PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Volume 56, Supplement II, pp. 1-324, 220 figs., 1 table, Appendices October 28, 2005 Atlas of Phylogenetic Data for Entelegyne Spiders (Araneae: Araneomorphae: Entelegynae) with Comments on Their Phylogeny Charles E. Griswold1, Martin J. Ramirez2, Jonathan A. Coddington 3> and Norman I. Platnick4 1 Department of Entomology, California Academy of Sciences, 875 Howard Street, San Francisco, CA 94103, USA; Email: cgriswold@calacademy.org; 2 Museo argentino de Ciencias naturales "Bernardino Rivadavia," Avenida Angel Gallardo 470, 1405 Buenos Aires, Argentina; 3 Systematic Biology - Entomology, Smithsonian Institution, P.O. Box 37012, NMNH E529, NHB-105, Washington, DC 20013-7012, USA; 4 Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, USA Table of Contents Abstract 2 Introduction 2 Taxon choice 3 Conventions 5 Materials and methods 5 Specimen preparation 6 Acknowledgments 7 Results 8 The Entelegyne exemplars 8 Agelenidae 8 Amaurobiidae 9 Amphinectidae 12 Araneoidea 14 Archaeidae 14 Austrochilidae 15 Ctenidae 17 Deinopidae 18 Desidae 19 Dictynidae 21 Eresidae 24 Filistatidae 27 Gradungulidae, Huttoniidae 28 Hypochilidae 29 Mimetidae 30 Neolanidae 31 Nicodamidae 32 Oecobiidae 33 Pararchaeidae 34 Phyxelididae 35 Psechridae 37 Segestriidae 39 Stiphidiidae 40 Tengellidae 41 2 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Titanoecidae 42 Uloboridae 43 Zorocratidae 44 Zoropsidae 46 Characters 47 Discussion 70 Character systems and homoplasy levels 71 Groups 71 Conclusions 76 Literature cited 76 Appendicies 83 Appendix 1. Taxa examined to provide exemplar data 85 Appendix 2. Data matrix 94 Appendix 3. Comments on data matrix 97 Illustrations 101 Index: Systematic and Geographic 319 We present a phylogenetic analysis of higher groups of entelegyne spiders, with representatives of all entelegyne families containing cribellate members and of Palpimanoidea. We examined 55 exemplar species of eresoids (Oecobiidae, Eresidae), Orbiculariae (Deinopidae, Uloboridae, Araneidae), Palpimanoidea (Archaeidae, Huttoniidae, Mimetidae, Pararchaeidae), titanoecoids (Phyxelididae, Titanoecidae), lycosoids and related groups (Ctenidae, Psechridae, Tengellidae, Zorocratidae, Zoropsidae), and several other entelegyne families of contentious rela- tionships (Agelenidae, Amphinectidae, Desidae, Neolanidae, Stiphidiidae, Amaurobiidae, Dictynidae, Nicodamidae), plus representatives of the relatively basal araneomorph groups Palaeocribellatae (Hypochilidae), Austrochiloidea (Austrochilidae and Gradungulidae), and Haplogynae (Filistatidae, Segestriidae). We also included the enigmatic cribellate genera Aebutina (Dictynidae?) and Poaka (Psechridae? Amaurobiidae?). This selection of taxa covers much of the morpholog- ical diversity found through major groups of araneomorph spiders. The 154 charac- ters in our dataset include all the classical character sources in higher level spider taxonomy. We present a large collection of labeled images to document character definitions and observations. The phylogenetic trees obtained with our dataset differ according to parameters of analysis (equal or implied weights), and diverge from previous results. We obtain a paraphyletic Araneoclada, excluding members of Haplogynae. At least some taxa previously assigned to Palpimanoidea appear to be nested within orb weavers. The outgroups to Orbiculariae remain an open question, and the monophyly of Nicodamidae, Amaurobiidae and Zorocratidae are ques- tioned. We corroborated Austrochiloidea, Eresoidea, and the Divided Cribellum, Oval Calamistrum and RTA Clades. The Entelegynae comprise the largest group of spiders with more than 38,000 described species (Platnick 2004). Ideas about entelegyne spider evolution were long dominated by the fauna of the northern hemisphere, especially Europe. This impoverished fauna includes distinct and dis- tantly related taxa, most of which are ecribellate or colulate. Spiders with cribella, the remarkable spinning plate that works in conjunction with a comb on the fourth leg to make a "hackled band" capture thread, form a small subset of this northern fauna. For nearly a century the higher spiders were divided into two groups: those with cribella, and those without. All this changed in the late 20th century with global taxonomic studies of higher spiders (Lehtinen 1967), intense studies of the spiders of the southern hemisphere (Forster 1970; Forster and Wilton 1973), and application of Pfennig's cladistic principles to spider taxonomy (e.g., Platnick and Gertsch 1976; Platnick 1977). GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 3 The cribellum was reinterpreted as a primitive feature common to all higher spiders, albeit, for most, in greatly modified form. The cribellum is clearly more ancient than even the entelegyne con- dition, i.e., that of having separate copulatory and fertilization ducts. Entelegynes have repeatedly been the subject of quantitative phylogenetic analysis. The earli- est studies focused on clades of special interest, e.g., orb weavers (Coddington 1990a, 1990b; Griswold et al. 1998, Fig. 211), haplogynes (Platnick et al. 1991, Fig. 209), lycosoids (Griswold 1993, Fig. 213). The first attempt at a comprehensive entelegyne phylogeny was by Griswold et al. (1999), who chose and coded cribellate members from families across the araneomorph spectrum. Their approach was guided by the words of Lehtinen (1967:202) who declared, "because of the central position of the Cribellate groups in Araneomorphae, a detailed revision of them is a short cut to a rough classification of the whole suborder." They chose exemplars from all cribellate fam- ilies, reasoning that taxa retaining this plesiomorphic feature are more likely to straddle the basal nodes of the phylogeny of higher groups than are their relatives that have lost the cribellum: there- fore they are most likely to reflect phylogenetic groundplans. Although phylogenetically ancient, the cribellum is a complex feature unlikely to have evolved more than once. Most major araneo- morph clades have cribellate members (exceptions are Palpimanoidea and Dionycha). A phyloge- ny of these basal taxa should mirror the relationships of the large clades they exemplify. The pro- visional phylogeny of Griswold et al. (1999) tested many suprafamilial hypotheses of the last 30 years and was the first attempt to relate them using quantitative phylogenetic techniques. The cladogram confirmed some accepted groupings, refuted others, and several novel phylogenetic and nomenclatural changes were proposed (Fig. 212). Confirmed were the monophyly of Neocribel- latae, Araneoclada, Entelegynae, and Orbiculariae. The Lycosoidea, Amaurobiidae and some included subfamilies, Dictynoidea, and Amaurobioidea (sensu Forster and Wilton 1973) appeared polyphyletic. Phyxelididae Lehtinen was raised to family level and Zorocratidae Dahl was revali- dated. A group including all other entelegynes other than Eresoidea was weakly supported as the sister group of Orbiculariae and several new, informative, informal clades were proposed or rede- fined: the "Canoe Tapetum Clade," "Divided Cribellum Clade," the "Titanoecoids," the "RTA Clade," the "Fused Paracribellar Clade," the "Stiphidioids" and the "Agelenoids." This slender paper (Griswold et al. 1999), constrained by publication in a congress volume, offered only the briefest outline of the data. Griswold and Wang (2001) presented a fuller account of the results, pre- senting character state trees for each of the 137 characters and figures depicting many of the char- acter states. In this study we will present and illustrate in detail the morphology and other characteristics of the exemplar taxa, explain the character coding, and discuss some implications of our trees for spider evolution. We hope that this paper, especially the new data presented herein, will provide a springboard to further, more detailed and more comprehensive analyses of araneomorph phyloge- ny. This paper is dedicated to the memory of Ray Forster. TAXON CHOICE We chose exemplars from all cribellate families, reasoning that taxa retaining this plesiomor- phic feature are more likely to straddle the basal nodes of the phylogeny of higher groups than are their relatives that have lost the cribellum: therefore they are most likely to reflect phylogenetic groundplans. We have also added 11 ecribellate representatives of clades that we believe are not adequately represented by their cribellate members only (e.g., Haplogynae, Nicodamidae), or where all their members lack cribella (Palpimanoidea, Araneoidea). Only the major group 4 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II TABLE 1. List of anatomical and institutional abbreviations used in the text and figures. A alveolus ML AC aciniform gland spigot(s) MJR AD vulval afferent duct MRAC AER anterior eye row MS AC, aggregate gland spigot(s) ALE anterior lateral eyes MTP ALS anterior lateral spinneret MUSM AME anterior median eyes AMNH American Museum of Natural History, New York N AN anneli of subtegulum NMSA AT epigynal atrium NS AX cribellate silk axial lines OAL BH basal haematodocha oLl BMNH The Natural History Museum, London oL4 C conductor OMD CB cymbium OQA CAS California Academy of Sciences, San Francisco OQP CF "cuticular finger" on PLS PC CG Charles Griswold PER CO copulatory opening PF CM cribellate silk mass PI CY cylindrical gland spigot(s) PLE DTA tibial dorsal process PLS E embolus PME EB embolar base PMS EF epigastric furrow PTA F fundus PY FD fertilization duct QMS FL flagelliform gland spigot(s) RMNH FMNH Field Museum of Natural History, Chicago RTA FO foundation line RW GV groove S HNU Hunan Normal University, Changsha SEM HS spermathecal head SR in inside leg 1 ss InBio Institute Nacional de Biodiversidad, San Jose ST ITC inferior tar sal claw STC JC Jonathan Coddington STP JGU Johannes Gutenberg University, Mainz T LNZ Landcare New Zealand, Wellington TA LL epigynal lateral lobe(s) TO L3 leg 3 TP L4 leg 4 TR MA median apophysis UE MACN Museo Argentino de Ciencias Naturales, Buenos Aires USNM MAP major ampullate gland spigot(s) mAP minor ampullate gland spigot(s) VTA MCZ Museum of Comparative Zoology, Harvard ZMUC MH epigynal median hood epigynal median sector or lobe Martin J. Ramirez Musee Royal de l'Afrique Centrale, Tervuren modified PLS spigots (including pseudoflagelliform gland spigot) membranous tegular process Museo de Historia Natural de La Universidad de San Marcos, Lima nubbin Natal Museum, Pietermaritzburg non sticky silk ocular area length outside leg 1 outside leg 4 Otago Museum, Dunedin ocular quadrangle, anterior ocular quadrangle, posterior paracribellar spigot(s) posterior eye row postepigastric fold piriform gland spigot(s) posterior lateral eyes posterior lateral spinneret posterior median eyes posterior median spinneret tibial proapical process paracymbium Queensland Museum, South Brisbane Rijksmuseum van Natuurlijke Historic, Leiden tibial retroapical process cribellate silk reserve warp epigynal scape scanning electron microscope sperm receptacle sticky silk subtegulum superior tarsal claw sclerotized tegular process tegulum tegular apophysis tarsal organ tartipore terminal apophysis of embolic division uterus externus National Museum of Natural History, Smithsonian Institution, Washington, D. C. tibial ventroapical process Zoological Museum, University of Copenhagen Dionycha (sensu Coddington and Levi 1991) is not represented. The scope and placement of the Dionycha were tangentially addressed by Silva Davila (2003; Fig. 214) and are currently under study (Ramirez, in prep.); we leave those problems to that study. Our dataset comprises 55 exemplars (Appendix 1) from all of the 22 araneomorph families currently (Platnick 2004) with cribellate members. As outgroups we included HYPOCHILIDAE (Hypochilus), GRADUNGULIDAE (Gradungula), AUSTROCHILIDAE (Hickmania and Thaida), FILISTATIDAE (Filistata and Kukulcania, Filistatinae), and SEGESTRIIDAE (Ariadna). In this group we included OECOBIIDAE (Oecobius and Uroctea) and ERESIDAE (Eresus and Stegodyphus) from the eresoids. From Orbiculariae we included DEINOPIDAE (Deinopis and Menneus), ULOBORIDAE (Octonoba and Uloborus) and ARANEIDAE (Araneus). Recent phy- GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 5 logenetic study of Araneoidea (Griswold et al. 1998; Fig. 211) gives us confidence that Araneus accurately reflects the primitive conditions of characters treated herein for this superfamily. From the Palpimanoidea we included ARCHAEIDAE (Archaea), HUTTONIIDAE (Huttonia), MIMETIDAE (Mimetus), and PARARCHAEIDAE {Pararchaea). Our choice of exemplars for the Palpimanoidea included both haplogyne (Archaea, Huttonia) and entelegyne (Mimetus, Pararchaea) taxa. In particular, Mimetidae is the most contentious of the taxa placed in Palpimanoidea (Platnick et al. 1991, Fig. 209; Schtitt 2002, Fig. 210). From the fused paracribel- lar clade (sensu Griswold et al. 1999; Fig. 212) we included AGELENIDAE (Neoramia), AMPHINECTIDAE (Maniho and Metaltella), DESIDAE (Desis, Badumna, Matachia and Phryganoporus, formerly Matachiinae), NEOLANIDAE (Neolana), and STIPHIDIIDAE (Pillara and Stiphidion). From the titanoecoids (sensu Griswold et al. 1999; Fig. 212) we included PHYX- ELIDIDAE (Phyxelida, Vytfutia, and Xevioso) and TITANOECIDAE (Goeldia and Titanoeca). From lycosoids and related groups we included CTENIDAE (Acanthoctenus), PSECHRIDAE (Psechrus and Poaka), the latter genus recently transferred to the Amaurobiidae, TENGELLIDAE (Tengella), ZOROCRATIDAE (Zorocrates, Raecius and Uduba), and ZOROPSIDAE (Zoropsis). Other families represented are the AMAUROBIIDAE (Amaurobius) and Callobius (Amaurobi- inae), Macrobunus, Retiro, and Pimus (Macrobuninae), DICTYNIDAE (Dictyna and Nigma, Lathys, and Tricholathys representing Dictyninae, Cicurininae, and Tricholathysinae, respectively, and the enigmatic Aebutina), and NICODAMIDAE (Megadictyna and Nicodamus). Voucher spec- imens for exemplars are listed in Appendix 1. CONVENTIONS Throughout the text, figures cited from previous papers are listed as "fig."; those appearing in this paper as "Fig." Abbreviations used in the text and figures are listed in Table 1. Exemplar local- ities in the text are represented by "Place, State, Country" except for those in Australia and the USA, which also include the State. All localities are listed completely in Appendix 1. MATERIALS AND METHODS Cladistic Analysis Analyses were performed with TNT 1.0 (Goloboff et al. 2003, Goloboff et al. 2004) and Nona 2.0 (Goloboff 1993b). All characters in this dataset were treated unordered. Under equal weights, both programs find 96 optimal trees of 483 steps, using either collapsing rule 4 (TNT collapse 4) or "min length = 0" rule (TNT collapse 2, Nona ambiguous-) (Coddington and Scharff 1994). This set expands to 224 dichotomous trees (no collapsing; TNT collapse 0, Nona poly=). Using the par- simony ratchet (Nixon 1999) as heuristic search, TNT and Nona find the optimal trees in 100% of the replicates, each with 100 iterations keeping up to 5 trees per iteration, using tree bisection- reconnection (TBR) (TNT ratchet: iter 100; mult = tbr replic 100 hold 5 ratchet; Nona hold/ 5 nixwts*100 100). With so many hits it is likely the optimal tree was found. We also analyzed the dataset under weighting regimes against homoplasy, using successive weighting (Farris 1969) and implied weighting (Goloboff 1993). The most recent method of implied weights was given priority over successive weighting because implied weights is not affected by starting points or ambiguities in weights from multiple trees. Successive weighting was calculated in Nona using the consistency index as a weighting function, using 100 random addition sequences followed by TBR swapping in each round (run[swt mu*100]). The searches stabilized in the second round, and the tree is very similar to the one found under implied weights; the minor 6 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II differences are shown in Figure 219. Analyses under implied weights were made with TNT, with integer values of the constant of concavity K = 1 to 6. Under this parameter, TNT finds the opti- mal tree under implied weights in 25% percent of the ratchet replicates. We present the unique, fully resolved optimal tree under a mild concavity with K = 6 (Fig. 217), and present the sensitiv- ity to concavity changes in Figures 218 and 219. Ramirez (2003) found that mild concavity values produced higher topological congruence indices. The sensitivity of groups to changes in the analy- sis parameters also provides an insight to the support of groups (Giribet 2003). We produced synapomorphy lists mapping the unambiguous changes (e.g., 0 ?> 1, but not 01 ?> 1; 01 ?? 2, but not 01 ?