Family ties: molecular phylogeny of crab spiders (Araneae

Family ties: molecular phylogeny of crab spiders(Araneae: Thomisidae)

Suresh P. Benjamina,b*�, Dimitar Dimitrova, Rosemary G. Gillespieb and GustavoHormigaa

aThe George Washington University, Department of Biological Sciences, 2023 G Street NW, Washington DC, 20052, USA; bUniversity of California,

Berkeley, Insect Biology Division—ESPM, 201 Wellman Hall 3112, Berkeley, CA 94720-3112, USA

Accepted 2 November 2007

Abstract

The first quantitative phylogenetic analysis of three sequenced genes (16S rRNA, cytochrome c oxidase subunit I, histone 3) of 25genera of crab spiders and 11 outgroups supports the monophyly of Thomisidae. Four lineages within Thomisidae are recovered.They are informally named here as the Borboropactus clade, Epidius clade, Stephanopis clade and the Thomisus clade, pendingdetailed morphology based cladistic work. The Thomisus clade is recovered as a strongly supported monophyletic group with aminimal genetic divergence. Philodromidae previously widely considered a subfamily of Thomisidae do not group within thomisidsand is excluded from Thomisidae. However, Aphantochilinae previously generally considered as a separate family falls within theThomisus clade and is included in Thomisidae. The recently proposed new family Borboropactidae is rejected, as it is paraphyletic.

� The Willi Hennig Society 2008.

Thomisidae Sundevall, 1833, commonly called crabspiders are often cryptically colored sit-and-wait preda-tors that do not build capture webs (Fig. 1). Thomisidaeis the sixth largest spider family. It includes 2062described species in 171 genera (Platnick, 2007) withmany more species remaining to be described. They aremainly active during the day and ambush insects withtheir well-adapted first and second legs (Homann, 1934;Comstock, 1948). Not surprisingly, they are an impor-tant component of terrestrial ecosystems (Riechert,1974). As predators of agricultural pests, thomisids playan important part in natural pest control (Riechert andLockley, 1984; Nyffeler and Benz, 1987; Uetz et al.,1999).

Some thomisids (e.g., Misumena, Diaea, Runcinia andThomisus) possess the ability to change color and blendinto their habitat, in most cases flowers (Packard, 1905;

Gabritschevsky, 1927; Comstock, 1948).Misumena vatia(Clerck, 1757) has a remarkable ability to change color,which takes place during migration to flowers ofdifferent color from spring to the early part of summer(Comstock, 1948). Crab spiders are attracted by fra-grance components of flowers (Aldrich and Barros,1995; Krell and Kraemer, 1998) and use visual andtactile cues for selecting flowers (Morse, 1988; Grecoand Kevan, 1994). They reach their ambush sites in astep-by-step process using several draglines and bal-looning events (Homann, 1934).

There are social crab spiders with maternal care in theEucalyptus forest of Australia (Main, 1988; Evans,1995). Mother Diaea ergandros Evans (1995), catchlarger prey for their own offspring, but not for theadopted offspring leading to large size and possiblybetter survival rates for the natural offspring (Evans,1998). Mothers further increase survival of naturaloffspring by producing trophic oocytes used in a systemof sacrificial care (Evans, 1998).

Myrmecomorphism is known in a number of thomis-ids: Strigoplus albostriatus Simon, 1885, Amyciaeaforticeps (O.P.-Cambridge, 1873), A. lineatipes

*Corresponding author:E-mail address: [email protected]�Present address: Department of Entomology, National Museum ofNatural History NHB 105, PO Box 37012, Smithsonian Institution,Washington, DC 20013-7012, USA.

� The Willi Hennig Society 2008

Cladistics 24 (2008) 708–722

Cladistics

10.1111/j.1096-0031.2008.00202.x

O.P.-Cambridge, 1901 and Aphantochilus rogersiO.P.-Cambridge, 1870, are known to be ant mimics.A. forticeps is of the same color as the ant Oecophyllasmaragdina and bears on the posterior part of theabdomen a pair of eye-like spots that correspond to eyesof the ant (Shelford, 1902). The similarity of A. rogersiito some spiny South American ants is striking, andforms a further instance of myrmecomorphism (Oliveiraand Sazima, 1984).

Given their ecological significance and appealingadaptations one would expect to see a plethora ofphylogenetic studies. However, no such studies ofthomisids exist. Moreover, understanding the exacttaxonomic limits of this large family has always beenproblematic. Thomisidae was proposed to accommo-date spiders with legs generally extended sideways(laterigrade), instead of being oriented towards the frontor back as in most other spiders. Originally all spiderswith laterigrade legs such as Sparassidae and Philodr-omidae were included. Simon (1892) was the first topropose a hypothesis of generic groups for all thomisidgenera recognized during his time. Although his‘‘groups’’ were neither evolutionary nor phylogeneticin the modern sense, he provided some information onThomisidae morphology and arguments supporting hisideas. His Stephanopsinae contained spiders with che-liceral teeth; Aphantochilinae and Strophiinae con-tained species with modifications such as elongatedmaxillae related to their ant mimicking habits; Stiphro-podinae included species with an enlarged tarsus; spiders

that did not fit into the above categories were includedin Misumeninae and Philodrominae. Species groupswere then formed within these subfamilies based on eyepattern and shape of prosoma.

Since Simon (1892) the understanding of genericrelationships has not changed greatly. Holm (1940), onthe grounds of embryological studies, and Homann(1975), on the grounds of eye morphology, excludedPhilodromids from Thomisidae. Philodromids werelater given family status (Ono, 1988), which has beenaccepted since (but see Roberts, 1995). Separately,family status was proposed for some thomisids, Steph-anopidae (Pickard-Cambridge, 1871) and Aphantochi-lidae (Thorell, 1873). Levi (1982) placed the Thomisidaein a superfamily along with Aphantochilidae, which wasnot accepted by Ono (1988). In the most current studyThomisidae was separated into seven subfamilies (Ono,1988): Stephanopinae, Thomisinae, Bominae, Stiphro-podinae, Dietinae, Strophiinae and Aphantochilinaeusing characters proposed by Simon (1892). Recently(Wunderlich, 2004b) proposed a new family ‘‘Borborop-actidae’’ to include parts of Thomisidae. Thus, to date,the monophyly of Thomisidae remains untested.

