Upload
claus-westphal-kroon
View
38
Download
0
Tags:
Embed Size (px)
Citation preview
Molecular phylogeny and diversification history ofProsopis (Fabaceae: Mimosoideae)
SANTIAGO ANDRÉS CATALANO1*, JUAN CÉSAR VILARDI1, DANIELA TOSTO1,2 andBEATRIZ OFELIA SAIDMAN1
1Departamento de Ecología Genética y Evolución, Facultad de Ciencias Exactas y Naturales,Universidad Nacional de Buenos Aires. Intendente Güiraldes 2160, Ciudad Universitaria,C1428EGA - Capital Federal, Argentina.2Instituto de Biotecnología, CICVyA INTA Castelar, CC 25 Castelar 1712, Argentina
Received 29 December 2006; accepted for publication 31 May 2007
The genus Prosopis is an important member of arid and semiarid environments around the world. To studyProsopis diversification and evolution, a combined approach including molecular phylogeny, molecular dating, andcharacter optimization analysis was applied. Phylogenetic relationships were inferred from five different molecularmarkers (matK-trnK, trnL-trnF, trnS-psbC, G3pdh, NIA). Taxon sampling involved a total of 30 Prosopis speciesthat represented all Sections and Series and the complete geographical range of the genus. The results suggest thatProsopis is not a natural group. Molecular dating analysis indicates that the divergence between SectionStrombocarpa and Section Algarobia plus Section Monilicarpa occurred in the Oligocene, contrasting with a muchrecent diversification (Late Miocene) within each of these groups. The diversification of the group formed by speciesof Series Chilenses, Pallidae, and Ruscifoliae is inferred to have started in the Pliocene, showing a highdiversification rate. The moment of diversification within the major lineages of American species of Prosopis iscoincident with the spreading of arid areas in the Americas, suggesting a climatic control for diversification of thegroup. Optimization of habitat parameters suggests an ancient occupation of arid environments by Prosopisspecies. © 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640.
ADDITIONAL KEYWORDS: ancestral environments – aridification – character optimization – moleculardating – penalized likelihood.
INTRODUCTION
The genus Prosopis L. (Fabaceae) has approximately45 species, which are important members of arid andsemiarid environments (Fig. 1). Some of these speciesare characteristic of the driest regions in the world.For example, Prosopis tamarugo F. Philippi, is one ofthe few tree species capable of surviving in theextremely arid climate of the Atacama desert inNorthern Chile. Other species are distinctive of thelarge deserts of North America (Tamaulipas, Sonora,Chiuahua) and of arid and semiarid regions of SouthAmerica (Monte, Patagonia, Puna, and Chaco).However, only a few representatives of the genus,
such as Prosopis africana (Guill., Perr., & Rich.)Taubert, are partially distributed in subhumid tropi-cal or subtropical regions. From a taxonomic point ofview, the genus is divided into five Sections, basedmainly on the presence and type of armature andshoot structure (Burkart, 1976). The two main Sec-tions (Algarobia and Strombocarpa) are native toNorth and South America and include approximately90% of all the species of the genus. Section Monili-carpa comprises only one species endemic to centralArgentina. Sections Anonychium and Prosopis haveexclusively Old World representatives. Prosopis afri-cana, the only species of Section Anonychium, isnative to the Soudano-Guinean zone and neighbour-ing areas of Africa, from Senegal in the west to Sudanand Kenya in the east (Pasiecznik et al., 2001).*Corresponding author. E-mail: [email protected]
Biological Journal of the Linnean Society, 2008, 93, 621–640. With 6 figures
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640 621
Section Prosopis comprises three species that arenative to North Africa and Asia, stretching east toIndia, north to Georgia and Turkmenistan, and westto Algeria along the North African coast (Pasieczniket al., 2001).
Different studies have evaluated the relationshipsamong Prosopis species (Saidman & Vilardi, 1987,1993; Saidman et al., 1996, 1998; Ramírez et al.,1999; Bessega, Saidman & Vilardi, 2005; Bessega,Vilardi & Saidman, 2006). However, these generallyinvolved phenetic analyses and included species ofonly some of the Series and Sections of the genus. Inaddition, none of these studies have evaluated thenature of the Prosopis generic limits as either out-groups were not included (Ramírez et al., 1999;Bessega et al., 2005), or these were distantly relatedto Prosopis species (Bessega et al., 2006). The onlycladistic analysis performed (Bessega et al., 2006)suggested that most of the groups taxonomically rec-ognized are not monophyletic.
Several ideas were proposed about the timing ofProsopis diversification (Pasiecznik et al., 2001).Burkart (1976), based on the presence of the genus inboth New and Old World, suggested a late Cretaceousor Early Tertiary origin for the genus, previous to theformation of the Atlantic Ocean. Following this idea,Ramírez et al. (1999) suggested that the genus wouldhave originated 130 Mya. However, this time frameis strongly contradicted by both palaeonthologicalevidence (Herendeen, Crepet & Dilcher, 1992)and molecular dating (Lavin, Herendeen &
Wojciechowski, 2005) because both support the notionthat the subfamily Mimosoideae originated between42–50 Mya. Burkart & Simpson (1977) also consid-ered that Prosopis is an old genus that diverged earlyinto several principal lineages but that, within someof these lineages, more recent episodes of expansionand isolation produced further speciation. To date, noattempt has been made to evaluate the divergencetimes within Prosopis using molecular dating.
The observation of hybrid formation between somespecies of Section Algarobia, together with their highmorphological similarity and partially overlappinggeographical distributions, led to the considerationthat this group is a syngameon (Palacios & Bravo,1981). However, recent molecular studies have sup-ported the biological meaning of the specific limitswithin this group (Saidman & Vilardi, 1987;Burghardt, 1995; Bessega et al., 2000) and indicatethe possible existence of isolating mechanisms thatrestrict the introgression between its species. Previ-ous cladistic analysis (Bessega et al., 2006) suggestedthat these species do not form a monophyletic group.Nonetheless, in that analysis, taxon sampling of non-hybridizing species within Algarobia was relativelyscarce. A more complete taxon sampling within Alga-robia would give the opportunity to study the timingand evolution of isolating barriers within this group.
During recent years, different studies have sup-ported the hypothesis that dry-adapted taxa in differ-ent regions of the world diverged concomitantly withthe expansion of arid environments. That is the case
Figure 1. Natural distribution of Prosopis after Pasiecznik et al. (2001). Number of native species are indicated for eachregion.
622 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
for Phylica (Richardson et al., 2001), Ruchoideae(Klak, Reeves & Hedderson, 2003), and Ehrharta(Verboom, Linder & Stock, 2003) in South Africa,Tiquillia (Moore & Jansen, 2006) and Agave (Good-Avila et al., 2006) in North America, and Rheum(Wang, Yang & Liu, 2005) in East Asia. Evidencesupporting this hypothesis is in all cases a temporalrelationship between the increase of arid environ-ments and the diversification of these groups. Due tothe affinity of Prosopis with arid environments, this isa promising hypothesis to be tested in this genus.
OBJECTIVES AND HYPOTHESIS
The present study aimed to investigate the mainpatterns of Prosopis diversification. We are particu-larly interested in answering the following questions.(1) Which are the relationships among the main lin-eages within Prosopis? (2) When did Prosopis diver-sification occur? (3) Has the spreading of aridenvironments driven Prosopis evolution, as suggestedfor other arid-adapted taxa? To answer these ques-tions, a combined approach was applied that includedmolecular phylogeny, molecular dating, and characteroptimization analyses.
MATERIAL AND METHODSMOLECULAR TECHNIQUES
Five different gene regions were sequenced includ-ing two nuclear markers: coding and noncodingsequences of the glyceraldehyde-3-phosphate dehy-drogenase gene (G3pdh), one intron of the nitratereductase gene (NIA) and three chloroplast markers:partial sequences of trnS-psbC, trnK-matK and trnL-trnF regions.
Total DNA was extracted from five day old cotyle-dons or silica dried leaves and/or stems with theDNeasy Plant Mini Kit (QIAGEN). All polymerasechain reaction reactions were conducted in a totalvolume of 50 mL, containing 80 ng of DNA template,1 ¥ Taq-Buffer, 2.5 mM MgCl2, 0.25 U Taq polymerase,0.2 mM of each DNTP and 0.04 mM of each primer. Theintergen between the psbC and trnS genes was ampli-fied using the primers psbC and trnS (Demesure,Sodzi & Petit, 1995) together with a new internalprimer psbCint (5′-CGTTCTTGCCAAGGCTGTAT-3′).Since a preliminary survey indicated that most of thevariation among sequences was located in the first500 bp from the 3′ end region of trnS gene, only thisregion was subsequently sequenced for the rest of thespecies. G3pdh region was amplified with primersGPDX7F and GPDXR9 (Strand, Leebens-Mack &Milligan, 1997) and a newly designed primerG3pdhintF (5′-GACTGGAGGGGTGGAAGAG-3′).NIA intron was amplified with a degenerate primer
pair: NIA3F-NIA3R (Howarth & Baum, 2002) andsequenced with a nondegenerate pair: NIAproF (5′-GAACCAGCAGTTGTTCATCAT-3′) and NIAproR (5′-ACTGGTGCTGGTGTTTTTGG-3′). trnL intron wasamplified and sequenced using the primers c, d, e andf (Taberlet et al., 1991). Partial sequences of trnKintron were amplified with Ac283R and trnK 3914Fprimers (Miller & Bayer, 2001). Partial sequences ofmatK were amplified and sequenced with primers685F and 1265R (Lavin et al., 2000). To amplify theregion trnS-psbC, trnK-matK and trnL-trnF, the fol-lowing cycling profile was used: 35 cycles of 1 min at94 °C, 1 min at 56 °C, and 1.5 min at 72 °C. In thecase of G3pdh and specific NIA amplification, thecycling profile was the same, except that annealingtemperature for these regions was set to 52 °C. Forthe amplification of NIA region with degenerateprimers, a stepdown programme was used thatincluded ten cycles in which melting temperaturedecreased 1 °C per cycle starting from 55 °C. Twentyadditional cycles were performed with 30 s at 94 °C,30 s at 45 °C, and 40 s at 72 °C. In all cases, theprogramme started with one cycle at 94 °C for 4 minand finished with 7 min at 72 °C.
