biome reconstruction shows multiple switches from closed to open biomes (Fig. 3b), the savanna species in Ziziphus have some of the youngest ages, reflecting the globally young age of the open biome.

Do functional trait shifts coincide with biome or geographical shifts?

Although the ancestral state for Ziziphus is hairless leaves (Suppl. material 1: fig. S4), leaf hair likely confers advantages under different evolutionary pressures. Hairy leaves are adaptations both against herbivory and also to reduce water loss (Woodman and Fernandes 1991; Ripley et al. 1999). Leaf hair scored higher for RI than biome, habit, or any other functional traits (Table 2), suggesting it is a key synapomorphy. Hairy leaves are a shared character for the African Ziziphus clade Z. zeyheriana, Z. abyssinica and Z. mucronata and evolved independently in Z. nummularia, Z. hamur, and Z. hajarensis, all of which live in open and desertic biomes (Suppl. material 1: fig. S4). This association between traits and biome is notable in other genera; life forms in for example Crassula and Monsonia L. are adaptations to arid biomes (Garcia-Aloy et al. 2017; Lu et al. 2021). While hairy leaves are not a synapomorphy for any closed biome clades, hairy leaves are evolved independently within closed biome clades for example by the lianas Z. apetala, and Z. havilandii. Indeed, the frequency of the hairy leaf adaptation indicates it is relatively easy to evolve.

Spines provide no grouping structure for the phylogeny (RI=0) (Table 2) which, since almost all Ziziphusspecies are spinescent, suggests that unarmed species evolved independently. Of the four species in Ziziphus which lack spines (Rickenback, Pennington and Lehmann 2022), only two were included in our phylogeny (Suppl. material 1: fig. S1) – Z. pubinervis, a closed biome central Asian liana, and Z. glabrata, an open biome tree from the Arabian Peninsula. The geographical, ecological and phylogenetic separation of Z. pubinervis and Z. glabrata (B.Heyne ex Shult.) B.Heyne ex Wight & Arn. (Suppl. material 1: fig. S1) further confirms the independence of spine losses. In open biomes spinescence is associated with defences against mammalian herbivory (Charles-Dominique et al. 2016). However, spinescence confers some benefits in the closed biomes too, where many lianas use them for climbing (Schnitzer and Bongers 2002). The loss of spines in forest genera such as Schinus L. sect. Myrtifolia Silva-Luz & J.D. Mitch. suggests that there is a cost associated with spinescence (da Silva-Luz et al. 2019). We were unable to sequence the other spineless species, two of which are from the closed biomes and include Z. ridleyana Rasingam & Karthig, a rainforest tree native to Borneo. Including more of these spineless species in future phylogenetic studies would confirm whether losses are independent or connected, therefore allowing inferences to be made about spine evolution.

Consistency and retention analyses suggest fruit hair has evolved multiple times without providing much grouping information (Table 2). While there does appear to be some ancestral association between hairy fruits and lianas (Suppl. material 1: fig. S5), data deficiency in three out of eight species of one liana clade, and one out of two species of the other liana clade makes inference unreliable. Of the eight liana species not included in the phylogeny (Suppl. material 1: table S5), two are data deficient, two have hairless fruit, and four have hairy fruit (Rickenback, Pennington and Lehmann 2022). Hairiness in fruit may reduce water loss, and protect against irradiation and pathogens (Fernandez et al. 2011). Lianas are mechanistically adapted to minimize water loss through deep roots and efficient vascular systems (Schnitzer 2005). If there is an association of hairy fruit with liana species it may form part of a suite of traits related to minimizing seasonal water stress.

Ziziphus demonstrates clear associations between biome and habit. We found that habit has a low CI (0.17) but one of the highest RIs (0.56) (Table 2) suggesting that habit is a trait which has changed multiple times independently, but that provides substantial grouping information. Liana is the dominant form of Ziziphus in closed biomes (Fig. 3b) where 18 of 32 closed biome species are lianas. Furthermore, one of the two biome shifts back into closed biomes at 10 mya is associated with a secondary evolution of the liana form (clade Z. kunstleri, Z. affinis, Z. attopensis, Z. elegans, Z. suluensis, Z. havilandii, and Z. calophylla) (Fig. 3b). Lianas are adapted to life in closed biomes, utilizing tree trunks and canopy to support a quest for light (Schnitzer and Bongers 2002). The tree and shrub forms are found in both open and closed biomes, mirroring the occasional physiognomic overlap of those two woody forms (Götmark et al. 2016; Rickenback et al. 2022). The geoxylic form evolved only once (Z. zeyheriana), in the open savannas of Africa, and is the youngest habit to evolve at 3 mya (Fig. 3b). This replicates patterns seen in other geoxylic congeners of rainforest species, where geoxylic radiations peaked around 2.3 mya reflecting the rise of woody fire-adapted species (Simon et al. 2009; Simon and Pennington 2012; Maurin et al. 2014).

Conclusion

At a global scale geography structures Ziziphus’ diversification more than ecology, as evidenced by higher phylogenetic signal in retention indices, delta statistic, and biogeographic analyses in BioGeoBEARS. Multiple ecological switches among biomes are associated with dispersal events, which structure Ziziphus’ diversification at smaller scales. Our findings suggest that Ziziphus evolved out of closed biomes in Asia during the Oligocene. Long distance dispersal from Asia to Africa resulted in the occupation of two novel biomes, the expanding open grasslands and savannas, and the desertic biome. Based on fossil dating the evolution into these biomes was associated with a global increase of aridity and an expansion of the grassy biome, creating opportunities for diversification. Transitions across continents but within-biome are far fewer but are demonstrated by a dispersal from Asia to Australia, which maintained occupation of the closed biome. Ziziphusevolved out of closed biomes with younger species found in open biomes, in keeping with the relative ages of these biomes. Hairy leaves are associated with transitions into open and desertic biomes, while one of two transitions back into closed biomes is associated with the development of the liana habit. The dispersal to Africa was also the driver for diversification of habit, with the development of the geoxylic life form around 3 mya. In general, a more complete phylogeny would increase the robustness of these results, and future work should prioritise incorporating endemic New Guinean species. Further, the incorporation of more spineless Ziziphus species into the phylogeny would enable inferences to be drawn regarding the costliness of spiney resource allocation, and whether loss of spinescence in species is independent or related.