A great part of our knowledge of evolutionary historyis derived from phylogenies, reconstructed by samplingand grouping characters. Harvey and Pagel (1991)illustrated the richness of evolutionary questions thatcan be approached with phylogenies. Our currentunderstanding of relationships within Thomisidae islargely based on Simon (1892) and modifications to his

A

C D

B

Fig. 1. Living thomisids sampled in this study. (A) ‘‘Monaeses’’ sp. A, Sri Lanka, Central Province; (B) Diaea placata, Sri Lanka, Western Province(not sampled); (C) Oxytate taprobane, Sri Lanka, Central Province; (D) ‘‘Lysiteles’’ sp. B, Sri Lanka, Central Province. All photos by SPB.

709S. P. Benjamin et al. / Cladistics 24 (2008) 708–722

ideas by different workers (Schick, 1965; Homann, 1975;Ono, 1988). However, they do not provide muchinformation, partly because the family has never beensubject to quantitative phylogenetic analysis. Recenttaxonomic work on tropical Asian crab spiders (Tikader,1980; Barrion and Litsinger, 1995) added more confu-sion, illustrating the fragile systematic state of thefamily.

Here we present the first cladistic analysis of Thom-isidae. This analysis of molecular data aims to test themonophyly of Thomisidae, including the validity of‘‘Borboropactidae’’ and the placement of some enig-matic taxa such as Epidius and Cebrenninus, which sharecharacters such as the presence of a conductor andmedian apophysis with ‘‘Borboropactidae’’. Epidius wasprovisionally placed in Thomisidae owing to its unusualmale palp (Benjamin, 2000). Owing to the considerablevolatility in thomisid systematics and paucity of mor-phological information for a large number of taxa, wedo not formally name any higher-level taxa here.However, we provide putative morphological synapo-morphies for most of the largest clades. The first authoris presently undertaking a detailed morphological revi-sion of Thomisidae.

Materials and methods

Ingroup taxa

We sampled 41 ingroup and 13 outgroup taxa forDNA sequencing (Table 1). Representatives of all majorthomisid subfamilies except for Stiphropodinae andStrophiinae are included. They are: Stephanopinae,Thomisinae, Bominae, Dietinae and Aphantochilinae.Additionally, several sequences of thomisids and out-groups available in GenBank were added to the anal-yses. Accession codes for all sequences are given inTable 1.

Outgroup taxa

Thomisids fall within the large clade Dionycha(Coddington and Levi, 1991), which are characterizedby the loss of the unpaired tarsal claw. The monophylyof Dionycha has, however, been elusive. Furthermore,the phylogenetic structure within Dionycha has not beenfully explored [however, see Davila (2003) who includeda small sample of Dionycha in her study of Ctenidaeinterrelationships]. Thus, the only character-based out-group hypothesis for thomisids is by Loerbroks (1984);in a functional study of the male palp of M. vatia, hefound characters to relate Salticidae to Thomisidae.Morphological comparisons of representatives of fam-ilies within Dionycha have suggested that Creugas(Corinnidae) might be the sister to thomisids

(P. Lehtinen, pers. comm.). We have included six speciesof Salticidae, one species each of Miturgidae andCorinnidae in our study. As Philodromidae have beenin the past included as a subfamily of Thomisidae, weinclude three philodromid taxa in our study.

DNA sequencing and alignment

Genomic DNA was extracted from either fresh orETOH-preserved leg tissue using the Qiagen DNeasyTissue Kits (Qiagen, Valencia, CA, USA). Otherwiseintact spiders, preserved in alcohol have been depositedas voucher specimens (Table 1). Partial fragments of themitochondrial genes cytochrome c oxidase subunit I(COI) and 16S rRNA (16S) and the nuclear gene histoneH3 (H3) were amplified using the following primer pairs:(COI) C1-J-1751 and C1-N-2191 (Simon et al., 1994)(16S) LR-N-13398 (Simon et al., 1994) and LR-J-12864(Arnedo et al., 2004) and (H3) H3aF and H3aR (Colganet al., 1998). PCR products were purified using theQIAquick PCR Purification Kit (Qiagen) and sequenceddirectly in both directions using an ABI 3730 automatedsequencer (Applied Biosystems, Foster City, CA, USA),in combination with the ABI PRISM Big Dye Termi-nator Cycle Sequencing Ready Reaction Kit. Thechromatograms produced for each DNA sequence wereedited in Sequencher 3.1 (Gene Codes, Ann Arbor, MI,USA).

The protein-coding COI and H3 sequences were easilyaligned unambiguously. For non-protein-coding se-quences there are two basic approaches to multiplesequences alignment. The first are the static alignmentmethods with fixed homology statements, in which thealignment is independent of the phylogenetic analysis.The second is the optimization alignment method wherealignment is an integral step of phylogenetic analysis(Wheeler, 1996). We explored our data using bothanalytical philosophies. In all analyses gaps were treatedeither as missing data (parsimony and Bayesian infer-ence) or as a fifth state (POY).

Direct optimization

Direct optimization (Wheeler, 1996; Wheeler et al.,2006) was implemented with the software package POYVersion 3.0.11. Unlike analysis of statically aligned data,in POY alignments are tree dependent and as a result thehomology statements are dynamic. In all the analysesthe protein-coding genes H3 and COI were treated asprealigned. This avoids the use of in ⁄del transformationsfor these fragments and saves computing time withoutalteration of results (Giribet, 2001). To further reducecomputational time and to minimize the probability ofunwanted artifacts in the alignment the 16S wasdivided into two homologous fragments (Giribet,2001). Homolog partitions in 16S were defined based

710 S. P. Benjamin et al. / Cladistics 24 (2008) 708–722

on a preliminary static multiple alignment withClustalW (Thompson et al., 1994) using the defaultparameters.

Sensitivity of the results to various parameters wasexplored using five different values of gap openingcosts. As an objective criterion to select amongdifferent parameter combinations, maximum congru-

ence between the 16S and the COI + H3 partitionswas used (Wheeler and Hayashi, 1998). Congruencewas measured using the incongruence length difference(ILD) test (Mickevich and Farris, 1981). Tables 2 and3 contain the information about the parameter com-binations investigated together with statistics from theanalyses. All POY searches were run on a

Table 1Specimens sequenced for molecular analyses. Fragments successfully sequenced and GenBank accession number are given

Family Genus Species Country Locality COI 16S H3 Depository

Philodromidea Pagiopalus nigriventris USA Maui, HI EU168155 EU168142 ⁄EU168143

EU157106 USNM

Proernus stigmaticus USA Hawaii, HI EU168156 EU168144 EU157107 USNMPhilodromus sp. USA Berkeley, CA EU168157 EU168145 EU157108 USNM

