ebook img

Phylogenetic Relationships of Whiptail Lizards of the Genus Cnemidophorus (Squamata: Teiidae): A Test of Monophyly, Reevaluation of Karyotypic Evolution, and Review of Hybrid Origins PDF

64 Pages·2002·2.7 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Phylogenetic Relationships of Whiptail Lizards of the Genus Cnemidophorus (Squamata: Teiidae): A Test of Monophyly, Reevaluation of Karyotypic Evolution, and Review of Hybrid Origins

PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, NY 10024 Number 3365, 61 pp., 7 figures, 3 tables May 17, 2002 Phylogenetic Relationships of Whiptail Lizards of the Genus Cnemidophorus (Squamata: Teiidae): A Test of Monophyly, Reevaluation of Karyotypic Evolution, and Review of Hybrid Origins TOD W. REEDER,1 CHARLES J. COLE,2 AND HERBERT C. DESSAUER3 CONTENTS Abstract ....................................................................... 3 Introduction .................................................................... 3 Cnemidophorus Background and Classification ................................... 3 Higher-Level Relationships and Cnemidophorus Monophyly ....................... 4 Objectives of the Present Study ................................................ 6 Materials and Methods .......................................................... 7 Choice of Taxa ............................................................... 7 Molecular Data ............................................................... 7 DNA Data ................................................................... 7 Allozyme Data ............................................................... 8 Morphological Data ........................................................... 8 Phylogenetic Analysis ......................................................... 8 Results ....................................................................... 10 1Research Associate, Division of Vertebrate Zoology (Herpetology), American Museum of Natural History; As- sistantProfessor,DepartmentofBiology,SanDiegoStateUniversity,SanDiego,CA92182–4614;e-mail:treeder@ sunstroke.sdsu.edu. 2Curator, Division of Vertebrate Zoology (Herpetology), American Museum of Natural History; e-mail: cole@ amnh.org. 3Research Associate, Division of Vertebrate Zoology (Herpetology), American Museum of Natural History;Pro- fessor Emeritus, Department of Biochemistry and Molecular Biology, Louisiana State University Medical Center, NewOrleans,LA70119-2799. Copyright(cid:113)AmericanMuseumofNaturalHistory2002 ISSN0003-0082 2 AMERICAN MUSEUM NOVITATES NO. 3365 Uniformly Weighted Analysis ................................................. 10 Successive Approximations Analysis ........................................... 11 Effects of Initial Starting Tree in Successive Approximations ..................... 13 Phylogenetic Placement of Cnemidophorus murinus and Cnemidophorus ocellifer ... 13 Discussion .................................................................... 14 ‘‘Cnemidophorus’’ Phylogeny ................................................. 14 ‘‘Cnemidophorus’’ Paraphyly .................................................. 14 ‘‘Cnemidophorus’’ lemniscatus Group .......................................... 14 North American ‘‘Cnemidophorus’’ Clade ...................................... 18 ‘‘Ameiva’’ Phylogeny ........................................................ 20 Evolution of Tongue Characters Traditionally Used in Cnemidophorine Systematics 20 Taxonomic Implications and Nomenclatural Recommendations .................... 21 Aspidoscelis Fitzinger, 1843 ................................................... 22 Maternal Ancestor of Kentropyx borckiana ..................................... 24 Unisexual Species: An Overview .............................................. 25 Teioid Unisexual Species ..................................................... 25 Knowing Ancestors .......................................................... 28 Extent and Origin of Parthenogenetic Cloning in Vertebrates ...................... 29 Successive Approximations and Initial Starting Trees ............................ 29 Karyotype Evolution Revisited ................................................ 30 Summary and Conclusions ...................................................... 34 Acknowledgments ............................................................. 35 References .................................................................... 36 Appendix 1: Specimens Examined ............................................... 41 Appendix 2: Mitochondrial DNA Data ........................................... 