> 12; option ambiguous- of Nona, only option in TNT). Because synapo- morphy lists for polytomies in consensus representations are dependent on the optimal resolutions, we calculated all optimal dichotomous trees and produced lists of synapomorphies that are com- mon to all dichotomous trees (Fig. 216; command apo[ of TNT and Nona). Character indices (Fig. 220) were calculated exporting basic values (steps, minimum and maximum possible character lengths) from Nona to a spreadsheet. Under equal weights, only the best scores over the 224 dichotomous trees are reported in Figure 220. Bremer support values were heuristically estimated by TBR swapping from the optimal trees, retaining suboptimal trees with increasing bounds, up to 50,000 (equal weights) or 30,000 trees (implied weights). Symmetric resampling frequencies (p = 0.33) are reported as GC values (GC = 100 is perfect support, GC = 0 is unsupported). The GC is the absolute frequency of a group, minus the frequency of the most frequent contradictory group, and has shown to be less biased than the traditional bootstrap or jackknifing estimations (Goloboff et al. 2003; Goloboff et al. 2004). We estimated the GC values with 1000 pseudoreplicates of five random sequence additions each fol- lowed by TBR swapping, keeping up to 10 trees, collapsing trees with a round of TBR SPECIMEN PREPARATION Male palpi were expanded for all taxa by immersing them overnight in a 10-15% solution of potassium hydroxide (KOFI) and transferring them to distilled water where expansion continued. Palpi were transferred back and forth between KOH and distilled water until expansion stopped. Small structures were examined in temporary mounts following the procedure described in Coddington (1983), or in excavated slides with clearing medium. Spinneret preparations were obtained most reliably when animals were quick-killed by sudden immersion in boiling water. Extension of the spinnerets provided a clear view of all spigots. If live material was unavailable, clean museum material was chosen; the specimen was ultrasonically cleaned, the abdomen squeezed with forceps to extend and separate the spinnerets (Coddington 1989:73) if necessary, and the specimen passed through serial concentrations from 75% to 100% ethanol. Prior to scanning electron microscope examination palpi and spinnerets were critical-point dried; all other structures were air dried. Cribellate silk preparation and examination were done by Robin Carlson and Martin Ramirez. All drawings were made with a camera lucida attached to Olympus, Leitz, or Leica stereo or compound microscopes. Trichobothria bases were considered "smooth" if their sculpturing did not differ from the surrounding leg cuticle. Vulvae were cleaned by immersion in a trypsin solu- tion for three to fives days at room temperature or by digestion with contact lens cleaner overnight (Sierwald 1990), or cleared with clove oil or with Chlorox? bleach. The tracheal system was exam- ined after digestion in KOH 10-20% in a double boiler. Specific methods are discussed under each character section. The preferred method of examining the tapetum was by microscopic examina- tion of live or recently-dead specimens. The tapetum is a shiny reflective surface that, when pres- ent, occurs only in the ontogenetically lateral eyes, and not in the AME. It remains clearly visible for 24?48 hours after a spider's death (after this time the vitreous body usually becomes cloudy, GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 7 obscuring the details of the tapetum). When live material was not available, preserved specimens were prepared by removing the chelicerae and most of the musculature from the anterior part of the cephalothorax. The cephalothorax was immersed in lactic acid for 1-5 hours. Frequently the reti- na cleared and details of the tapetum became visible, though, for unknown reasons, this was not always successful. ACKNOWLEDGMENTS The following institutions and curators kindly lent essential specimens: Field Museum of Natural History (Petra Sierwald), Hunan Normal University (Changmin Yin), Instituto Nacional de Biodiversidad, Costa Rica, Johannes Gutenberg University (Peter Jager), Landcare New Zealand (Grace Hall), Musee Royal de l'Afrique Centrale (Rudy Jocque), Museo de Historia Natural de La Universidad de San Marcos (Diana Silva Davila), Museum of Comparative Zoology (Herbert Levi and Laura Leibensperger), Natal Museum (Peter Croeser), The Natural History Museum, London (Paul Hillyard and Janet Beccaloni), Otago Museum, Dunedin (Ray Forster), Queensland Museum (Robert Raven), Rijksmuseum van Natuurlijke Historic (Christa Deeleman), and Zoological Museum University of Copenhagen (Nikolaj Scharff). Other specimens are from the American Museum of Natural History, California Academy of Sciences, Museo Argentino de Ciencias Naturales, and National Museum of Natural History (Smithsonian Institution). Partial support for all authors came from NSF grant EAR-0228699 (Assembling the Tree of Life: Phylogeny of Spiders) (W. Wheeler PL). Griswold wishes to acknowledge financial support from National Science Foundation grants BSR-9020439 and DEB-9020439 (Systematics and Biogeography of Afromontane Spiders), DEB-0072713: (Terrestrial Arthropod Inventory of Madagascar), the Exline-Frizzell and In-house Research Funds of the CAS, and post-doctoral fel- lowships from the Smithsonian Institution and the American Museum of Natural History. Coddington wishes to acknowledge financial support from National Science Foundation grants DEB-9712353 and DEB-9707744, and the Neotropical Lowlands Program and Biotic Surveys and Inventory Program from the Smithsonian Institution. Ramirez wishes to acknowledge financial support and post-doctoral fellowships from a Fessenden Research Fellowship from AMNH, a post- doctoral fellowship and grant PEI 6558 from CONICET, PICT03/14092 from FONCYT, and UBA research grant X019. Lucrecia Nieto shared new data on the internal anatomy of Kukulcania. Robin Carlson shared new data on the fine structure of cribellate silk; her research was enabled by the CAS Summer Systematics Institute, itself supported by NSF grant BIR-9531307. Specimens and observations essential to the completion of this study were obtained through fieldwork in Australia, Cameroon, Madagascar, New Zealand, South Africa, Chile, and Tanzania. This fieldwork was made possible through permits issued by a variety of government bodies and traditional authorities. Permits to do research in Australia were made available from the Queensland National Parks and Wildlife Service, facilitated by Dr. R. Raven of the Queensland Museum. Permits to do research in Cameroon were made available from the Institute of Zootechnical Research, Ministry of Higher Education and Scientific Research of the Republic of Cameroon. Griswold especially thanks Dr. John T. Banser, Director of the Institute, and Dr. Chris Wanzie, for facilitating his research in Cameroon. Fieldwork in Cameroon was also made possible by the permission and assistance from the following traditional authorities: Their Royal Highnesses the chiefs of Batoke, Etome, and Mapanja, H.R.H. R.M. Ntoko, Paramount Chief of the Bakossi, Chiefs Jonas Achang, Albert Ekinde, and Peter Epie of Nyassosso, and H.R.H. the Fon of Oku. Permits to do research in and export specimens from Madagascar were obtained from the Association Nationale pour le Gestion des Aires Protegees (ANGAP) and Direction des Eaux et Forets of the Ministre d'Etat a L'Agriculture et au Development Rural, under Accordes de 8 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Collaboration of the Xerces Society (facilitated by Dr. C. Kremen, Mr. C. Ramilison, and Ms. B. Davies of that organization), and Institute for Conservation of Tropical Environments (ICTE) (facilitated by Drs. R. van Berkum and Benjamin Andriamihaja). Permits to collect specimens in New Zealand protected areas were obtained from Te Papa Atawhai (Department of Conservation). We thank Lyn and the late Ray Forster and Brian Patrick for advice, help with logistics, and hospitality in New Zealand. Research in Tanzania was made possible through a Research Permit from the Tanzania Commission for Science and Technology (COSTECH) and Residence Permit Class C from the Tanzanian Department of Immigration, and export of specimens made possible by a CITES Exemption Certificate from the Wildlife Division of the United Republic of Tanzania, facilitated by Professor Kim M. Howell of the University of Dar-es-Salaam. Research in the East Usambaras (Tanzania) was made possible by accommodation at the East Usambara Conservation and Agricultural Development Project, Dr. J.K. Ningu, Project Manager, and facilitated by Mr. Massaba I.L. Katigula, East Usambara Catchment Forest Office, Tanga, and Mr. Bruno Samuel Mallya, Kwamkoro. Research in the West Usambaras (Tanzania) was made possible by Dr. SAO. Chamshama, Dean of Forestry, Sokoine University, Morogoro, and Mr. Modest S. Mrecha, Officer in Charge, Mazumbai Forest Reserve. Research in South Africa was made possible through a Research Permit from KwaZulu-Natal Nature Conservation Service. Research in Chile was made possible through a Research Permit from the Corporation Nacional Forestal (CONAF), and a grant-in-aid of research from Sigma-Xi to MJR. Karin Schtitt kindly supplied access to her thesis and manuscripts in press. We thank Leticia Aviles, Per de Place Bj0rn, L. Joy Boutin, Fred Coyle, Crista Deeleman, Niall Doran, the late Ray Forster, Mark Harvey, Gustavo Hormiga, Andrew McLachlan, Lara Lopardo, the late Bill Peck, Barbara and the late Vincent Roth, Evert Schlinger, Diana Silva Davila, Darrell Ubick and Carlos Viquez for providing crucial specimens. Roy Harimer assisted with digital photography of silk. Darrell Ubick (CAS), Susan Braden (NMNH), Angela Klaus and Kevin Frischmann (AMNH), and Fabian Tricarico (MACN) assisted with scanning electron microscopy. Cristian Grismado assisted with text editing. Photographs were kindly provided for reproduction by Rollin Coville, Pat Craig, Gustavo Hormiga, Joel Ledford, Teresa Meikle, Nikolaj Scharff, Lenny Vincent and Hannah Wood. Drafts of the manuscript were read and criticized by Robert Raven, Nikolaj Scharff, Petra Sierwald and Darrell Ubick, and the prepress version was proofed in its entirety by Dr. Michele L. Aldrich. We are grateful for their time, effort, and suggestions. RESULTS The results of this study are presented in two complementary forms. The first section, The Entelegyne Exemplars, contains a detailed description of the characters and their states in each of the exemplar taxa. The descriptions are organized in a consistent order: taxonomic details of the family, web and silk structures, eyes, chelicerae, legs (including setae, trichobothria and tarsal organ), spiracles and tracheae, spinneret and spigot data, with a description of male and female copulatory organs at the end of each family description. All stated characters are referenced by fig- ures. The second section, Characters, contains a detailed discussion of all characters, references to figures and an explanation of the delineation of states. THE ENTELEGYNE EXEMPLARS Agelenidae C.L. Koch, 1837 The Agelenidae are a large, worldwide family of 39 genera and 487 species (Platnick 2004; Ramirez et al. 2004), particularly rich in Eurasia, Africa and North America. A few monotypic gen- GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 9 era occur in South America, and at least the synanthropic Tegenaria domestica appears to be cos- mopolitan. Agelenids are widely known as funnel web spiders, building non-sticky sheets across which the spider runs quickly, hiding in a funnel shaped retreat at one end. Most are solitary but some African species are social. A bipartite colulus comprising two hairy patches (first noted by Lehtinen 1967:342) may be a synapomorphy for the ecribellate taxa. New Zealand is home to sev- eral endemic genera, including the only cribellate agelenids. We have chosen the New Zealand Neoramia sana as our exemplar. Neoramia sana were observed near Dunedin, New Zealand, where they live in a retreat surrounded by an appressed, radiating web upon which they move (Figs. 204H, I). The fine structure of Neoramia cribellate silk is unknown. The eight eyes in two straight rows (Fig. 2041) have a canoe-shaped tapetum, the body is cov- ered with plumose hairs, trichobothrial bases are smooth (Fig. 155H) and the capsulate tarsal organ has a teardrop- or keyhole-shaped orifice (Fig. 153F). The chelicerae have a large boss, teeth on the fang furrow and thickened setae near the fang base. We could not observe a chilum. Metatarsi III and IV of both sexes have apical preening combs of several setae. There are three claws but claw tufts, serrate accessory setae and scopulae are absent. Forster and Wilton (1973) recorded four sim- ple tracheal tubes. We describe Neoramia spinning organs here for the first time. The cribellum is divided into two fields of strobilate spigots (Fig. 82F). The anterolateral spinneret (ALS) has a wide, bare mar- gin around the spinning field (Fig. 73B). The female ALS has two major ampullate spigots (MAP) clustered at the mesal margin, a large tartipore mesad of these and a piriform (PI) field of more than 30 spigots with rounded base margins interspersed with tartipores (Fig. 73B). Males retain only the anterior MAP, with the posterior one replaced by a nubbin (Figs. 74A-B). The female posterior median spinneret (PMS) (Fig. 73C) has two aciniform gland (AC) spigots along the anterior mar- gin, a large mesal minor ampullate gland spigot (mAP) posterad of these, and two posterior cylin- drical gland spigots (CY). The paracribellar spigots (PC) are remarkable: at midfield two thick bases each give rise to bundles of 10-12 strobilate PC shafts. The male PMS (Fig. 74C) lacks the CY, and only large nubbins remain of the PC, though one vestigial PC shaft remains in one male (Fig. 82B). The conical apical segment of the female posterolateral spinnerets (PLS) (Figs. 73D, 82A) has an apical modified spigot (MS) flanked by two slender nubbins, similar to those found in some phyxelidids (e.g., Xevioso), and are probably the bases of PC that have lost their shafts. There are 7-10 acinifirm gland (AC) spigots interspersed by tartipores and three large CY spigots along the mesal margin (Fig. 73D) of the PLS. The apex of the male PLS retains only the nubbins of the MS and its two accompanying spigots (Fig. 74D). Males lack epiandrous spigots (Fig. 161F). The male pedipalpus has a complex retrolateral tibial apophysis (RTA) consisting of a broad apical and three slender subapical processes (Figs. 179C, 189B). The bulb (Figs. 179A-B, 189B) has two processes in addition to the embolus: a sclerotized, cup-shaped conductor (C) that oppos- es the tip of the embolus and a flexibly attached, attenuate median apophysis (MA). The form of the tibia and bulb resembles many ecribellate agelenids, e.g., Calilena (Chamberlin and Ivie 1941) from California. The epigynum has teeth posteriad of the copulatory openings and the entelegyne vulva is simple. Amaurobiidae Thorell, 1870 Amaurobiids are a large, worldwide family of 68 genera and 626 described species (Platnick 2004) including cribellates and ecribellates. Our exemplars are the cribellate Amaurobius fenes- tralis from Denmark, Callobius bennetti and Callobius pictus from the USA, Pimm pitus from the USA and a Retiro sp. from Peru. We also included an ecribellate representative, Macrobunus cf. multidentatus, from Chile. Recently Raven and Stumkat (2003) transferred Poaka from the 10 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Psechridae into the Amaurobiidae. We discuss Poaka under the Psechridae (below) and note that our analysis suggests that placement of that genus in either the Amaurobiidae or Psechridae is prob- lematic. We have observed Callobius and Pimus in California, both in the field and in captivity, and have observed Amaurobius in California and Denmark. These spiders make irregular webs of cribellate silk that radiate out from a retreat, usually in a cavity (Figs. 206D, G). The cribellate silk carding leg is braced with a mobile leg IV. The spider moves on the web. We never saw amauro- biids wrap their prey. We have not observed living Retiro. Macrobunus cf. multidentatus was found in loose silken cells under logs. Our amaurobiid exemplars have two nearly straight rows of eyes (Fig. 206B) with canoe-shaped tapeta. Tarsal organs are capsulate (Figs. 15 ID, 153I-K) and there is a single row of tarsal trichobothria that increase in length distally. Trichobothrial bases are trans- versely ridged in Amaurobius, Callobius and Pimus (Figs. 147B, 156D), but smooth to longitudi- nally striate in Macrobunus and Retiro (Fig. 151C, 156E). There are three tarsal claws without tenent or accessory claw setae (Fig. 136G, 138E). The chilum is median (but absent in Macrobunus), and the chelicerae have teeth, thickened setae near the base of the fang furrow, and a large boss. In our cribellate representatives the cribellum is divided into two fields of strobilate spigots (Figs 8 8A, 96A-B) and the spinneret cuticle is ridged but typical cribellate macrobunines have an entire cribellum. Male pedipalpal tibiae have multiple processes, with at least a retrolater- al (RTA), dorsal (OTA) and prolateral (PTA) tibial apophysis. The palpal bulb has a tegular apoph- ysis (TA) in addition to the hyaline C and flexibly attached MA, which we code as a sclerotized tegular process (STP), which in most species is broad and blunt. Female genitalia are entelegyne. Details of individual genera, where they differ from this general outline, are discussed below. Amaurobius fenestralis has apical preening combs of several setae on metatarsi III and IV of both sexes. The respiratory system comprises four simple tracheal tubes (Lamy 1902; Wang 2000). We have examined the spinnerets of a female in detail (Fig. 88A). The ALS (Fig. 