Salticidae Onomastus sp. A Sri Lanka Central Province EU168158 EU168146 NA USNMOnomastus sp. B Sri Lanka Central Province EU168159 EU168148 EU157109 USNMPortia labiata NA NA AY297361 AY296653 NA NALyssomanes virides NA NA AY297360 AY296652 NA NANeon nelli NA NA AF327988 AF327959 NA NAHelvetia cf. zonata NA NA AY297394 AY296685 NA NA

Corinnidae Castianeira sp. NA NA AY297419 AY296710 NA NAMiturgidae Cheiracanthium sp. NA NA AY297421 NA NA NAThomisidae Borboropactus cinerascens Singapore Bukit Timah NR NA EU168138 EU157126 MHNG

Borboropactus sp. South Africa Limpopo EU168187 EU168151 NA MHNGEpidius parvati Sri Lanka Western Province EU168163 EU168141 EU157114 MHNGCebrenninus rugosus Thailand Chumphon Province EU168175 EU168139 EU157134 MHNGCebrenninus rugosus Malaysia Kelantan EU168177 EU168140 NA MHNG‘‘Epicadinus’’ sp. Argentina Misiones NA EU168153 NA MACN-Ar‘‘Stephanopis’’ sp. A Chile Region IX EU168167 EU168137 EU157117 CAS‘‘Stephanopis’’ sp. B Australia Tasmania EU168185 NA EU157138 USNM‘‘Stephanopis’’ sp. C Australia Tasmania NA NA EU157137 USNM‘‘Stephanopis’’ sp. D Australia Western Australia NA NA EU157139 USNMPseudoporrhopis granum Madagascar Fianarantsoa EU168170 EU168131 EU157120 CASOxytate taprobane Sri Lanka Central Province EU168161 EU168133 EU157112 MHNGAmyciaea forticeps Thailand Montha Tarn NA NA EU157135 MHNG‘‘Lysiteles’’ sp. A Sri Lanka Central Province EU168184 EU168128 EU157133 MHNG‘‘Lysiteles’’ sp. B Sri Lanka Central Province EU168183 EU168129 NA MHNGCoriarachne versicolor USA VA NA NA EU157136 USNMXysticus californicus USA CA EU168181 EU168115 EU157131 USNMXysticus sp. USA CA EU168182 EU168116 EU157132 USNM‘‘Monaeses’’ sp. A Sri Lanka Central Province EU168172 EU168117 EU157122 MHNG‘‘Monaeses’’ sp. B Australia Western Australia EU168186 EU168150 NA CASTmarus angulatus USA CA EU168179 EU168120 EU157110 USNMTmarus angulatus USA CA EU168180 EU168121 EU157111 USNMTalaus sp. Vietnam Kien Gian Province NA EU168135 EU157127 MHNGRuncinia albostriata Sri Lanka Central Province EU168178 EU168125 EU157130 MHNGRuncinia albostriata Myanmar Bago Division EU168165 EU168124 EU157116 CASRuncinia acuminata Myanmar Bago Division EU168166 EU168126 NA CASDiaea subdola Sri Lanka Central Province EU168174 EU168118 EU157124 MHNGAphantochilus sp. Argentina Misiones NA EU168152 EU157140 MACN-ArDiaea nr subdola Myanmar Yangon Division EU168169 EU168127 EU157119 CASCyriogonus sp. Madagascar Antananarivo EU168168 EU168130 EU157118 CASThomisus granulifrons Sri Lanka Western Province EU168162 EU168122 EU157113 MHNGThomisus sp. A Sri Lanka Sabaragamuwa Province EU168164 NA EU157115 MHNGThomisus sp. B Thailand Phuket Province EU168176 EU168123 EU157129 MHNGThomisops piger Sri Lanka Western Province EU168171 EU168134 EU157121 MHNGHaplotamarus sp. Sri Lanka Western Province EU168173 EU168132 EU157123 MHNGCamaricus sp. Sri Lanka Central Province NA EU168119 EU157125 MHNGBoliscus sp. Malaysia Kelantan NA EU168136 EU157128 MHNGMecaphesa a semispinos NA NA DQ174382 DQ174339 NA NAMecaphesa naevigera NA NA DQ174344 DQ174387 NA NAXysticus sp. B NA NA AY297423 AY296714 NA NA


computer cluster at The George WashingtonUniversity.

Initial trees were built using an approximation short-cut (command –approxbuild). Thirty replicates ofrandom addition of sequences were performed (com-mand –replicates 30) holding up to 300 trees(command –holdmaxtrees 300). Tree fusing andtree drifting (TBR) strategies were used for tree search-ing (commands –tbr –treefuse –drifttbr). Inorder to increase the search efficiency and guarantee thatPOY will save only disparate trees the command-fitchtrees was used. To reduce error from heuristicoperations suboptimal trees with 1% length distancefrom the shortest trees were submitted to a furtherround of TBR (command –checkslop 10). Indicesand topology specific implied alignments were savedusing the commands –indices and –impliedalignment.

Static alignments

Static alignments were built using ClustalW (Thomp-son et al., 1994) with the following gap opening ⁄gapextension costs: 8 ⁄2, 8 ⁄4, 15 ⁄6.66 (default), 20 ⁄2, 24 ⁄4and 24 ⁄6 and transition weight fixed to 0.5. Owing to thelack of an a priori basis for deciding on one alignmentover the other and to avoid having to use all sixalignments, we adopted phylogenetic congruence toselect an ‘‘optimal’’ alignment. Phylogenetic congruenceworks by comparing phylogenetic results from eachalignment matrix with results of a concatenation of all

matrices (elision matrix; Wheeler et al., 1995; Hedin andMaddison, 2001). In principle, an elision matrix shouldgenerally recover relationships that are robust to align-ment differences. This was done as suggested by Hedinand Maddison (2001).

Topological congruence between the results from theanalysis of each particular cost scheme and the elisionmatrix was measured by the number of nodes incommon between the consensus tree, which they pro-duced and the average symmetric-difference distancesbetween all trees from the individual and the elisionmatrices (Hedin and Maddison, 2001; Swofford, 2002).Based on these criteria the alignment built with gapopening ⁄gap extension cost of 20 ⁄2 was selected andused in all further analyses.

Phylogenetic analysis of static alignments

Both parsimony and likelihood methods were used asoptimality criteria for the phylogenetic analyses. Bayes-ian inference was preferred over maximum likelihoodbecause performing Bayesian analysis is computationallyless demanding than traditional likelihood methods.Manipulations of data matrices and trees were performedwith MacClade (Maddison and Maddison, 2001),PAUP* (Swofford, 2002) and WinClada (Nixon, 2002).