Acknowledgements

JR gratefully acknowledges financial support both from the SYNTHESYS Project http://www.synthesys.info/ which is financed by European Community Research Infrastructure Action under the FP7 “Capacities” Program, and from the Davis Expedition Fund at the University of Edinburgh. CL received funding from NERC under the project ‘How do global change and functional traits influence savannah woody plant encroachment?’ (NE/T000759/1).

We are particularly grateful to the herbariums at RBGE, RBGK, and UCPH for facilitating sampling of their collections. We received specimen assistance from Olof Ryding at UCPH, Suzanne Cubey at RBGE, and Melissa Islam. Help with species identifications was provided by Sabine Knees and Mark Newman at RBGE, and Tim Utteridge and Daniel Cahen at RBGK. Michelle Hart and Laura Forrest at RBGE supported lab work. Guidance on analyses was provided by Edeline Gagnon, with support on divergence time estimates from Renske Onstein.

For the purposes of Open Access, the author has applied a Creative Commons Attribution (CC-BY) Licence to any Author Accepted Manuscript version arising from this submission.

Author Contributions

JR, TP and CL designed the research question. JR performed the research. FH contributed data and sequencing guidance. JR analysed the data with support from FP, MH, TP. JR wrote the manuscript with support from TP and CL. All authors read and approved the manuscript.

Data Availability Statement

All genetic sequences generated for this study are available on GenBank (accessions PP855017 to PP855058 and PP934635 to PP934658). Trait data are available as Suppl. material 1: tables S3, S4, and previously published as part of Rickenback, Pennington and Lehmann (2022).

References

Akaike H (1974) A new look at the statistical model identification. IEEE Transactions on Automatic Control 19: 716–723. https://doi.org/10.1109/TAC.1974.1100705

Angyalossy V, Angeles G, Pace MR, Lima AC, Dias-Leme CL, Lohmann LG, Madero-Vega C (2012) An overview of the anatomy, development and evolution of the vascular system of lianas. Plant Ecology and Diversity 5: 167–182. https://doi.org/10.1080/17550874.2011.615574

Antonelli A, Zizka A, Silvestro D, Scharn R, Cascales-Miñana B, Bacon CD (2015) An engine for global plant diversity, highest evolutionary turnover and emigration in the American tropics. Frontiers in Genetics 6: 130. https://doi.org/10.3389/fgene.2015.00130

Antonelli A, Zizka A, Carvalho FA, Scharn R, Bacon CD, Silvestro D, Condamine FL (2018) Amazonia is the primary source of Neotropical biodiversity. Proceedings of the National Academy of Sciences USA 115: 6032–6039. https://doi.org/10.1073/pnas.1713819115

Armbruster WS (1997) Exaptations link evolution of plant-herbivore and plant pollinator interactions: a phylogenetic inquiry. Ecology 78: 1661–1672. https://doi.org/10.1890/0012-9658(1997)078[1661:ELEOPH]2.0.CO;2

Armstrong KE, Stone GN, Nicholls JA, Valderrama E, Anderberg AA, Smedmark J, Gautier L, Naciri Y, Milne R, Richardson JE (2014) Patterns of diversification amongst tropical regions compared: a case study in Sapotaceae. Frontiers in Genetics 5: 362. https://doi.org/10.3389/fgene.2014.00362

Bateman RM, Crane PR, DiMichele WA, Kenrick PR, Rowe NP, Speck T, Stein WE (1998) Early evolution of land plants: phylogeny, physiology, and ecology of the primary terrestrial radiation. Annual Review of Ecology and Systematics 29: 263–292. https://doi.org/10.1146/annurev.ecolsys.29.1.263

Boer HJ de, Steffen K, Cooper WE (2015) Sunda to Sahul dispersals in Trichosanthes (Cucurbitaceae): a dated phylogeny reveals five independent dispersal events to Australasia. Journal of Biogeography 42: 519–531. https://doi.org/10.1111/jbi.12432

Bond W (2019) Open ecosystems: ecology and evolution beyond the forest edge. Oxford University Press, 192 pp. https://doi.org/10.1093/oso/9780198812456.001.0001

Borges R, Machado JP, Gomes C, Rocha AP, Antunes A (2019) Measuring phylogenetic signal between categorical traits and phylogenies. Bioinformatics 35: 1862–1869. https://doi.org/10.1093/bioinformatics/bty800

Bouchenak-Khelladi Y, Maurin O, Hurter J, Van der Bank M (2010) The evolutionary history and biogeography of Mimosoideae (Leguminosae): an emphasis on African acacias. Molecular Phylogenetics and Evolution 57: 495–508. https://doi.org/10.1016/j.ympev.2010.07.019

Burge DO, Manchester SR (2008) Fruit morphology, fossil history, and biogeography of Paliurus (Rhamnaceae). International Journal of Plant Sciences 169: 1066–1085. https://doi.org/10.1086/590453

Calitz W, Potts AJ, Cowling RM (2015) Investigating species-level flammability across five biomes in the Eastern Cape, South Africa. South African Journal of Botany 101: 32–39. https://doi.org/10.1016/j.sajb.2015.07.005

Cancio I, Gonzalez-Robles A, Bastida JM, Isla J, Manzaneda AJ, Salido T, Rey PJ (2017) Landscape degradation affects red fox (Vulpes vulpes) diet and its ecosystem services in the threatened Ziziphus lotus scrubland habitats of semiarid Spain. Journal of Arid Environments 145: 24–34. https://doi.org/10.1016/j.jaridenv.2017.05.004