88B) has a wide, bare margin with two MAP clustered mesally and a piriform field of more than 30 spigots with rounded base margins and interspersed with tartipores. The PMS has only two AC spigots, a medi- an mAP, one posteromedian CY and two anterior strobilate PC spigots (Fig. 88C). The PLS has an apical MS flanked by two PC. One PC shares a common base with the MS (Fig. 88D). There are ten AC spigots interspersed by tartipores and two CY, one subapical and one subbasal. Wang's (2000) observations agree with ours except that his female specimen had only a single AC spigot on the PMS. Males have epiandrous spigots in two bunches. The cribellate silk of a Californian species of Amaurobius has axial fibers and reserve warp (Fig. 122C). The male pedipalpal tibia has four processes: RTA, OTA PTA and VTA. A broad TA extends retrolaterad obscuring the base of the concave MA. The margin of the tegulum has a lobe but there is no corresponding lobe on the subtegulum. The epigynum has small teeth posteriad of the copulatory openings near the epigas- tric furrow. Callobius (Fig. 206B), like Amaurobius, has apical preening combs of several setae on metatarsi III and IV of both sexes and a respiratory system of four simple tubes. We have exam- ined the spinnerets of a female in detail (Fig. 89A). The ALS has a wide, bare margin with two mesally clustered MAP, a large tartipore mesad of these and a piriform field of more than 30 spig- ots with rounded base margins and interspersed with tartipores (Fig. 89B). The PMS (Fig. 89C) has several median to posterior AC spigots, a median mAP, and a retrolateral and a posterior CY. There are three anterior PC spigots with strobilate shafts. The PC spigots arise from single bases, but one Callobius individual examined had a pair of shafts arising from a common PC base (Fig. 96C). Abnormal, duplicate shafts are occasionally found in spiders (e.g., Figs. 107C, 111A), hence we have coded Callobius PMS PC as arising from a single base. The PLS has an apical MS flanked GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 11 by two separate PC. There are more than 20 AC spigots interspersed by tartipores and two CY, one subapical and one subbasal (Fig. 89D). Wang's (2000) observations are similar to ours. Males have epiandrous spigots in two bunches. Cribellate silk has axial fibers and reserve warp (Figs 123A-B). The male pedipalpal tibia has four processes: RTA, DTA, PTA and VTA. The VTA and PTA are short and conical (Figs. 182A, 193A), the RTA is a backward curving hook, and the DTA is bifid with a long, slender process and a short, conical one (Fig. 181C). The MA is convex and hatchet- shaped and the TA is hemispherical (Figs. 182A-B). Unlike Amaurobius the epigynum lacks teeth. Pimus has a prolateral series of thorn-like setae on the male palpal femur (Fig. 147C). Preening combs are lacking. The respiratory system comprises four simple tracheal tubes. We examined the spinnerets of a male and female (Figs. 90A, 91A). The ALS has a narrow, bare margin (Fig. 90B). The female ALS has two mesally clustered MAP and a nearby tartipore (Figs. 90B, 96F). The male (Fig. 91B) has one functional MAP plus nearby nubbin and tartipore. The piriform gland spigots have rounded bases and are interspersed with tartipores. The female PMS has two mAP accompa- nied by a TP, a single anterior PC, several median AC spigots, and a single posterior CY (Figs. 90C, 96G). The male PMS has a nubbin in place of the PC, and lacks the CY (Fig. 91C). Wang (2000) recorded only one mAP in Pimus, but his specimen was somewhat damaged and difficult to inter- pret. The female PLS has an apical MS flanked by two separate PC (Figs. 90D, 96D, H). There are about 15 AC spigots interspersed by tartipores and two retrolateral CY (Figs. 90D, 96H). The male retains only the AC spigots. The CY are absent and the MS and PC are replaced by apical nubbins (Fig. 91D). Males have epiandrous spigots dispersed along the epigastric furrow (Fig. 161E). The cribellate silk has axial fibers and reserve warp. Reserve warp fibers are visible in Figures 123C-D; the axial fibers are hidden, but present. The male pedipalpal tibia has long, slender and pointed DTA and RTA, and a short conical PTA (Fig. 18ID). The embolus is a slender, curved spine, the C hyaline, and the TA is flattened and arched, covering the base of the MA (Fig. 182C). The epigy- num lacks teeth. Retiro differs in many features from the other amaurobiid exemplars, especially in the tracheal system, epiandrous region, and trichobothrial bases. The tracheal system comprises two thick median trunks that give rise to bunches of fine lateral tracheoles in the abdomen, and penetrate the cephalothorax where they give rise to two lateral and a terminal bunch of fine tracheoles. The lat- eral tracheae comprise a fine, simple tube on each side, which is confined to the abdomen. Male epiandrous spigots are absent. The trichobothrial bases are smooth to longitudinally striate (Fig. 156E), the cheliceral boss is small, and preening combs are lacking. We have examined the spin- nerets of a female (Fig. 92A) and male (Fig. 93A). The female ALS (Fig. 92B) has a narrow field margin and a single MAP and posterior nubbin at the edge of the spinning field. A second appar- ent nubbin mesad of these appears to be a PI spigot with a broken shaft. The piriform spigots have rounded base margins. The male ALS is the same (Fig. 93B). The female PMS has five AC spig- ots, an anterior mAP and two lateral CY (Fig. 92C). Unlike other amaurobiids, PMS PC are absent. The male PMS has the mAP and five AC but lacks the CY spigots (Fig. 93C). The female PLS (Fig. 92D) has several AC spigots, two median CY and an apical MS but no PC (Fig. 96E). The male PLS lacks the CY and the MS is replaced by an apical nubbin (Fig. 93D). The male pedipalpus has a forward-curving RTA and an elongate, transverse VTA (Figs. 182D-E), a short, stout PTA and a short, curved, bifid DTA (Fig. 18IB). The bulb has a broad, trifid TA that obscures the C base and a slender, apically hooked MA (Figs. 181 A, 182D-E). The epigynum lacks teeth posteriad of the copulatory openings but has small pockets. Macrobunus has simple lateral tracheae, and slightly branched median tracheae, limited to the abdomen. Male epiandrous spigots are absent (Fig. 158H). The chelicerae have a very long fang, three promarginal teeth, and a long retromarginal series of 8 teeth plus 6-10 tightly packed denti- 12 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II cles. The tarsal organ has an oval orifice (Fig. 15 ID), the proximal hood of the trichobothrial base is longitudinally striate (Fig. 151C), and the cuticle is smooth with some ridged areas. The male femur II has a ventral, basal, conical process. We present scans of male and female spinning organs (Figs. 94-95). The colulus is broad and covered by setae (Figs. 94A-B, 95A). The ALS has about 27 PI spigots with rounded bases and thin shafts; the female has two MAP spigots with a tartipore (Fig. 94C), while the male has the posterior MAP replaced by a nubbin (Fig. 95B). The PMS has one mAP spigot without traces of nubbin or tartipore, and many AC spigots with long, thin shafts (Figs. 94D, 95C); in addition to these, the female has two larger CY spigots on the mesal and pos- terior margins. The PLS has several AC and no traces of MS or nubbins, and the female has one CY spigot on the anterior external margin (Figs. 94E, 95D). The male pedipalpus has a complex RTA with several processes and ridges (Figs. 183A, 193B-C), a simple OTA, and a small, round- ed VTA. The OTA has a membranous area on its base. One of the processes of the RTA is flat, translucent, closely apressed to a concave area of the cymbium (Fig. 183A); on closer inspection this concave area has regularly disposed ridges, apparently a stridulatory apparatus (Fig. 183B). Similar stridulatory fields were found in other macrobunine genera as well (Anisacate, Ramirez pers. obs.; Rubrius and Emmenomma, Fig. 183C-D). The bulb (Fig. 193B-C) has an apical hya- line C, a large, articulate sclerite that we identify as a MA, and a small sclerotized plate surround- ed by a membranous area that we identify as a TA. The embolus is fused with the tegulum and has a complex basal process; the embolar base is projected in a lobe, which has no corresponding lobe in the subtegulum. The epigynum lacks teeth. Amphinectidae Forster and Wilton, 1973 Amphinectidae comprise 34 genera and 181 described species from Australia, New Zealand and South America (Platnick 2004). At least Metaltella simoni is introduced to the USA. There are cribellate and ecribellate amphinectids, and there are several cases in which cribellates and ecribel- lates are closely related, e.g., Maniho and Amphinecta (Forster and Wilton 1973). Cribellate amphinectids build small space webs for prey capture, whereas ecribellate amphinectids are wan- dering hunters. Our amphinectid exemplars are Maniho ngaitahu from New Zealand and Metaltella simoni (Fig. 206A) from California (this species also occurs in the southern USA and in South America). We have also examined Metaltella rorulenta from Chile. We have observed Maniho species in the wild in New Zealand and Metaltella simoni in captivity. Maniho occured beneath logs and stones. Metaltella simoni built irregular space webs in captivity and spent most of the time hanging upside down in these webs. The cribellate silk carding leg is braced with a mobile leg IV. We never observed prey wrapping. Metaltella cribellate silk has reserve warp and axial fibers (Figs. 124A-C). Our amphinectid exemplars have two nearly straight rows of eyes with canoe-shaped tapeta. The chelicerae have teeth, thickened setae near the base of the fang furrow, and a large boss. Tarsal organs are capsulate (Figs. 147D, 153E) and there is a single row of tarsal trichbothria with smooth bases (Figs. 1551, 156A). There are three claws and preening combs occur at the apices of metatar- si III and IV, but serrate accessory setae and scopulae are lacking. Male epiandrous spigots are absent (Fig. 160D). The cribellum is divided into two fields of strobilate spigots and the spinneret cuticle is ridged. Female genitalia are entelegyne, the epigynum has lateral teeth (Fig. 180D) and the vulva is complex with convoluted ducts (Fig. 164F). Male pedipalpal tibiae have a simple, api- cal RTA and a proximal DTA (Fig. 180C, 189C). The cymbium lacks chemosensory scopulae but has trichobothria. The palpal bulb has a sclerotized C, an MA and an additional TA that arises near GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 13 the embolic base, but details differ dramatically between Maniho (Figs. 180A-B, 189C) and Metaltella (Fig. 179D). Amphinectids are heterogeneous in several other features. Maniho has deeply notched trochanters, whereas those of Metaltella are unnotched. Tarsal trichobothria form a single row and are irregular in length in Maniho but increase in length distally in Metaltella. Maniho has feathery scales (Fig. 147D) but Metaltella lacks them. The chilum is median in Metaltella but bilateral in Maniho. We examined the spinning organs of both sexes of Maniho (Figs. 75A, 76A). The ALS has a wide bare margin. The female ALS has two MAP clustered at the mesal margin and a piriform field of more than 30 spigots with rounded base margins and interspersed with tartipores (Fig. 75B). The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. 76B). The female PMS (Fig. 75C) has several AC spigots, one anterior mAP, and a median and lateral CY. There is a median row of 10-11 PC with strobilate shafts. Most PC have multiple shafts emerging from a common base (Fig. 82D). The male PMS lacks the CY, and the PC are replaced by a median row of nubbins (Fig. 76C). The female PLS (Fig. 75D) has two basal CY, more than 20 AC, and an api- cal MS flanked by two PC (Fig. 82E). The male PLS lacks the CY and has the MS and PC replaced by nubbins (Fig. 76D). The male pedipalpus (Figs. 180A-C, 189C) has a trapezoidal tegulum, a transverse, curved MA, and a ribbon-like embolus that spirals to encircle the tegulum. The large C opposes the embolus tip. Near the embolic base is a short, semicircular TA (Figs. 180B, 189C). Metaltella spinnerets resemble those of Maniho (Figs. 77A, 78A). The ALS has a wide bare margin. The female ALS has two MAP clustered at the mesal margin and a piriform field of more than 30 spigots with rounded base margins and interspersed with tartipores (Fig. 77B); the male posterior MAP is replaced by a nubbin (Fig. 78B). The female PMS (Fig. 77C) has one large ante- rior mAP, two posterior CY and several AC. There is a median group of 10-11 PC with strobilate shafts. Most PC have single shafts but a few have multiple shafts emerging from a common base. The PC are replaced by a median group of nubbins in the male (Fig. 78C). The female PLS (Fig. 77D) has a basal CY, more than 20 AC, and an apical MS flanked by two PC (Fig. 82C). The MS and PC are replaced by nubbins in the male (Fig. 78D). The male palpal bulb appears relatively simple (Fig. 179D), but has an astonishing internal complexity (Figs. 191, 192). The MA is elon- gate, with a concavity facing retrolaterally (identified as primary conductor by Davies 1998). The broad shaft of the C (identified as secondary conductor by Davies 1998) arises retrolaterally on the tegulum and is inrolled to form a near cylinder that contains the threadlike embolus; the hypertro- phied C base forms most of the visible part of the tegulum. We have distinguished the approximate limit between tegulum and conductor by the origin of the embolus, and by a furrow that seems to mark the suture line between them (Fig. 192B-C, tegulum grayed). The origin of the embolus is totally concealed by the C (E* in Fig. 192A); after digestion of tissues, the long embolus can be seen describing several internal loops before emerging from a slit in the C shaft; a dissection of the tegulum shows that the embolus runs through convoluted cuticular foldings of the C (C* in Fig. 192A). The embolar base has a sclerotized process that arises apically from the tegulum (E** in Fig. 192A); there is no subtegular lobe opposing this embolar lobe. The female vulva has long, con- voluted copulatory ducts (cf. Fig. 164F), an indication of a correspondingly long intromittent embolus. This suggests that a significant part of the embolus comes out through the conductor slit during mating. A high hydrostatic pressure in the bulb may produce the membranous internal fold- ings of the conductor to push the embolus out of the copulatory bulb. This unique conductor form is characteristic of metaltellines (Davies 1998); a similar disposition of the embolus origin internal to the conductor is reported here also for Desis, although not as dramatically developed as in met- altellines. 