Parsimony

Parsimony searches were conducted using TNT versus1.0 (Goloboff et al, 2003a). All multistate characterswere treated as non-additive (Fitch, 1971). In allparsimony analyses, heuristic searches under the ‘‘tra-ditional search’’ option were performed using 1000replicates, holding 10 trees per replicate to a maximumof 10 000 trees. Clade support was assessed by means ofnon-parametric bootstrap analysis (Felsenstein, 1985) asimplemented in TNT with 1000 pseudoreplicates ofheuristic searches with 100 interactions of randomaddition of taxa and holding 10 trees per interaction.The same parameters were used to perform a jackknifeanalysis (Farris et al., 1996). Additionally, Poissonweighting bootstrap and symmetric re-sampling supportvalues (Goloboff et al., 2003b) were calculated using thesame search parameters. These methods try to avoiderrors where high support values are found for groupsnot supported by the original data (Goloboff et al.,2003b). While bootstrap and Poisson weighting boot-strap re-sample characters by building a matrix of thesame size as the original, jackknife and symmetric re-sampling removes part of the characters, constructing asmaller matrix where character replacement does notoccur (Goloboff et al., 2003b). In both cases theanalyses were performed using the default setting inTNT (probability of character removal in the pseudore-plicates was set to 0.36).

Table 2Statistics resulting from the exploration of the effect of a range of gapopening ⁄gap extension cost on the data set used to generate the treeshown in Fig. 1

Gapcost

16Slength

COI + H3length

Combinedlength ILD

1 1362 1834 3279 0.0253125952 1489 1834 3414 0.026654953 1569 1834 3501 0.0279920024 1642 1834 3578 0.0285075465 1709 1834 3649 0.029049055

Table 3Statistics resulting from the exploration of the effect of a range of gapopening ⁄gap extension cost on the data set used to generate the treeshown in Fig. 2

Gapcost

16Slength

COI + H3length

Combinedlength ILD

1 1423 1772 1774 0.0538939892 1539 1772 1798 0.0195439743 1591 1772 1807 0.0271912064 1631 1772 1809 0.0354308395 1671 1772 1828 0.041481069


In addition to the parsimony analysis with equalcharacter weights, analyses using implied weighting wereconducted in TNT. Implied weighting was used to buildtrees using differential character weighting (Goloboff,1993a). It is superior to the successive weighting as itprovides an optimality criterion (maximum fit) to buildtrees and weight characters (Goloboff, 1993b). Thishelps to avoid iterative searches from preliminary trees,which were one of the most criticized aspects of thesuccessive weighting method [Swofford and Olsen(1991), but see Steel (1994) and Farris (2001)]. Fit iscalculated as a function in which concavity depends onthe constant K. With the increase of K the penalizationagainst homoplasy decreases (Goloboff, 1993b). Impliedweighting analyses, for K-values from 1 to 100 weredone in TNT for 500 replicates.

Bayesian inference (likelihood approach)

MrBayes 3.4 (Ronquist and Huelsenbeck, 2003) wasused to estimate topologies and to calculate posteriorprobabilities of inferred clades. The best-fit model ofsequence evolution was selected using Modeltest 3.6using the Akaike information criterion (Posada andCrandall, 1998). Four replicate analyses were runsimultaneously from random starting trees using defaultpriors. Metropolis-coupled MCMC (one cold and threeheated) were run for 10 million generations. Topologieswere sampled at intervals of 1000 generations withineach chain. Average likelihood scores (–ln) were exam-ined to assess convergence. Sampled trees were plottedagainst generations in order to determine point ofstability. Trees under the stability value were discardedas ‘‘burn-in’’. There is some disagreement on thereliability of the posterior probability, which has beencriticized by Suzuki et al. (2002) for overestimatingstatistical confidence when concatenated gene sequencesare used. However, others have countered: Wilcox et al.(2002) based on simulations have shown that Bayesiansupport values represent much better estimates ofphylogenetic accuracy than do non-parametric boot-strap support values. However, when model misspecifi-cations are present (which is absent in simulation-basedstudies) inflation of posterior probabilities may occur.

Results

The lengths of the sequenced fragments excluding theprimers were as follows: 16S, 430 bp; H3, 328 bp; andCOI, 557 bp. All sequences have been deposited inGenBank and their acquisition numbers are listedin Table 1. The ILD test (Farris et al., 1994), done inNONA, did not find significant incongruence betweenthe three gene fragments. Thus, all three gene fragmentswere combined into a single matrix.

Direct optimization

The results of the analysis with the combination ofgap opening ⁄gap extension cost of 2 ⁄1 produced impliedalignments with maximum congruence among partitions(Tables 2 and 3). One of the two most parsimonioustrees (L ¼ 3377) shown in Fig. 2, recovers a monophy-letic Thomisidae. The strict consensus of the two mostparsimonious trees is given in Fig. 3. Thomisids arerendered possibly paraphyletic in the strict consensus, asit unites Cesonia sp. (Gnaphosidae) and Hibana sp.(Anyphaenidae) in a basal tricotomy with ingroup taxa(Fig. 3). However, this arrangement of Cesonia sp. andHibana sp. within thomisids is not supported and is verysensitive to taxon sampling. Adding sequence fragmentsof four taxa (oneMisumenoides, twoMecaphesa and oneXysticus species; see Table 1) and ⁄or excluding theoutgroup taxon Portia labiata (Thorell, 1887) recoversa monophyletic Thomisidae, which excludes Cesoniaand Hibana (Fig. 4). The analysis of this modifiedmatrix found six most parsimonious trees (L ¼ 3279).The combination of gap opening ⁄gap extension cost of1 ⁄1 resulted in higher overall congruence for this matrix.The strict consensus of this analysis is shown in Fig. 4.In all analyses under dynamic optimization Anyphaeni-dae (represented by Hibana) plus Gnaphosidae (repre-sented by Cesonia) is recovered as the closest relative ofThomisidae.

Parsimony analyses with static homology

Analyses of the static alignment under equal weightsfound a single most parsimonious tree (L ¼ 3234, CI ¼0.32, RI ¼ 0.477) shown in Fig. 5. Thomisids form amonophyletic family. The main lineages within Thom-isidae and their relationships are the same as in the treesresulting from dynamic optimization. However, only theEpidius clade and to a lesser extent the Thomisus cladeare well supported. Hibana sp., but not Cesonia sp.appears as closest relative to Thomisidae.

Parsimony analyses under the criterion of impliedweights constantly recover the monophyly of Thomis-idae. Relationships within Thomisidae are as in uweight-ed parsimony and dynamic optimization analysessupporting the existence of the same four main lineages.The only difference refers to the composition of theStephanopis clade, which was always recovered asmonophyletic, but for the majority of the concavities italso included Epicadinus sp. (Fig. 6).