Cardillo M, Weston PH, Reynolds ZKM, Olde PM, Mast AR, Lemmon EM, Lemmon AR, Bromham L (2017) The phylogeny and biogeography of Hakea (Proteaceae) reveals the role of biome shifts in a continental plant radiation. Society for the Study of Evolution 8: 1928–1943. https://doi.org/10.1111/evo.13276

Carvalho MR, Jaramillo C, de la Parra F, Caballero-Rodríguez D, Herrera F, Wing S, Turner BL, D'Apolito C, Romero-Báez M, Narváez P, Martínez C, Gutierrez M, Labandeira C, Bayona G, Rueda M, Paez-Reyes M, Cárdenas D, Duque Á, Crowley JL, Santos C, Silvestro D (2021) Extinction at the end-Cretaceous and the origin of modern Neotropical rainforests. Science 372: 63–68. https://doi.org/10.1126/science.abf1969

Chandan C (2021) Ziziphus jujuba or common plum (বগৰী) is being dried in sunlight on a Dola (bamboo equipment) in Assam. https://commons.m.wikimedia.org/wiki/File:Ziziphus_Jujuba_or_Common_plum_%28%E0%A6%AC%E0%A6%97%E0%A7%B0%E0%A7%80%29_is_being_dried.jpg [Accessed 22.01.2024]

Charles-Dominique T, Staver AC, Midgley GF, Bond WJ (2015) Functional differentiation of biomes in an African savanna/forest mosaic. South African Journal of Botany 101: 82–90. https://doi.org/10.1016/j.sajb.2015.05.005

Charles-Dominique T, Davies TJ, Hempson GP, Bezeng BS, Daru BH, Kabongo RM, Maurin O, Muasya AM, van der Bank M, Bond WJ (2016) Spiny plants, mammal browsers, and the origin of African savannas. Proceedings of the National Academy of Sciences USA 113: E5572–E5579. https://doi.org/10.1073/pnas.1607493113

Chen, S, Yao H, Han J, Liu C, Song J, Shi L, Zhu Y, Ma X, Gao T, Pang X, Luo K, Li Y, Li X, Jia X, Lin Y, Leon C (2010) Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species. PLoS ONE 5: p.e8613. https://doi.org/10.1371/journal.pone.0008613

Chen Y-S, Meseguer AS, Godefroid M, Zhou Z, Zhang J-W, Deng T, Kim J-H, Nie Z-L, Liu Y-S, Sun H (2017) Out-of-India dispersal of Paliurus (Rhamncaeae) indicated by combined molecular phylogenetic and fossil evidence. Taxon 66: 78–90. https://doi.org/10.12705/661.4

Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW (2016) GenBank. Nucleic Acids Research 44: 67–72. https://doi.org/10.1093/nar/gkv1276

Coley PD, Endara M-J, Ghabash G, Kidner CA, Nicholls JA, Pennington RT, Mills AG, Soule AJ, Lemes MR, Stone GN, Kursar TA (2019) Macroevolutionary patterns in overexpression of tyrosine: An anti-herbivore defence in a speciose tropical tree genus, Inga (Fabaceae). Journal of Ecology 107: 1620–1632. https://doi.org/10.1111/1365-2745.13208

Correa E, Jaramillo C, Manchester S, Gutierrez M (2010) A fruit and leaves of Rhamnaceous affinities from the late Cretaceous (Maastrichtian) of Colombia. American Journal of Botany 97: 71–79. https://doi.org/10.3732/ajb.0900093

Couvreur TLP, Kissling WD, Condamine FL, Svenning J-C, Rowe NP, Baker WJ (2015) Global diversification of a tropical plant growth form: environmental correlates and historical contingencies in climbing palms. Frontiers in Genetics 5: 103552. https://doi.org/10.3389/fgene.2014.00452

Couvreur TL, Dauby G, Blach‐Overgaard A, Deblauwe V, Dessein S, Droissart V, Hardy OJ, Harris DJ, Janssens SB, Ley AC, Mackinder BA, Sonké B, Sosef MSM, Stévart T, Svenning JC, Wieringa JJ, Faye A, Missoup AD, Tolley KA, Nicolas V, Ntie S, Fluteau F, Robin C, Guillocheau F, Barboni D, Sepulchre P (2021) Tectonics, climate and the diversification of the tropical African terrestrial flora and fauna. Biological Reviews 96: 16–51. https://doi.org/10.1111/brv.12644

Cox B (2001) The biogeographic regions reconsidered. Journal of Biogeography 28: 511–523. https://doi.org/10.1046/j.1365-2699.2001.00566.x

Crayn DM, Costion C, Harrington MG (2015) The Sahul-Sunda floristic exchange, dated molecular phylogenies document Cenozoic intercontinental dispersal dynamics. Journal of Biogeography 42: 11–24. https://doi.org/10.1111/jbi.12405

Crisp M, Cook L, Steane D (2004) Radiation of the Australian flora, what can comparisons of molecular phylogenies across multiple taxa tell us about the evolution of diversity in present-day communities? Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 359: 1551–1571. https://doi.org/10.1098/rstb.2004.1528

Crisp MD, Cook LG (2007) A congruent molecular signature of vicariance across multiple plant lineages. Molecular Phylogenetics and Evolution 43: 1106–1117. https://doi.org/10.1016/j.ympev.2007.02.030

Crisp MD, Arroyo MT, Cook LG, Gandolfo MA, Jordan GJ, McGlone MS, Weston PH, Westoby M, Wilf P, Linder HP (2009) Phylogenetic biome conservatism on a global scale. Nature 458: 754–756. https://doi.org/10.1038/nature07764

Crisp MD, Cook LG (2013) How was the Australian flora assembled over the last 65 million years? A molecular phylogenetic perspective. Annual Review of Ecology, Evolution, and Systematics 44: 303–324. https://doi.org/10.1146/annurev-ecolsys-110512-135910

da Silva-Luz CL, Pirani JR, Mitchell JD, Daly D, do Valle Capelli N, Demarco D, Pell SK, Plunkett GM (2019) Phylogeny of Schinus L. (Anacardiaceae) with a new infrageneric classification and insights into evolution of spinescence and floral traits. Molecular Phylogenetics and Evolution 133: 302–351. https://doi.org/10.1016/j.ympev.2018.10.013