45 Appendix 3: Allozyme Data ..................................................... 60 Appendix 4: Morphological Data ................................................ 61 2002 REEDER ET AL.: PHYLOGENETICRELATIONSHIPSOF CNEMIDOPHORUS 3 ABSTRACT Phylogenetic relationships of the whiptail lizards of the genus Cnemidophorus are inferred based on a combined analysis ofmitochondrial DNA,morphology,andallozymes.Withinthe Teiini, Teius and Dicrodon are the most basal lineages, and these two taxa form a graded series leading to a cnemidophorine clade containing Ameiva, Cnemidophorus, and Kentropyx. Cnemidophorus monophyly is not supported, with members of the neotropical ‘‘C.’’ lemnis- catus species group (except ‘‘C.’’ longicaudus) being more closely related to species in other neotropical cnemidophorine taxa (Ameiva and Kentropyx). Ameiva is also paraphyletic. The‘‘Cnemidophorus’’lemniscatusspeciesgroupisalsoparaphyletic,witha‘‘C.’’murinus (cid:49) ‘‘C.’’ lemniscatus complex clade being more closely related to Kentropyx than to ‘‘C.’’ lacertoides, ‘‘C.’’ longicaudus, and/or ‘‘C.’’ ocellifer. Although the ‘‘C.’’ lemniscatus species group is paraphyletic, the three remaining bisexual ‘‘Cnemidophorus’’ species groups(deppii, sexlineatus, and tigris species groups) are each monophyletic. Together, these three groups form a clade ((cid:53) North American ‘‘Cnemidophorus’’ clade), with the deppii and tigrisspecies groups being sister taxa. Within the ‘‘Cnemidophorus’’ deppii species group, the Baja Cali- fornia ‘‘C.’’ hyperythrus is the sister species to a more exclusive mainland Mexico clade containing ‘‘C.’’ deppii and ‘‘C.’’ guttatus. Except for a ‘‘C.’’ inornatus (cid:49) ‘‘C.’’ sexlineatus clade and a monophyletic ‘‘C.’’ gularis complex, the inferred inter- and intraspecificrelation- ships within the sexlineatus species group are weakly supported. In none of the inferred phy- logenies are the ‘‘C.’’ costatus populations (‘‘C.’’ c. costatus and ‘‘C.’’ c. griseocephalus) represented as each other’s closest relatives. Because of Cnemidophorus paraphyly, nomenclatural changes are recommended. Aspidos- celisFitzinger,1843,isresurrectedfortheNorthAmerican‘‘Cnemidophorus’’cladecontaining thedeppii,sexlineatus,andtigrisspeciesgroups(andtheunisexualtaxaassociatedwiththem). LizardsofthegenusAspidoscelisdifferfromallothercnemidophorinelizardsbythecombined attributesofabsenceofbasaltonguesheath,posteriorportionoftongueclearlyforked,smooth ventral scutes, eight rows of ventral scutes at midbody, absence of anal spurs in males, me- soptychial scales abruptly enlarged over scales of gular fold (more anterior mesoptychials becoming smaller), three parietal scales, and three or four supraocular scales on each side. Previous studies using morphology and allozymes have determined that the unisexual Ken- tropyxborckianaoriginatedfromahistoricalhybridizationeventbetweenthebisexualspecies K. calcarata and K. striata. In this study mitochondrial DNA confirms K. striata as the ma- ternal ancestor of K. borckiana. Areviewofourcurrentknowledgeofteioidunisexualsandtheirhybridoriginsisprovided. Also, a reevaluation of teiine chromosomal evolution is presented from a phylogenetic per- spective. These reviews elucidate the paradox that the capability of instantly producing par- thenogeneticclonesthroughonegenerationofhybridizationhasexistedforapproximately200 million years, yet the extant unisexual taxa are of very recent origins. Consequently, these lineages must be ephemeral compared to those of bisexual taxa. INTRODUCTION approximately 50 species known (for recent summaries see Maslin and Secoy, 1986; CNEMIDOPHORUS BACKGROUND AND Wright, 1993), with new species continuing CLASSIFICATION tobefound(e.g.,Markezichetal.,1997;Ro- Teiid whiptail lizards of the genus Cnem- cha et al., 1997, 2000). idophorus range widely in the New World, Because of their abundance and conspic- extending from the northern United States uousnature,whiptailsareanecologicallyim- southward to Argentina, and occupy many portant squamate lizard clade, which is re- diverse ecological communities. However, flectedbythegreatnumberofecologicaland while exhibiting this extensive distribution, life history studies conducted on this group their greatest diversity occurs in North (reviewed in Wright and Vitt, 1993). Cnem- America, where they areaconspicuouscom- idophorushasbeen(andcontinuestobe)one ponent of the herpetofauna of the arid and of the mostextensivelystudiedgeneraofliz- semiarid regions of the southwestern U.S. ards, third only to Sceloporus and Anolis andMexico.Byconservativecount,thereare (Dunham et al., 1988). Besides their abun- 4 AMERICAN MUSEUM NOVITATES NO. 3365 TABLE1 Cnemidophorus Species Groupsa dance and geographic proximity to North cies (C. lemniscatus) extending into Central American biologists, one of the reasons America. Two of the northern Cnemidopho- whiptails have been so intensively studied is rus species groups (cozumela and tesselatus) the occurrence of parthenogenetic all-female are composed entirely of parthenogenetic species (of interspecific hybrid origin; see species. The origins of the unisexual species below) within this diverse clade. Approxi- in both of these groupsinvolvehybridization mately one-third of the described speciesare between bisexual species from different spe- unisexual, with the majority of these all-fe- cies groups (i.e., sexlineatus group (cid:51) deppii male species occurring in the southwestern group (cid:53) cozumela group; sexlineatus group U.S. and northern Mexico (Wright, 1993). (cid:51) tigris group (cid:53) tesselatus group). The lem- Diploid and triploid unisexual species have niscatus and sexlineatus groups each possess evolved many times in Cnemidophorus, in bisexualandunisexualspecies.However,un- each instance the switch from sperm-depen- like the aforementioned completely unisex- dent to sperm-independent reproduction oc- ual groups, the unisexuals in the lemniscatus curring in one generation in an F interspe- andsexlineatusgroupsarederivedexclusive- 1 cific hybrid (for reviews, seeDarevskyetal., ly from hybridizations betweenspecieswith- 1985;DessauerandCole,1989;Moritzetal., in theirrespectivegroups(intragrouphybrid- 1989a, 1992a; Darevsky, 1992; Cole and izations). Dessauer, 1995), and dynamic hybridization presentlyoccursinnature(e.g.,Walkeretal., HIGHER-LEVEL RELATIONSHIPS AND 1989; Dessauer et al., 2000; Taylor et al., CNEMIDOPHORUS MONOPHYLY 2001). Consequently, whiptail lizards are used broadly in research, particularly in re- WhileCnemidophorushasbeenextensive- productive biology, population genetics, ly studied and much is known about its bi- physiological ecology, and evolutionary bi- ology, ecology, and natural history, the spe- ology, often with emphasis on the instanta- cific phylogenetic placement of Cnemido- neous, multiple and independent origins of phorus within the Teiidae, as well as the parthenogenetic cloning. higher-level relationships within Cnemido- The species of Cnemidophorus are cur- phorus, has received little attention. Presch rently allocated to six species groups (table (1974) provided osteological evidence that 1). Based on external morphology and kar- the macroteiids consisted of two major yology, these groups were erected by Lowe groups: Teiini (including Ameiva, Cnemido- et al. (1970), who modified Burt’s (1931) ar- phorus,Dicrodon,Kentropyx,andTeius)and rangement. All except the lemniscatus group Tupinambini (including Callopistes, Crocod- are confined to North and Central America. ilurus, Dracaena, and Tupinambis). Within The lemniscatus group is largely a South the Teiini, Ameiva, Cnemidophorus, and American radiation, with only a single spe- Kentropyx shared the most similarities,lead- 2002 REEDER ET AL.: PHYLOGENETICRELATIONSHIPSOF CNEMIDOPHORUS 5 ing Presch to hypothesize that these three taxa were more closely related to each other than any were to Dicrodon or Teius. How- ever,therewerenoderivedosteologicalchar- acters provided to resolve the relationships amongAmeiva,Cnemidophorus,andKentro- pyx. Informally we refer to these three very similar taxa as the cnemidophorines. Using external morphology and intuition, Burt(1931)wasthefirsttohypothesizehigh- er-level relationships within Cnemidophorus (fig. 1A). Because members of the South American lemniscatus group shared some characteristicswithotherSouthAmericante- iids (e.g., Ameiva), Burt (1931) postulated that the lemniscatus group was the most primitive lineage within Cnemidophorus. The ancestor of the North American groups was hypothesized to have been derived from the lemniscatus group, with this lineage giv- ing rise to the deppii (excluding C. hypery- thrus) and sexlineatus groups. Burt (1931) also proposed that his tesselatus group (in- cluding the as-yet-to-be-described tigris group) and hyperythrus groups were derived from the sexlineatus group. Based onkaryology,externalmorphology, and knowledge of the existence of unisexual species, Lowe et al. (1970) modified the higher-level classification and hypothesized relationships within Cnemidophorus. The evolutionary scenario (fig. 1B) proposed by Lowe et al. (1970) was largely influencedby their assumption that the