Bayesian inference

The best fitting model for the fragment of 16S wasGTR + I + C; for H3 it was HKY + I + C and forCOI it was GTR + I + C. The results of the first150 000 generations were discarded as burn-in. The


resulting tree is shown in Fig. 7. Results practicallymirror those from the implied weight parsimony anal-yses and are highly congruent with the tree topologiesfound by the unweighted parsimony and direct optimi-zation analyses. Thomisidae is again monophyletic andincludes the same four main lineages as in all previousanalyses. The Borboropactus clade is sister to all otherthomisids. Support for thomisid monophyly is higherthan in other analyses. Most of the main groups withinThomisidae and their relationships are also well sup-ported. Further, the Bayesian inference results suggestthat Anyphaenidae (represented by Hibana sp.) is thesister lineage of the family Thomisidae.

Discussion

Monophyly of Thomisidae

The taxonomy of Thomisidae is challenging. Almostall genera await revision. Despite past attempts (Simon,1892; Petrunkevitch, 1928; Roewer, 1954; Ono, 1988;Lehtinen, 2005), the monophyly of Thomisidae wasnever established. Our study, based on molecular dataprovides strong support for the monophyly of Thomis-idae for the first time. Monophyly of Thomisidae issupported by three molecular synapomorphies in thisstudy, all from the 16S gene fragment. They are three

Fig. 2. One of the two most parsimonious trees (L ¼ 3377) found under direct optimization. recovers a monophyletic Thomisidae. Gap opening ⁄gapextension cost of 2 ⁄1. Jackknife values greater than 60 are shown above the branches.


changes to T optimized as synapomorphies at thealigned positions 181, 183, 230, based on the preferredtopology from POY (Fig. 2).

Of the 11 outgroup taxa of the families Philodromi-dae, Salticidae, Miturgidae, Corinnidae, Gnaphosidaeand Anyphaenidae included in our study, Hibana sp.(Anyphaenidae) or Hibana sp. plus Cesonia sp. (Gnaph-osidae) were placed as sister to Thomisidae. This wasunexpected for two reasons: first, Salticidae was previ-ously considered sister to Thomisidae (Loerbroks, 1984)

and second, Philodromidae was considered a subfamilyof Thomisidae (Simon, 1892). Philodromids are stillconsidered by some contemporary arachnologist asderived thomisids (Tikader, 1980; Roberts, 1995). Inthis study they do not group within Thomisidae, andmay fall at the root of Dionycha. In contrast, Cheir-acanthium sp. is placed closer to the root of Thomisidae.Cheiracanthium was previously a member of Clubioni-dae, a family also proposed as sister to Thomisidae(Ono, 1988).

Fig. 3. The strict consensus of the two most parsimonious trees found under direct optimization. Gap opening ⁄gap extension cost of 2 ⁄1. Jackknifevalues greater than 60 are shown above the branches.


All phylogenetic analyses independent of gap open-ing ⁄gap extension cost or taxon sampling, recover fourwell-supported lineages within Thomisidae. They areinformally named here as the Borboropactus clade,Epidius clade, Stephanopis clade and the Thomisusclade, pending detailed morphology-based cladisticwork. The Borboropactus clade is sister to all remainingthomisids. All ‘‘derived’’ thomisids are grouped inthe Thomisus clade. Epidius and Stephanopis cladesare more closely related to each other than to theprevious two groups. None of these clades recoveredin our study correspond strictly to any currentlyaccepted subfamily groupings. In the most currentstudy Thomisidae was separated into seven sub-

families: Stephanopinae, Thomisinae, Bominae,Stiphropodinae, Dietinae, Strophiinae and Aphantoc-hilinae (Ono, 1988). The current study shows thesegroupings to be paraphyletic and should thus betreated with caution.

The following morphological synapomorphies areproposed to define Thomisidae:

1 Legs 1 and 2 longer and stronger then legs 3 and 4(Ono, 1988; Wunderlich, 2004b; Jocque and Dippenaar-Schoeman, 2006). However, this might not be obviousfor some African genera such as Thomisops Karsch,1879 (Dippenaar-Schoeman, 1989).

2 Lateral eyes on tubercles, larger and much moredeveloped than the median eyes (Ono, 1988).

Fig. 4. The strict consensus with gap opening ⁄gap extension cost of 1 ⁄1 under direct optimization: with the addition of the ingroups Misumenoidessp., two Mecaphsesa and one Xysticus species and excluding the outgroup Portia labiata. See text for details. Jackknife values greater than 60 areshown above the branches.


3 Presence of a group of setae instead of a colulus(Homann, 1975).

Borboropactus clade

Borboropactus Simon, 1884 is one of the few thomisidgenera represented in amber. It was dubbed a ‘‘relict’’and elevated to family rank (Wunderlich, 2004a,b). Thisstudy shows that there is no phylogenetic justificationfor such a family (unless of course Thomisidae is split tofour families). Borboropactidae would be paraphyleticas it includes the Epidius clade. As Wunderlich (2004b)himself mentions Borboropactus shares characters suchas the presence of a conductor and median apophysiswith other thomisids such as Epidius and Cebrenninus(Benjamin, 2000; Benjamin unpublished data). Further,

the Malagasy endemic Geraesta Simon, 1889 should beincluded in this clade based on the three characters givenbelow (Benjamin unpublished data).

The following morphological synapomorphies mightdefine the Borboropactus clade: (1) flexibly attached cup-shaped median apophysis (Fig. 8A,B); (2) hyaline con-ductor (Fig. 8A–C); and (3) epigynal teeth (Fig. 8D–E).Wunderlich (2004b, figs 4–8) mentions a specializedgland or ‘‘tarsal pit organ’’ of tarsi of legs 1–4 assynapmorphic for ‘‘Borboropactidae’’. However, thissensory organ appears to be autapomorphic forBorboropactus, as it is absent in Geraesta (Benjaminunpublished data). Furthermore, his ‘‘Borboropactidae’’should include the genera Cebreninnus, Cupa, Epidiusand Stephanopis, which then is essentially Simon’s‘‘Stephanopinae’’. Homann (1934) suggested that

Fig. 5. The single most parsimonious tree (L ¼ 3234, CI ¼ 0.32, RI ¼ 0.477): static alignment, analyzed under unweighted parsimony. Supportvalues are show as follows: above branches Bootstrap ⁄Poisson Bootstrap; below branches Jackknife ⁄Symmetric resampling. Support values bellow50 are omitted.


thomisids should be placed in Lycosoidea as they have agrate-shaped tapetum. Interestingly, Borboropactus hasepigynal teeth as do some Lycosoidea (Fig. 7D,E;Griswold, 1993). However, Borboropactus has a canoe-shaped tapetum and not a grate-shaped tapetum(Benjamin unpublished data). This apparent characterconflict needs to be addressed with a broader taxonsample in future studies.