PHYLOGENETIC ANALYSIS
Thirty species of Prosopis were included in this analy-sis. Taxon sampling within Prosopis was planned toinclude representatives from the entire geographicalrange of the genus (Appendix 1). Moreover, taxonomicdiversity of the genus was totally represented asspecies of all Sections and Series were included.Three different analyses that involved different com-binations of taxa and molecular markers were per-formed. A first analysis was performed comprisingtrnL-trnF and trnK-matK (two-marker analysis)sequences of 80 legume species downloaded fromGenBank (see Supplementary material) plus fiveProsopis species that represented each Section of thegenus. In the second analysis (three-marker analy-sis), G3pdh, NIA, and trnS-psbC sequences from allthe sampled species of Prosopis plus Prosopidastrumangusticarpum, Acacia caven, Xerocladia viridiramis,and Lotus japonicus were analysed in a total evidencecontext (Kluge, 1989). As more than one sequence ofeach species was sampled in Algarobia clade, the totalnumber of terminals was 41. This was due to the highlevel of shared polymorphism found in previousanalyses within this group (Saidman & Vilardi, 1987;Bessega et al., 2000). A third analysis (five-markeranalysis) involved a combination of the first twodatasets.
Sequences were edited with Bioedit software (Hall,1999) and aligned with Clustal X (Thompson et al.,1997) with a posteriori minor manual changes. Par-
PROSOPIS PHYLOGENY AND EVOLUTION 623
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
simony based searches were performed using the TNTprogram (Goloboff, Farris & Nixon, 2003). Charactersand state transformations were considered as equallyweighted. Gaps were alternatively treated as a fifthstate, missing entries or as separate charactersSimmons & Ochoterena (2000). Since the results werevery similar, only those based on fifth state coding areshown and discussed. Within trnL-trnF region, a dele-tion of approximately 300 bp was found in someMimosoid species. Individual gaps representing thisdeletion were treated as missing data. As a model-based analysis alternative to the parsimony analysis,we also conducted a Bayesian analysis. Search strat-egies in parsimony and Bayesian analysis are indi-cated in Appendix 2.
ESTIMATING DIVERGENCE TIMES
To estimate divergence times, G3pdh and trnSsequences were concatenated and analysed together.Simultaneous analysis of gene sequences from mul-tiple loci was performed because the penalized likeli-hood method used for age estimation (see below) isprone to errors when dealing with zero lengthbranches or very short branches (Sanderson, 2003).The congruence between both datasets in branchlengths was evaluated by the likelihood ratio testdescribed by Lewis (1998) and Xiang et al. (2005). Thestatistics of this test is -2 [ln L - (ln L1 + ln L2)], whereL1 is the likelihood of the tree obtained from the firstgene, L2 is the likelihood of the tree obtained from thesecond gene and L is the likelihood obtained from thecombined analysis of both genes. The value of thisstatistic was compared to a c2 distribution withdegrees of freedom equal to the sum of 2N + 3 - P,where N is the number of terminals and P is thenumber of free parameters of the model used in thelikelihood calculation. NIA sequences were notincluded because it was not possible to obtain thesequence of this region from P. angusticarpum, one ofthe taxa that defines the unique fixed node in thecalibration step. Furthermore, X. viridiramis was notincluded because only a part of G3pdh was possible tobe sequenced. Branch lengths were estimated byMaximum Likelihood using PAUP* (Swofford, 2002).The evolutionary model was chosen with the hierar-chical likelihood ratio test as implemented inModeltest, version 3.06 (Posada & Crandall, 1998).Penalized likelihood implemented in the R8s software(Sanderson, 2003) was used for age estimation usingthe Powell algorithm and defining smoothing param-eter values with a cross validation procedure (Sander-son, 2002).
Divergence times were derived from molecular datacombining two calibration points. The time obtainedin Lavin et al. (2005) for the divergence between one
species of Section Algarobia (Prosopis pallida) andone species of genus Prosopidastrum (Prosopidastrummexicanum) was used to fix the node subtendingProsopidastrum and the American species of Prosopis(ASP). To take into account the error associated withthe estimation of Lavin et al. (2005), we repeated theanalysis considering the mean ± SD. A second calibra-tion point was used to give minimum age to the ASPdivergence. In this case, this fossil corresponds topollen grains obtained from Early Oligocene sedi-ments of British Columbia, Canada (Piel, 1971). Asthis Epoch extends from 28.4 and 33.7 Mya, theformer age was used to define the minimum age.Since the sister group of ASP + Xerocladia clade wasnot clearly defined in our phylogenetic analysis, twodifferent topologies were considered, one of them cor-responding to the results of the three-marker analysisand the other corresponding to the two-marker analy-sis. Error in age estimates was evaluated by boot-strapping following Sanderson (2002).
DIVERSIFICATION ANALYSIS
To study the possible changes in diversification rateover time a lineages through time (LTT) analysis (Neeet al., 1995) was performed. LTT analysis was repeat-edly performed with extreme values of divergencetimes obtained in the penalized likelihood analysis.The lack of resolution in the mesquite clade (seeResults) was considered in two ways for this analysis:(1) as a hard polytomy with all the lineages divergingsimultaneously and (2) as if the diversificationoccurred at a constant rate. Consequently, PHYLO-GEN program, version 1.1 (Rambaut, 2002), was usedto simulate trees under a Yule Model of diversification(Yule, 1924) with the final number of species equal tothe present in the mesquite group. Subsequently, thissubtree was grafted in the original tree, replacing thepolytomy in the mesquite clade. This was repeated for100 different simulated subtrees. As the generalpattern obtained was the same, LTT of one randomlychosen tree is shown. To examine the effects ofmissing species in LTT analysis, these were placed inthe most likely position on the tree on the grounds oftaxonomic treatment (Burkart, 1976) in accordnacewith Barraclough & Vogler (2002). To compare diver-gence rates within Prosopis with those obtained inother plant genera, per lineage net diversificationrate (NDR) sensu Coyne & Orr (2004) was calculated.As in LTT analysis, missing species were included intheir most likely place on the tree.
ESTIMATION OF ANCESTRAL HABITAT
To study the historical relationship of Prosopis witharid environments, maximum and minimum value ofhumidity index (HI) were gathered for the natural
624 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
geographical range of the species of Prosopis includedin the phylogenetic analysis plus X. viridiramis. HI isdefined as the ratio of precipitation to potentialevapotranspiration (UNEP, 1992). Species distribu-tions were defined from the bibliography (Ross, 1975;Burkart, 1976; Rzedowski, 1988; Roig, 1993) andfrom herbarium specimens (BAFC, SI, LIL, CTES).Climate data for each species (see Supplementarymaterial) were extracted from FAO World ClimateDatabase using the New LocClim software (FAO,2005). Regions were classified as a function of HIvalues following UNEP (1992): 0.05 < Hyper-Arid;0.05–0.20 Arid; 0.20–0.50 Semi-Arid; 0.50–0.65Dry-Subhumid; > 0.65 Humid. The phrase ‘aridenvironments’ is used here to collectively representhyperarid, arid and semiarid classes. Maximum andMinimum of HI (MaHI and MiHI, respectively) wereindependently optimized on the tree as in theMinMax coding (Hardy & Linder, 2005). Once thesevalues were optimized, they were used to delimit thetotal range of HI for each ancestral node. We consid-ered that MinMax coding is the most suitable for thischaracter because maximum drought tolerance mightbe partial or totally decoupled from maximum humid-ity tolerance. MaHI and MiHI were optimized usingthe implementation in TNT (Goloboff et al., 2003) ofFarris (1970) optimization to deal with continuouscharacters (Goloboff, Mattoni & Quinteros, 2006). Inthis case, linear changes are minimized along eachbranch. To evaluate how uncertainty in the state ofthe root can affect the results, the analysis wasrepeated with two extreme values: hyperarid andhumid. In addition, the effect of topological uncer-tainty within the mesquite clade (see Results) in theresults was evaluated by considering 10 000 differentresolutions of this clade. A strict association of aparticular node with arid environments was deter-mined when the value of MaHI was lower thanthe limit between humid and arid environments(HI < 0.50). A nonstrict association with arid environ-ments corresponded to a value of MiHI lower andMaHI higher than this limit. No association with arid
environment occurred when MiHI value was higherthan this limit.
RESULTSPHYLOGENETIC RESULTS
Two-marker analysisStatistics of the different phylogenetic analyses basedon parsimony are shown in Table 1. Strict consensussuggests that Prosopis is not monophyletic (Fig. 2).Forcing the monophyly of the genus gave a tree ninesteps longer than the optimum (Table 2). ASP formeda highly supported clade (BS = 4; J = 95) with X. viri-diramis (subsequently named ASP + Xerocladia). ASPdid not appear as monophyletic because X. viridira-mis formed a monophyletic group with P. tamarugo.However, this clade is not well supported and (BS = 1;J = 66) and forcing the monophyly of ASP gave a treeonly one step longer than the optimum. Prosopiscineraria appeared as sister group of this cladethough with low support (BS = 1; J < 50). Prosopisnigra of Section Algarobia formed a strongly sup-ported group (BS = 5; J = 98) with Prosopis argentinaof Section Monilicarpa. Bayesian analysis gave iden-tical results concerning the lack of monophyly ofProsopis and the relationships among APS. By con-trast, P. cineraria does not appear forming a mono-phyletic clade with ASP + Xerocladia. Posteriorprobabilities for the groups found in the Bayesiananalysis are given in Figure 2.
Three-marker and five-marker analysesStrict consensus obtained in this analysis agreed withtwo-marker analysis in rejecting Prosopis monophyly(Fig. 3). Forcing the monophyly of the genus gave atree six steps longer than the optimum (Table 2). Asin two-marker analysis, ASP formed a monophyleticgroup with X. viridiramis (APS + Xerocladia clade),though with lower support (BS = 1; J = 64). Inaddition, forcing the monophyly of the ASP gave atree only one steps longer than the optimum. Two
Table 1. Statistics of the different phylogenetic analyses based on parsimony
Two-marker analysis Three-marker analysis Five-marker analysis
Number of MPT 224 9 4680Length 5079 1552 6710Number of characters 4031 (1444–2587)† 1764 (648–701–415)* 5795Number of informative characters 1357 (609–748)† 391 (193–183–15)* 1458CI/RI 0.496/0.655 0.695/0.846 0.535/0.700
*Statistics for G3pdh, NIA and trnS-psbC characters, respectively.†Statistics for trnL-trnF and trnK-matK characters, respectively.CI, consistency index (excluding uninformative characters); MPT, most parsimonious trees; RI, retention index.