chromosomes of vertebrates evolve primarily by means of Robertsonian centric fusion, thus resultingin the reduction of diploid chromosome num- ber.Membersofthedeppiigrouppossessthe highest diploid number (2n (cid:53) 52) within Cnemidophorus. Given this, Lowe et al. (1970) suggested that the deppii group (in- cluding the cozumela group) represented the Fig. 1. Previous phylogenetic hypotheses of most ‘‘primitive’’ lineage within Cnemido- higher-level relationships within Cnemidophorus. phorus, possessing a karyotype essentially A. Modified hypothesis of Burt (1931). B. Mod- identical to that of the hypothesized ancestor ified hypothesis of Lowe et al. (1970). of Cnemidophorus. Such a conclusion dif- fered from Burt (1931), who suggested that the lemniscatus group was ancestral to the type (2n (cid:53) 50) from the deppii group/ances- remaining Cnemidophorus species groups. tral karyotype. The sexlineatus and tigris Lowe et al. (1970) postulated that the lem- groups were proposed to be sister taxa, with niscatus group evolved from a deppii-like their common ancestor being derived from a ancestor, requiring only a single centric fu- deppii-like ancestor (via three centric fu- sion to derive the lemniscatus group karyo- sions). 6 AMERICAN MUSEUM NOVITATES NO. 3365 Based on mitochondrial DNA restriction eses suggesting that various lineages of site data, Moritz et al. (1992a) provided the Cnemidophorus were independently derived first explicit phylogenetic analysis of higher- from ancestral South American ‘‘stocks’’ level relationships withinCnemidophorus.In (e.g., Burt, 1931; Lowe et al., 1970) suggest that study, C. lemniscatus was used to root that Cnemidophorus monophyly is in ques- the resulting phylogeny. This outgroup tion and should be rigorously tested. Al- choicewasbasedonBurt(1931)andthefact though taxon sampling was limited (ingroup that the greatest observed genetic distances taxa (cid:53) three Cnemidophorus species groups, were between C. lemniscatus and the re- Ameiva, and Kentropyx), a phylogenetic maining Cnemidophorus species (see also study using allozymes by Dessauer and Cole Dessauer and Cole, 1989). Moritz et al. (1989) provided support for Cnemidophorus (1992a) provided strong support for a sister paraphyly, with the lemniscatus group hy- group relationship between the sexlineatus pothesizedtobemorecloselyrelatedtoKen- and tigris groups, corroborating the hypoth- tropyx than to a clade containing the sexli- esis of Lowe et al. (1970). These mitochon- neatus and tigris groups. drialdataalsosupportedtheplacementofthe deppii group as the sister taxon to the sexli- OBJECTIVES OF THE PRESENT STUDY neatus group (cid:49) tigris group clade. Mono- phyly of the deppii and sexlineatus groups As the use of Cnemidophorus increases in was also supported by Moritz et al. (1992a). research and the literature mushrooms, it be- However, because of the limited sampling, comes increasinglyimportanttoestablishthe these conclusions could only be considered validity of this taxon as a monophyletic preliminary. Even so, the relatively large es- group, if indeed it is. Dessauer and Cole timated sequence divergences between C. (1989) provided preliminary evidence sug- lemniscatus and the remaining Cnemidopho- gesting Cnemidophorus paraphyly.However, russpeciesaresuggestiveofarelativelybas- their taxon sampling was limited and/or in- al position for the lemniscatus group. How- complete (e.g., absence of the deppii group ever, this study cannot be viewed as a rig- and other critical cnemidophorine lineages). orous test of the basal relationships within Thus,itistimelytomorerigorouslyexamine Cnemidophorus (e.g., hypotheses of Burt, the phylogenetic relationships between 1931 vs. Lowe et al., 1970). Such a test Cnemidophorus and other teiine taxa (Amei- would require the inclusion of other closely va, Dicrodon, Kentropyx, and Teius), partic- related teiine taxa (e.g., Ameiva, Kentropyx) ularly now that the necessary samples are as outgroups. available.Theinferredphylogeneticrelation- While there have been previous attempts ships presented below are based on diverse to organize Cnemidophorus into species types of data. The bulk of these data are de- groups and hypothesize on the interrelation- rived from mitochondrial ribosomal RNA ships of these groups, there has never been a (rRNA) genes, but these data are augmented rigorous attempt to demonstrate the mono- with previously published allozyme data phyly of this group of lizards. All previous (Dessauer and Cole, 1989; Cole and Des- studies generally assumed that Cnemidopho- sauer, 1993; Cole