Monophyly of the Epidius and Stephanopis clades

No morphological characters supporting these twoclades are known. This is because Stephanopis andrelated genera are poorly studied. It is clear thatStephanopis is paraphyletic and badly in need ofrevision. Preliminary data indicate that Cebrenninusmight be monophyletic. Epidius is the only genus that iswell defined, due to its elongated tibia (Benjamin, 2000).

Monophyly of the Thomisus clade

Our study revealed the Thomisus clade, morpholog-ically the most homogeneous, to be monophyletic(Figs 1–5). It is defined by several morphologicalsynapomorphies: (1) scopula hairs circular in cross-section (Homann, 1975); (2) bulbus subequal in lengthand width (Benjamin, 2000); (3) tegulum disc shaped(Ono, 1988); (4) sperm duct follows a circular peripheral

course through the tegulum (Schick, 1965; Ono, 1988;Benjamin, 2000); (5) conductor absent (Benjamin, 2000;Schick, 1965); and (6) median apophysis absent.

Within the Thomisus clade, although there is somephylogenetic structure, none of the groupings are wellsupported. This might be due to either an inadequatetaxon sample, a suboptimal character sample (shorter orless informative gene fragments) or both. On the otherhand it might well be due to paraphyletic genera. This ismost likely the case. The apparent non-monophyly ofseveral thomisid genera is not surprising, and wassuspected earlier (Benjamin, 2001, 2002; Lehtinen,2005). The study of generic relationships within theThomisus clade will require additional data.

Future studies

Although our hypothesis of thomisid relationshipsforms the basis for future phylogenetic studies, it alsoemphasizes that little progress has been made inunderstanding the phylogeny of this key dionychanfamily since Simon’s seminal work almost a century ago.The confusion in thomisid systematics is probably dueto the emphasis placed on single character systems, suchas male genital morphology, relative size of eyes or thepresence of cheliceral teeth. Thomisid genitalia arerelatively simple and uniform, thus less informative.

Fig. 6. Majority rule consensus tree from the trees found by the implied weight analyses with K ranging from 4 to 100. Results with K-values below 4were excluded from the consensus as suggested by Goloboff (1993b, 1995).


Further, it also may be due to the placement of undueemphasis on taxonomic work based on Holarctic faunaas in Loerbroks (1984), ignoring the fact that thomisiddiversity is biased towards tropical habitats, followingthe typical latitudinal gradient. Furthermore, cladisticmethodology was not used in previous studies.

Addition of more taxa and morphology to futurestudies is vital for understand evolutionary relationshipswithin the four clades named here. Species of mostdescribed genera could be collected easily for molecular

work, and many more are available in museums formorphological studies. Unfortunately, some enigmaticgenera, such as Heterogriffus Platnick, 1976 and Stiph-ropella Lawrence, 1952 (Stiphropodinae) or SimorcusSimon, 1895 (Strophiinae) are known only from a fewmuseum specimens. There is an immediate need toinclude them in future molecular and morphologicalstudies.

Finally, thomisid outgroup relationships need atten-tion. The phylogenetic relationships of Dionycha

Fig. 7. Tree recovered from the Bayesian analysis. Clade posterior probabilities are reported. Values <95% are omitted.


(a clade that includes 15 families in addition to Thomis-idae) have got to be resolved. Understanding theplacement of Thomisidae within Dionycha is a necessaryfirst step to resolve this major issue in spider evolution.

Acknowledgments

Thanks to the following individuals for spider spec-imens: Fernando Alvarez, Cristian Grismado, CharlesGriswold, Charles Haddad, Stephen Lew, Lara Lop-ardo, Martin Ramirez and Peter Schwendinger. Jona-than Coddington improved earlier drafts of thismanuscript. We thank Fernando Alvarez for analyticalsupport and Charles Griswold and Pekka Lethinen fordiscussion. A. H. Sumanasena (Department of Wild LifeConservation, Colombo) provided a research permit tocollect in Sri Lanka. Funding for this research has beenprovided by grants from the US National ScienceFoundation (DEB-0328644 to G. Hormiga andG. Giribet and EAR-0228699 to W. Wheeler, J.Coddington, G. Hormiga, L. Prendini and P. Sierwald)and by a REF grant from The George WashingtonUniversity to GH. Additional funding to SPB camefrom the Schlinger Foundation, the University ofCalifornia, Berkeley and a Smithsonian Institutionpostdoctoral fellowship.

References

Aldrich, J.R., Barros, T.M., 1995. Chemical attraction of male crabspiders (Araneae, Thomisidae) and kleptoparasitic flies (Diptera,Milichiidae and Chloropidae). J. Arachnol. 23, 212–214.

Arnedo, M.A., Coddington, J., Agnarsson, I., Gillespie, R.G., 2004.From a comb to a tree: phylogenetic relationships of the comb-footed spiders (Araneae, Theridiidae) inferred from nuclear andmitochondrial genes. Mol. Phylogenet. Evol. 31, 225–245.

Barrion, A.T., Litsinger J.A., 1995. Riceland Spiders of South andSoutheast Asia. CAB International, Wallingford, UK.

Benjamin, S.P., 2000. Epidius parvati sp. n., a new species of the genusEpidius from Sri Lanka (Araneae: Thomisidae). Bull. Br. Arachnol.Soc. 11, 284–288.

Benjamin, S.P., 2001. The genus Oxytate L. Koch 1878 from SriLanka, with description of Oxytate taprobana sp. n. (Araneae:Thomisidae). J. South Asian Nat. Hist. 5, 153–158.

Benjamin, S.P., 2002. Smodicinodes schwendingeri sp. n. from Thailandand the first male of Smodicinodes Ono, 1993, with notes on thephylogenetic relationships in the tribe Smodicini (Araneae: Thom-isidae). Rev. Suisse Zool. 109, 3–8.

Coddington, J.A., Levi, H.W., 1991. Systematics and evolution ofspiders (Araneae). Annu. Rev. Ecol. Syst. 22, 565–592.

Colgan, D.J., MacLauchlan, A., Wilson, G.D.F., Livingston, S.P.,Edgecombe, G.D., Macaranas, J., Cassis, G., Gray, M.R. 1998.Histone H3 and U2 snRNA DNA sequences and arthropodmo-lecular evolution. Aust. J. Zool. 46, 419–437.

Comstock, J.H., 1948. The Spider Book. Comstock Publishing Co.,Inc., Ithaca, New York.