PROSOPIS PHYLOGENY AND EVOLUTION 625
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
Zapoteca tetragona
Xylia africana
Xerocladia viridiramis
Tetrapleura tetraptera
Prosopis tamarugo
Schleinitzia insularum
Samanea saman
Pseudosamanea guachapele
Pseudoprosopis gilletii
Prosopis cineraria
Prosopis africana
Prosopidastrum mexicanumProsopidastrum angusticarpum
Plathymenia reticulata
Piptadeniopsis lomentifera
Piptadenia viridiflora
Piptadeniastrum africanum
Pentaclethra eetveldtiana
Parkia biglandulosa
Paraserianthes lophantha
Pararchidendron pruinosum
Parapiptadenia rigida
Prosopis nigra
Newtonia buchananii
Neptunia gracilis
Mimozyganthus carinatus
Mimosa aculeaticarpaMicrolobius foetidus
Lysiloma divaricatum
Leucaena greggiiKanaloa kahoolawensis
Inga edulis
Havardia albicans
Gagnebina bernieriana
Fillaeopsis discophora
Faidherbia albida
Enterolobium contortisiliquum
Entada abyssinicaElephantorrhiza elephantina
Ebenopsis ebano
Dinizia excelsa
Dichrostachys paucifoliolata
Desmanthus acuminatus
Cylicodiscus gabunensis
Chloroleucon mangens
Cercidium andicola
Cedrelinga cateniformis
Cathormion umbellatum
Calpocalyx dinklagei
Calliandropsis nervosus
Calliandra carbonaria
Prosopis argentina
Anadenanthera colubrina
Amblygonocarpus andongensis
Albizia kalkora
Alantsilodendron alluaudianum
Adenanthera pavonina
Acacia visco
Acacia tortilis
Acacia spinescens
Acacia senegal
Acacia schweinfurthii
Acacia schottii
Acacia roemeriana
Acacia pulchella
Acacia platycarpa
Acacia pennatula
Acacia neovernicosa
Acacia modesta
Acacia melanoxylon
Acacia macracantha
Acacia karroo
Acacia hindsii
Acacia glomerosa
Acacia elata
Acacia drummondii
Acacia dolichostachyaAcacia coulteri
Acacia constricta
Acacia collinsiiAcacia cochlicantha
Acacia chiapensis
Acacia caven
Acacia boliviana
Acacia berlandieri
Acacia ampliceps
Lotus japonicus
2/66
>12/99
3/-
1/52
12/100
12/1006/98
12/100
>12/1001/-
2/-
9/98
9/99
1/-
1/55
3/76
1/-
1/-4/98
4/95
1/-
1/66
5/98
1/-
8/100
2/55
4/87
1/-
1/74
1/90
>12/100
11/1003/73
2/78
2/78
8/1001/72
3/93
1/58
2/796/99
>12/100
6/74
1/-4/92
1/-3/75
8/-6/97
1/-
2/-
10/100
9/1001/-
1/-5/75
1/-1/-
7/921/68
1/-3/-
4/-
4/91
5/1005/95
1/-
>>
1.000.55
0.981.00
0.68
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.001.00
0.53
1.00
1.00
1.00
1.000.99
1.00
0.991.00
1.00
1.00
1.00
1.001.00
1.00
0.93
1.00
1.00
1.00
0.79
1.00
1.00
0.99
1.00
1.00
0.98
0.78
0.96
1.00
1.00
0.68
1.00
626 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
subclades were recognized in ASP + Xerocladia clade.One of them is formed by species of Section Stromb-ocarpa plus X. viridiramis whereas the other (subse-quently named Algarobia s.l.) is formed by species ofSection Algarobia plus P. argentina of Section Monili-carpa (BS = 6; J = 100). Section Algarobia appeared asparaphyletic because P. argentina formed a well sup-ported clade (BS = 5; J = 100) with species of SeriesHumiles, Sericanthae, and Denudantes of SectionAlgarobia (Prosopis kuntzei–P. argentina clade).Forcing the monophyly of Section Algarobia gave atree ten steps longer. Series Denudantes (representedhere by Prosopis denudans, Prosopis ruizleali andProsopis castellanosii) appeared as monophyletic butSeries Sericanthae (P. kuntzei and Prosopis serican-tha) would be paraphyletic. The sister group of theP. kuntzei–P. argentina clade was a very well sup-ported clade (BS = 6, J = 100) formed by species ofseries Chilenses, Pallidae, and Ruscifoliae (mesquiteclade). However, the relationships among the speciesof this clade are unclear because the strict consensusis highly unresolved and individuals of the samespecies in most of the cases did not form monophyleticgroups. Section Strombocarpa was monophyletic withX. viridiramis as its sister clade. Within this section,two groups were well supported. One of them corre-sponded to Series Cavenicarpae (Prosopis ferox andP. tamarugo) whereas the other was formed by twoNorth American species of Series Strombocarpae:Prosopis pubescens and Prosopis palmeri. Bayesiananalysis showed very similar results (Fig. 3). The
main difference was related with the sister group ofAPS + Xerocladia. In this analysis, the sister groupwas formed by A. caven and P. angusticarpum.
The results obtained in the five-marker analysiswere very concordant with those obtained in the othertwo analyses (see Supplementary material) andthe support increased in several relevant nodes:ASP + Xerocladia clade (BS = 9; J = 96); Stromb-ocarpa clade (BS = 2; J = 83); Strombocarpa +X. viridiramis clade (BS = 2; J = 77). However, differ-ing from the two-marker analysis, P. cineraria did notappear as sister group of APS + Xerocladia. Strictconsensus indicated identical relationships withinAPS + Xerocladia clade than those given by the three-marker analysis, except that the groupings within themesquite clade are less resolved.
MOLECULAR DATING
The optimal model chosen by the hierarchical likeli-hood ratio test was Hasegawa–Kishino–Yano plusGamma (HKY + G). A likelihood ratio test signifi-cantly rejected the hypothesis of evolutionary rateconstancy for the molecular regions analysed (Clockconstrained –lnL = 4394.64; unconstrained –lnL =4363.50; d.f. = 39; P < 0.05). Therefore, divergencetimes were estimated in a penalized likelihoodapproach that does not assume rate constancy. Tostudy the possible effect of short branches withinmesquite clade in the age estimated for the othernodes outside this group, an analysis leaving only one
Figure 2. Strict consensus of the 224 optimal trees based two-marker analysis (trnL-trnF, trnK-matK) with parsimonyas optimality criteria. Absolute Bremer support (left) and jackknifing values over 50% (right) are indicated abovebranches. Posterior probabilities estimated in Bayesian analysis are indicated below branches. Prosopis species areunderlined.�
Table 2. Extra steps obtained in searches conducted with positive constrains for some groupings not appearing in mostparsimonious trees
Two-marker analysis(5079)
Three-marker analysis(1552)
Five-marker analysis(6710)
Section Strombocarpa +Prosopis cineraria
6 63 68
Prosopis + Prosopidastrum 14 3 14Prosopis + Xerocladia 4 5 5Genus Prosopis 9 6 11Section Algarobia NA 10 10OSP + Xerocladia 8 6 11OSP 4 5 11ASP 1 1 3
In parentheses: optimal lengths in unconstrained searches. ASP, Americans species of Prosopis; NA, not applicable; OSP,Old World species of Prosopis.
PROSOPIS PHYLOGENY AND EVOLUTION 627
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
terminal was performed. The results differ onlyslightly (data not shown).
The common and most relevant pattern observed inthe different analyses was (Fig. 4; Table 3): (1) anancient divergence between the two main lineages ofASP within the Oligocene; (2) a more recent diver-gence, starting in the late Miocene, within Stromb-ocarpa and Algarobia s.l.; and (3) a divergence amongmesquite species starting in the Pliocene.
DIVERSIFICATION ANALYSIS
Phylogenetic analyses indicated that Old Worldspecies of Prosopis are not closely related with theASP. This was true for P. africana in all the analysesand for P. cineraria in two out of three analyses.
Taking this into account, the study of diversificationand climate affinity evolution was restricted to theASP + Xerocladia clade.
Lineage through time analysis clearly suggests anincrease in the diversification rate of the ASP groupfrom the Late Miocene (Fig. 5A, B) until the present.Within this period, a more detailed examination sug-gests two different moments of rate acceleration: onein the Late Miocene and the other in the Pliocene.This general pattern is neither affected by the inclu-sion of missing species in the analysis (Fig. 5B), norby the different topologies used to derive the chrono-grams (data not shown). The NDR estimated for dif-ferent clades within Prosopis and for other taxa isshown in Table 4.
P.vinalillo 2
P.vinalillo 1
P. torquata
P. tamaulipana
P. tamarugo
P. strombulifera
P. sericantha
P. ruscifolia
P. ruizleali
P. reptans
P.pubescens
P.africana
Prosopidastrum angusticarpum
P.palmeri
P.nigra 2
P.nigra 1
Lotus japonicus
P.kuntzei 2P.kuntzei 1
P. juliflora 2
P. juliflora 1
P.humilis
P.glandulosa
P. flexuosa 2
P. flexuosa 1
P. ferox
P.denudans
P.chilensis 2P.chilensis 1
P.castellanosii
P.campestris 2P.campestris 1
P.caldenia 2
P.caldenia 1
P.articulata
P.alpataco
P.albaP.affinis
Acacia caven
-
100
100
100
80
87
-
61
62
65
72100
99
64
100
5476
57
1
1
1
1
1
1
2
2
2
3
2/ -
3
3
5
4
4
6
1
6
5
P.k
untz
eiP
. arg
entin
a- clad
e
P.cineraria
2
13
3/ -
60
Alg
arob
ia s
.l cl
ade
Str
ombo
carp
acl
ade
Am
eric
ansp
ecie
sof
clad
eP
roso
pis
+X
eroc
ladi
a
3
4
P.argentina
Mes
quit
ecl
ade
//
/
/
/
/57
/
/
/
/
/
/
/
/
/
/
//
702 /965/
/
6
/
1.000.58
0.96
1.001.00
0.690.63
0.83
0.99
0.910.99
0.64
0.96
0.99
1.00
0.61
0.85
0.90
0.55
Xerocladia viridiramis
7620.58
/
Str
ombo
carp
a+
clad
eX
eroc
ladi
a
Figure 3. Strict consensus of the nine optimal trees obtained in the three-marker analysis (trnS-psbC, G3pdh, NIA) withparsimony as optimality criteria. Absolute Bremer support (left) and jackknifing values over 50% (right) are indicatedabove branches. Posterior probabilities estimated in Bayesian analysis are indicated below branches. Rectangles delimiteach Section: 1, Prosopis; 2, Anonychium; 3, Strombocarpa; 4, Algarobia; 5, Monilicarpa. Black bars indicate the cladeswhose names are used throughout the manuscript.