14 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Araneoidea Simon, 1895 This wordwide superfamily comprises the ecribellate orb weavers and their kin, including the families Anapidae, Araneidae, Cyatholipidae, Linyphiidae, Mysmenidae, Nesticidae, Pimoidae, Symphytognathidae, Synaphridae, Synotaxidae, Tetragnathidae, Theridiidae, and Theridiosomat- idae, and includes at least 11022 described species (Platnick 2004). This is a large and important taxon comprising 29% of described spider species and including the second (Linyphiidae) and third (Araneidae) largest families. Griswold, Coddington, Hormiga and Scharff (1998, Fig. 211) pro- posed a comprehensive phylogeny for the superfamily. Recent work by Karin Schtitt (Schiitt 2000, 2002, Fig. 210, 2003) suggests that the minute Micropholcommatidae, previously included in Palpimanoidea, may belong in Symphytognathoidea, a relatively derived group within Araneoidea. Araneus (Araneidae), our selected exemplar taxon, is, for the characters treated herein, a good rep- resentative of the groundplan of Araneoidea, as optimized at the base of the superfamily. We also illustrate some characters as seen in other araneoid taxa. In Araneus the eyes have a canoe-shaped tapetum and the chelicerae have a small boss and teeth on the fang furrow but lack thickened setae near the fang base. The chilum is bilateral. The legs lack tarsal trichobothria and have only one subapical metatarsal trichobothrium, trichobothri- al bases are smooth and the capsulate tarsal organ has a round orifice (Fig. 149A), and there are three claws with conspicuous serrate accessory setae (sometimes called "false claws") associated (Fig. 137C, arrow). Hairs are serrate (Figs. 148F, 149A). Preening combs, scopulae and claw tufts are absent. The posterior respiratory system comprises four simple tracheal tubes. Araneoids are ecribellate, and lack characters associated with the cribellum, calamistrum and cribellate silk and silk spinning. The ALS has a narrow, bare margin surrounding the spinning field. The female ALS has a single MAP accompanied by a nubbin and a tartipore, all at the edge of the spinning field, separated by a deep furrow from the PI field (Figs. 20E-F). The piriform spigots have rounded bases and their field is interspersed with tartipores. Some derived Araneoidea have lost their PI spigot bases (Fig. 20F) but round bases optimize as the groundplan state. The male's ALS is iden- tical to the female. The female PMS has a single posterior mAP and associated nubbin and tarti- pore, several AC spigots and 1-2 CY. The female PLS has a peripheral triplet of a flagelliform gland (FL) and two aggregate gland (AC) spigots (Figs. 38C, E) responsible for the gluey silk cap- ture line (Fig. 119C-E). There are also numerous AC spigots and two CY on the PLS. Appendage cuticle is squamate (Fig. 149A). Males have epiandrous spigots evenly distributed along the epi- gastric furrow. The male pedipalpal tibia has simple, rounded ventral processes and the cymbium has a paracymbium (Fig. 17IF) but lacks trichobothria or chemosensory scopulae. The male pal- pal tarsus is rotated so that the cymbium is prolateral and the bulb retrolateral. The bulb has a scle- rotized C of the "uloborid" type (see character 118 state 3 below) and a flexibly attached, convex MA (Fig. 17IE). The entelegyne female genitalia comprise an epigynum that lacks teeth and has a simple vulva. Araneoids are orb builders and all behaviors that build the orb web apply, i.e., con- struction of a frame, radius, hub, temporary spiral and sticky spiral. Araneoids hang beneath the web and wrap their prey before biting. Archaeidae C.L. Koch and Berendt, 1854 Archaeidae comprise three genera and 25 species from Africa, Madagascar, and Australia (Platnick 2004). Our exemplar is Archaea workmani from Madagascar. Archaea make no webs for prey capture but use silk for drag lines and to wrap their eggs. They prey upon other spiders (Fig. 195D). The eight eyes are in two nearly straight rows and have a canoe-shaped tapetum. All but the GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 15 AME are very small. The pars cephalica is extremely prolonged into a "neck" and sclerotized com- pletely around the base of the elongate chelicerae (Figs. 127A, 195D). The chelicerae lack a boss or stout setae near the fang base but have peg teeth and a few true teeth (Fig. 127B) and a patch of stridulatory striae on the outer margin (Fig. 127D). The cheliceral gland opens on a mound (Fig. 127C). The labrum has lateral extensions (Fig. 127E). Two faint sclerotizations above the chelicer- al base suggest a bilateral chilum. The elongate legs are spineless, all hairs are plumose and the cuticle is squamate (Figs. 134D, 149C). Tarsal trichobothria are absent and there is only a single subapical trichobothrium on the metatarsus. The trichobothrial base hood is transversely ridged (Fig. 149H) and the capsulate tarsal organ has round orifice (Fig. 149C). There are three claws but claw tufts, preening combs and scopulae are absent. The setae below the claws resemble serrate accessory setae (Fig. 134E-F). Forster and Platnick (1984) and Platnick et al. (1991) did not record serrate accessory setae in Archaea, but we code them as present. The median claw is elongated, remarkably similar to that of symphytognathoid araneoids (Griswold et al. 1998: character 63). A pair of posterior spiracles (Fig. 127F) each leads to a single tracheal tube (Forster and Platnick 1984, fig. 305); lateral tracheae are absent. The male lacks epiandrous spigots. The spinnerets of Archaea were described by Platnick et al. (1991, figs. 228-233). We examined male, female and immature Archaea workmani from Madagascar (Figs. 20A-D, 21-22). Archaea are ecribellate. The ALS of both sexes have a wide spinning field margin, one MAP spigot plus a tartipore and a reduced MAP, all on the mesal margin, and a field of more than 25 PI spigots with short, sharp bases and interspersed with tartipores (Figs. 21B, 22B). The immature has two normally developed MAP, suggesting that the posterior reduced MAP in adults is homologous with the nubbin found in other spiders (Fig. 20B). The MAP field is separated from the PI field by a deep furrow (Figs. 20B, 22B). The female PMS has an anteromedian mAP, a median AC, and a lateral and posterior CY (Fig. 21C). The male retains only the AC and mAP (Fig. 22C), but the immature seems to have two mAP and two AC (Fig. 20C). The female PLS has a median row of five AC spigots and three large CY spigots in the anterior, mesal and posterior positions (Fig. 21D). The male retains only the five AC (Fig. 22D), while the immature has only two AC (Fig. 20D). The male palpus lacks tibial processes, and the cymbium lacks trichobothria or chemosensory scopulae (Figs. 168A-C). The apex of the Archaea tegulum has two sclerotized ridges that spiral around a central pit that con- tains the E and MA (Fig. 168D). We code these ridges as an apical C. The haplogyne female gen- italia lack an epigynum (Fig. 165); there is a large, membranous median seminal receptacle with patches of gland ductules on its dorsal side. Internally, the epigastric fold bears two strong apodemes for muscle insertion (Fig. 165 A-B). Austrochilidae Zapfe, 1955 Austrochilidae comprise three genera: the monotypic Hickmania from Tasmania, and Austrochilus and Thaida with eight described species between them from Argentina and Chile (Platnick 2004). The family was established by Gertsch and Zapfe (in Zapfe 1955) and considered a senior synonym of Thaididae and Hickmaniidae by Forster et al. (1987:25), contra Lehtinen (1967:299) and Marples (1968:30). Our exemplars are Thaida peculiaris from Chile and Hickmania troglodytes from Tasmania. Austrochilids make and hang beneath extensive sheet webs (Figs. 198A-F) (Thaida: Forster et al. 1987: fig. 118; Hickmania: Morrison and Morrison 1990:148). Lopardo, Ramirez, Grismado and Compagnucci (2004) report that in Thaida peculiaris and Austrochilus forsteri the cribellate silk carding leg is braced with a mobile leg IV (an "advanced" entelegyne behavior) (Fig. 198B) and that they occasionally wrap prey after biting (Fig. 198F). We examined the cribellate silk of Hickmania and Thaida. In both cases the cribellate mass is puffed, and the fibrils have the regular- 16 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II ly spaced nodules reported by Eberhard and Pereira (1993) for entelegyne spiders (Figs. 118A-D, 119F). Carlson (in lit.) also examined the fine structure of Hickmania cribellate silk, which has axial fibers but lacks reserve warp (Figs. 120A-C). Although there are wavy fibers in the cribellate mass (Fig. 120C), these lack the characteristic spiral of reserve warp and are probably axial fibers. She also noted nodules on the cribellar fibers. Austrochilids have primitive tapeta and chelicerae without a boss but with teeth and thickened setae along the fang furrow. The chilum is absent at least in Thaida. The clypeal hood makes exam- ination of the chilum difficult. Tarsal trichobothria are lacking and there is only a single, subapical trichobothrium on the metatarsus: the base is smooth and has a distal notch. The tarsal organ is exposed (Figs. 148D, 150A, 152A). There are three claws and serrate accessory setae (Figs. 133A-D); scopulae and claw tufts are lacking. The palpal femora of Thaida have probasal thick- ened setae modified as thorns; such thorns are lacking in Hickmania. Thaida has some feathery scales (Fig. 148D) whereas those of Hickmania are only plumose. Males have numerous epiandrous spigots in two bunches (Fig. 158A). The respiratory systems of austrochilids vary. Hickmania has four booklungs. Thaida has a wide posterior spiracle that leads to modified organs that resemble vestigial booklungs in the hatching immatures, which become elongate and similar to tracheae through successive stages (Forster and Platnick 1984; Ramirez 2000). We reflect this ambiguity as a polymorphic scoring. Hickmania and Thaida have similarly projecting female gen- ital regions that we code as an epigynum, but Thaida has in addition a sclerotized area posterior to the genital opening. The haplogyne female genitalia of Thaida and Hickmania have a genital open- ing anterior to the epigastric fold, which leads to a median sperm receptacle (in Thaida) or to four slender spermathecae (in Hickmania) and also to the uterus extemus (Figs. 163A-B). In Thaida the genital opening is also exposed in the male (Fig. 158A), whereas that of Hickmania is hidden by the epigastric fold. The epigastric fold leads to a blind invagination that serves as a muscle attach- ment. Forster, Platnick and Gray (1987) and Platnick et al. (1991) described the spinning organs of Austrochilus melon, Thaida peculiaris and Hickmania troglodytes, but some of our interpretations here differ from those previously published. We scanned the spinning organs of both sexes of Thaida peculiaris from Argentina and Chile. Like hypochilids they have entire cribella (Fig. 13 A), but in other aspects they resemble entelegynes. The male and female ALS of Thaida have two large MAP at the median edge of the spinning field, flanked by a huge tartipore (Figs. 11B, 12B, 14A). The margin of the spinning field is narrow. There are more than 50 piriform spigots with rounded bases. The piriform spinning field is interspersed with tartipores and is separated from the MAP and large tartipore by a wide, semicircular bare area (Fig. 11B). The female PMS has about 6 AC spigots, a large median mAP accompanied by a tartipore, and 11 large posterior spigots (Fig. 11C) that we code as CY. A row of 10 PC spigots encircles the anterior side of the spinneret. Each PC base gives rise to a single shaft. The PC shafts have numerous closely-spaced annulations (Figs. 11C, 13D). The female PLS is flattened (Fig. 11D), has numerous AC spigots, an anterior-external line plus a basal group of large spigots, and an apical MS spigot flanked by one PC and one very small nubbin (Figs. 13E-F). Close to this group, there is a spigot with a shaft of intermediate mor- phology between AC and PC (Fig. 13E, asterisk), which we code as a second PLS PC. The male PMS retains two PC spigots, but the other PC are replaced by long nubbins, and the large posteri- or spigots are absent (Fig. 12C). The male PLS has only AC spigots, plus the apical nubbins of the MS, its tiny accompanying nubbin, and the nubbin of the PC spigot (Fig. 12D). Because the numer- ous large spigots on the female PMS and PLS are absent in the male, while all the other spigots (or their corresponding nubbins) occur in the same relative position, we identified the large spigots as cylindrical (Figs. 11C-D, 13D-F). These spigots fulfill our ontogenetic and morphological crite- GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 17 ria for CY. This conclusion is novel: previous morphological (Forster et al. 1987) and phylogenet- ic (Platnick et al. 1991, Griswold et al. 1999) studies coded CY spigots as absent in austrochilids. We scanned the spinning organs of both sexes of Hickmania troglodytes from Tasmania. The male and female ALS has two large MAP at the median edge of the spinning field, flanked by a huge tartipore (Figs. 7B, 8B, 10D). The ALS PI spigots gradually increase their size towards the external border (Figs. 7B, 8B, IOC). The PMS PC encircle the spinneret anteriorly (Fig. 7C), and, like Thaida, the shafts have closely-spaced annulations (Fig. 10F). In the male PMS, several PC spigots are replaced by nubbins, but about 15 still retain their shafts (Fig. 8C). There is one medi- an mAP spigot, with a short conical base, flanked by a tartipore (Figs. 7C, 8C, 10E). A second, sim- ilar but more posterior spigot is probably a mAP as well. There are about 20 small AC spigots. A group of large CY spigots with thicker shafts encircle the PMS posteriorly and at the sides. The PLS are very flat, have numerous small AC spigots and an apical MS spigot, but PC are lacking (Figs. 7D, 8D, 9A-D). The female PLS has an anterior line of CY spigots. As in Thaida, we iden- tified these as CY spigots because they are absent in the male. A further class of spigots occurs on the PMS and PLS of both males and females (Figs. 7C, 9A-D). They have large bases, and the shafts are intermediate in size between those of AC and CY. We tentatively identified them as a sec- ond class of AC spigots (marked with '?' on the plates). These large spigots occur on a median and a posterior line on the PLS. The male pedipalpi of austrochilids lack tibial processes and cymbial chemosensory scopulae and trichobothria, but the bulbs are diverse. The bulb of Hickmania is simple, lacking C and MA (Forster et al. 1987, figs. 343-346). Nevertheless, the tegulum and subtegulum are distinguishable, and the bulb is not spindle-shaped, so we do not code it as piriform. Thaida has a complex bulb with several processes (Figs. 166C, 187A). The embolus is a broad, twisted flange that contains the membraneous sperm duct. The MA is a slender spine that arises from soft cuticle and the C is scle- rotized and has a serrate trip. The embolus and C, the latter with a serrate apex, are both elongate cones that are in close association in the unexpanded bulb (Fig. 166C) but arise far from each other in the expanded bulb (Fig. 187A). The subtegulum has a sclerotized hook, noted with an arrow in Figure 166C. Our interpretation of bulb processes differs from the interpretation of Forster et al. (1987), who considered the slender MA spine to be the embolus and the broad embolic flange to beaC. Ctenidae Keyserling, 1877 Ctenidae comprise a large, worldwide family of 39 genera and 450 described species (Platnick 2004). Most are ecribellate, but Acanthoctenus and three other genera retain the cribellum. Ctenids, or "tropical wolf spiders," are fast moving, running hunters that make little use of silk. Traditional synapomorphies for this family are the 2-4-2 eye pattern and claw tufts. Although the monophyly of Ctenidae including Acanthoctenus is dubious (e.g, Griswold 1993, Fig. 213; Silva Davila 2003, Fig. 214), we accept the current broad limits of the family and choose a few species of Acanthoctenus (Fig. 208C) as our exemplar. The eyes have a grate-shaped tapetum and the chelicerae have a large boss, teeth on the fang furrow and thickened setae near the fang base. The chilum is bipartite. The legs are very spinose (more than 7 pairs of ventral spines on the first tibia), hence Acanthoctenus. The capsulate tarsal organ has an oval orifice (Fig. 153M) and the trichobothrial bases have transverse ridges (Figs. 148C, 156H). Two to three dorsal rows of trichobothria occur on the tarsi. The short, basal calamistrum is oval (Figs. 1451?J). The ITC is absent and there are well-developed claw tufts (Fig. 139E), and at least the posterior tarsi of females have scopulae. The respiratory system consists of 4 simple tracheal tubes. 18 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II We describe the spinning organs of male and female Acanthoctenus specimens from Panama and Peru. The narrow, deep cribellum is divided into two fields of strobilate spigots that are clumped in short, longitudinal linear rows (Figs. 97A, G). The ALS has a narrow, bare margin, a pair of large MAP and numerous PI spigots interspersed with tartipores (Figs. 115B, 116B, 117A). There is a large tartipore close to the MAP (Figs. 115B, 117A). The PMS (Figs. 115D, 117B-C) lacks a paracribellum, and has two mAP spigots with short, stout bases and cylindrical shafts, with a large tartipore in between (Figs. 115D, 116C). Posteriad of these are more than 20 small AC spig- ots with cylindrical bases and slender shafts. The female has in addition three large CY spigots with conical bases and long, conical shafts, interspersed posteriorly among the AC (Figs. 115C-D, 117C). The conical apical segment of the PLS has numerous AC and an apical anterior MS (Figs. 115E, 117D) which is replaced by a nubbin in the male (Fig. 116D). The female has in addition three large CY spigots similar to those on PMS. Males lack epiandrous spigots. The structure of Acanthoctenus cribellate silk is unknown. The male pedipalpus has a simple, triangular, apical RTA and the cymbium has a dorsal chemosensory scopula. The bulb has retrolateral locking lobes on the tegulum and subtegulum and two processes in addition to the embolus: an apical, hyaline C that opposes the tip of the embolus and a flexibly attached, concave MA. The epigynum lacks teeth and vulva is simple. Deinopidae C.L. Koch, 1851 Deinopidae comprise four genera and 57 species, and occur worldwide except in New Zealand (Platnick 2004). All are cribellate. Our deinopid exemplars are Deinopis spinosus from Costa Rica (Figs. 200D-E) and Florida in the USA and Menneus camelus from Kenya and South Africa (Fig. 200C). We have not examined a male of Menneus, so our data come from the only published description (Tullgren 1910), which is not complete. We have observed Deinopis in many parts of the world, especially Florida and Costa Rica, and have observed Menneus in South Africa. The fine structure of the cribellate silk of Menneus was studied by Akerman (1926) and that of Deinopis by Kullman (1975) and Peters (1992a) and sum- marized by Eberhard and Pereira (1993). The puffed cribellate band has both reserve warp and axial fibers, and the cylindrical cribellar fibrils have nodules. Deinopids build a characteristic, highly modified orb web (Figs. 200C-D) (Coddington 1986b) that we term "deinopid web archi- tecture." In addition to the suite of orb web building behaviors, deinopids locate sticky silk with leg IV (SS localization with L4). Like other Orbiculariae, deinopids brace the silk carding leg with a mobile leg IV and wrap their prey after biting it. Deinopid eyes lack a tapetum and the posterior eye row is strongly recurved. At least in Deinopis the PME are greatly enlarged, hence the common name, "ogre-faced" spiders (Fig. 200E). The tarsi lack trichobothria and there is but a single trichobothrium near the apex of the metatar- sus. The trichobothrial bases are smooth to weakly ridged (Figs. 135B, 154E) and the capsulate tarsal organ has a round (Figs. 148A, 152H) orifice. The legs have numerous feathery scales (Figs. 135A, 148A) as well as plumose hairs with short barbs (Fig. 147F, 148A). These latter have been called "pseudoserrate" (Green 1970; Coddington 1986a; Griswold et al. 1998) but differ only slightly from typical plumose hairs. Deinopis have a characteristic line of stout setae on the hind tarsi, the "deinopoid tarsal comb" (Figs. 141B-C). Only a few such setae are found in Menneus (Fig. 1406). There are three claws and serrate accessory setae (Figs. 135C-E) but the legs lack scopulae, preening combs or claw tufts. The large chelicerae have teeth on the fang furrow but lack a thickened seta near the fang base or a boss. We have been unable to find and score the chilum. Lamy (1902) recorded a respiratory system comprising four simple tracheal tubes in Deinopis. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 19 Males have epiandrous spigots arranged in several groups sunk into pits along the epigastric fur- row (Figs. 159C-D). The spinning organs of Deinopis have been discussed previously (Coddington 1989; Peters 1992a). We here illustrate the spinnerets of Menneus as well as some details of Deinopis. In most details Menneus and Deinopis are alike. The spigots have ridged cuticle (Figs. 44D-E) and the cribellum is entire with strobilate spigots (Fig. 43A). The female ALS has five MAP clustered at the inner edge of the spinning field (Fig. 43B), two large tartipores proximad of these (Fig. 44C) and a large, bare semicircular region separating these from the rest of the spinning field. Similar ALS bare regions are found in uloborids (Fig. 45B) and Thaida (Fig. 11B). The piriform spinning field is interspersed with tartipores and comprises more than 50 piriform spigots with flat base mar- gins. The female PMS (Fig. 43C) has numerous anterior to median AC spigots, an anterior mAP, and several posterior and posterobasal CY. Peters (1992a) found two PMS mAP spigots in a Deinopis, but we have found only one in our Deinopis and Menneus specimens. However, one of the mAP found by Peters is much smaller, and our images of Deinopis do not permit certain iden- tification. We scored two mAP in Deinopis, following Peters, but one in Menneus. There are numerous PC bunched along the anterior margin of the PMS, each base supporting a single shaft. The shafts have numerous, closely-spaced annulations, hence the term "deinopoid" PC spigots (e.g, in Deinopis, Fig. 45D) used in previous phylogenetic studies (e.g., Griswold et al. 1999). The con- ical PLS apical segment (Fig. 43D) has several basal CY, several median to apical AC, and an api- cal MS segregated from the spinning field (Fig. 44D). Both AC and CY spigots have columnar bases with flat margins, but the AC shafts are short cones whereas the CY shafts are long cylinders (Fig. 44E). The male pedipalpus of Deinopis lacks tibial processes, and the cymbium lacks trichobothria or chemosensory scopulae. The bulb has a central lobate C with the embolus spiralling around it, and there is no MA (Fig. 171D). The published figure of the male palpus of Menneus (Tullgren 1910: fig. 2) shows a similar organ. The entelegyne female genitalia comprise an epigynum that lacks teeth posteriad of the copulatory openings. Desidae Pocock, 1895 Desidae comprise 38 genera and 180 described species (Platnick 2004), mostly occurring in Australia, New Zealand, New Caledonia and southeast Asia. Porteria occurs in Chile, and the intertidal genera Desis and Paratheuma occur on Indopacific coastlines. At least Phryganoporus candidus may be subsocial. Badumna longinqua is a synanthropic invader in California, New Zealand and Uruguay. Desidae are a heterogeneous family with cribellate and ecribellate genera, the former occurring primarily in Australia and New Zealand. Our desid exemplars are the cribellate Phryganoporus candidus from Australia, Badumna longinqua (Figs. 205B-C) from California, New Zealand and Uruguay (this species is probably native to Australia), Matachia australis and M. marplesi from New Zealand (Figs. 205D-G), and the ecribellate Desis formidabilis (Fig. 205 A) from South Africa. Our desid exemplars are alike in having eight eyes with canoe-shaped tapeta in two nearly straight rows, plumose setae, capsulate tarsal organs (Figs. 151H, 153G-H), and three tarsal claws but no claw tufts, serrate accessory hairs or scopulae (Forster 1970 reported scopulae in Desis, but they have only a dense cover of plumose setae). The chelicerae have teeth and a large basal boss, epiandrous spigots are lacking, and the spinnerets have ridged or smooth cuticle, tartipores, and when present, strobilate PC spigot shafts. The respiratory systems have highly branched median and lateral tracheae (Forster 1970: Matachia and Badumna longinqua; Gray 1983: Phryganoporus [as Badumna Candida]). The male palpus has a MA and a typical C that arises laterally on the bulb near the embolic base and cradles the embo- 20 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II lus in a groove that extends to the embolic apex and embraces half to all of the embolus (Figs. 178A, 189D), totally hiding the embolus in Desis (Fig. 190). The female genitalia are entelegyne and the epigynum has teeth. The fine structure of the cribellate silk has been studied by Eberhard and Pereira (1993) for Paramatachia and Phryganoporus (as Badumna Candida) and by Carlson {in lit.) for Badumna longinqua. We use Eberhard and Pereira's data from Paramatachia to code Matachia in our matrix. The Paramatachia cribellate band is puffed and the fine structure lacks both axial lines and reserve warp whereas the band of Badumna lacks puffs but has both reserve warp and axial fibers (Fig. 122D). We have observed Badumna longinqua in the field and lab. This species builds webs (Fig. 205C) with sheets of cribellate silk radiating out from a central, funnel-like retreat. The spider walks on these sheets, braces the cribellate silk carding leg with a mobile leg IV (Fig. Fig. 205B), and has not been observed to wrap prey. Matachia builds and walks on a similar web (Figs. 205D-G), but other details of its behavior are unknown. Desis is an intertidal spider (Fig. 205A) that actively hunts its prey (Robson 1878; Lamoral 1968). Badumna and Phryganoporus are alike in numerous details, but differ in numerous details from Matachia and Desis. Badumna, Phryganoporus and Matachia have one row of tarsal tri- chobothria, while Desis has two; in Badumna and Phryganoporus the tarsal trichobothria are of irregular lengths whereas in Matachia and Desis they increase in length distally; Badumna, Phryganoporus and Matachia have smooth trichobothrial bases (Figs. 156B-C), but those of Desis have transverse ridges (Fig. 151G); Desis has reduced leg spination, elongated trochanters, and characteristically pointed maxillae (Forster 1970: fig. 35); Matachia and Desis have smooth tarsal cuticle, whereas that of Badumna and Phryganoporus is ridged (the tibial cuticle in Desis is squa- mate); Badumna, Phryganoporus and Desis have unnotched trochanters whereas those of Matachia are notched; Badumna and Phryganoporus have a bilateral chilum whereas those of Matachia and Desis are median; Badumna, Phryganoporus and Desis lack preening combs on the hind metatarsi but these occur in Matachia; Badumna and Phryganoporus cribella are divided but that of Matachia is entire; Badumna and Phryganoporus PMS PC are median and have several shafts arising from a common base whereas those of Matachia are anterior and single; and Badumna and Phryganoporus have simple vulvae whereas Matachia and Desis vulvae have con- voluted ducts. The spinning organs of Phryganoporus candidus comprise a divided cribellum in the female (Fig. 84A) which is still visible as two bare plates in the male (Fig. 85A) and ALS with a wide, bare margin. The female ALS has two MAP clustered at the mesal edge of the spinning field plus a large tartipore, and a field of more than 25 piriform spigots with flat base margins (Fig. 84B). Males have the posterior MAP replaced by a nubbin (Fig. 85B). The female PMS has only two identifiable AC spigots at the anterior margin of the spinneret, an anteromedian mAP and three pos- terior CY. The PC arise in the middle of the spinning field: 7 to 15 shafts arise from each of two large PC spigot bases (Fig. 84C). The male PMS lacks the CY, and three posteromedian nubbins replace the PC (Fig. 85C). The female PLS (Fig. 84D) has at least one basal CY, several AC, and the apex has an MS and two PC, one of which shares a common base with the MS (Fig. 87B). The apex of the male PLS has a large and a small nubbin, the former probably representing the fused MS and PC of the female (Fig. 85D). Badumna longinqua spinnerets (Figs. 86A-D, 87C-D, F) are like those of Phryganoporus candidus in most details except that the female ALS piriform spigots have their bases more round- ed (Fig. 86B), the PMS has at least six AC and seven CY spigots and the numerous PC spigot shafts arise from five thick common bases (Fig. 86C). There is a huge anteromedian mAP, visible in Figure 87C but hidden in Figure 86C. The PLS has at least four CY spigots (Fig. 86D). GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 21 We have studied only the female spinnerets of Matachia. Unlike Badumna and Phryganoporus the cribellum is entire (Figs. 83A, 87E), but like these genera the ALS has a wide, bare margin and two MAP clustered at the mesal edge of the spinning field but the piriform spigots have rounded base margins (Fig. 83B). The PMS has a median mAP and more than 20 posterior spigots with long, cylindrical bases and short conical shafts (Fig. 83C). As these are all similar to one another we cannot distinguish AC from CY and so code these all as AC spigots. Numerous PC spigots, each with a shaft arising from a single base, are bunched at the anterior margin of the spinneret (Fig. 83C). As on the PMS, the PLS has numerous spigots with long, cylindrical bases and slender con- ical shafts (Fig. 83D); again, these are all coded as AC. The PLS apex has an MS and an single PC (Fig. 87A). We present scans of male and female spinning organs of Desis formidabilis (Figs. 79-81). The colulus is broad and covered by setae (Figs. 79A, 80A). The ALS is densely covered by setae, which must be removed to expose the spigots. On the prolateral external side of the terminal arti- cle there is a field of short, curved setae with bare smooth tips (Fig. 79B). There is one MAP spig- ot with one tartipore and no nubbin; the placement of the MAP is unusual, on the anterior margin (Figs. 79B, 80B). There are many PI spigots with long, smooth shafts (Fig. 79B). At the side of the MAP there is a subtle mound that could be a vestige of a nubbin of a posterior MAP (Figs. 81C-D). The PMS are very large, and together with the PLS, have a dense cover of small AC spigots, as occur in other intertidal spiders that build dense, waterproof silken retreats (for example, the cybaeid Argyroneta aquatica and the anyphaenid Amaurobioides; Ramirez 2003:53). There is one mAP spigot with a short base, without traces of nubbin or tartipore (Figs. 79C, 80C, 8 IE), and the female also has three posterior CY spigots. The PLS (Figs. 79D, 8IF) has one median spigot sim- ilar to the mAP, that we identify as a MS, and the female also has two CY spigots on its prolateral mesal margin (Figs. 79D, 8IF). The male pedipalpal tibiae vary among desids. Matachia and Desis have a bifid RTA, formed by a wide more ventral blade and a dorsal spine (Fig. 177E; Forster 1970: fig. 52), Phryganoporus has a simple process (Fig. 189D), and Badumna has several processes (Figs. 178B-C). Phryganoporus also has a basal OTA (Fig. 189D). Our desid representatives have the characteris- tic "desid-amphinectid" C and a MA. The latter is concave in Badumna and Phryganoporus (Figs. 178A, 189D) and Desis (Fig. 177B-D) but hooked in Matachia (Forster 1970: figs. 52-54). The copulatory bulb of Desis is very particular. The origin of the embolus is completely hid- den among the complex foldings of the C, even in the expanded bulb (Fig. 190), suggesting that the basal part of the C became integrated with the tegulum, embracing the embolus from its origin, as occurs with the metaltellines (Amphinectidae). There is a fleshy membranous tegular sclerite (MTP) in addition to the MA and C (Figs. 177B, 190B). Davies (1998:212, fig. 7) illustrated the expanded bulb, but missed the long loop of the sperm reservoir before the entrance to the embo- lus. Dictynidae O.P.-Cambridge, 1871 Dictynidae is a large, worldwide family of 48 genera and 555 described species (Platnick 2004). It is relatively poor in species in Australia and South America. The family is heterogeneous, containing cribellate and ecribellate forms, and the placement in this family of most of the ecribel- lates is uncertain. We include exemplars from each of the subfamilies that include cribellate species: Cicurininae (Lathys), Dictyninae (Dictyna and Nigma, and some silk data from Mallos), and Tricholathysinae (Tricholathys), which we refer to as "typical dictynids." We also include the enigmatic genus Aebutina, currently placed in Dictynidae. Aebutina differs from typical dictynids in so many ways that we discuss it separately at the end of the family treatment. 22 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Typical dictynids make webs with a central, funnel-like retreat and multiple cribellate sheets in different planes on which they walk (Fig. 200B). In Dictyna the cribellate silk carding leg is braced with a mobile leg IV, and we have never observed them to wrap captured prey. The fine structure of the cribellate silk of Dictna was studied by Eberhard and Pereira (1993) and that of Mallos by Carlson (in lit.). We add the Mallos observations to our matrix to code Nigma. The cylin- drical cribellar fibrils have nodules. The cribellate band is puffed (Fig. 120E) and axial fibers are absent (Fig. 120F) but the reserve warp may be absent (Dictyna) or present (Mallos: Fig. 120F). Common characteristics of our typical dictynid exemplars are the distinctive respiratory sys- tem and male palpal bulb. The dictynid tracheal system comprises thick median tracheal trunks that extend into the cephalothorax and have many fine lateral branches; lateral tracheae are absent (Lamy 1902; Forster 1970; Griswold and Ramirez, pers. obs.). The male palpal bulb lacks a medi- an apophysis and has a characteristic sclerotized conductor that arises laterally on the bulb near the embolic base and cradles the embolus in a groove. The conductor extends distad but the apex turns back proximad, in most species extending past the base of the bulb (Figs. 175A-B, D) to the tibia. The conductor apex of Tricholathys forms a spiral (Figs. 175C-D) and that of Lathys fits through a notch behind the RTA (Fig. 176C). Our dictynid exemplars are also alike in having canoe-shaped tapeta, capsulate tarsal organs (Figs. 152K, 153A-B), smooth trichobothrial bases (Figs. 155B-E), no palpal trichobothria or cymbial chemosensory scopulae (Fig. 175C), few spines on the legs, a linear calamistrum (Figs. 145D-E), only plumose hairs, and three claws but no serrate accessory setae, claw tufts, preening combs or scopulae (Fig. 136F). The chelicerae have teeth (Figs. 131C, E), thickened setae near the fang base and a basal boss (Figs. 129C-D). The chilum is absent or impossible to discern in the specimens that we examined. Male epiandrous spigots are absent (Fig. 160C), the cribellum is entire (though at least some Mallos have divided cribella) with strobilate cribellar spigots (Figs. 59A, 66D-E) and ridged spigot cuticle (Figs. 63C, 65C). The female ALS has only one MAP spigot accompanied by a large tartipore and nubbin (Fig. 59B), which is very small in Lathys (Fig. 63B). All spinning fields are interspersed with tartipores. Female genitalia are entelegyne and the epigyna lack teeth. Below we describe individually the peculiarities of our "typical dictynid" exemplars Dictyna, Nigma, Lathys and Tricholathys, including spinning organs and male genitalia. Dictyna has two nearly straight rows of eyes (Fig. 200A), strongly bowed male chelicerae (Fig. 129D) and legs lacking tarsal trichobothria and having only a single distal metatarsal. We scanned both female (Fig. 59A) and male (Fig. 60A) spinnerets. The female ALS has a single MAP at the mesal edge of the spinning field flanked posteriorly by a nubbin and mesally by a tartipore (Fig. 59B) and a field of eighteen piriform spigots with flat base margins. The male ALS is simi- lar, with several large tartipores in the piriform field (Fig. 60B), more conspicuous in the male. The female PMS has only three or four AC spigots, an anteromedian mAP and a posterior CY. The CY shape is unusual: the base is short and concave and the shaft long and nearly cylindrical. The PC encircle the spinnerets anteriorly and laterally, and the PC spigot bases may give rise to one to many shafts (Fig. 59C). The male PMS lacks the CY, has the median mAP and several AC, and the PC spigots are replaced by an encircling series of nubbins (Fig. 60C). The female PLS (Fig. 59D) has two median CY, again with short, concave