Dantas VDL, Pausas JG (2013) The lanky and the corky: fire-escape strategies in savanna woody species. Journal of Ecology 101: 1265–1272. https://doi.org/10.1111/1365-2745.12118

Dantas VL, Pausas JG (2020) Megafauna biogeography explains plant functional trait variability in the tropics. Global Ecology and Biogeography 29: 1288–1298. https://doi.org/10.1111/geb.13111

Darbyshire I, Kiel CA, Astroth CM, Dexter KG, Chase FM, Tripp EA (2020) Phylogenomic study of Monechma reveals two divergent plant lineages of ecological importance in the African savanna and succulent biomes. Diversity 12: 237. https://doi.org/10.3390/d12060237

Darriba D, Taboada GL, Doallo R, Posada D (2012) jModelTest 2: More models, new heuristics and parallel computing. Nature Methods 9: 772. https://doi.org/10.1038/nmeth.2109

Davies TJ, Daru BH, Bezeng BS, Charles-Dominique T, Hempson GP, Kabongo RM, Maurin O, Muasya AM, van der Bank M, Bond WJ (2020) Savanna tree evolutionary ages inform the reconstruction of the paleoenvironment of our hominin ancestors. Scientific Reports 10: 1–8. https://doi.org/10.1038/s41598-020-69378-0

Davis CC, Bell CD, Fritsch PW, Mathews S (2002) Phylogeny of Acridocarpus-Brachylophon (Malpighiaceae), Implications for tertiary tropical floras and Afroasian biogeography. Evolution 56: 2395–2405. https://doi.org/10.1111/j.0014-3820.2002.tb00165.x

Davis CC, Webb CO, Wurdack KJ, Jaramillo CA, Donoghue MJ (2005) Explosive radiation of Malpighiales supports a mid‐Cretaceous origin of modern tropical rain forests. The American Naturalist 165: E36–E65. https://doi.org/10.1086/428296

De Groot GA, During HJ, Maas JW, Schneider H, Vogel JC, & Erkens RHJ (2011) Use of rbcl and trnL-F as a two-locus DNA barcode for identification of NW-European ferns: An ecological perspective. PLoS ONE 6: e16371. https://doi.org/10.1371/journal.pone.0016371

Donkpegan AS, Doucet JL, Hardy OJ, Heuertz M, Piñeiro R (2020) Miocene diversification in the Savannahs precedes tetraploid rainforest radiation in the African tree genus Afzelia (Detarioideae, Fabaceae). Frontiers in Plant Science 11: 798. https://doi.org/10.3389/fpls.2020.00798

Donoghue MJ, Edwards EJ (2014) Biome shifts and niche evolution in plants. Annual Review of Ecology, Evolution, and Systematics 45: 547–572. https://doi.org/10.1146/annurev-ecolsys-120213-091905

Douzery EJP, Pridgeon AM, Kores P, Linder HP, Kurzweil H, Chase MW (1999) Molecular phylogenetics of disease (Orchidaceae): A contribution from nuclear ribosomal ITS sequences. American Journal of Botany 86: 887–899. https://doi.org/10.2307/2656709

Dunning LT, Liabot AL, Olofsson JK, Smith EK, Vorontsova MS, Besnard G, Simpson KJ, Lundgren MR, Addicott E, Gallagher RV, Chu Y, Pennington RT, Christin PA, Lehmann CER (2017) The recent and rapid spread of Themeda triandra. Botany Letters 164: 327–337. https://doi.org/10.1080/23818107.2017.1391120

Dunning LT, Olofsson JK, Papadopulos AS, Hibdige SG, Hidalgo O, Leitch IJ, Baleeiro PC, Ntshangase S, Barker N, Jobson RW (2022) Hybridisation and chloroplast capture between distinct Themeda triandra lineages in Australia. Molecular Ecology 31: 5846–5860. https://doi.org/10.1111/mec.16691

Echeverria-Londono S, Sarkinen T, Fenton IS, Purvis A, Knapp S (2020) Dynamism and context-dependency in diversification of the megadiverse plant genus Solanum (Solanaceae). Journal of Systematics and Evolution 58: 767–782. https://doi.org/10.1111/jse.12638

Edwards EJ, Smith SA, Donoghue MJ (2010) Phylogenetic analyses reveal the shady history of C4 grasses. Proceedings of the National Academy of Sciences USA 107: 2532–2537. https://doi.org/10.1073/pnas.0909672107

Eiserhardt WL, Couvreur TLP, Baker WJ (2017) Plant phylogeny as a window on the evolution of hyperdiversity in the tropical rainforest biome. New Phytologist 214: 1408–1422. https://doi.org/10.1111/nph.14516

Gagnon E, Ringelberg JJ, Bruneau A, Lewis GP, Hughes CE (2019) Global succulent biome phylogenetic conservatism across the pantropical Casealpinia Group (Leguminosae). New Phytologist 222: 1994–2008. https://doi.org/10.1111/nph.15633

Fan P-F, Fei H-L, Scott MB, Zhang W, Ma C-Y (2011) Habitat and food choice of the critically endangered cao vit gibbon (Nomascus nasutus) in China: Implications for conservation. Biological Conservation 144: 2247–2254. https://doi.org/10.1016/j.biocon.2011.05.016

Fernández V, Khayet M, Montero-Prado P, Heredia-Guerrero JA, Liakopoulos G, Karabourniotis G, Del Río V, Domínguez E, Tacchini I, Nerín C, Val J, Heredia A (2011) New insights into the properties of pubescent surfaces: Peach fruit as a model. Plant Physiology 156: 2098–2108. https://doi.org/10.1104/pp.111.176305