628 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
ANCESTRAL CLIMATE ESTIMATION
Optimization analysis unambiguously indicated thatthe range of the most recent common ancestor(MRCA) of the clade formed by the ASP plus X. viri-diramis extended to semiarid or more xeric conditions
(Fig. 6). A strict association of this node with aridenvironments was obtained for all the reconstructionswhen the state of the root was defined as hyperaridand, in some of the reconstructions, when the state ofthe root was defined as humid. In the case of Strom-bocarpa clade, the different assignments of states to
Early Middle LatePLIOCENE
OLIGOCENEEarly Late
MIOCENEPLE
1.85.311.216.423.828.433.7
angusticarpum
P.ruizlealiP. strombulifera
P. reptansP. torquataP.palmeriP.pubescens
P. tamarugo
P.argentinaP. sericanthaP.humilisP.kuntzei 1P.kuntzei 2P.denudansP.castellanosii
P. ferox
P. flexuosa 1P.articulataP.caldenia 1P. juliflora 1P. juliflora 2P.glandulosaP. tamaulipanaP.vinalillo 2P.campestris 2P.nigra 2P.caldenia 2P. ruscifoliaP.alpatacoP.albaP.affinisP.chilensis 2P.campestris 1P.chilensis 1P.nigra 1P.vinalillo 1
P. flexuosa 2
Prosopidastrum
Figure 4. Chronogram resulting from penalized likelihood analysis. The topology corresponds with one of the mostparsimonious trees obtained in three-marker analysis (Fig. 3) and calibrated according to divergence times derived fromLavin et al. (2005). The approximate temporal spreading of arid zones in America is shown over the timescale with darkerzones representing the moment of its maximum historical extension. Grey bars over the nodes represent 95% confidenceinterval for node ages obtained by bootstrap techniques. PLE, Pleistocene.
Table 3. Divergence times (Mya) obtained in the penalized likelihood analysis
MRCA
Outgroup placement
Three-marker analysis Two-marker analysis
ASP + Xerocladia 28.96 (26.25–31.68) 29.37 (27.39–33.05)Algarobia s.l. 7.89 (7.15–8.62) 7.96 (7.58–9.14)Mesquites 3.65 (3.31–3.99) 3.69 (3.38–4.08)Strombocarpa 9.21 (8.35–10.07) 9.31 (9.18–11.07)
Ages estimated fixing the node subtending ASP + Xerocladia clade and Prosopidastrum to 33.2 Mya following the meanvalue obtained in Lavin et al. (2005). Ages estimated considering the mean ± one standard deviation are given inparentheses. When the topology corresponded to the two-marker analysis the node subtending ASP + Xerocladia node wasconstrained to a minimum age of 28.4 Mya following the fossil evidence.ASP, American species of Prosopis; MRCA, most recent common ancestor.
PROSOPIS PHYLOGENY AND EVOLUTION 629
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
the root did not affect the results as, in all cases, itsMRCA appear to be restricted to arid environments.The same result was obtained for the MRCA ofX. viridiramis + Strombocarpa clade. In the case ofthe MRCA of Algarobia s.l., P. kuntzei–P. argentinaand mesquite clades, optimal reconstructions in-cluded in some cases ranges that fell completelywithin arid conditions whereas, in others, the rangesextended from arid to humid conditions.
DISCUSSIONSYSTEMATICS
The phylogenetic analysis performed in the presentstudy is the first to include species of all the Sectionsand Series of the genus and intends to test Prosopismonophyly by incorporating several species ofrelated genera. The results of this analysis suggestthat Prosopis would not be monophyletic. ASPappear to be more related with X. viridiramis, theonly species of this southern Africa genus, than withOld World representatives of Prosopis. A close rela-
tionship among Prosopis species and X. viridiramishas been previously indicated by Luckow et al.(2004). However, considering that only species ofSection Algarobia were included in that analysis, acloser relationship of ASP with X. viridiramis thanwith Old World Prosopis species is a novel and inter-esting result. Prosopis monophyly was also rejecteddue to the position of P. africana and, in some of theanalyses, the position of P. cineraria. It is suggestivethat many of the characters used to define the genusby Burkart (1976) would be, as stated by him, primi-tive within the whole subfamily (symplesiomorphic).If this is confirmed, the idea of Prosopis as naturalgroup might also be questioned at morphologicalgrounds.
A closer relationship among species of the Ameri-can sections of Prosopis (Strombocarpa, Algarobia,and Monilicarpa) than with species of the Old Worldsections agrees with the subgeneric classification pro-posed by (Guinet & Bessedik, 1984) based mainly onpalynological evidence. Particularly, they included allthe ASP in the subgenus Neoprosopis, whereas theother two subgenera corresponded to Section Proso-pis and Anonychium, the two Old World sectionsconsidered in Burkart (1976). However, the corre-spondence between the groups obtained in our analy-ses and the Subgenus Neoprosopis is not strictbecause ASP formed a monophyletic group withX. viridiramis. This close relationship among Ameri-can Sections of Prosopis contrasts with the ideastated by Burkart (1976) that Section Strombocarpais closely related to Section Prosopis because speciesof both Sections present comparatively smallerlegumes and share the ability to spread by means ofrootsuckers.
Species of Section Algarobia and P. argentinaformed a highly supported clade (Algarobia s.l.) thatis subsequently divided into two highly supportedclades. One of them corresponded to species of SeriesPallidae, Chilenses, and Ruscifoliae (mesquite clade),whereas the other corresponded to species of seriesSericanthae, Humiles, and Denudantes plus P. argen-tina of Section Monilicarpa (P. kuntzei–P. argentinaclade). This latter clade, endemic to southern SouthAmerica, includes all the subaphyllous and aphyllousspecies sampled for the phylogenetic analysis. Thisresult suggests that this condition might have origi-nated only once in Prosopis history.
The results obtained in the present study are con-sistent with the only previous cladistic analysisinvolving Prosopis species (Bessega et al., 2006) inshowing that the Section Algarobia is not monophyl-etic, and that P. argentina (Section Monilicarpa) isclose to species of Section Algarobia. However, severaldisagreements between these two studies can beobserved. Within Section Algarobia, the clade formed
Ln
(num
ber
ofli
neag
es)
Ln
(num
ber
ofli
neag
es)A
B
MIOCENEOLIGOCENE
23.828.433.7 16.4 11.2 5.3 1.8
0.5
1.0
1.5
2.0
2.5
3.0
3.5
1.0
1.5
2.5
3.0
3.5
2.0
E L E M LPLIOC PLE
Figure 5. Lineages-through-time plot (LTT) derived fromthe chronogram of Fig. 4. A, LTT including missingspecies. B, LTT with species included in the three-markeranalysis. Dotted lines represent LTT when the polytomy inthe mesquite clade is considered as hard. Black arrow-heads indicate the beginning of the diversification withinAlgarobia s.l. and Strombocarpa clades. Empty arrow-heads indicate the start of mesquite diversification. E,Early; M, Medium; L, Late; PLIOC, Pliocene; PLE, Pleis-tocene. The approximate temporal spreading of arid zonesin America is shown over the timescale with darker zonesrepresenting the moment of its maximum historicalextension.
630 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
Table 4. Comparison of net diversification rates (NDR) estimated within Prosopis with those obtained for other planttaxa
NDR Life forms
ProsopisMesquite clade 0.58–0.72* Trees†Prosopis kuntzei–Prosopis argentina clade 0.16–0.21* Trees and shrubsStrombocarpa clade 0.27–0.37* Trees and shrubsASP + Xerocladia clade 0.12–0.14* Trees and shrubs
Others taxaInga1 0.50 TreesLupinus2 1.92–3.84 Postrate herbs to treeletRuschioideae3 0.77–1.75 Stone plants to tree-like shrubsClarkia, section Peripetasma4 0.62 HerbsGentianella5 1.7–3.2 Herbs and small shrubsValeriana6 0.8–1.3 HerbsYucca7 0.21–0.27 Succulent rossete plants to arborescentEhrharta8 0.12–0.57 HerbsGaertnera9 0.71–0.83 Shrubs and small treesHawaiian silverwoods10 0.55 Vines and shrubs to treesAgave7 0.51 Succulent rossete plants to arborescent
*Range of rates values estimated with times derived from the different calibration strategies.†95% of the species are trees or present arboreal forms.1Richardson et al. (2001); 2Hughes & Eastwood (2006); 3Klak et al. (2003); 4Hey (1992); 5Von Hagen & Kadereit (2001);6Bell & Donoghue (2005); 7Good-Avila et al. (2006); 8Verboom et al. (2003); 9Malcomber (2002); 10Baldwin & Sanderson(1998).ASP, Americans species of Prosopis.
AAR-SA
SA-SH-HU
AR-SASA-SH-HU
AR-SASA-SH-HU
AR-SASA-SH-HU
B
Mes
quite
sS
trom
boca
rpa
AR-SA
AR-SA
AR-SA
HA-AR
HA-AR
P.ku
ntze
i-P.
arge
ntin
a
Xerocladia
ARAR-SA
Xerocladia
ARSA
HA-ARSA
SA-SH-HU
SA
SA-SH-HU
SA-SH-HU
SA
Mes
quite
sS
trom
boca
rpa
P.ku
ntze
i-P.
arge
ntin
a
Figure 6. Reconstruction of ancestral states of relevant nodes for maximum (above) and minimum (below) humidityindex (MaHI and MiHI, respectively). Instead of HI values, classes that represent each value are shown. A, optimizationresult when the state of the root was defined as hyperarid. B, optimization result when the state of the root was definedas humid. The presence of ambiguities in the optimization step and/or the different possible resolutions within themesquite clade produced, in some nodes, more than one optimal value for MiHI and MaHI.
PROSOPIS PHYLOGENY AND EVOLUTION 631
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
by species of Series Ruscifoliae, Chilenses, and Pal-lidae (mesquite clade) is not monophyletic in theanalysis of Bessega et al. (2006). A similar situation isobserved in Section Strombocarpa, which is mono-phyltic in our analysis (Fig. 3) but not in that ofBessega et al. (2006). In addition, in the presentanalysis, we found that American Prosopis species aremore closely related among themselves than with OldWorld Prosopis species, whereas in the analysis ofBessega et al. (2006), P. cineraria, an Old Worldspecies, appears intermingled with New Worldspecies. The groupings obtained in our analysis aregenerally more in agreement with previous analyses(Ramírez et al., 1999; Bessega et al., 2005), and withtraditional taxonomy (Fig. 3). Although several ad hocexplanations could be given to account for the dis-agreement between these two analyses, it is possiblethat the wider sampling of species and molecular-markers performed in the present study might be thebest explanation for these differences.