bases, several AC, and an MS and PC at the apex (Fig. 66C). The male PLS has only AC spigots, several tartipores, and apical nubbins representing the MS and PC of the female (Fig. 60D). Like Dictyna, Nigma has two nearly straight rows of eyes and strongly bowed male chelicer- ae and the legs lack tarsal trichobothria and have only a single distal metatarsal. We scanned both female (Fig. 61 A) and male (Fig. 62A) spinnerets. Like Dictyna the female ALS has a single MAP at the mesal edge of the spinning field flanked posteriorly by a nubbin and a tartipore (Figs. 6113). GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 23 There are only nine piriform spigots with flat base margins. The male ALS is similar (Fig. 62B). The female PMS (Fig. 61C) has three AC spigots, but the mAP is posterior. Posteromesad of the mAP is a spigot slightly larger than the AC. There is no spigot in the corresponding position in the male, therefore we code the female spigot as a CY. The PC encircle the spinnerets anteriorly and laterally, and the PC spigot bases may give rise to one to many shafts (Fig. 66B). The male PMS has the posterior mAP and three AC, and the PC spigots are replaced by an encircling series of nub- bins (Fig. 62C). The female PLS (Fig. 6ID) has a single basomedian CY, several AC, and the apex has an MS and PC (Fig. 61D). The CY morphology is normal for entelegynes and contrasts with that of Dictyna: the CY base is long and tapers slightly, as does the shaft. The male PLS has only AC spigots and apical nubbins representing the MS and PC of the female (Fig. 62D). Lathys humilis has two nearly straight rows of eyes, whereas L. delicatula has only six eyes (the AME are absent). Lathys has normal male chelicerae (Fig. 129C) and rows of several tri- chobothria on the tarsus (Fig. 147E) and one or two on the metatarsus. We scanned both female (Fig. 63A) and male (Fig. 64A) spinnerets. Like other dictynids the female ALS has a single MAP at the mesal edge of the spinning field and a tartipore, but the posterior MAP nubbin is very small (Fig. 63B). The eight piriform spigots have rounded base margins. The male ALS is similar (Fig. 64B). The female PMS (Fig. 63C) has three AC spigots, a posteromedian mAP, and two posterior CY. The four PC encircle the spinnerets anteriorly, and the PC spigot bases give rise to only one shaft each (Fig. 63C). The male PMS retains the posterior mAP and has three AC, and the PC spig- ots are replaced by an encircling series of four nubbins (Fig. 64C). The female PLS (Fig. 63D) has a mesal CY, five AC, and the apex has an MS but lacks a PC. The CY base and shaft are long and tapering. The male PLS has only four AC spigots and an apical nubbin representing the MS of the female (Fig. 64D). Like Lathys but unlike Dictyna and Nigma, Tricholathys has normal male chelicerae and rows of trichobothria on the tarsus and metatarsus. There are two nearly straight rows of eyes. The tarsal trichobothria increase in length distally. We present a complete description of the female spinnerets (Fig. 65 A). Like other dictynids the ALS has a single MAP at the mesal edge of the spinning field, a large MAP nubbin posteriad of this and a mesal tartipore (Fig. 65B) The eight piriform spigots have rounded base margins. The PMS (Figs. 65C, 66F) has six or seven AC spigots, a median mAP, and three posterior CY. A single, large central PC base gives rise to several strobilate shafts (Fig. 66A), and is replaced by a large nubbin in the male (Fig. 66G). The PLS (Fig. 65D) has only AC and a lateral CY: we are unable to discern MS or PC. Typical dictynid male palps exhibit a characteristic C and lack a MA (Figs. 175A-D), but the tibia is variable. All except Nigma have an RTA, i.e, Tricholathys (Fig. 175C), Lathys (Fig. 176C), and Dictyna (Figs. 176D-E). Nigma has a blade-shaped, apical OTA and a patellar process as well (Fig. 176A), whereas Dictyna has a basal OTA surmounted by two short spines (Figs. 176B, D-E). Aebutina, a monotypic genus known from Ecuador and Brazil, is currently placed in the Dictynidae (Platnick 2004). The placement of Aebutina has long been uncertain. Simon (1892) considered its external morphology intermediate between Uloboms and Dictyna, with the balance of affinity with the former. Petrunkevitch (1928) placed it in the Dictynidae, where it has remained ever since. Millot (1933a) examined the internal anatomy of Aebutina and found well developed, endocephalic venom glands, refuting placement in Uloboridae and supporting Dictynidae. Aviles (1993) has recently described social behavior in A. binotata from Ecuador. This species maintains communal nests and cooperatively captures and feeds on prey. Cribellate silk fine structure, silk carding behavior and prey wrapping are unknown. The spiders spend much of the time beneath leaves, so we code their web posture as inverted. Our exemplars of A. binotata are from Ecuador. Aebutina has eight eyes with canoe-shaped 24 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II tapeta in two nearly straight rows. The faint chilum is median. The chelicerae have a large basal boss and teeth on the fang furrow, but lack the characteristic stout seta at the retromargin of the fang (Fig. 130E). Spines are absent from the legs, all leg setae are plumose, the calamistrum is lin- ear and extends for nearly the whole length of the fourth metatarsus (Fig. 142E), and the leg metatarsi and tarsi have short rows of two trichobothria each. The trichobothrial base hood has transverse ridges (Fig. 150C), and the capsulate tarsal organ has a round opening. There are three claws but preening combs, serrate accessory setae, claw tufts and scopulae are absent (Fig. 142F). Male epiandrous spigots are absent (Fig. 157C). Millot (1933a) found no tracheae, but our prepa- rations revealed four simple tracheae. The median tracheae are flat, wide, until a point where they are abruptly truncated, then extending into a thin tube. The truncate border has the ragged texture typical of muscle attachments. We report on Aebutina spinning organs for the first time. We scanned both female (Figs. 56A-C, 57A) and male (Fig. 58A) spinnerets. The cribellum is divided and covered with evenly spaced strobilate spigots (Fig. 157A) and the spinneret cuticle texture is ridged. The female ALS has a narrow spinning field margin, a pair of MAP at the mesal edge of the spinning field, a large tartipore mesad of these, and a field of twenty-four piriform spigots with flat base margins and interspersed with tartipores (Fig. 57B). The male ALS is similar, but the posterior MAP is replaced by a flat-topped nubbin (Fig. 58B). The female PMS lacks PC spigots, and has a large anterome- dian spigot with a short cylindrical base and long, tapering shaft, and two posterior spigots with tapering bases and cylindrical shafts (Fig. 57C). Comparison with the male PMS reveals that the large posterior spigots are missing, but the larger anteromedian one remains (Fig. 58C). We sug- gest that the former are CY and the latter is a mAR The female PMS also has eight small spigots, seven with relatively long bases and short shafts and another, close to the mAP, that has a relative- ly long shaft. Comparison to the male reveals that the seven small spigots are present but that the one near the mAP is replaced by a flat-topped nubbin similar to that of the posterior MAP on the ALS. We suggest that this spigot is a second mAP, and that the others are AC. The domed apical segment of the female PLS has more than 15 apical AC spigots, two lateral CY, and an apical spig- ot with a long, cylindrical shaft that we code as an MS (Fig. 57D). Two additional spigots occur anterobasolaterally on the female PLS (arrows in Figs. 56B, 57D) well separated from the main spinning field. These spigots differ from AC, and from each other, but at least the basal one resem- bles the apical MS (Fig. 56C). The CY and MS are absent in the male, the latter replaced by an api- cal nubbin, and two other nubbins replace the basolateral spigots (Fig. 58D). Because these spig- ots are replaced by nubbins in the male we suggest that these are the two flanking spigots of the triad that are displaced to a basal position (see character 96). The male pedipalpus is unique among dictynids and lacks the characteristic dictynid C (Fig. 169D). The tibia has a pointed apical RTA (Fig. 169C). The cymbium lacks trichobothria, chemosensory scopulae or processes. The bulb has a slender embolus that arises laterally, and a retroventral, bifid process. This process is flexibly attached, and because of the position, far from the embolus, and type of attachment, we code this as an MA. This in turn suggests that the C is absent. The entelegyne female genitalia have a simple vulva and an epigynum with median and lat- eral lobes but lacking teeth. Eresidae C.L. Koch, 1851 Eresidae comprise 10 genera and 102 species (Platnick 2004). Their center of richness is in Africa but eresids occur across Eurasia. They are absent from the Americas except for two species of Stegodyphus from Brazil, and they are absent from Australia and New Zealand. All are cribel- GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 25 late except Wajane. Our exemplars are Eresus cinnaberinus and E. sandaliatus from Eurasia and Stegodyphus mimosamm from Africa. Eresids are both solitary and social (three species of Stegodyphus). The solitary species make webs with with a central, funnel-like retreat and an irregular cribellate mass appressed to the sub- strate or multiple cribellate sheets in different planes on which they walk. The social species make a large nest of silk, plant debris and chitinous remains of their insect prey, and large sheets of cribel- late silk may extend out in several directions (Figs. 199A, C-D). The spiders walk upon or hang beneath these sheets: we code them as walking on the silk like the solitary species. The cribellate silk carding leg is braced with a mobile leg IV, and the margins of the cribellate band are entire. We have never observed them to wrap captured prey. The fine structure of the cribellate silk of Stegodyphus was studied by Kullman (1975) who recorded both axial fibers and reserve warp and noted that the cribellar fibrils are cylindrical in cross section. Eresids have eight eyes lacking tapeta, with the posterior row strongly recurved. The anterior margin of the carapace is truncated, giving it a "square" look when viewed from above (Fig. 199B). The chelicerae may have a small boss (Fig. 129A) and small teeth near the fang furrow, but the elongate thickened seta behind the fang is absent (Figs. 131A-B, D). The paturon is extended slightly toward the fang tip, but we do not score this as homologous to the chela of filistatids and other haplogynes. We could not detect a chilum beneath the characteristic clypeal hood (Fig. 129A). The legs are stout and spines are few. Tarsal trichobothria are lacking and there is only a single, subapical trichobothrium on the metatarsus: the base has transverse ridges (Fig. 154D). There are only plumose hairs and the tarsal organ is capsulate (Fig. 152C). Eresids have a linear calamistrum and a dorsal patch of smaller calamistral setae (i.e., with lines of teeth, Figs. 144D-F). In some specimens of Eresus the line of larger setae is not clearly distinguishable from the dorsal patch. There are three claws but serrate accessory setae, scopulae, claw tufts and preening combs are lacking (Fig. 136B). A series of partially serrated hairs occur at the sides of the superior claws (Fig. 136A) but these differ in position from typical serrate accessory setae. Males have numerous scattered epiandrous spigots (Fig. 159A). The posterior respiratory system comprises four simple tracheal tubes (Lamy 1902, Griswold, pers. obs. Stegodyphus). Males lack apophyses on the palpal tibia, and the cymbium lacks processes, trichobothria, or chemosensory scopulae. The male palpal bulb has only an apical, sclerotized conductor that embraces the embolus (Fig. 170D). The enig- mantic ecribellate eresid Wajane is reported to have a tibial apophysis and MA (Lehtinen 1967). Female genitalia are entelegyne and the epigynum has median and lateral lobes but lacks teeth. The silk glands and spinnerets have been discussed previously but their interpretation remains controversial. Kovoor and Lopez (1979) studied the silk glands of the eresids Eresus cinnaberinus (as E. niger) and Stegodyphus dufouri. Peters (1992b) studied the spigots of two species of Stegodyphus and traced the origin of the fibers that composed the cribellate strands. Eresid spin- nerets have been studied with scanning electron microscopy and coded in matrices by Coddington (1990b), Platnick et al. (1991), Griswold et al. (1999) and Schiitt (2002). In this paper we make several changes from codings in previous phylogenetic studies. Kovoor and Lopez (1979) identi- fied pseudoflagelliform glands of the sort that we assume to serve the PLS MS. These spigots were overlooked in eresids by Griswold et al. (1999), but in this new study we recognize MS spigots in eresids. Kovoor and Lopez (1979) further asserted that eresids have numerous ampullate and cylin- drical glands but lack aciniform glands. Previous phylogenetic studies (Griswold et al. 1999) relied upon these gland data to code eresids as having numerous MAP, mAP, and CY and lacking AC, but here we rely upon our ontogenetic data and recognize AC spigots as present. Schiitt (2002) also coded eresids as having a brush of AC spigots. We here illustrate the spinnerets of an Eresus cinnaberinus male from Greece (Figs. 32A-D, 26 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 33E-F) and provide data for females from Morocco (Figs. 31A-D, 33A-D), and Switzerland (Figs. 34B, E). We also illustrate the female spinnerets of Eresus sandaliatus from Denmark (Figs. 34A, C). We also discuss the male and female spinnerets of Stegodyphus mimosarum from Malawi (Figs. 34D, F, 35A-D, 36A-D) and of a female S. mimosarum from South Africa (Figs. 33G-J, 37A-D). In most details Eresus and Stegodyphus are alike. The spigot cuticle lacks the characteristic "fin- gerprint" ridges of most entelegynes. Eresids are unique in that their ampullate shafts have small papillae or imbricate protrusions (Figs. 33C, E, 34B, D, 37C). The cribellum is divided (Fig. 34A) into two parts (Dresserus has the cribellum transversely divided into four parts) with strobilate spigots (Fig. 34C) and paracribellar spigots are absent from both the PMS and PLS. The female ALS of Stegodyphus has at least five or more MAP scattered throughout the spin- ning field of more than thirty PI spigots with rounded bases (Figs. 35B, 37B). A large nubbin may be present near the MAP that are clustered at the inner edge of the spinning field (Fig. 35B), and other nubbins and tartipores occur amongst the PI spigots. The MAP are conspicuously larger than the PI, and their bases are squat and larger in relation to their shafts than those of the PI. The ALS of the male is similar, though with fewer spigots of both types (Fig. 36B). The spigots on the PMS and PLS are difficult to interpret and females from Malawi and South Africa differ in details. The Malawian female (Fig. 35C) and male (Fig. 36C) each have at least one large spigot with a squat base and nearly cylindrical shaft: these resemble the MAP of the ALS, and we code these as mAP The South African female has two mAP (Fig. 37C). The female PMS has a large tartipore and smaller spigots of two types. There are 15-25 small spigots with narrow bases and cylindrical shafts and at least two spigots of intermediate size with conical bases and shafts (Figs. 35C, 37C). The former also occur in males but the intermediate size spigots are absent (Fig. 36C). The females have more than twenty additional spigots of various sizes, whereas the male has fewer than twelve. The female PLS also has more spigots than the male. Ontogenetically this resembles a pattern in which the female has both AC and CY spigots, but the male retains only the AC. Although Kovoor and Lopez (1979) state that eresids lack the glands that serve AC spigots, but have multiple ampul- late and tubuliform (cylindrical) glands, in this case we accept the ontogenetic evidence and sug- gest that female Stegodyphus have AC, mAP and CY spigots, with the latter absent in males. It is difficult to differentiate these types: the largest spigots with squat bases are probably mAP (Figs. 35C, 36C), the intermediate size with cylindrical bases may be CY, and the smallest spigots may be AC (Figs. 35C, 37C). On the anterior, basal margin of the apical segment of the male and female PLS, well separated from the rest of the spinning field, there is a triad of slender spigots, one of them larger (Figs. 37D, 33J, inset in 36A, D); of this triad, we coded the larger one as a MS, in agreement with Peters' (1992b) interpretation, who also found that these spigots produce the axial fibers in the cribellate strands. In the female from South Africa the PLS are also unusual in having some nubbins interspersed among the AC spinning field (Fig. 331). Eresus female spigots (Figs. 31A-D) are like those of Stegodyphus except that there are eight to eleven recognizable ampullate gland spigots (Fig. 3 IB). The ALS appears to have more MAP with squat bases (Fig. 34B), includ- ing six to eight concentrated near the median margin and two to four more scattered through