et al., 1995; Markezich et rus was monophyletic, based on the phenetic al., 1997) and morphological characters tra- similarity between Cnemidophorusandother ditionally used in Cnemidophorus systemat- South American teiid lizards (i.e., Ameiva, ics. Dicrodon, Kentropyx, and Teius). Historical- The following questions are addressed in ly, Cnemidophorus has been defined by the this paper: (1) Is Cnemidophorus a mono- absence of presumably derived character phyletic group? (2) If not, what nomencla- states exhibited by these other South Amer- tural changes are needed and appropriate at ican genera (i.e., laterally compressed teeth this time? (3) What are the relationships be- in Dicrodon and Teius, keeled ventral scales tween Cnemidophorus and the other teiinine in Kentropyx, basal tongue sheath in Amei- genera? (4) Are the traditionally recognized va). The long recognition that Cnemidopho- bisexual species groups within Cnemidopho- rus lacked apomorphies, and earlier hypoth- rus monophyletic, and what is their relation- 2002 REEDER ET AL.: PHYLOGENETICRELATIONSHIPSOF CNEMIDOPHORUS 7 ship to each other? Finally, (5) Do thenewly from the 16S rRNA gene. The primers and inferred higher-level relationships requirere- PCR parameters used to amplify these frag- examination of past hypotheses of chromo- ments are described in Reeder (1995). Puri- somal evolution within Cnemidophorus? In fication of amplified DNA and automated additionwecommentbrieflyonthereticulate DNA sequencing were performed following phylogeny of unisexual clones of hybrid or- methods described in Wiens and Reeder igin and determination of the maternal an- (1997). The DNA sequences for Acantho- cestor of Kentropyx borckiana, a unisexual dactyluscantorisandLacertaagiliswereob- species of hybrid origin. tained from GenBank (accession numbers AF080298, AF080300, AF080344, and MATERIALS AND METHODS AF080346). The mitochondrial rDNA sequences (ap- CHOICE OF TAXA pendix 2) were aligned under varying gap costs(openinggapcostof6,9,and12)using Twenty-seven recognized Cnemidophorus the multiple sequence alignment program taxa were included in the present study, rep- ClustalW(Thompsonetal.,1994).Sequence resenting all currently recognized bisexual alignment procedures and parameters arede- species groups (deppii, lemniscatus, sexli- scribed in Wiens and Reeder (1997). It has neatus, and tigris species groups; Wright, been demonstrated that rRNA secondary 1993). This sample allows a preliminary test structure models can be useful in the align- of the monophylyofthesegroups.Also,sev- ment of these gene sequences (Kjer, 1995; eral additional non-Cnemidophorus teiine Titus and Frost, 1996). Following the pro- species were included in order to test Cnem- cedure outlined in Wiens and Reeder(1997), idophorusmonophyly.Inall,41ingrouptaxa ((cid:53) Ameiva, Cnemidophorus, Dicrodon,Ken- rRNA secondary structure information was used to assist in DNA sequence alignment. tropyx, and Teius) were included (appendix Regions of sequence were considered align- 1). ment-ambiguous if nucleotide positional ho- The following five outgroup taxa (succes- mologies differed among the different gap sivelymoredistant)wereincludedalso:Tup- cost alignments (Gatesy et al., 1993). Am- inambis (Teiidae), Pholidobolus (Gymnoph- biguously aligned regions were excluded thalmidae), Acanthodactylus and Lacerta from phylogenetic analysis. In all, 1072 nu- (Lacertidae), and Eumeces (Scincidae). The cleotide positions were aligned (49112Sand relationships of these outgroups to the in- 581 16S; appendix 2), with 61 positions (25 group are fairly wellunderstood(Estesetal., 12Sand3616S)excludedfromphylogenetic 1988;Lee,1998).However,tominimizeout- analysis. Gaps ((cid:53) insertion/deletion events) group assumptions, a global parsimony root- were coded as a fifth character state, as de- ing approach was taken (Maddison et al., scribed in Wiens and Reeder (1997). All 1984), with Eumeces (assumed to be the DNA sequences are deposited in GenBank most distantly related outgroup) being used (accession numbers AY046420–AY046503, to root the overall resulting tree(s). AF080344, AF080346, AF080298, and AF080300). Upon request, the PAUP* ma- MOLECULAR DATA trix is available from one of us (T.W.R.). DNA DATA: Total genomic DNA was iso- We followed Dessauer et al. (1996) in us- lated from small amounts of liver or eryth- ing allele-specific oligonucleotide probes to rocytes ((cid:59)100 mg) following the phenol- screen multiple individuals of Kentropyx chloroformextractionprotocolofHillisetal. borckianatodeterminethematernalancestor (1996). Two