Davila, D.S., 2003. Higher-level relationships of the spider familyCtenidae (Araneae: Ctenoidea). Bull. Am. Mus. Nat. Hist. 274, 1–86.

Dippenaar-Schoeman, A.S., 1989. The crab-spiders of Africa (Ara-neae: Thomisidae). 8. The genus Thomisops Karsch, 1879. Phyto-phylactica, 21, 319–330.

Evans, T.A., 1995. Two new species of social crab spiders of the genusDiaea from Eastern Australia, their natural history and distribu-tion. Rec. West. Aust. Mus. 52(Suppl.), 151–158.

Evans, T.A., 1998. Offspring recognition by mother crab spiderswith extreme maternal care. Proc. R. Soc. Lond. B, 265, 129–134.

Farris, J.S., 2001. Support weighting. Cladistics 17, 389–394.Farris, J.S., Albert, V.A., Kallersjo, M., Lipscomb, D., Kluge, A.,

1996. Parsimony jackknifing outperforms neighbor-joining. Cla-distics 12, 99–124.

Farris, J.S., Kallersjo, M., Kluge, A.G., Bult, C., 1994. Testingsignificance of incongruence. Cladistics 10, 315–319.

Felsenstein, J., 1985. Confidence limits on phylogenies: an approachusing the bootstrap. Evolution 39, 783–791.

Fitch, W.M., 1971. Towards defining the course of evolution:minimal change for a specific tree topology. Syst. Zool. 20, 406–416.

Gabritschevsky, E., 1927. Experiments on color changes and regen-eration in the crab-Spider, Misumena vatia. J. Exp. Zool. 47, 251–266.

Giribet, G., 2001. Exploring the behavior of POY, a program for directoptimization of molecular data. Cladistics 17, 60–70.

A B C D E

Fig. 8. Borboropactus cinerascens. Male palp: (A) ventral view; (B) prolateral view; (C) retrolteral view. Female: (D) epigynum, ventral view; (E)vulva, ventral view. Scale bars: 1 mm (D,E), 2 mm (A–C). Abbreviations: C conductor; E embolus; ET epigynal teeth; MA median apophysis; RTAretrolateral tibial apophysis.


Goloboff, P.A., 1993a. Estimating character weights during treesearch. Cladistics 9, 83–91.

Goloboff, P.A., 1993b. Pee-Wee, ver. 3.0. Program available at http://www.cladistics.com.

Goloboff, P.A., 1995. Parsimony and weighting: a reply to Turner andZandee. Cladistics 11, 91–104.

Goloboff, P.A., Farris, J.S., Nixon, K., 2003a. T.N.T. Tree AnalysisUsing New Technology, Version 1.0. Program and documentationavailable at http://www.zmuc.dk/public/phylogeny/.

Goloboff, P.A., Farris, J.S., Kallersjo, M., Oxelman, B., Ramırez,M.J., Szumik, C.A., 2003b. Improvements to resampling measuresof group support. Cladistics 19, 324–332.

Greco, C.F., Kevan, P.G., 1994. Contrasting patch choosing byanthophilous ambush predators: vegetation and floral cues fordecisions by a crab spider (Misumena vatia) and males and femalesof an ambush bug (Phymata americana). Can. J. Zool. 72, 1583–1588.

Griswold, C.E., 1993. Investigations into the phylogeny of thelycosid spiders and their kin (Arachnida: Araneae: Lycosoidea).Smithson. Contrib. Zool. 539, 1–39.

Harvey, P.H., Pagel, M.D., 1991. The Comparative Method inEvolutionary Biology. Oxford University Press, Oxford.

Hedin, M.C., Maddison, W.P., 2001. A combined molecular approachto phylogeny of the jumping spider subfamily Dendryphantinae(Araneae: Salticidae). Mol. Phylogenet. Evol. 18, 386–403.

Holm, A., 1940. Studien uber die Entwicklung und Entwicklungs-biologie der Spinnen. Zool. Bidr. Uppsala 19, 1–214.

Homann, H., 1934. Beitrage zur Physiologie der Spinnenaugen. IV.Das Sehvermogen Thomisiden. Z. Vergl. Physiol. 20, 420–429.

Homann, H., 1975. Die Stellung der Thomisidae und der Philodromi-dae im System der Araneae (Chelicerata, Arachnida). Z. Morph.Okol. Tiere 80, 181–202.

Jocque, R., Dippenaar-Schoeman, A.S., 2006. Spider Families of theWorld. Royal Museum of Central Africa, Tervuren.

Krell, F., Kraemer, F., 1998. Chemical attraction of crab spider(Araneae, Thomisidae) to a flower fragrance component. J.Arachnol. 26, 117–119.

Lehtinen, P.T., 2005. Taxonomic notes on the Misumenini (Araneae:Thomisidae: Thomisinae), primarily from the Palaearctic andOriental regions. In: Logunov, D.V., Penney, D. (Eds.), EuropeanArachnology 2003. Arthropoda Selecta (Special Issue no. 1, 2004).

Levi, H.W., 1982. Araneae. In: Parker, S.P. (Ed.), Araneae. McGrawHill, New York, pp. 77–95.

Loerbroks, A., 1984. Mechanik der Kopulationsorgane von MisumenaVatia. Clerck, 1757) (Arachida: Araneae: Thomisidae). Abh.Naturw, Ver. Hamburg (NF) 27, 383–403.

Maddison, W.P., Maddison, D.R., 2001. MacClade: Analysis ofPhylogeny and Character Evolution, ver. 4.0. Sinauer Associates,Sunderland, MA.

Main, B.Y., 1988. The biology of a social thomisid spider. In: Austin,A.D., Heather, N.W. (Eds.), Australian Arachnology. AustralianEntomological Society, Brisbane, pp. 55–73.

Mickevich, E., Farris, J.S., 1981. The implications of congruence inMenidia. Syst. Zool. 30, 351–370.

Morse, D.H., 1988. Cues associated with patch-choice decisions byforaging crab spiders Misumena vatia. Behaviour 107, 297–313.

Nixon, K.C., 2002. WinClada. Published by the Author.Nyffeler, M., Benz, G., 1987. Spiders in natural pest control: a review.

J. Appl. Entomol. 103, 321–339.Oliveira, P.S., Sazima, I., 1984. The adaptive bases of ant-mimicry in a

neotropical aphantochilid spider (Araneae: Aphantochilidae). Biol.J. Linn. Soc. 22, 145–155.

Ono, H., 1988. A Revisional Study of the Spider Family Thomisidae(Arachnida, Araneae) of Japan. National Science Museum, Tokyo.