the PI spinning field (Fig. 3 IB). Males have fewer MAP and PI (Fig. 32B). The PMS of both females and males have at least four large mAP along the anterior and median margins (Figs. 31C, 32C, 34E). The female PMS has more than 30 small spigots of varying sizes, probably representing both CY and AC (Fig. 31C), whereas the male has only five or six small AC spigots (Fig. 32C). The AC and CY spigots of Eresus are not so clearly differentiated as in Stegodyphus. The difference in spigot number between female and male suggests that there may be several CY in the female but we have not attempted to label these. The female PLS has a field of more than 40 small spigots and a basal triad of spigots with large, squat bases (Fig. 3ID); as in Stegodyphus, we coded the larger one as a GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 27 MS. In the male, this basal group has the two (Fig. 32D) or three (Fig. 33F) accompanying spigots vestigial or reduced to nubbins. Filistatidae Ausserer, 1867 Filistatidae are a nearly cosmopolitan family of 16 genera and 108 species, absent only from New Zealand (Platnick 2004). All are cribellate. Our exemplars are Filistata insidiatrix from Spain and Italy (Fig. 196A) and Kukulcania spp. from Argentina (Figs. 197A-C) and from Florida and Californa in the USA. Filistatids make webs of cribellate silk radiating out from a retreat and appressed to the substrate (Figs. 196A-B). The spider walks on top of the web. Filistatids have the primitive cribellate silk carding behavior using a mobile leg III to support the combing leg IV (Fig. 196E). We have observed Kukulcania to bite prey and then wrap it using slow alternating movements of legs IV. Cribellate silk of Filistata was studied by Lehmensick and Kullman (1956) and that of Kukulcania by Eberhard and Pereira (1993) using a transmission electron microscope (TEM). The cribellate band is highly folded (Figs. 118E, 196C). Filistatid cribellate silk is peculiar: axial fibers and reserve warp are present but the cribel- lar fibrils lack nodules. Eberhard and Pereira (1993) observed that the cribellar fibrils are flattened but our SEM preparations of Kukulcania silk do not show flattened fibrils (Fig. 118F). Filistatids are peculiar in many ways, combining primitive and derived character states. There are eight eyes on a mound (Figs. 196D, F, 197B); the tapetum is "primitive" (Homann 1971), and sigilla are found on the sternum. Lucrecia Nieto {in lit.) found that the intestine of Kukulcania hibernalis is M-shaped, as in other primitive Araneomorphae (Marples 1968). The chelicerae (Figs. 126A-C) are fused together at the base and lack a basal boss, teeth or a stout seta at the fang base, but the paturon is prolonged to meet and form a chela with the fang tip (Fig. 126B). We could not detect a chilum. Tarsal trichobothria are lacking but there is a row of several trichobothria on the metatarsus: the base is smooth (Figs. 154A-B). There are only plumose hairs, autospasy is at the patella-tibia jont, and the tarsal organ is capsulate (Fig. 152B). There are three claws but serrate accessory setae, scopulae, claw tufts and preening combs are lacking (Fig. 136C). Filistatids have the calamistral setae in three staggered rows (Fig. 143B). The calamistral setae have multiple rows of large teeth. Males have numerous scattered epiandrous spigots. The posterior respiratory system comprises a wide spiracle leading to two stiff median tubes, and two lateral flat extensions (Lamy 1902; Ramirez and Grismado 1997; Griswold, pers. obs.); Ramirez found that the hatching instars of Filistata and Kukulcania have about three booklung lamella in place of the flat extensions. We scored both genera as having reduced posterior booklungs. Males lack apophyses on the palpal tibia, and the cymbium lacks processes, trichobothria, or chemosensory scopulae. The male palpal bulb is piriform: spindle-shaped with the subtegulum and tegulum fused (Figs. 166D, 167D), and it retains the claw flexor and claw extensor muscles M29 and M30 (Fig. 167D). The female is hap- logyne and lacks an epigynum (Fig. 164E). The spinning organs of filistatids were described in detail by Platnick et al. (1991) and we sup- plement those observations with new scans of Filistata. Filistatid cribella are divided (Figs. 3A, 5A, 14D), tartipores are absent, and the cribellate sigots are "claviform": nearly smooth and thick- er apically (Figs. 5B-C, 14E). The female ALS of Filistata has numerous PI spigots with rounded bases and three MAP spigots: two near the outer margin and one within the PI spinning field (Figs. 3B, 5F). The ALS has three segments and a posterior fan of large setae with teeth (Figs. 5D-E). The male ALS is similar, though there are fewer PI spigots (Fig. 4B). The PMS has a median mAP spigot with a squat base and slender shaft, at least three slender spigots with cylindrical bases and shafts that we code as AC, and three peculiar posterior spigots with cylindrical bases and flattened, transversely-ridged, "floppy" shafts, which we code as PC (Figs. 3C, E, 4C). These presumed PC 28 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II are like the PC of other neocribellate spiders in that they occur on both the PMS and PLS, and the morphology resembles the cribellar spigots of the same species: shafts of these spigots may be pointed or claviform (Fig. 14C) These differ from typical PC in being posterior on the PMS (Figs. 3C, 4C). Like some other "lower" araneomorphs (e.g., Thaida, Hickmania), some PC may occur in the male as well as the female. The female PLS has more than fifty AC spigots and two floppy PC spigots at the apex (Figs. 3D, F); the male is similar (Fig. 4D). Kukulcania females resemble Filistata in having three ALS MAP, two on the field margin and one within the PI spinning field (Fig. 6B), three AC on the PMS and many AC on the PLS, and floppy PC posteriorly on the PMS (Figs. 6A, C) and apex of the PLS (Fig. 6D). Gradungulidae Forster, 1955 Gradungulidae comprise seven genera and 16 species (Platnick 2004). They occur only in Australia and New Zealand. All are ecribellate except Macrogradungula moonya from Queensland and two species of Progradungula from New South Wales and Victoria, Australia. At least the cribellate Progradungula carraiensis wraps prey after ensnaring it (Forster et al. 1987:59). The cribellate species are very rare and we cannot add to the data presented by Forster et al. (1987) and the analysis of Platnick et al. (1991). We included as exemplar the ecribellate Gradungula sorenseni from New Zealand. The colulus is represented by a pilose area. All spigots are interspersed with tartipores and small orifices (asterisks in Figs. 17A, D-E). The ALS has three segments (Figs. 15B, 16B). The mesal sector of the ALS spinning field has 12-14 MAP spigots and several tartipores (Fig. 17B). Two slightly larger MAP are set apart with a nearby tartipore (Fig. 17C). The PI closer to the mar- gins have longer bases and shafts. The PMS (Figs. 15C, 16C) has a central area with small AC spig- ots, more numerous in the female, and encircled by larger spigots with thicker shafts (Figs. 17D-E). Because these larger spigots are present in both sexes, we tentatively identified them as a second class of AC (marked with '?' in Figs. 15C-D, 16C-D). We cannot recognize CY spigots. The PLS has a conical terminal segment, with spigots as in the PMS, and lacks an MS (Figs. 15D, 16D, 17F). Gradungulids have characteristically modified tarsi I and II, with the raptorial proclaw much larger than the retroclaw (Forster et al. 1987: figs. 185-189). The male palp of Gradungula lacks an apophysis on the tibia and the bulb has a complex, divided embolus and a second small, flexi- bly-attached process on the bulb that Forster et al. (1987) classify as a MA. We code the MA pres- ent for Gradungula and code the C as absent. The female genitalia of Gradungula were described by Forster et al. (1987). As in the austrochilids, the epigastric fold leads to a blind fold, which has the attachment of abdominal muscles. The female genital area is protruding, with some sclerotiza- tions, and the genital opening is exposed on the posterior face, well out of the epigastric fold. The genital opening leads to the spermathecae, and to the uterus externus. The male has numerous epiandrous spigots dispersed in a transverse line, and the genital opening is exposed, in the same way as in the female. Huttoniidae Simon, 1893 This monotypic family is endemic to New Zealand (Platnick 2004). In addition to Huttonia palpimanoides O.R-Cambridge, there appear to be several undescribed species. Long considered a member of the Stenochilidae, Huttoniidae was elevated to family status by Forster and Platnick (1984). Our exemplar is Huttonia palpimanoides from near Dunedin, New Zealand. We choose the Huttoniidae for this phylogenetic study because of all the families representing the "classic" GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 29 Palpimanoidea (Huttoniidae, Palpimanidae and Stenochilidae) only huttoniids retain all six spin- nerets. The eight eyes are in two nearly straight rows and have a canoe-shaped tapetum. The cheliceae lack a boss, stout setae near the fang base or true teeth, but have peg teeth. The chilum is absent. The legs are spineless, the first tarsus, metatarsus, and tibial apex have a prolateral scopula (like palpimanids and stenochilids), all hairs are plumose and the cuticle is ridged (Figs. 134A-C). Tarsal trichobothria are absent and there is only a single subapical trichobothrium on the metatar- sus. The trichobothrial base hood is transversely ridged (Fig. 1491) and the capsulate tarsal organ has a round orifice (Fig. 149D). There are three claws but serrate accessory setae and claw tufts are absent. Metatarsi III have apicoventral preening combs. Forster and Platnick (1984: fig. 304) recorded a single thick median trunk and two simple, lateral tracheal tubes. Epiandrous spigots are absent from the male. The spinnerets of Huttonia palpimanoides O.P.-Cambridge were described by Platnick et al. (1991: figs. 246-248, 305-310). We scanned the spinning organs of both sexes of a probably undescribed Huttonia species from Orongorongo, New Zealand. Huttonia are ecribellate (Figs. 23A, 24A) and lack a colulus. The female ALS has a narrow field margin, a sin- gle MAP spigot on the mesal margin, a small adjacent nubbin and tartipore, and a field of about 10 PI spigots with short, rounded bases and interspersed with tartipores (Fig. 23B). The male ALS is similar (Fig. 24A). The posterior spinnerets have numerous long hairs that make the identification of spigots difficult. The female PMS is fused basally and has 13-15 AC spigots with long, filiform shafts, and two posterior CY spigots with long, thicker shafts (Fig. 23C). Platnick et al. (1991) identified a larger anterior spigot in the female PMS of Huttonia palpimanoides as an mAP, but it is absent in both sexes of our specimens. Because they did not find a similar spigot in the male, it might be a CY spigot instead. We scored the mAP absent. The male retains only 10-12 AC (Fig. 24C). The female PLS has numerous AC spigots, and four basal CY spigots (Fig. 23D). The male retains only the AC (Fig. 24D). Both sexes have a central papillate mark on the PLS. The male pal- pus lacks tibial processes, and the cymbium lacks trichobothria or chemosensory scopulae. Huttonia has an almost piriform bulb. The embolus is a short apical spine accompanied by a small pointed process that we code as the C (Forster and Platnick 1984: figs. 350-352). The haplogyne female genitalia lack an epigynum. Hypochilidae Marx, 1888 Hypochilidae comprise two genera: the Chinese Ectatosticta (one species) and the North American Hypochilus (ten species) (Platnick 2004). Hypochilus makes a peculiar, "lampshade" web (Figs. 195A-B). The spider hangs beneath a central retreat surrounded by a circular curtain of cribellate silk. Shear (1969), Eberhard (1988) and Catley (1994) have studied the behavior of Hypochilus. Cribellate silk carding behavior is primi- tive: a mobile leg III supports the combing leg IV. They have never been observed to wrap prey. Cribellate silk of Hypochilus was studied by Eberhard and Pereira (1993) who found that axial fibers and reserve warp are present but that the cribellar fibrils lack nodules. Hypochilids are peculiar in many ways, retaining primitive and exhibiting derived character states. There are eight eyes with "primitive" tapeta (Homann 1971); the posterior eye row is recurved (Fig. 195C). Sigilla are found on the labium and sternum, and the serrula consists of a plate bearing several rows of teeth. Internally, the venom glands are confined to the chelicerae, the coxal glands have highly convoluted ducts, diverticula of the proximal portion of the midgut (the thoracenteron) extend anteriorly into the base of the chelicerae, the pharyngial dilators originate on an apodeme of the rostrum, the posterior midgut is M-shaped, and they retain four heart ostia, two 30 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II pairs of booklungs and the fifth abdominal endosternite, which is lost in the remaining araneo- morphs. The large chelicerae (Fig. 130A) have teeth but lack a basal boss or a stout seta at the fang base. Hypochilids have distinct concavities on the inner subbasal face of the paturon into which the fang tips fit when closed (Fig. 130B). We could not detect a chilum. Tarsal trichobothria are lack- ing and there is only a single subdistal trichobothrium on the metatarsus: the base has a smooth proximal hood with a marginal ridge and is embedded distally (Fig. 150B). There are only plumose hairs, and the tarsal organ is exposed. There are three claws (Fig. 132A) but serrate accessory setae, scopulae, claw tufts and preening combs are lacking. Hypochilids have the calamistral setae form- ing two rows (Fig. 143A). Males have numerous scattered epiandrous spigots. Males lack apophy- ses on the palpal tibia, and the cymbium lacks trichobothria or chemosensory scopulae but has a lateral lobe that we code as a paracymbium (Fig. 166A). The male palpal bulb retains a clear sep- aration between the subtegulum and tegulum, an apical C embraces and spirals with the embolus, and there is a lobe on the tegulum which, in keeping with Catley's (1994: fig. 4) interpretation, we code as an MA (Fig. 166B). The female is haplogyne and lacks an epigynum (Figs. 164A-B). The spinning organs of Hypochilus pococki were described in detail by Platnick et al. (1991), and we supplement those findings here with our own scans of a female H. pococki from North Carolina. The cribellum is entire with strobilate spigots (Figs. 1A, 2A-B), tartipores and paracribellar spigots are lacking and the spigot cuticle is annulate (Figs. 1C, 2C). The ALS field margin is narrow and at least ten MAP spigots with tapering shafts are clustered at the mesal mar- gin of the spinning field (Fig. IB). The two anterior MAP are larger (Fig. 2C). The PMS has only AC spigots (Fig. 1C). The PLS has numerous AC spigots and at the apex are two or three (one shaft is broken in all our scans) large spigots with slender shafts (Fig. ID). The wide-shafted spigots depicted in Platnick et al. (1991: figs. 11, 12) were broken. Though they resemble the MAP spig- ots on the ALS, we code these as PLS modified spigots (MS), homologous to the similar spigots in neocribellates and the pseudoflagelliform and flagelliform gland spigots of Orbiculariae. It is noteworthy that the Chinese Ectatosticta seems to have two MS on the PLS, but those are smaller than the AC (Fig. 2E). There are no CY spigots. Mimetidae Simon, 1881 This worldwide family comprises twelve genera and 154 species (Platnick 2004). They are not known to build webs. So far as is known, all invade the webs of other spiders and prey on them. Our exemplar is Mimetus hesperus from North America. The eight eyes are in two nearly straight rows and have a canoe-shaped tapetum (Figs. 128B-C). The cheliceae are fused at the base (Fig. 128C), lack a boss (Fig. 128A) or stout setae near the fang base but have peg teeth and a few true teeth (Figs. 128D, 130D). The cheliceral gland opens through a very low mound. The chilum is absent. There is an evident diastema (Fig. 128A, Schiitt 2002), i.e., a wide space between chelicerae and endites. The legs have spines, and legs I and II have characteristic series of raptorial spines along the anterior prolateral surfaces of tibiae and metatarsi (Fig. 142C). All hairs are serrate and the cuticle squamate (Fig. 149B). Tarsal tri- chobothria are absent and there is only a single subapical trichobothrium on the metatarsus. The trichobothrial base hood is smooth (Fig. 149G) and the capsulate tarsal organ has a round orifice (Fig. 149B). There are three claws and serrate accessory setae but claw tufts and scopulae are absent (Fig. 142B). Forster and Platnick (1984: fig. 306) recorded four simple tracheal tubes. Male epiandrous spigots are scattered anteriad of the epigastric furrow (Fig. 157E). We describe the spin- nerets of a female (Fig. 25A) and male (Fig. 26A). Mimetids are ecribellate and have a triangular colulus surmounted by a few setae (Fig. 26A). The female ALS (Fig. 25B) has a broad spinning GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 31 field margin with a single MAP accompanied by a posterior nubbin at the inner mesal