MOLECULAR DATING AND FOSSIL RECORD
Molecular dating analysis suggests that the diver-gence between the two main groups within ASP +Xerocladia clade is remarkably ancient (26.2–33 Mya)considering that the diversification of the extantmembers of the whole subfamily Mimosoideae wouldhave occurred during the last 40 Myr (Lavin et al.,2005). However, most of the process of diversificationwithin each of the major clades occurred morerecently, in the Late Miocene. This contrast betweenearly divergence of the main groups and late diver-gence within these groups had already been estab-lished on the grounds of morphological evidence(Burkart & Simpson, 1977; Pasiecznik et al., 2001).
The number of fossil remains assigned to Prosopisis particularly abundant (Catalano et al. unpubl.data). However, as none of the fossils have beenanalysed in a phylogenetic context and some of thedescriptions are poorly detailed, their assignation toProsopis and its subgroups is, in some cases, verytentative. The oldest fossil that could be attributed tothe ASP clade (Guinet & Bessedik, 1984), which wasused as one of the calibration points in our moleculardating analysis, belongs to pollen grains (Prosopisquesneli) of the early Oligocene from British Colum-bia, Canada (Piel, 1971). A similar age has beenestimated for fruit remains (Prosopis lazarii) found inthe palaeoflora of Puebla, Mexico (Magallon-Puebla &Ceballos-Ferriz, 1994). Nevertheless, the affinity ofthis fossil with extant groups within the genus isuncertain. Recently, Anzótegui & Herbst (2004) havedescribed, based on leaflets remains, a new fossilspecies of Prosopis from the Middle Miocene of Argen-tina that appears to be related with extant members
of Section Strombocarpa. The age of this fossil iscompatible with the times estimated in the moleculardating analysis as both indicate that the divergencebetween the to major groups of Prosopis occurredbefore the middle Miocene. A more detailed compari-son of divergence times will require the combinedinclusion of fossil and extant species in a phylogeneticanalysis.
MESQUITES DIVERSIFICATION
The results obtained in the present study stronglysupport a close relationship of species of Series Chil-enses, Pallidae, Ruscifoliae (mesquite clade). Interest-ingly, many of these species were supposed to form asyngameon (Palacios & Bravo, 1981) because of theirability to hybridize naturally, their high morphologi-cal similarity and partially overlapping geographicaldistributions. However, several recent analysessuggest that hybrid formation does not produce sig-nificant gene flow among these species. First, molecu-lar studies showed that, in spite of their high geneticsimilarities, populations formed groupings coincidentwith their specific origin (Palacios & Bravo, 1981;Saidman & Vilardi, 1987; Burghardt, 1995; Bessegaet al., 2000). This was true even for populations ofdifferent species that were sympatric. In addition,Bessega et al. (2000) showed that sympatric speciesare not genetically more similar than allopatric ones.Second, the only analysis of hybrid viability availableto date (Naranjo, Poggio & Enus Zeiger, 1984) sug-gests that, at least for some species combinations,hybrids are partially or totally sterile. In the light ofour results, the high genetic similarity among thesespecies might now be explained by their recent diver-gence (< 4 Mya) without the need to invoke introgres-sion among them.
The average per lineage diversification rate esti-mated for the mesquite group is comparable to otherknown rapid plant radiations (Table 4). In particular,the diversification rate estimated for this group isfound to be higher than in Inga, another genus fromthe same subfamily that also comprises tree species.It is possible that this rapid and recent radiation ofmesquites might be the cause of the lack of resolutionobtained in the mesquite clade in the present study. Asimilar explanation for the lack of resolution has beenproposed in diverse plant genera (Hodges & Arnold,1995; Baldwin & Sanderson, 1998; Richardson et al.,2001; Hughes & Eastwood, 2006).
EVOLUTION OF REPRODUCTIVE ISOLATING
MECHANISM WITHIN ALGAROBIA
Molecular dating and phylogenetic results obtained inthe current study provides an improved insight into
632 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
the evolution of isolating mechanism within thegenus. In the case of the mesquite clade, as previouslyindicated, gene flow among its species would behighly restricted. This suggests the existence of iso-lating mechanism(s) that prevent introgressionamong them. The existence of hybrid between thesespecies in nature indicates that prezygotic reproduc-tive isolation barriers are probably weak (Palacios &Bravo, 1981) and stresses the potential importance ofpostzygotic isolating mechanisms within this group.The observation that hybrids between mesquitespecies are generally found in disturbed environ-ments (Palacios & Bravo, 1981; Verga, 1995) supportsthe idea that hybrids are not intrinsically inviable,but that their development is limited by the lack ofsuitable environmental conditions (ecological inviabil-ity; Coyne & Orr, 2004). Evidence for a second postzy-gotic isolating barrier has been obtained by Naranjoet al. (1984) who observed diminished fertility in somehybrid combinations. Given the times derived frommolecular dating, these postzygotic isolating mecha-nisms would have developed in less than 4 Myr, thetotal span for the diversification of this group(Table 3).
The phylogenetic results suggest that the ability tohybridize is not a characteristic extended throughoutthe Algarobia s.l. group. Indeed, the species of the twomain clades found within this group (Fig. 3) presentstrikingly different behaviour. Whereas, as previouslyindicated, mesquites do frequently hybridize, hybridformation between species of the P. kuntzei–P. argentina clade has not been documented. Althoughspecies of the P. kuntzei–P. argentina clade mighthave developed isolating mechanisms that preventhybrid formation, we consider that the reduced levelof sympatry between most of these species is a suffi-cient explanation for the difference. Furthermore, nohybrids have been observed between mesquite speciesand species of P. kuntzei–P. argentina clade even if inthis case they frequently appear in sympatry. Thissuggests the development of another barrier eitherintrinsic postzygotic or prezygotic in less than 8 Myr(Table 3), the estimated divergence time betweenthese two clades.
ANCESTRAL CLIMATE RECONSTRUCTION
During recent years, optimization techniques havebeen increasingly used to infer ancestral ecologicalconditions (Verboom et al., 2003; Graham et al., 2004;Hardy & Linder, 2005; Schrire, Lavin & Lewis, 2005).In the present study, an optimization analysis wasperformed to evaluate the historical association ofProsopis with arid environments. In particular, wewere interested in defining since when Prosopisspecies have occupied this kind of environment. The
analysis performed suggested an ancient associationof Prosopis with arid conditions (Fig. 6), alreadyestablished before the splitting of the two maingroups within ASP + Xerocladia clade (26.7–33 Mya).This result disagrees with Roig (1993) who proposedthat extant species of Prosopis occupying arid regionsin America have originated from ancestral lineages ofthis genus which were restricted to wetter regions.Palaeoenvironmental data suggest that, although thedevelopment of large arid areas in America started inthe late Miocene (see below), there were local aridconditions in North America (Axelrod, 1979a) andSouth America (Volkheimer, 1971; Jordan, Schluneg-ger & Cardozo, 2001) during much of the Tertiary. Itis possible therefore that ancestral lineages of Proso-pis had occupied these sites until the later spreadingof arid conditions.
DIVERSIFICATION PATTERN AND CAUSES
One of the predictions derived from the hypothesis ofaridity-driven diversification of Prosopis was a tem-poral correlation between the spreading of arid areasin the Americas and the diversification of the genus.Molecular clock dating showed that, except for thesplitting between the two main groups, the diver-gence among extant lineages within ASP clade wouldhave occurred in the Late Miocene or more recently,coincident with the estimated spreading of arid envi-ronments in America. Although with distinctive char-acteristics in each case, the development of aridenvironments occur simultaneously in North andSouth America starting in the Late Miocene andcontinuing in the Pliocene and Pleistocene. In NorthAmerica, the development of regional arid areas isthought to have started in the late Miocene toPliocene (Graham, 1999; Riddle & Hafner, 2006) asso-ciated with the uplift of the plateaus in WesternNorth America. The Sonora desert, where somespecies of the genus are present today, would havedeveloped from 15 to 8 Mya (van Devender, 2000).This process of aridification continued until the Pleis-tocene, reaching a considerable distribution in theinterglacials (Axelrod, 1979a,b). In South America,the spreading of large areas of semiarid-arid condi-tions possibly started in the late Miocene concomi-tantly with the uplift of the Andes in the Quechuadistrophic phase when it reached half of its presentelevation (Gregory-Wodzicki, 2000) and started to acteffectively as a barrier to moisture laden winds.During the ‘Diaguita’ diastrophic Phase, the finaluplift of the Pampean Mountain Range and theCentral Andes produced a rain-shadow effect thatresulted in the extremely xeric conditions existing atpresent on the areas located between them (Pascual,Ortiz-Jaureguizar & Prado, 1996; Alberdi. Bonadonna
PROSOPIS PHYLOGENY AND EVOLUTION 633
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
& Ortiz-Jaureguizar, 1997). These climatic changeswould have produced the establishment of the biomeswhere Prosopis is established today (Axelrod, KalinArroyo & Raven, 1991; Pascual et al., 1996).
The possible correlation of the ASP diversificationwith the expansion of arid environments is also sup-ported by the analysis of the temporal changes ofdiversification rates. LTT analysis indicates a clearincrease in diversification rates from the late Mioceneuntil the present compared to previous times. Whenthis pattern is analysed in detail, two different shiftstowards higher rates may be distinguished. The firstof those occurred in the late Miocene and is coincidentwith the start of the divergence within each of the twomajor lineages of the ASP. The second shift appears tobe associated with the start of mesquites divergencein the Pliocene and is in agreement with the highdiversification rate estimated for this group (Table 4).A wrong assignment of missing taxa could havepotentially affected the conclusion derived from thisanalysis. However, as most of the species not sampledbelong to Series Chilensis, Ruscifoliae, and Pallidae,which formed the highly supported mesquite clade, awrong assignment seems to be improbable.
The combined assessment of the results obtained inthe present study is in agreement with the hypothesisthat Prosopis evolved and diversified concomitantlywith the spreading of arid areas in the Americas. Thisresult is coincident with those obtained for otherdry-lover plant groups distributed almost exclusivelyin North America, which is suggested to have under-gone a parallel radiation in the last 10 Myr (Good-Avila et al., 2006). To our knowledge, no previousstudy found this pattern for a group with mainlySouth American distribution, as is the case of Proso-pis. Consequently, the results of the present studysuggest an extension of this parallel radiation to acontinental scale.