Garcia-Aloy S, Sanmartin I, Kadereit G, Vitales D, Millanes AM, Roquet C, Vargas P, Alarcon M, Aldasoro JJ (2017) Opposite trends in the genus Monsonia (Geraniaceae): Specialization in the African deserts and range expansions throughout eastern Africa. Scientific Reports 7: 9872. https://doi.org/10.1038/s41598-017-09834-6

Gao Q-H, Wu C-S, Wang M (2013) The Jujube (Ziziphus jujuba Mill) Fruit: a review of current knowledge of fruit composition and health benefits. Journal of Agricultural and Food Chemistry 61: 3351–3363. https://doi.org/10.1021/jf4007032

Gianolo E (2004) Evolution of a climbing habit promotes diversification in flowering plants. Proceedings of the Royal Society of London B 271: 2011–2015. https://doi.org/10.1098/rspb.2004.2827

Gomes AL, Revermann R, Meller P, Goncalves FM, Aidar MP, Lages F, Finckh M (2021) Functional traits and symbiotic associations of geoxyles and trees explain the dominance of detarioid legumes in miombo ecosystems. New Phytologist 230: 510–520. https://doi.org/10.1111/nph.17168

Götmark F, Götmark E, Jensen AM (2016) Why be a shrub? A basic model and hypotheses for the adaptive values of a common growth form. Frontiers in Plant Science 7: 1095. https://doi.org/10.3389/fpls.2016.01095

Hall TA (1999) BioEdit, a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acid Symposium Series 41: 95–98.

Hall, R (2002) Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific, computer-based reconstructions, model and animations. Journal of Asian earth sciences 20: 353–431. https://doi.org/10.1016/S1367-9120(01)00069-4

Hall R (2009) Southeast Asia’s changing palaeogeography. Blumea-Biodiversity, Evolution and Biogeography of Plants 54: 148–161. https://doi.org/10.3767/000651909X475941

Hammer T, Davis R, Thiele K (2015) A molecular framework phylogeny for Ptilotus (Amaranthaceae), evidence for the rapid diversification of an arid Australian genus. Taxon 64: 272–285. https://doi.org/10.12705/642.6

Hauenschild F, Matuszak S, Muellner-Riehl AN, Favre A (2016a) Phylogenetic relationships within the cosmopolitan buckthorn family (Rhamnaceae) support the resurrection of Sarcomphalus and the description of Pseudoziziphus gen.nov. Taxon 65: 47–64. https://doi.org/10.12705/651.4

Hauenschild F, Favre A, Salazar GA, Muellner-Riehl AN (2016b) Analysis of the cosmopolitan buckthorn genera Frangula and Rhamnus s.l. supports the description of a new genus Ventia. Taxon 65: 65–78. https://doi.org/10.12705/651.5

Hauenschild F, Favre A, Schnitzler J, Michalak I, Freiberg M, Muellner-Riehl AN (2017) Spatio-temporal evolution of Allium L. in the Qinghai-Tibet-Plateau region: Immigration and in situ radiation. Plant Diversity 39: 167–179. https://doi.org/10.1016/j.pld.2017.05.010

Hauenschild F, Favre A, Michalak I, Muellner‐Riehl AN (2018) The influence of the Gondwanan breakup on the biogeographic history of the ziziphoids (Rhamnaceae). Journal of Biogeography 45: 2669–2677. https://doi.org/10.1111/jbi.13448

Herold N, Huber M, Greenwood DR, Muller RD, Seton M (2011) Early to Middle Miocene monsoon climate in Australia. Geology 39: 3–6. https://doi.org/10.1130/G31208.1

Higgins SI, Buitenwerf R, Moncrieff G (2016) Defining functional biomes and monitoring their change globally. Global Change Biology 22: 3583–3593. https://doi.org/10.1111/gcb.13367

Ho SY, Phillips MJ (2009) Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Systematic Biology 58: 367–380. https://doi.org/10.1093/sysbio/syp035

Holstein N, Renner SS (2011) A dated phylogeny and collection records reveal repeated biome shift in the African genus Coccinia (Cucurbitaceae). BMC Evolutionary Biology 11: 1–16. https://doi.org/10.1186/1471-2148-11-28

Holt BG, Lessard J-P, Borregaard MK, Fritz SA, Araújo MB, Dimitrov D, Fabre PH, Graham CH, Graves GR, Jønsson KA, Nogués-Bravo D, Wang Z, Whittaker RJ, Fjeldså J, Rahbek C (2013) An Update of Wallace’s Zoogeographic Regions of the World. Science 339: 74–78. https://doi.org/10.1126/science.122828

Huang J, Spicer RA, Li S-F, Liu J, Van Do T, Ba Nguyen H, Zhou Z-K, Su T (2022) Long-term floristic and climatic stability of northern Indochina, Evidence from the Oligocene Ha Long flora, Vietnam. Palaeogeography, Palaeoclimatology, Palaeoecology 593: 110930. https://doi.org/10.1016/j.palaeo.2022.110930

Huelsenbeck JP, Ronquist F (2001) Mr Bayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. https://doi.org/10.1093/bioinformatics/17.8.754

Islam MB, Simmons MP (2006) A thorny dilemma, testing alternative intrageneric classifications within Ziziphus (Rhamnaceae). American Society of Plant Taxonomists 31: 826–842. https://doi.org/10.1600/036364406779695997

Johnston MC (1972) Rhamnaceae. In: Milne-Redhead E, Polhill RM (Eds) Flora of Tropical East Africa. Royal Botanic Gardens, Kew, 40 pp.