portions of the mitochondrial of this unisexual species. genomewereamplifiedusingthepolymerase ALLOZYME DATA: Data on 31 phylogenet- chain reaction (PCR) in Perkin-Elmer 2400 ically informative protein loci ((cid:53) characters) or Ericomp TwinBlock thermocyclers. One were scored for 19 taxa of teiid lizards. The PCR product was a (cid:59)380 bp fragment from entire allozyme database was produced in the 12S ribosomal RNA (rRNA) gene. The one laboratory (H.C.D.’s), so there is com- other PCR product was a (cid:59)500 bp fragment plete internal consistency across the data set. 8 AMERICAN MUSEUM NOVITATES NO. 3365 Data are the alleles detected at individual MORPHOLOGICAL DATA geneloci.Forphylogeneticanalysis,eachlo- Data on the 10 morphological characters cus was interpreted as the character and the were recorded for 42 taxa of teioid lizards alleles present in a taxon as character states (including Pholidobolus and twopopulations (Buth, 1984). All allozyme characters were of Kentropyx altamazonica). These taxa in- analyzed unordered. clude all of the teiids for which DNA se- The gene loci and codes for phylogenetic quencedatawereanalyzed.Becauseofprob- analysisoftheallozymesarepresentedinap- lemswithhomologyassessment,morpholog- pendix 3. The data were published previous- ical data were not coded for any of the non- ly in the following reports: Dessauer and teioid taxa. Data were recorded from Cole, 1989 (Ameiva, Cnemidophorus, Ken- museum specimens, which are specified in tropyx, and Tupinambis); Cole and Dessauer, appendix 1 (Specimens Examined). 1993 (South American Cnemidophorus); These characters have historically been Coleetal.,1995(Kentropyx);andMarkezich useful in recognizing generic and subgeneric et al., 1997 (South American Cnemidopho- species groups within the Teiidae, as sug- rus).However,thisisthefirstreportinwhich gested by previous authors (Burt, 1931; all of these data have been cross-correlated, Loweetal.,1970;PetersandDonoso-Barros, so the individual alleles as specified in this 1970; Hoogmoed, 1973). While not a large report (appendix 3) will not necessarily bear set of characters, we felt it was better to in- the same letter designation as in those orig- clude these traditional characters than to ex- inalpapers,someofwhichwerealphabetized clude them. It has been demonstrated that even a small number of morphological char- only on the basis of the alleles being com- acters (within the context of a large com- pared within the individual report. bineddatasetlargelyconsistingofmolecular Methodsof collecting,preparing,andstor- characters) can have an effect on a phylo- ing tissue samples, and methods of conduct- genetic analysis (e.g., Titus and Larson, ing protein electrophoresis, identifying loci, 1996). The character descriptions, coding, and determining allele products present in and matrix are presented in appendix 4. All the various species are detailed in the papers were discrete characters that could be scored cited above and relevant references therein unambiguouslyandforwhichtherewaslittle (also see Dessauer et al., 2000). The data are intraspecific variation. of discrete characters that could be scored unambiguously. Although most loci for each PHYLOGENETIC ANALYSIS taxonshownointraspecificvariationorpoly- morphism, some do. In cases where two or The mtDNA, allozymic, and morphologi- cal data were combined into a single data more alleles were recorded for a taxon, each matrix for phylogenetic analysis. Taxa miss- allele was recorded as present at that locus ing a particular subset of the total data (e.g., for that taxon. We did not attempt to usefre- allozymes) were coded as missing (?) those quency data (we used only presence or ab- data. Phylogenetic analyses were performed sence of allele character states) because de- with PAUP* 4.0b2 (Swofford, 1999). The gree of variability varies widely among loci, heuristic tree search routine was used (with it can vary geographically, and becausesam- TBRbranchswappingand100randomtaxon ple sizes vary widely among the taxa. For additions).Whenmultipleshortesttreeswere example, we examined only onespecimenof discovered, the trees were summarized with Tupinambis teguixin and more than 35 of a strict consensus tree (Sokal and Rohlf, Cnemidophorus inornatus. The problems as- 1981),thusdepictingonlythoserelationships sociated with geographic variation and sam- shared among all shortest trees. A character ple size are illustrated by Dessauer et al. state change was considered to unambigu- (2000), who examined more than 650 indi- ously support a clade if it was placed along viduals of Cnemidophorus tigris. We did not a branch by both ACCTRAN (Farris, 1970) trytointegratealloftheirdataonrarealleles and DELTRAN (Swofford and Maddison, into this report. 