ACKNOWLEDGEMENTS
We thank P. Steibel, H. Troiani, A. Vilela, andG. Humphreys for helping in sampling; N. Iusem andL. Maskin for helping in laboratory issues; F. Prevostifor helpful comments on the manuscript; E. Ulibarrifor helping in SI herbarium; P. Harris for P. africanamaterial; O. Ruiz for L. japonicus material; J. JuarezMuñoz for P. palmeri, Prosopis tamaulipana andProsopis articulata material; C. Pometti for A. cavenmaterial; M. Luckow for lending P. nigra and X. viri-diramis sequences and X. viridiramis DNA; A. Gott-lieb for helping in different steps of this study. S.A.C.is fellow of Consejo Nacional de InvestigacionesCientíficas y Técnicas (CONICET, Argentina). B.O.S.and J.C.V. are members of Carrera del InvestigadorCientífico of CONICET. This study was supported by
grants to B.O.S. from University of Buenos Aires(UBACYT_X321) Agencia Nacional de PromociónCientífica y Tecnológica (PICT 32064/05), and a grantto J.C.V. from CONICET (PIP 5122/04) grants. Theauthors wish to dedicate this manuscript to thememory of our mentor and advisor Professor CarlosA. Naranjo. Santiago A. Catalano wishes also to dedi-cate this manuscript to the memory of his father.
REFERENCES
Alberdi MT, Bonadonna FP, Ortiz-Jaureguizar E. 1997.Chronological correlation, paleoecology, and paleobiogeogra-phy of the Late Cenozoic South American ‘Rionegran’ land-mammal fauna: a review. Revista Española de Paleontología12: 249–255.
Anzótegui LM, Herbst R. 2004. Megaflora (hojas y frutos)de la Formación San José (Mioceno Medio) en río Seco,departamento Santa María, provincia de Catamarca, Argen-tina. Ameghiniana 41: 423–436.
Axelrod DI. 1979a. Age and origin of Sonoran desert vegeta-tion. Occasional Papers of the California Academy of Sci-ences 132: 1–74.
Axelrod DI. 1979b. Desert vegetation, its age and origin. In:Goodin JR, Northington DK, ed. Arid land plant resources.Lubbock, TX: International Center for Arid and SemiaridLand Studies, 1–72.
Axelrod D, Kalin Arroyo M, Raven P. 1991. Historicaldevelopment of temperate vegetation in the Americas.Revista Chilena de Historia Natural 64: 413–446.
Baldwin BG, Sanderson MJ. 1998. Age and rate of diver-sification of the Hawaiian silversword alliance (Composi-tae). Proceedings of the National Academy of Sciences of theUnited States of America 95: 9402–9406.
Barraclough TG, Vogler AP. 2002. Recent diversificationrates in North American tiger beetles estimated from adated mtDNA phylogenetic tree. Molecular Biology andEvolution 19: 1706–1716.
Bell CD, Donoghue MJ. 2005. Phylogeny and biogeographyof Valerianaceae (Dipsacales) with special reference to theSouth American valerians. Organisms, Diversity and Evo-lution 5: 147–159.
Bessega C, Ferreyra LI, Vilardi JC, Saidman BO. 2000.Unexpected low genetic differenatiation among allopatricspecies of Section Algarobia of Prosopis (Leguminoseae).Genetica 109: 255–266.
Bessega C, Saidman BO, Vilardi JC. 2005. Genetic rela-tionships among American species of Prosopis (Legumino-sae) based on enzyme markers. Genetics and MolecularBiology 28: 277–286.
Bessega C, Vilardi JC, Saidman BO. 2006. Genetic rela-tionships among American species of the genus Prosopis(Mimosoideae, Leguminosae) inferred from ITS sequences:evidence for long-distance dispersal. Journal of Biogeogra-phy 33: 1905–1915.
Bremer K. 1994. Branch support and tree stability. Cladis-tics 10: 295–304.
634 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
Burghardt AD. 1995. La identidad de cuatro especies deProsopis L. expresada a travéz de sus patrones electroforeti-cos. Mendeliana 12: 38–50.
Burkart A. 1976. A monograph of the genus Prosopis(Leguminosae subFam Mimosoideae). Journal of the ArnoldArboretum 57: 219–249.
Burkart A, Simpson BB. 1977. The genus Prosopis andannoted key to the species of the world. In: Simpson B, ed.Mesquite: its biology in two desert ecosystems. Stroudsburg,PA: Dowden, Hutchinson & Ross, 201–215.
Coyne JA, Orr HA. 2004. Speciation. Sunderland, MA:Sinauer Associates.
Demesure B, Sodzi N, Petit RJ. 1995. A set of universalprimers for amplification of polymorphic non-coding regionsof mitochondrial and chloroplast DNA in plants. MolecularEcology 4: 129–131.
van Devender TR. 2000. The deep history of the SonoranDesert. In: Phillips SJ, Comus PW, eds. A natural history ofthe Sonoran desert. Tucson, AZ: Arizona-Sonora DesertMuseum Press; Berkeley, CA: University of CaliforniaPress, 61–69.
FAO. 2005. New LocClim 1.06 local climate estimator. Agrom-eteorology group. Rome: FAO/SDRN.
Farris S. 1970. Methods for computing Wagner trees. Sys-tematic Zoology 34: 21–24.
Goloboff PA. 1999. Analyzing large data sets in reasonabletimes: solutions for composite optima. Cladistics 15: 415–428.
Goloboff PA, Farris JS, Nixon K. 2003. TNT: tree analysisusing new technology. Software and documentation avail-able at http://www.zmuc.dk/public/phylogeny
Goloboff PA, Mattoni CI, Quinteros AS. 2006. Continuouscharacters analyzed as such. Cladistics 22: 589–601.
Good-Avila SV, Souza V, Gaut BS, Eguiarte LE. 2006.Timing and rate of speciation in Agave (Agavaceae). Pro-ceedings of the National Academy of Sciences of the UnitedStates of America 103: 9124–9129.
Graham A. 1999. Late cretaceous and cenozoic history ofNorth American vegetation. New York, NY: Oxford Univer-sity Press.
Graham CH, Ron SR, Santos JC, Schneider CJ, MoritzC. 2004. Integrating phylogenetics and environmental nichemodels to explore speciation mechanisms in Dendrobatidfrogs. Evolution 58: 1781–1793.
Gregory-Wodzicki KM. 2000. Uplift history of the centraland northern Andes: a review. Geological Society of AmericaBulletin 112: 1091–1105.
Guinet PH, Bessedik M. 1984. Presence de Genre Prosopis(Leguminosae-Mimosoideae) a l’Aquitanien Basal dansl’aude (Languedoc-France). Pollen et Spores 26: 101–108.
Hall TA. 1999. BioEdit: a user-friendly biological sequencealignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98.
Hardy CR, Linder HP. 2005. Intraspecific variability andtiming in ancestral ecology reconstruction: a test case fromthe Cape flora. Systematic Biology 54: 299–316.
Herendeen PS, Crepet WL, Dilcher DL. 1992. The fossilhistory of the leguminosae: phylogenetic and biogeographic
implications. In: Hendereen PS, Dilcher DL, eds. Advance inlegume systematics. Kew: The Royal Botanical Garden, 313–316.
Hey J. 1992. Using phylogenetic trees to study speciation andextinction. Evolution 46: 627–640.
Hodges SA, Arnold ML. 1995. Spurring plant diversifica-tion: are floral nectar spurs a key innovation? Proceedings ofthe Royal Society of London Series B, Biological Sciences262: 343–348.
Howarth DG, Baum DA. 2002. Phylogenetic utility of anuclear intron from nitrate reductase for the study of closelyrelated plant species. Molecular Phylogenetics and Evolu-tion 23: 525–528.
Hughes CE, Eastwood R. 2006. Island radiation on a con-tinental scale: exceptional rates of plant diversification afteruplift of the Andes. Proceedings of the National Academy ofSciences of the United States of America 103: 10334–10339.
Jordan TE, Schlunegger F, Cardozo N. 2001. Unsteadyand spatially variable evolution of the Neogene AndeanBermejo foreland basin, Argentina. Journal of South Ameri-can Earth Sciences 14: 775–798.
Klak C, Reeves G, Hedderson T. 2003. Unmatched tempoof evolution in Southern African semi-desert ice plants.Nature 427: 63–65.
Kluge AG. 1989. A concern for evidence and a phylogenetichypothesis of relationships among Epicrates (Boidae, Ser-pentes). Systematic Zoology 38: 7–25.
Lavin M, Thulin M, Labat J, Pennington RT. 2000.Africa, the odd man out: molecular biogeography of Dalber-gioid legumes (Fabaceae) suggests otherwise. SystematicBotany 25: 449–467.
Lavin M, Herendeen PS, Wojciechowski MF. 2005. Evo-lutionary rates analysis of Leguminosae implicates a rapiddiversification of lineages during the Tertiary. SystematicBiology 54: 575–594.
Lewis PO. 1998. Maximum likelihood as an alternative toparsimony for inferring phylogeny using nucleotidesequence data. In: Sotis DE, Soltis PS, Doyle JJ, eds.Molecular systematics of plants II DNA sequencing lluwer.Boston, MA: Academic Publisher, 132–163.
Luckow M, Miller JT, Jobson R, Murphy DJ. 2004.Generic relationships in the Mimosoideae (Fabaceae).Botany 2004, Alpine Diversity: Adapted to the Peaks, SaltLake City, Utah, USA. Session 19 (Colloquium) RupertBarneby and his legume legacy. Abstract 114, 29.
Magallon-Puebla S, Ceballos-Ferriz SRS. 1994. Fossillegume fruits from tertiary strata of Puebla, Mexico. Cana-dian Journal of Botany 72: 1027–1038.
Malcomber S. 2002. Phylogeny of Gaertnera Lam. (Rubi-aceae) based on multiple DNA markers: evidence of a rapidradiation in a widespread, morphologically diverse genus.Evolution 56: 42–57.
Miller JT, Bayer RJ. 2001. Molecular phylgenetics of Acacia(Fabaceae: Mimosoideae) based on the chloroplast matKconding sequence and flanking trnK intron spacer regions.American Journal of Botany 88: 697–705.
Moore MJ, Jansen RK. 2006. Molecular evidence for theage, origin, and evolutionary history of the American desert
PROSOPIS PHYLOGENY AND EVOLUTION 635
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
plant genus Tiquilia (Boraginaceae). Molecular Phylogenet-ics and Evolution 39: 668–687.