Packard, A.S., 1905. Change of color and protective coloration in aflower-spider (Misumena vatia Thorell). J. N.Y. Entomol. Soc. 13,85–96.

Petrunkevitch, A., 1928. Systema Araneorum. Trans. Connect. Acad.Arts Sci. 29, 1–270.

Pickard-Cambridge, O., 1871. Arachnida. Zool. Rec. 7, 207–224.Platnick, N.I., 2007. The World Spiders Catalog, ver. 7.5. American

Museum of Natural History, New York. http://research.amnh.org/entomology/spiders/catalog81-87/index.html/.

Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model ofDNA substitution. Bioinformatics 14, 817–818.

Riechert, S.E., 1974. Thoughts on the ecological significance of spiders.Bioscience 24, 352–356.

Riechert, S.E., Lockley, T.C., 1984. Spiders as biological controlagents. Annu. Rev. Entomol. 29, 299–320.

Roberts, M.J., 1995. Spiders of Britain and Northern Europe. HarperCollins, London.

Roewer, C.F., 1954. Katalog der Araneae von 1758 bis 1940, bzw1954. Institut Royal des Sciences Naturelles de Belgique, Bruxelles.

Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phyloge-netic inference under mixed models. Bioinformatics 19, 1572.

Schick, R.X., 1965. The crab spiders of California (Araneida:Thomisidae). Bull. Am. Mus. Nat. Hist. 129, 1–180.

Shelford, R., 1902. Observations on some mimetic insects and spidersfrom Borneo and Singapore. Proc. Zool. Soc. Lond. 1902, 230–284.

Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., Flook, P.,1994. Evolution, weighting, and phylogenetic utility of mitochon-drial gene sequences and a compilation of conserved polymerasechain reaction primers. Ann. Entomol. Soc. Am. 87, 651–701.

Simon, E., 1892 (1892–1895). Histoire naturelle des Araignees. Roret,Paris. pp. 949–1066.

Steel, M.A., 1994. The maximum likelihood point for a phylogenetictree is not unique. Syst. Biol. 43, 560–564.

Suzuki, Y., Glazko, G.V., Nei, M., 2002. Overcredibility of molecularphylogenies obtained by Bayesian phylogenetics. Proc. Acad. Nat.Sci. Phila. 99, 16138–16143.

Swofford, D.L., 2002. PAUP*. Phylogenetic Analysis Using Parsi-mony (*and Other Methods), ver. 4. Sinauer Associates, Sunder-land, MA.

Swofford, D.L., Olsen, 1991. Phylogeny reconstruction. In: Hillis,D.M., Moritz, C. (Eds.), Phylogeny Reconstruction. Sinauer,Sunderland, MA, pp. 407–514.

Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W:improving the sensitivity of progressive multiple sequence align-ment through sequence weighting, positions-specific gap penaltiesand weigh matrix choice. Nucleic Acids Res. 22, 4673–4680.

Thorell, T., 1873. Remarks on synonyms of European spiders. Part IV,C.J. Lundstrom, Uppsala, pp. 375–645.

Tikader, B.K., 1980. Thomisidae (Crab-spiders). Zoological Survey ofIndia, Calcutta.

Uetz, G.W., Halaj, J., Cady, A.B., 1999. Guild structure of spiders inmajor crops. J. Arachnol. 27, 270–280.

Wheeler, W.C., 1996. Optimization alignment: the end of multiplesequence alignment in phylogenetics? Cladistics 12, 1–9.

Wheeler, W.C., Hayashi, C.Y., 1998. The phylogeny of the extantchelicerate orders. Cladistics 14, 173–192.

Wheeler, W.C., Gatesy, J., DeSalle, R., 1995. Elision: a method foraccommodating multiple molecular sequence alignments withalignment-ambiguous sites. Mol. Phylogenet. Evol. 4, 1–9.

Wheeler, W.C., Aagesen, L., Arango, C.P., Faivovich, J., Grant, T.,D’Haese, C., Janies, D., Smith, W.L., Varon, A., Giribet, G., 2006.Dynamic Homology and Phylogenetic Systematics: A UnifiedApproach Using POY. American Museum of Natural History,New York.

Wilcox, T.P., Zwickl, D.J., Heath, T.A., Hillis, D.M., 2002. Phyloge-netic relationships of the dwarf boas and a comparison of Bayesianand bootstrap measures of phylogenetic support. Mol. Phylogenet.Evol. 25, 361–371.

Wunderlich, J., 2004a. Fossil crab spiders (Araneae: Thomisidae) inBaltic and Dominican amber. Beitr. Araneol. 3, 1747–1760.


Wunderlich, J., 2004b. The new spider (Araneae) family Borboropac-tidae from the tropics and fossil in baltic amber. Beitr. Araneol. 3,1737–1746.

Appendix 1

Taxonomic changes, Fig. 7(A–E).Genus: Borboropactus Simon, 1884.Borboropactus cinerascens (Doleschall, 1859).B. bangkongeus Barrion and Litsinger (1995). New synonomy.B. mindoroensis Barrion and Litsinger (1995). New synonomy.B. umaasaeus Barrion and Litsinger (1995). New synonomy.B. mindoroensis is a juvenile female; penultimate female thomisids

some times have partly sclerotized genitalia. B. bangkongeus andB. umaasaeus are one and the same species. Their illustrations (Barrionand Litsinger, 1995) leave no doubt that they are synonymous withB. cinerascens. Vouchers are deposited at the Natural History Museumof Geneva.

Appendix 2

Glossary of terms

Terms used in the discussion taken from Jocque and Dippenaar-Schoeman (2006).

Conductor: a sclerite of the male copulatory palp that accompaniesand supports the embolus; a hyaline conductor is a conductor that istransparent and very thin (Fig. 8C).

Colulus: short median protuberance in front of spinnerets, consid-ered a modification of the spinning plate or cribelum.

Embolus: embolus is the intromitted part of the male copulatorypalp (Fig. 8B).

Epigynum: a usually chitinous section on the ventral side of theabdomen of adult female spiders on which the genital openings arefound. Epigynes of some thomisids have lateral sclerotized projectionstermed teeth (Fig. 7D,E).

Male palp: modified tarsus of the palp in male spiders. It is acopulatory organ of great complexity and is probably the mostimportant character system in spiders (Fig. 7A,B).

Median apophysis: an apophysis of the male palpal bulb (Fig. 8B).Scopula hairs: specialized hairs found on the tip of legs in some

thomisids. Thought to improve grip.Tegulum: part of the male copulatory palp that houses the sperm

duct.


Documents

Family ties: molecular phylogeny of crab spiders (Araneae