1987) optimizations. 2002 REEDER ET AL.: PHYLOGENETICRELATIONSHIPSOF CNEMIDOPHORUS 9 Initial phylogenetic analyses were per- garding the use of biological processes or formed with uniformly weighted characters models of evolution are debated (e.g., Swof- (i.e., all character state transformationshada ford et al., 1996), one should always be cau- weight of 1, irrespective of data type). How- tious of the assumptions that are being made ever, it is fairly well understood that verte- inanyphylogeneticanalysis.Inourstudywe brate mtDNA exhibits substitution biases use successive approximations to test the (e.g., transitions occurring more rapidly than sensitivity of the most parsimonious un- transversions), and different sites or regions weighted trees(s) to differential character (e.g., third codon positions, stem vs. loop re- weighting based on inferred levels of ho- gions) evolve at different rates. Thus, differ- moplasy (Farris, 1969; Kluge, 1997a). Clade ential weighting of nucleotide substitutions stability during successive approximations and/orsitesmaybewarranted.Seeminglyre- ((cid:53) clades congruent with tree(s) based on alistic and justifiable weighting schemes can uniformweighting)instillsuswithadditional be devised for the DNA data at hand (e.g., confidence for those relationships inferred in Arevalo et al., 1994; Cunningham, 1997; the uniformly weighted analysis (Carpenter Wiens et al., 1999). However, philosophical et al., 1993). and methodological difficulties arise within A common criticism or concernofsucces- the contextof acombinedphylogeneticanal- sive approximations is that the final inferred ysis (e.g., what weight is applied to morpho- tree may be largely dependent on the initial logical characters vs. the differentially starting tree from which weights were deter- weighted nucleotide substitutions?). Also, mined (Swofford et al., 1996). To test how different genes within a combined analysis robust inferred clades were to initial starting may have different substitution properties trees, we generated 20 random trees in (e.g., are the best character state transfor- MacClade v3.07 (Maddison and Maddison, mations of gene A equivalent to those of 1992) and performed successive approxima- gene B?). An objective way to differentially tions on each of the random trees. Congru- weight characters within the context of a ence among the final trees from the 20 com- combined analysis is to use the a posteriori pleted successive approximation analyses methodofsuccessiveapproximations(Farris, was summarized with a 50% majority-rule 1969; Carpenter, 1988). Such a weighting consensus tree. strategy differentially weights all the char- ThenumberoftaxascoredforthemtDNA acters based on their relative degrees of ho- (n (cid:53) 44) and morphological (n (cid:53) 43) data moplasy. Those characters most consistent far exceeded the number available for the al- with the initial starting tree are given the lozyme (n (cid:53) 19) data. Thus, some taxa are greatest weights, regardless of data partition incompleteforasubsetofthetotalcombined (i.e., DNA, allozymes, morphology). In our data. However, these incomplete taxa (miss- study, the initial tree(s) for successive ing(cid:59)8%ofinformativecharacters)werestill weightingwasthatinferredfromauniformly included in the phylogenetic analyses (see weighted combined data analysis. Reweight- Wiens and Reeder, 1995; Reeder and Wiens, ing characters was performed in PAUP*, us- 1996). Also, two taxa (Cnemidophorus mu- ing the maximum rescaled consistencyindex rinus and C. ocellifer) were coded for only (rci; Farris, 1989) (base weight (cid:53) 100; the 10 morphological characters (represent- weights truncated instead of being rounded ing (cid:59)3% of the total informativecharacters), [as in Hennig86]). since we lacked tissue samplesformolecular While originally envisioning means ofob- analysis. While these highly incomplete taxa jectively determining character weights, (missing (cid:59)97% of informative characters) Kluge (1997a, 1997b) has recently argued were included in certain phylogenetic anal- that all character weighting (a priori and a yses, their impact on tree stability was as- posteriori) should be rejected. Kluge states sessedbybootstrapping(seebelow)thecom- that all forms of differential character bined data with and without these two spe- weighting invoke additional background cies. The phylogenetic placement of C. mu- knowledge about biological processes that rinus is of special significance because it is are untestable. While such an affirmation re- the type species of Cnemidophorus. 