Naranjo CA, Poggio L, Enus Zeiger S. 1984. Phenolchromatography, morphology and cytogenetics in threespecies and natural hybrids of Prosopis (Leguminosae-Mimosoideae). Plant Systematic and Evolution 144: 257–276.
Nee S, Holmes EC, Rambaut A, Harvey PH. 1995. Infer-ring population history from molecular phylogenies. Philo-sophical Transactions of the Royal Society of London, SeriesB 349: 25–31.
Palacios RA, Bravo LD. 1981. Hibridación natural en Proso-pis (Leguminosae) en la region chaqueña argentina. Eviden-cias morfologicas y cromatográficas. Darwiniana 23: 3–35.
Pascual R, Ortiz-Jaureguizar E, Prado JL. 1996. Landmammals: paradigm of Cenozoic South American geobioticevolution. In: Arratia G, ed. Contribution of Southern SouthAmerica to Vertebrate Paleontology. Munich: MünchnerGeowissenschaftliche Abhandlungen (A), 265–319.
Pasiecznik NM, Felker P, Harris PJC, Harsh LN, CruzG, Tewari JC, Cadoret K, Maldonado LJ. 2001. Theprosopis juliflora-prosopis pallida complex: a monograph.Coventry: HDRA.
Piel KM. 1971. Palynolgy of Oligocene sediments from BritishColumbia. Canadian Journal of Botany 49: 1885–1920.
Posada D, Crandall KA. 1998. Modeltest: testing the modelof DNA substitution. Bioinformatics 14: 817–818.
Rambaut AP. 2002. Phylogen, Version 1.0. Oxford: Depart-ment of Zoology, University of Oxford.
Ramírez L, De La Vega A, Razkin N, Luna V, HarrisPJC. 1999. Analysis of the relationships between species ofthe genus Prosopis revealed by the use of molecularmarkers. Agronomie 19: 31–43.
Richardson JE, Pennington RT, Pennington TD,Hollingsworth PM. 2001. Rapid diversification of aspecies-rich genus of Neotropical rain forest trees. Science293: 2242–2245.
Riddle BR, Hafner DJ. 2006. A step-wise approach to inte-grating phylogeographic and phylogenetic biogeographicperspectives on the history of a core North American warmdeserts biota. Journal of Arid Environments 66: 435–461.
Roig FA. 1993. Informe nacional para la selección de germo-plasma en especies de Prosopis en la República Argentina.Mendoza, Argentina: IADIZA-CRICYT, 1–37.
Ronquist FR, Huelsenbeck JP. 2005. Mr Bayes 3. Baye-sian phylogenetics inference under mixed models. Bioinfor-matics 19: 1572–1574.
Ross JH. 1975. Flora of Southern Africa. Vol. 16, Part I. Mimo-soideae, Fabaceae. Pretoria: Deptartment of Agriculture.
Rzedowski J. 1988. Analisis de la distribución geográficadel complejo Prosopis (Leguminosae, Mimosoideae) enNorteamerica. Acta Botanica Mexicana 3: 7–19.
Saidman B, Vilardi J. 1987. Analysis of the genetic simi-larities among seven species of Prosopis (Leguminoseae:Mimosoideae). Theoretical and Applied Genetics 75: 109–116.
Saidman BO, Vilardi JC. 1993. Genetic variability andgermplasm conservation in the genus Prosopis. In: Puri S,
ed. Nursery technology of forest tree species of arid andsemiarid regions. New Delhi: Winrock-Oxford & IBH Pub-lishing Co. PVT. Ltd., 187–198.
Saidman BO, Vilardi JC, Pocoví MI, Acreche N. 1996.Isozyme studies in Argentine species of the Section Strom-bocarpa, Genus Prosopis (Leguminosae). Journal of Genetics75: 139–149.
Saidman BO, Vilardi JC, Montoya S, Dieguez MJ, HoppHE. 1998. Molecular markers: a tool for the understandingof the relationships among species of Prosopis (Legumino-sae, Mimosoideae). In: Puri S, ed. Tree improvement: appliedresearch and technology transfer. Enfield, NH: Science Pub-lishers, Inc., 324.
Sanderson MJ. 2002. Estimating absolute rates of molecularevolution and divergence times: a penalized likelihoodapproach. Molecular Biology and Evolution 19: 101–109.
Sanderson MJ. 2003. r8s: inferring absolute rates of molecu-lar evolution and divergence times in the absence of amolecular clock. Bioinformatics 19: 301–302.
Schrire BD, Lavin M, Lewis GP. 2005. Global distributionpatterns of the Leguminosae: insights from recent phylog-enies. In: Friis I, Balslev H, eds. Plant diversity and com-plexity patterns: local, regional and global dimensions.Biologiske Skrifter 55: 375–422.
Simmons MP, Ochoterena H. 2000. Gaps as characters insequence-based phylogenetic analyses. Systematic Biology49: 369–381.
Strand AE, Leebens-Mack J, Milligan BG. 1997. NuclearDNA-based markers for plant evolutionary biology. Molecu-lar Ecology 6: 113–118.
Swofford DL. 2002. PAUP* phylogenetic analysis using par-simony (*and other methods), Version 4. Sunderland, MA:Sinauer Associates.
Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universalprimers for amplification of three non-coding regions ofchloroplast DNA. Plant Molecular Biology 17: 1105–1109.
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F,Higgins DG. 1997. The CLUSTAL–windows interface: flex-ible strategies for multiple sequence alignment aided byquality analysis tools. Nucleic Acids Research 25: 4876–4882.
UNEP. 1992. World atlas of desertification. London: EdwardArnold.
Verboom GA, Linder HP, Stock WD. 2003. Phylogenetics ofthe grass genus Ehrharta: evidence for radiation in thesummer-arid zone of the South African cape. Evolution 57:1008–1021.
Verga AR. 1995. Genetische untersuchungen an Prosopis chil-ensis und Prosopis flexuosa (Mimosaceae) im trockenenChaco Argentiniens. Göttingen: Universität Göttingen.
Volkheimer W. 1971. Aspectos paleoclimáticos del Terciarioargentino. Revista del Museo Argentino de Ciencias Natu-rales Bernardino Rivadavia 1: 243–262.
Von Hagen KB, Kadereit JW. 2001. The phylogeny of Gen-tianella (Gentianaceae) and its colonization of the southernhemisphere as revealed by nuclear and chloroplast DNAsequence variation. Organisms Diversity and Evolution 1:61–79.
636 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
Wang A, Yang M, Liu J. 2005. Molecular phylogeny, recentradiation and evolution of gross morphology of the rhubarbgenus Rheum (Polygonaceae) inferred from chloroplast DNAtrnL-F sequences. Annals of Botany 96: 489–498.
Xiang Q, Manchester SR, Thomas DT, Zhang W, Fan C.2005. Phylogeny, biogeography, and molecular dating of
cornelian cherries (Cornus, Cornaceae): traking tertiaryplant migration. Evolution 59: 1658–1700.
Yule GU. 1924. A mathematical theory of evolution based onthe conclusions of Dr. J. C. Willis. Philosophical Transac-tions of the Royal Society of London, Series B 213: 21–87.
SUPPLEMENTARY MATERIAL
The following material is available for this article online:
Figure S1. Strict consensus of 4680 most parsimonious trees obtained in the five-marker analysis (trnS-psbC,G3pdh, NIA, trnL-trnF, trnK-matK).Table S1. Sequences downloaded from GenBank.Table S2. Maximum and minimum humidity index for species included in the ancestral climate estimationanalysis.
This material is available as part of the online article from:http://www.blackwell-synergy.com/doi/abs/10.1111/j.1095-8312.2007.00907.x(This link will take you to the article abstract).
Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementarymaterials supplied by the authors. Any queries (other than missing material) should be directed to thecorresponding author for the article.
PROSOPIS PHYLOGENY AND EVOLUTION 637
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
AP
PE
ND
IX1
Lis
tof
spec
ies
sequ
ence
din
the
pres
ent
stu
dy,
indi
cati
ng
the
Sec
tion
and
Ser
ies
tow
hic
hth
eybe
lon
g,vo
uch
ersp
ecim
en,
the
nat
ura
ldi
stri
buti
onan
dac
cess
ion
nu
mbe
rs
Spe
cies
Sec
tion
–S
erie
sV
ouch
ersp
ecim
en/s
eed
ban
kac
cesi
onN
atu
ral
dist
ribu
tion
Gen
Ban
kA
cces
ion
nu
mbe
rsG
3pdh
trn
S-p
sbC
NIA
trn
L-t
rnF
trn
K/m
atK
Pro
sopi
sci
ner
aria
(L.)
Dru
ceP
roso
pis
DA
NID
AN
°01
089/
82A
FE
F16
5221
EF
1653
09E
F16
5269
EF
1652
92E
F16
5287
/E
F16
5248
Pro
sopi
saf
rica
na
(Gu
ill.,
Per
r.,
&R
ich
.)Ta
ube
rt
An
onyc
hiu
mH
DR
AN
°63
AF
EF
1652
09E
F16
5297
EF
1652
53E
F16
5291
EF
1652
86/
EF
1652
51
Pro
sopi
sst
rom
buli
fera
(Lam
.)B
enth
amS
trom
boca
rpa–
Str
ombo
carp
aeH
un
zike
r&
Gam
erro
1135
9(S
I)S
AE
F16
5241
EF
1653
27E
F16
5281
––
Pro
sopi
sre
ptan
sB
enth
amS
trom
boca
rpa–
Str
ombo
carp
aeS
aidm
an36
(BA
FC
)S
A,
NA
EF
1652
37E
F16
5333
––
–
Pro
sopi
sto
rqu
ata
(Cav
anil
les
exL
agas
ca)
Str
ombo
carp
a–S
trom
boca
rpae
Hu
nzi
ker
9733
(SI)
SA
EF
1652
44E
F16
5330
––
–
Pro
sopi
spu
besc
ens
Ben
tham
Str
ombo
carp
a–S
trom
boca
rpae
Eva
ns
N°
15(U
SD
A-U
SA
)N
AE
F16
5236
EF
1653
23–
––
Pro
sopi
spa
lmer
iS
.W
atso
nS
trom
boca
rpa–
Str
ombo
carp
ae–
NA
EF
1652
34E
F16
5322
EF
1652
78–
–
Pro
sopi
sfe
rox
Gri
seba
chS
trom
boca
rpa–
Cav
enic
arpa
eH
un
zike
ret
al.