Kalwij JM, Medan D, Kellermann J, Greve M, Chown SL (2019) Vagrant birds as a dispersal vector in transoceanic range expansion of vascular plants. Scientific Reports 9: 4655. https://doi.org/10.1038/s41598-019-41081-9

Keith DA, Ferrer-Paris JR, Nicholson E, Kingsford RT (2020) The IUCN global ecosystem typology 2.0: Descriptive profiles for biomes and ecosystem functional groups. IUCN, Gland. https://doi.org/10.2305/IUCN.CH.2020.13.en

Kissling WD, Eiserhardt WL, Baker WJ, Borchsenius F, Couvreur TLP, Balslev H, Svenning J-C (2012) Cenozoic imprints on the phylogenetic structure of palm species assemblages worldwide. Proceedings of the National Academy of Sciences USA 109: 7379–7384. https://doi.org/10.1073/pnas.1120467109

Kitching IJ, Forey PL, Humphries CJ, Williams DM (1998) Cladistics: the theory and practice of parsimony analysis, 2^nd edn. Oxford University Press, Oxford.

Kluge AG, Farris JS (1969) Quantitative phyletics and the evolution of anurans. Systematic Zoology 18: 1–32. https://doi.org/10.2307/2412407

Kong H, Condamine FL, Yang L, Harris AJ, Feng C, Wen F, Kang M (2022) Phylogenomic and macroevolutionary evidence for an explosive radiation of a plant genus in the Miocene. Systematic Biology 71: 589–609. https://doi.org/10.1093/sysbio/syab068

Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG (2007) ClustalW and ClustalX version 2. Bioinformatics 23: 2947–2948. https://doi.org/10.1093/bioinformatics/btm404

Li Y, Dressler S, Zhang D, Renner SS (2009) More Miocene dispersal between Africa and Asia – the case of Bridelia (Phyllanthaceae). Systematic Botany 34: 521–529. https://doi.org/10.1600/036364409789271263

Lu M, Fradera-Soler M, Forest F, Barraclogh TG, Grace OM (2021) Evidence linking life-form to a major shift in diversification rate in Crassula. American Journal of Botany 109: 272–290. https://doi.org/10.1002/ajb2.1797

Macphail M (2007) Australian palaeoclimates, Cretaceous to Tertiary a review of palaeobotanical and related evidence to the year 2000. Cooperative Research Centre for Landscape Environments and Mineral Exploration, Australia.

Magallon S, Sanchez-Reyes LL, Gomez-Acevedo SL (2019) Thirty clues to the exceptional diversification of flowering plants. Annals of Botany 123: 491–503. https://doi.org/10.1093/aob/mcy182

Matzke NJ (2013) Probabilistic historical biogeography, new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. Frontiers of Biogeography 5: 242–248. https://doi.org/10.21425/F5FBG19694

Matzke NJ (2014) Model Selection in Historical Biogeography Reveals that Founder-event Speciation is a Crucial Process in Island Clades. Systematic Biology 63: 951–970. https://doi.org/10.1093/sysbio/syu056

Maurin O, Davies TJ, Burrows JE, Daru BH, Yessoufou K, Muasya AM, van der Bank M, Bond WJ (2014) Savanna fire and the origins of the ‘underground forests’ of Africa. New Phytologist 204: 201–214. https://doi.org/10.1111/nph.12936

The Mesquite Project Team (2021) Documentation for Mesquite, a modular system forevolutionary analysis. http://www.mesquiteproject.org

Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Proceedings of the Gateway Environments Workshop (GCE), New Orleans 1–8. https://doi.org/10.1109/GCE.2010.5676129

Mitchell N, Carlson JE, Holsinger KE (2018) Correlated evolution between climate and suites of traits along a fast-slow continuum in the radiation of Protea. Ecology and Evolution 8: 1853–1866. https://doi.org/10.1002/ece3.3773

Monthe FK, Migliore J, Duminil J, Bouka G, Demenou BB, Doumenge C, Blanc-Jolivet C, Mensah Ekue MR, Hardy OJ (2019) Phylogenetic relationships in two African Cedreloideae tree genera (Meliaceae) reveal multiple rain/dry forest transitions. Perspectives in Plant Ecology, Evolution and Systematics 37: 1–10. https://doi.org/10.1016/j.ppees.2019.01.002

Morrone JJ (2018) The spectre of biogeographical regionalization. Journal of Biogeography 45: 282–288. https://doi.org/10.1111/jbi.13135

Mucina L (2019) Biome: evolution of a crucial ecological and biogeographical concept. New Phytologist 222: 97–114. https://doi.org/10.1111/nph.15609

Nascimento LFD, Guimarães PR, Onstein RE, Kissling WD, Pires MM (2020) Associated evolution of fruit size, fruit colour and spines in Neotropical palms. Journal of Evolutionary Biology 33: 858–868. https://doi.org/10.1111/jeb.13619

Nge FJ, Biffin E, Thiele KR, Waycott M (2020) Extinction pulse at Eocene-Oligocene boundary drives diversification dynamics of two Australian temperate floras. Proceedings of the Royal Society B, 287: 20192546. https://doi.org/10.1098/rspb.2019.2546

Nge FJ, Kellermann J, Biffin E, Waycott M, Thiele KR (2021) Historical biogeography of Pomaderris (Rhamnaceae): Continental vicariance in Australia and repeated independent dispersals to New Zealand. Molecular Phylogenetics and Evolution 158: 107085. https://doi.org/10.1016/j.ympev.2021.107085

Onstein RE, Carter RJ, Xing Y, Linder HP (2014) Diversification rate shifts in the Cape Floristic Region, The right traits in the right place at the right time. Perspectives in Plant Ecology, Evolution and Systematics 16: 331–340. https://doi.org/10.1016/j.ppees.2014.08.002

Onstein RE, Carter RK, Xing Y, Richardson JE, Linder HP (2015) Do Mediterranean-type ecosystems have a common history? Insights from the Buckthorn family (Rhamnaceae). Evolution 69: 756–771. https://doi.org/10.1111/evo.12605

Orwa C, Mutua A, Kindt R, Jamnadass R, Simons A (2009) Agroforestree Database, a tree reference and selection guide version 4.0. http://www.worldagroforestry.org/af/treedb/

Paradis E, Schliep K (2019) ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics 35: 526–528. https://doi.org/10.1093/bioinformatics/bty633