10 AMERICAN MUSEUM NOVITATES NO. 3365 Support for individualcladeswasassessed analysis. However, teiid paraphyly is only by nonparametric bootstrapping(Felsenstein, weakly supported, with the gymnophthalmid 1985).Bootstrapanalyseswerebasedon500 ((cid:53) microteiid) Pholidobolus being placed heuristic tree searches (with TBR branch with Tupinambis (bootstrap (cid:53) 57%). Teiini swapping). Because of computational con- (Clade 1) monophyly is strongly supported straints, only three random taxon additions (80%)by11synapomorphies,withTeiusand per pseudoreplicate were performed in each Dicrodon representing the most basal line- of the heuristic tree searches. Bootstrapping ages. Within the teiine clade, 18 of the 34 wasperformedinboththeuniformlyweight- unambiguously resolved clades are strongly ed and successive approximation analyses. supported (bootstraps (cid:36)70%) by the com- Sullivanetal.(1997)havenotedthatweight- bined data. Within the Teiini, the cnemido- ed parsimony analyses often significantlyin- phorine taxa are also supported as a clade crease bootstrap values (relative to their val- (Clade 3). However, cnemidophorine mono- ues in uniformly weighted analyses of the phyly is only weakly supported ((cid:44)70%). samedata).However,becauseoftheinherent While cnemidophorine monophyly is sup- properties of parsimony, the elevated boot- ported, monophyly of Cnemidophorus is re- strap values in weighted parsimony analyses jected. All of the South American Cnemi- probably represent overestimates of the dophorusspecies(exceptC.longicaudus)are amount of support for the inferred clades more closely related to species of other gen- (Yang et al., 1995; Sullivan et al., 1997). era of Central and South American cnemi- Therefore, we cautiously interpret the boot- dophorines(i.e.,AmeivaandKentropyx)than strap results of the successively weighted to the North American species of Cnemido- data, and base most of our conclusions of phorus. However, this neotropical clade relative support from the unweighted boot- (Clade 4) is only weakly supported by these strap analysis. For the uniformly weighted data.WithinClade4C.lacertoidesisweakly data, clades with bootstrap values of (cid:36)70% placed as the sister species of the remaining were considered strongly supported (follow- taxa. Monophyly of the lemniscatuscomplex ing Hillis and Bull, 1993). (i.e., C. arenivagus, C. gramivagus, and C. lemniscatus; Clade 10) is strongly supported RESULTS (100%), with this clade being placed as the sister taxon of a strongly supported Kentro- UNIFORMLY WEIGHTED ANALYSIS pyx (100%; Clade 13). In addition to 12 Phylogenetic analysis of the 317 uniform- mtDNA synapomorphies, the lemniscatus ly weighted phylogenetically informative complex is also supported by one morpho- characters (235 informative characters logical synapomorphy (basal tongue sheath among teiine taxa) resulted in four shortest absent [character state 1.b]). Kentropyx trees (L (cid:53) 1539; CI (cid:53) 0.39; RI (cid:53) 0.61).The monophyly is supported by 19 synapomor- strict consensus of these four trees is shown phies: 12 mtDNA, four morphological in figure 2A. The numbers of unambiguous (keeled ventral scutes [3.b], 14 rows of ven- synapomorphies supporting the unambigu- tral scutes [4.c], two enlarged anal spurs per ously resolved branches of the strict consen- side in males[6.c],abruptlyenlargedmesop- sus tree are given in table 2. All inferred tei- tychial scales [8.c]), and three allozymes. ine clades were supported byunambiguously Within Kentropyx, it is equally parsimonious placed synapomorphies. However, the vast to place K. calcarata as the sister taxon of majority of the clades were supported only all remaining Kentropyx, or as the sisterspe- by mtDNA character state transformations. cies to the K. altamazonica (cid:49) K. pelviceps In all, only four of the 36 teiine clades were clade. unambiguously supported by mtDNA, mor- Analysis of these data also rejects the phological, and allozymic synapomorphies monophyly of Ameiva. Within Clade 4, A. (table 2), possibly because allozyme data undulata is more closely related to the lem- were coded for only 19 taxa. niscatusgroup(cid:49)Kentropyxcladethantothe Monophyly of the Teiidae (excluding small clade containing A. ameiva, A. bifron- Gymnophthalmidae) is not supported by this tata,andA.quadrilineata.Also,theWestIn-

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.