1045
1(S
I)S
AE
F16
5223
EF
1653
11–
––
Pro
sopi
sta
mar
ugo
F.P
hil
ippi
Str
ombo
carp
a–C
aven
icar
pae
DA
NID
AN
°01
215/
83S
AE
F16
5242
EF
1653
28–
EF
1652
93E
F16
5289
/E
F16
5250
Pro
sopi
sar
gen
tin
aB
urk
art
Mon
ilic
arpa
BN
GP
roso
pis
N°
15-8
8.S
AE
F16
5212
EF
1653
00E
F16
5261
EF
1652
90E
F16
5288
/E
F16
5249
Pro
sopi
sse
rica
nth
aG
ille
sex
Hoo
ker
&A
rnot
t
Alg
arob
ia–
Ser
ican
thae
Sai
dman
&V
ilar
di96
1(B
AF
C)
SA
EF
1652
40E
F16
5326
EF
1652
80–
–
Pro
sopi
sku
ntz
eiH
arm
sA
lgar
obia
–S
eric
anth
aeS
aidm
an&
Vil
ardi
516
(BA
FC
);S
aidm
an&
Vil
ardi
521
(BA
FC
)
SA
EF
1652
30E
F16
5231
EF
1653
18E
F16
5319
EF
1652
76E
F16
5277
––
Pro
sopi
sru
scif
olia
Gri
seba
chA
lgar
obia
–R
usc
ifol
iae
Sai
dman
&V
ilar
di42
0(B
AF
C)
SA
EF
1652
39E
F16
5325
EF
1652
57–
–
Pro
sopi
svi
nal
illo
Stu
cker
tA
lgar
obia
–R
usc
ifol
iae
Sai
dman
&V
ilar
di50
5(B
AF
C);
Sai
dman
&V
ilar
di51
1(B
AF
C)
SA
EF
1652
45E
F16
5246
EF
1653
31E
F16
5332
EF
1652
83E
F16
5284
––
Pro
sopi
sd
enu
dan
sB
enth
amA
lgar
obia
–D
enu
dan
tes
Cat
alan
o33
(BA
FC
)S
AE
F16
5222
EF
1653
10E
F16
5270
––
Pro
sopi
sru
izle
ali
Bu
rkar
tA
lgar
obia
–D
enu
dan
tes
Cat
alan
o13
(BA
FC
)S
AE
F16
5238
EF
1653
24E
F16
5279
––
638 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
Pro
sopi
sca
stel
lan
osii
Bu
rkar
tA
lgar
obia
–D
enu
dan
tes
Cat
alan
o8
(BA
FC
)S
AE
F16
5218
EF
1653
06E
F16
5267
––
Pro
sopi
sh
um
ilis
Gil
les
exH
ooke
r&
Arn
ott
Alg
arob
ia–
Hu
mil
esC
atal
ano
1(B
AF
C)
SA
EF
1652
27E
F16
5315
EF
1652
75–
–
Pro
sopi
sca
mpe
stri
sG
rise
bach
Alg
arob
ia–
Pal
lida
eC
atal
ano
18(B
AF
C);
Cat
alan
o20
(BA
FC
)S
AE
F16
5216
EF
1652
17E
F16
5304
EF
1653
05E
F16
5265
EF
1652
66–
–
Pro
sopi
saf
fin
isS
pren
gel
Alg
arob
ia–
Pal
lida
eS
aidm
anet
al.
526
(BA
FC
)S
AE
F16
5208
EF
1652
96E
F16
5256
––
Pro
sopi
sar
ticu
lata
S.
Wat
son
Alg
arob
ia–
Pal
lida
e–
NA
EF
1652
13E
F16
5301
EF
1652
62–
–
Pro
sopi
sta
mau
lipa
na
Bu
rkar
tA
lgar
obia
–P
alli
dae
–N
AE
F16
5243
EF
1653
29E
F16
5282
––
Pro
sopi
sch
ilen
sis
(Mol
ina)
Str
un
tzem
end.
Bu
rkar
t
Alg
arob
ia–
Ch
ilen
ses
Sai
dman
&V
ilar
di91
1(B
AF
C);
Sai
dman
&V
ilar
di92
2(B
AF
C)
SA
EF
1652
19E
F16
5220
EF
1653
07E
F16
5308
EF
1652
68E
F16
5285
––
Pro
sopi
sju
lifl
ora
(Sw
artz
)D
CA
lgar
obia
–C
hil
ense
sH
un
zike
r10
039
(SI)
;H
un
zike
r10
040
(SI)
NA
,CA
,SA
EF
1652
28E
F16
5229
EF
1653
16E
F16
5317
EF
1652
58E
F16
5274
––
Pro
sopi
sn
igra
(Gri
seba
ch)
Hie
ron
ymu
s
Alg
arob
ia–
Ch
ilen
ses
Sai
dman
&V
ilar
di43
5(B
AF
C);
Sai
dman
315
(BA
FC
)
SA
EF
1652
32E
F16
5233
EF
1653
20E
F16
5321
EF
1652
59E
F16
5260
––
Pro
sopi
sca
lden
iaB
urk
art
Alg
arob
ia–
Ch
ilen
ses
Sai
dman
&V
ilar
di35
0(B
AF
C);
Sai
dman
&V
ilar
di13
8(B
AF
C)
SA
EF
1652
14E
F16
5215
EF
1653
02E
F16
5303
EF
1652
63E
F16
5264
––
Pro
sopi
sfl
exu
osa
DC
Alg
arob
ia–
Ch
ilen
ses
Sai
dman
&V
ilar
di30
2(B
AF
C);
Cat
alan
o10
(BA
FC
)
SA
EF
1652
24E
F16
5225
EF
1653
12E
F16
5313
EF
1652
71E
F16
5272
––
Pro
sopi
sgl
and
ulo
saTo
rrey
Alg
arob
ia–
Ch
ilen
ses
Eva
ns
N°
5(U
SD
A-U
SA
)N
AE
F16
5226
EF
1653
14E
F16
5273
––
Pro
sopi
sal
pata
coR
.A.
Ph
ilip
piA
lgar
obia
–C
hil
ense
sC
atal
ano
34(B
AF
C)
SA
EF
1652
11E
F16
5299
EF
1652
55–
–
Pro
sopi
sal
baG
rise
bach
Alg
arob
ia–
Ch
ilen
ses
Sai
dman
&V
ilar
di41
3(B
AF
C)
SA
EF
1652
10E
F16
5298
EF
1652
54–
–
Pro
sopi
das
tru
man
gust
icar
pum
Pal
acio
s&
Hoc
NA
Pet
er24
4(L
P)
SA
EF
1652
35E
F16
5295
––
–/E
F16
5252
Xer
ocla
dia
viri
dir
amis
Tau
b.N
AG
erm
ish
uiz
en82
44(P
RE
)A
FE
F58
5492
EF
5854
93–
Su
pp.
data
Su
pp.
data
Aca
cia
cave
n(M
ol.)
Mol
ina
NA
Sai
dman
etal
.78
9(B
AF
C)
SA
EF
1652
07E
F16
5294
EF
1652
06S
upp
.da
taS
upp
.da
ta
Lot
us
japo
nic
us
L.
NA
–S
AE
F16
5247
Su
pp.
data
–S
upp
.da
taS
upp
.da
ta
BN
GP
roso
pis,
Ban
coN
acio
nal
deG
erm
opla
sma
deP
roso
pis
(Cor
doba
,A
rgen
tin
a);
DA
NID
A,
Dan
ida
For
est
See
dC
entr
e,D
enm
ark;
GR
S-U
SD
A-U
SA
,U
nit
edS
tate
sD
epar
tmen
tof
Agr
icu
ltu
re;
HD
RA
,H
enry
Dou
bled
ayR
esea
rch
Ass
ocia
tion
,U
nit
edK
indo
m;
SI,
Dar
win
ion
Her
bari
um
,S
anIs
idro
,A
rgen
tin
a;B
AF
C,
Her
bari
um
ofF
acu
ltad
deC
ien
cias
Exa
ctas
yN
atu
rale
sU
BA
,Bu
enos
Air
es,A
rgen
tin
a;A
F,A
fric
a;A
S,A
sia;
CA
,Cen
tral
Am
eric
a;N
A,N
orth
Am
eric
a;S
A,S
outh
Am
eric
a;S
upp
.da
ta,
acce
ssio
nn
um
ber
incl
ude
din
Su
pple
men
tary
mat
eria
l(T
able
S1)
.
PROSOPIS PHYLOGENY AND EVOLUTION 639
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640
APPENDIX 2SEARCH STRATEGIES IN PARSIMONY AND
BAYESIAN ANALYSES
In three-marker analysis, parsimony searchesinvolved 100 Random Addition Sequences (RAS) fol-lowed by ten cycles of Tree Drifting. In the two-marker and five-marker analyses, the strategyinvolved 200 RAS followed by Tree Drifting andRandom Sectorial Searches (Goloboff, 1999). Thesetwo algorithms intend to overcome the problem oflocal optima during tree searches. In Tree Drifting,suboptimal solutions are accepted with a certainprobability in different moments of the swappingwhereas, in Sectorial Searches, reduced datasets arecreated for different parts of the tree which are analy-sed by tree-bisection-reconnection (TBR). Then, thebest tree for each sector is replaced in the originaltree. In all cases, trees found using these strategieswere used as starting trees for a final round of TBRkeeping a maximum of 10 000 trees. Constrainedsearches forcing groups not found in the most parsi-monious trees were run with the same settings.
Support was evaluated with jackknifing procedurewith an independent character removal probability of0.36. For each jackknifing replicate, search strategywas the same than previously described except thatthe number of RAS was decreased to 10. AbsoluteBremer Support (Bremer, 1994) was calculated savingup to 6000 suboptimal trees in successive steps up totrees 12 steps longer than the optimal length.
Bayesian phylogenetic analyses were performedwith the Metropolis-coupled Markov chain MonteCarlo algorithm implemented in MrBayes, version3.1.2 (Ronquist & Huelsenbeck, 2005). The optimalmodel was selected with Modeltest (Posada & Cran-dall, 1998) and the values of each parameter weredetermined during each run. The analysis involvedtwo independent runs of 2 ¥ 106 generations sampledevery 100 generations after a burnin period of0.50 ¥ 106 generations. Each run involved threeheated chains. Convergence was assessed evaluatinglikelihood plots and standard deviation of the splitfrequencies.
640 S. A. CATALANO ET AL.
© 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 93, 621–640