Pennington RT, Hughes CE (2014) The remarkable congruence of New and Old World savanna origins. New Phytologist 204: 4–6. https://doi.org/10.1111/nph.12996

Pennington RT, Lavin M (2015) The contrasting nature of woody plant species in different neotropical forest biomes reflects differences in ecological stability. New Phytologist 210: 25–37. https://doi.org/10.1111/nph.13724

Pezzini FF, Dexter KG, de Carvalho-Sobrinho JG, Kidner CA, Nicholls JA, de Queiroz LP, Pennington RT (2021) Phylogeny and biogeography of Ceiba Mill. (Malvaceae, Bombacoideae). Frontiers of Biogeography 13: e49226 https://doi.org/10.21425/F5FBG49226

Poorter L, Rose SA (2005) Light-dependent changes in the relationship between seed mass and seedling traits: a meta-analysis for rain forest tree species. Ecophysiology 142: 378–387. https://doi.org/10.1007/s00442-004-1732-y

R Core Team (2021) R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA (2018) Posterior summarisation in Bayesian phylogenetics using Tracer 1.7. Systematic Biology 67: 901–904. https://doi.org/10.1093/sysbio/syy032

Ratnam J, Tomlinson KW, Rasquinha DN, Sankaran M (2016) Savannahs of Asia: Antiquity, biogeography, and an uncertain future. Philosophical Transactions of the Royal Society B: Biological Sciences 371: 20150305. https://doi.org/10.1098/rstb.2015.0305

Raven PH, Gereau RE, Phillipson PB, Chatelain C, Jenkins CN, Ulloa CU (2020) The distribution of biodiversity richness in the tropics. Science Advances 6: eabc6228. https://doi.org/10.1126/sciadv.abc6228

Revell LJ (2012) phytools, An R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution 3: 217–223. https://doi.org/10.1111/j.2041-210X.2011.00169.x

Richardson JE, Fay MF, Cronk QCB, Chase MW (2003) Species delimitation and the origin of populations in island representatives of Phylica (Rhamnaceae). Society for the Study of Evolution 57: 816–827. https://doi.org/10.1111/j.0014-3820.2003.tb00293.x

Richardson JE, Costion CM, Muellner AN (2012) The Malesian floristic interchange, plant migration patterns across Wallace’s Line. In: Gower D, Johnson K, Richardson J, Rosen B, Ruber L, Williams S (Eds) Biotic evolution and environmental change in southeast Asia. Systematics Association Special Volume Series. Cambridge University Press, UK, 138–163. https://doi.org/10.1017/CBO9780511735882.008

Richardson JE, Bakar AM, Tosh J, Armstrong K, Smedmark J, Anderberg AA, Slik F, Wilkie P (2014) The influence of tectonics, sea-level changes and dispersal on migration and diversification of Isonandreae (Sapotaceae). Botanical Journal of the Linnean Society 174: 130–140. https://doi.org/10.1111/boj.12108

Rickenback J, Pennington RT, Lehmann CER (2022) Diversity in habit expands the environmental niche of Ziziphus (Rhamnaceae). Biotropica 54: 1285–1299. https://doi.org/10.1111/btp.13152

Ritchie AM, Lo N, Ho SY (2017) The impact of the tree prior on molecular dating of data sets containing a mixture of inter-and intraspecies sampling. Systematic Biology 66: 413–425. https://doi.org/10.1093/sysbio/syw095

Ripley BS, Pammenter NW, Smith VR (1999) Function of leaf hairs revisited: the hair layer on leaves Arctotheca populifolia reduces photoinhibition, but leads to higher leaf temperatures caused by lower transpiration rates. Journal of Plant Physiology 155: 78–85. https://doi.org/10.1016/S0176-1617(99)80143-6

Ronquist F, Teslenko M, Van Der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61: 539–542. https://doi.org/10.1093/sysbio/sys029

Sanmartín I, Meseguer AS (2016) Extinction in phylogenetics and biogeography, from timetrees to patterns of biotic assemblage. Frontiers in Genetics 7: 35. https://doi.org/10.3389/fgene.2016.00035

Sarver BA, Pennell MW, Brown JW, Keeble S, Hardwick KM, Sullivan J, Harmon LJ (2019) The choice of tree prior and molecular clock does not substantially affect phylogenetic inferences of diversification rates. PeerJ 7: e6334. https://doi.org/10.7717/peerj.6334

Schnitzer SA (2005) A mechanistic explanation for global patterns of liana abundance and distribution. The American Naturalist 166: 262–276. https://doi.org/10.1086/431250

Schnitzer SA, Bongers F (2002) The ecology of lianas and their role in forests. Trends in Ecology and Evolution 17: 223–230. https://doi.org/10.1016/S0169-5347(02)02491-6

Schwarz G (1978) Estimating the dimension of a model. The Annals of Statistics 6: 461–464. https://doi.org/10.1214/aos/1176344136

Simon MF, Pennington T (2012) Evidence for adaptation to fire regimes in the tropical savannas of the Brazilian Cerrado. International Journal of Plant Sciences 173: 711–723. https://doi.org/10.1086/665973

Simon MF, Grether R, de Queiroz LP, Skema C, Pennington RT, Hughes CE (2009) Recent assembly of the Cerrado, a neotropical plant diversity hotspot, by in situ evolution of adaptations to fire. Proceedings of the National Academy of Sciences USA 106: 20359–20364. https://doi.org/10.1073/pnas.0903410106

Stamatakis A (2014) RAxML Version 8, A tool for Phylogenetic Analysis and Post-Analysis of Large Phylogenies. Bioinformatics 30: 1312–1313. https://doi.org/10.1093/bioinformatics/btu033

Suchard MA, Lemey P, Baele G, Ayres DL, Drummond AJ, Rambaut A (2018) Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10 Virus Evolution 4: vey016. https://doi.org/10.1093/ve/vey016

Taberlet P, Gielly L, Pautou G, Bouvet J (1991) Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105–1109. https://doi.org/10.1007/BF00037152

Tian Q, Stull GW, Kellermann J, Medan D, Nge FJ, Liu SY, Kates HR, Soltis DE, Soltis PS, Guralnick RP, Folk RA, Onstein RE, Yi TS (2024) Rapid in situ diversification rates in Rhamnaceae explain the parallel evolution of high diversity in temperate biomes from global to local scales. New Phytologist 241: 1851–1865. https://doi.org/10.1111/nph.19504

Torfstein A, Steinberg J (2020) The Oligo-Miocene closure of the Tethys Ocean and evolution of the proto-Mediterranean Sea. Scientific Reports 10: 13817. https://doi.org/10.1038/s41598-020-70652-4

Tosso F, Hardy OJ, Doucet J-L, Dainou K, Kaymak E, Migliore J (2018) Evolution in the Amphi-Atlantic tropical genus Guibourtia (Fabaceae, Detarioideae), combining NGS phylogeny and morphology. Molecular Phylogenetics and Evolution 120: 83–93. https://doi.org/10.1016/j.ympev.2017.11.026

Toussaint EF, Hall R, Monaghan MT, Sagata K, Ibalim S, Shaverdo HV, Vogler AP, Pons, J, Balke M (2014) The towering orogeny of New Guinea as a trigger for arthropod megadiversity. Nature Communications 5: 4001. https://doi.org/10.1038/ncomms5001

Vaidya G, Lohman DJ, Meier R (2010) SequenceMatrix: Concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics 27: 171–180. https://doi.org/10.1111/j.1096-0031.2010.00329.x

van Ufford AQ, Cloos M (2005) Cenozoic tectonics of new Guinea. AAPG bulletin 89: 119–140. https://doi.org/10.1306/08300403073

Van Welzen PC, Parnell JAN, Slik JWF (2011) Wallace’s Line and plant distributions: two or three phytogeographical areas and where to group Java? Biological Journal of the Linnean Society 103: 531–545. https://doi.org/10.1111/j.1095-8312.2011.01647.x

Veranso-Libalah MC, Kadereit G, Stone RD, Couvreur TLP (2018) Multiple shifts to open habits in Melastomateae (Melastomaceae) congruent with the increase of African Neogene climatic aridity. Journal of Biogeography 45: 1420–1431. https://doi.org/10.1111/jbi.13210

Vitales D, Garnatje T, Pellicer J, Vallès J, Santos-Guerra A, Sanmartín I (2014) The explosive radiation of Cheirolophus (Asteraceae, Cardueae) in Macaronesia. BMC Evolutionary Biology 14: 1–15. https://doi.org/10.1186/1471-2148-14-118

Wallace AR (1876) The Geographical Distribution of Animals. Cambridge University Press, Cambridge.

Wiens JJ, Donoghue MJ (2004) Historical biogeography, ecology and species richness. Trends in Ecology and Evolution 19: 639–644. https://doi.org/10.1016/j.tree.2004.09.011

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs. Proceedings of the National Academy of Sciences USA 84: 9054–9058. https://doi.org/10.1073/pnas.84.24.9054

Woodman RL, Fernandes GW (1991) Differential mechanical defense, herbivory, evapotranspiration, and leaf-hairs. Oikos 60: 11–19. https://doi.org/10.2307/3544986

Yin X, Jarvie S, Guo WY, Deng T, Mao L, Zhang M, Chu C, Qian H, Svenning JC, He F (2021) Niche overlap and divergence times support niche conservatism in eastern Asia-eastern North America disjunct plants. Global Ecology and Biogeography 30: 1990–2003. https://doi.org/10.1111/geb.13360

Zietsman PC (1990) Pollination of Ziziphus mucronata subsp. mucronata (Rhamnaceae). South African Journal of Plant and Soil 56: 350–355. https://doi.org/10.1016/S0254-6299(16)31064-X

Zizka A, Carvalho-Sobrinho JG, Pennington RT, Queiroz LP, Alcantara S, Baum DA, Bacon CD, Antonelli A (2020) Transitions between biomes are common and directional in Bombacoideae (Malvaceae). Journal of Biogeography 47: 1310–1321. https://doi.org/10.1111/jbi.13815

Supplementary material

Supplementary material 1

Supplementary figures (figs S1–S6) and tables (tables S1–S5) (.docx)

https://doi.org/10.21425/fob.17.133817.suppl1

Download file (2.03 MB)

﻿Abstract

﻿Highlights

Keywords

﻿Introduction

﻿Questions

﻿Methods

﻿Taxon sampling

﻿DNA extraction

﻿PCR

﻿Alignment and phylogenetic analysis

﻿Divergence time estimates

﻿Attribution of biome, geography, and traits

﻿Analyses

﻿Phylogenetic signal in traits, geography and ecology

﻿Trait evolution and ancestral range estimation

﻿Results

﻿Phylogenetic signal in traits, geography and ecology

﻿Ancestral range estimation

﻿Discussion

﻿How does geography and ecology structure Ziziphus’ phylogeny?

﻿How old are Ziziphus’ transitions into biomes- are open biome species younger than species in other biomes?

﻿Do functional trait shifts coincide with biome or geographical shifts?

﻿Conclusion

﻿Acknowledgements

﻿Author Contributions

﻿Data Availability Statement

﻿References

Supplementary material

Abstract

Highlights

Introduction

Questions

Methods

Taxon sampling

DNA extraction

PCR

Alignment and phylogenetic analysis

Divergence time estimates

Attribution of biome, geography, and traits

Analyses

Phylogenetic signal in traits, geography and ecology

Trait evolution and ancestral range estimation

Results

Phylogenetic signal in traits, geography and ecology

Ancestral range estimation

Discussion

How does geography and ecology structure Ziziphus’ phylogeny?

How old are Ziziphus’ transitions into biomes- are open biome species younger than species in other biomes?

Do functional trait shifts coincide with biome or geographical shifts?

Conclusion

Acknowledgements

Author Contributions

Data Availability Statement

References