margin, and a tartipore mesad of these. The MAP field is separated from the PI field by a deep furrow. There are more than 40 PI spigots with very short bases and interspersed with tartipores. The male is sim- ilar (Fig. 26B). The PMS has a median row of four AC spigots, a posterior mAP with squat base and tapering shaft accompanied by a nubbin and a tartipore, and a remarkable anterior spigot with very broad base and hemispherical, grooved shaft (Fig. 25C). Comparison with the male, which lacks this spigot, suggests that this is a CY spigot. The male PMS also lacks the posterior tartipore (Fig. 26C). The female PLS has 14 AC spigots and a large anterior CY spigot of similar morphol- ogy to that on the PMS (Fig. 25D). The male retains only the AC spigots (Fig. 26D). The male pal- pal tibia has simple, rounded ventral processes and the cymbium lacks trichobothria or chemosen- sory scopulae. There is a retrolateral paracymbium and dorsal and lateral slender spinous process- es as well (Figs. 169A-B). Like araneids, the palpal tarsus is rotated so that the cymbium is pro- lateral and the bulb retrolateral. The tegulum of Mimetus has several sclerotized processes includ- ing an anterior ridge with groove for the embolus, a subapical hook, two small median teeth, and two broad retrolateral flanges. We code the C, MA and TA's as present although we cannot easily specify which is which. The entelegyne female genitalia have an epigynum. Neolanidae Forster and Wilton, 1973 This monogeneric family comprises three cribellate species of Neolana from New Zealand (Platnick 2004). We have chosen Neolana dalmasi (Figs. 204C, G) as our exemplar. Neolana dal- masi were observed in the Waipoua Forest of North Island, where they make characteristic webs on the trunks of huge Kauri (Agathis australis) and other large trees (Figs. 204A-B). These webs comprise a vertical curtain of cribellate silk placed on the outside of a concavity in the bark (Fig. 204B; Forster and Wilton 1973: fig. 951). The spiders hang head down on the inside of this cur- tain. We have not observed their silk carding or prey capture behavior, and the fine structure of their cribellate silk is unknown. The eight eyes are in two nearly straight rows (Figs. 204C, G) and have a canoe-shaped tape- turn. There are two rows of metatarsal and one of tarsal trichobothria (including the palpal tarsus), which have smooth to longitudinally ridged trichobothrial bases (Fig. 155G). Tarsi have three claws but no claw tufts, serrate accessory hairs or scopulae (Fig. 132B). The capsulate tarsal organ has a teardrop-shaped orifice (Fig. 153D). The chelicerae have a large boss, teeth on the fang fur- row and thickened setae near the fang base. The chilum is bilateral. Forster and Wilton (1973) recorded four simple tracheal tubes. We could not discern epiandrous spigots in the male. We describe Neolana spinning organs for the first time. We had only females suitable for scan- ning (Fig. 67A). The wide, short cribellum is divided into two fields of strobilate spigots. The ALS has a wide, bare margin surrounding the spinning field, two mesal MAP plus a tartipore, and a PI field of more than 40 spigots interspersed with tartipores (Fig. 67B). The anterior margin of the PMS (Fig. 67C) has two thick shafts each giving rise to bundles of 10-12 strobilate PC spigots. There are five AC spigots, a large mesal mAP posterad of these, and a posterior CY. The mAP spig- ot has a shorter, stouter base than does the CY. The conical apical segment of the PLS has an api- cal MS flanked by two PC spigots (Fig. 68A). There are 7-10 AC spigots interspersed by tartipores and one or two large CY spigots along the mesal margin (Fig. 67D). The male pedipalpus has a simple apical RTA and a proximal OTA (Fig. 178E). The RTA and DTA are connected by a ridge. The bulb has a central, sclerotized knob that we score as the C and a small, probasal protuberance that we score as a MA (Fig. 178D). The epigynum has a simple median lobe and lateral lobes without teeth, and the vulva is simple. 32 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Nicodamidae Simon, 1898 Nicodamidae comprise nine genera and 29 species from Australia, New Guinea and New Zealand (Platnick 2004). Only the two monotypic genera from New Zealand, Forstertyna and Megadictyna, are cribellate. Megadictyna was long classified as a dictynid, while the ecribellate Australian Nicodamus made up the Nicodamidae. Forster (1970) was the first to place Megadictyna in Nicodamidae. Harvey (1995) revised the family and provided a phylogeny. We chose Megadictyna thilenii from New Zealand (Figs. 203A-B) and Nicodamus mainae from Australia as exemplars. Our Megadictyna thilenii specimens were from the Nelson and Marlborough regions of northern South Island, New Zealand, and behavioral observations were made in the field and on captive individuals. These spun sheets of cribellate silk beneath concavities, and hung beneath the webs (Figs. 203C-D). The cribellate silk carding leg is braced with a mobile leg IV and prey was wrapped with legs IV after being bitten. Carlson (in lit.) provides data on the cribellate capture line (Figs. 122A-B). The cribellate band contains a pair of reserve warp fibers and a pair of slightly thinner axial lines. The cribellate mass is not puffed, and its fibers are cylindrical, with nodules. Nicodamids (Figs. 203A-B) have eight eyes with a canoe-shaped tapetum arranged in two nearly straight rows. The large chelicerae lack a boss or thickened setae near the fang base but have teeth on the fang furrow. We did not observe a chilum. Leg cuticle is smooth in Megadictyna (Fig. 154F) but squamate in Nicodamus (Fig. 149J). Tarsal trichobothria are lacking and there is only a single, subapical trichobothrium on the metatarsus; trichobothrial bases have a nearly smooth hood but may be smooth (Nicodamus, Fig. 149J) or longitudinally ridged distally (Megadictyna, Fig. 154F), and the capsulate tarsal organ has a round orifice (Figs. 149E, 152G). The legs have three pectinate claws (Figs. 137A, 139A). Megadictyna has thick accessory claw setae with weaker ser- rations than in Austrochilidae and Orbiculariae, but we code them as homologous. In Nicodamus (Fig. 139A) the accessory setae are similar to those of Archaea (Figs. 134E-F) but weaker, and we code them as uncertain. Tarsus IV has a series of stout ventral setae (Figs. 140D, 141A), but claw tufts, preening combs and scopulae are lacking. Forster (1970) recorded a respiratory system com- prising a pair of highly branched median tracheae and a pair of simple lateral tracheae in Megadictyna; the lateral tracheae are absent in Nicodamus. Male Megadictyna have numerous epiandrous spigots scattered in bunches along the epigastric furrow (Figs. 161A, C). We describe the spinning organs of Megadictyna (Figs. 39, 40) and Nicodamus (Figs. 41, 42) in detail. In Megadictyna the cribellum is entire, wide and short, and set with numerous strobilate spigots (Fig. 38D). The ALS has a narrow, bare margin (Figs. 39B, 40B) and spigot cuticle is ridged (Fig. 38B). Both males and females have a pair of large MAP at the mesal margin and 90-100 PI spigots interspersed with tartipores. The anterior margin of the female PMS is encircled by more than a dozen single-shaft PC spigots (Fig. 38B); males have a series of nubbins in this position (Fig. 40C). The mAP spigot is posterior, as in Araneoidea (Fig. 39C). The female PLS has a conical api- cal segment with a MS and single PC spigot at the tip (Fig. 38A): males have a large and small nubbin in this position (Fig. 40D). The PMS and PLS of males and females have numerous iden- tical spigots with long cylindrical shafts and short tips that we presume to be AC spigots. We are not able to distinguish any potential CY spigots, although the greater number of these uniform spig- ots in females relative to males (PMS - 90:65; PLS - 110:70) suggests that AC and CY spigots may be externally indistinguishable. We have coded the CY spigot characters as unknown in Megadictyna. Nicodamus mainae is ecribellate with a well defined, triangular colulus (Figs. 41A, 42A). The ALS has two segments, and there is a pair of mesal MAP spigots and a large tartipore, with slender shafts in the male (Figs. 41B, 42B). The PI spigots have a short base with sharp edge. The PMS has a very atypical complement of spigots, which we can only homologize tentatively GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 33 (Figs. 41C, 42C). One median spigot with a thick, cylindrical base, present in both male and female, we identified as mAR The nubbin posterior to the male mAP in Figure 42C is not present on the opposite PMS. A lateral spigot with a long base and thick shaft, on the female PMS, we iden- tified as CY. There is a central group of spigots with large, blunt shafts, and numerous spigots with thin shafts, encircling the PMS except in the median line. We tentatively identified those spigots as two different classes of AC. They occur in females and males. The female PLS (Fig. 41D) has a group of four CY spigots (the most anterior has two shafts on the same base) and lacks an MS. The male is similar but lacks the CY spigots (Fig. 42D); the apical nubbin is absent on the other PLS. The male pedipalpus of both Megadictyna and Nicodamus has a characteristic large, curved, proximal OTA (Figs. 171A, C, 172A-C). The bulb of Megadictyna is simple with only one process in addition to the embolus. We code this as a sclerotized, central C (Figs. 171A-B). There is no MA. Nicodamus has a similar spiral embolus but differs in having a large, distad-projecting C and similar MA that originate near the center of the bulb (Figs. 172B, D). Nicodamus has the retromar- gins of the tegulum and subtegulum with interlocking lobes (Figs. 172A, D); such lobes are absent in Megadictyna. Nicodamids are entelegyne with simple vulvae and epigyna with scape-like pro- jections. Oecobiidae Blackwall, 1862 Oecobiidae comprises six genera and 101 species (Platnick 2004) occurring worldwide, although the wide distribution is due in part to cosmopolitan synanthropic species of Oecobius. Oecobiids are both cribellate and ecribellate. The ecribellate genera are restricted to Africa and Eurasia. Our exemplars are Oecobius navus from the USA and undetermined species of Uroctea from India, Namibia (Figs. 199E-F) and South Africa. Oecobiids make silken retreats with upper and lower sheets and trip lines radiating out in all directions. Prey may be wrapped during "whirligig" behavior, i.e., with the spider running rapidly around the prey swathing it with silk from the elongate PLS, and Eberhard (1967) recorded Oecobius wrapping prey with alternate movements of legs IV. The fine structure of Oecobius cribellate silk was reported by Zimmerman (1975), who found cylindrical cribellar fibrils with nod- ules and a cribellate band with reserve warp (summary in Eberhard and Pereira 1993). The cribellate Oecobius and ecribellate Uroctea share some striking features. The carapace is heart-shaped to round in dorsal view and the eight eyes are arranged in a tight group (Figs. 199E-F). The chelicerae are small, not extending vertically past the endites (Fig. 129B), and lack a basal boss, teeth on the fang furrow, and thickened setae near the fang base (Fig. 130C). The anal tubercle is enlarged, nearly as long as the elongate PLS, and set with a marginal ring of setae (Figs. 27A, 30A). The male palpal bulb has several peculiar processes that are difficult to homologize with those of other spiders (Figs. 170A, C, 187B). Oecobiid legs have only plumose hairs, tri- chobothria are lacking from the tarsi and there is only a single subdistal trichobothrium on the metatarsi, and the capsulate tarsal organ has a round orifice (Fig. 152D). The legs have three claws but lack claw tufts, scopulae, serrate accessory claw setae or preening combs (Figs. 136E, 140C). Apart from differences in spinning organs (see below), Oecobius and Uroctea differ in several other features. Oecobius has a primitive tapetum, whereas the tapetum of Uroctea combines fea- tures of the primitive and canoe-shaped type (Homann 1971); the trichobothrial bases of Uroctea are smooth (Fig. 154C), whereas those of Oecobius have transverse ridges on the hood; Uroctea males have epiandrous spigots (Fig. 159B) whereas Oecobius males lack them. Uroctea has three ALS segments, whereas Oecobius has two. Kovoor (1980) made an extensive study of the silk glands of Uroctea and Oecobius, but close examination of the spinnerets and spigot morphology has not been presented before. We scanned 34 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Oecobius females from California and Georgia and a male from Washington DC. in the USA, and female and male Uroctea from Garies, South Africa. Oecobius is cribellate and has a divided cribellum (Fig. 27A). The female ALS has a single large MAP spigot at the inner margin and a field of eight PI spigots with round bases and inter- spersed by tartipores (Fig. 27B); the male ALS is similar (Fig. 28B). Interestingly, Millot (1938) described two ampullates for the ALS of Oecobius. Kovoor also refers to two pairs of ampullates, but it is not clear if she was referring to both major and minor ampullates. We find only one spig- ot on the ALS that has the larger size and peripheral position characteristic of MAP. The female PMS has a central series of thirteen small spigots with slender, barrel-shaped bases and narrow shafts, and single larger spigots anteriorly and posteriorly (Fig. 27C). Comparison with the male (Fig. 28C) shows that the median series and anterior large spigot remain but the large posterior spigot is absent, suggesting that the anterior spigot is a mAP and the medians AC. Although Kovoor (1980) states that "tubuliform" glands seem absent in Oecobius, the size and ontogeny of the pos- terior spigot suggest that it is a CY. The female PLS has an elongate apical segment with numer- ous small spigots resembling the AC of the PMS and three larger spigots along the outer margins of the AC row (Fig. 27D). These larger spigots are absent from the male (Fig. 28D), and, again, we suggest that these may be CY spigots. Uroctea differs, of course, in lacking the cribellum, but in many other characteristics as well. The female ALS has several large and small spigots (Fig. 29A), and the male ALS has a similar array of spigots, only fewer (Fig. 30B). Kovoor (1980) recorded 9-18 pairs of ampullate glands and 20 pairs of piriform glands in Uroctea: we consider the large ALS spigots to serve MAP glands and the small to serve PI. Kovoor recorded only CY and AC glands serving the PMS of Uroctea. Our scans of the female PMS reveal numerous small spigots with slender, conical shafts, and a few larger spigots with cylindrical shafts (Fig. 29B). The male PMS has only the smaller type of spig- ot (Fig. 30C), which we code as AC. The larger spigots are probably CY. The female PLS has the same spigot types as on the PMS (Figs. 29C-D). Numerous small AC spigots form a broad longi- tudinal band, and eight larger CY spigots occur along the outer margin of this band. Males lack the CY (Fig. 30D). The male palpus of oecobiids lacks tibial processes and has no cymbial processes, trichoboth- ria, or chemosensory scopulae. The palpal bulbs have a readily recognizable embolus and three additional processes that are difficult to homologize. Shear (1970) recognized a conductor, stipes and radix on the oecobiid tegulum, whereas Coddington (1990a) referred to an oecobiid embolic apophysis (OEA), probably Shear's conductor, an oecobiid tegular apophysis (OTA) and two lobes (OTLII and II). Because there are three proceses in addition to the embolus, by default we code a C, MA, and TA, but cannot determine which process is homologous to those processes in other spi- ders. All are listed as TA (Figs. 170A, C, 187B). The female genitalia of oecobiids are also remark- able (Baum 1972). The copulatory duct leads to an anterior large, membranous sac, from where another long, sclerotized duct runs to the posterior margin, where the fertilization ducts discharge. Oecobius has in addition a membranous sac at the base of the fertilization duct. Pararchaeidae Forster and Platnick, 1984 This family comprises one genus and seven species from Australia and New Zealand (Platnick 2004). Our exemplar is a Pararchaea species from Fiordland on South Island, New Zealand with additional data from Forster and Platnick (1984), Platnick et al. (1991), and Schiitt (2000). There are eight eyes and the tapetum is canoe-shaped. The pars cephalica is prolonged into a short "neck" and sclerotized completely around the base of the chelicerae. The chelicerae lack a GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 35 boss, have no stout setae near the fang base nor true teeth but have peg teeth. There is an evident diastema between the chelicera and endite. The chilum could not be observed. The legs are spine- less, all hairs are serrate and the cuticle is squamate. Tarsal trichobothria are absent and there is only a single subapical trichobothrium on the metatarsus. The trichobothrial base hood is smooth and the capsulate tarsal organ has a round orifice. There are three claws and serrate accessory setae but claw tufts, preening combs and scopulae are absent. Forster and Platnick (1984: fig. 307) recorded four simple tracheal tubes and two bunches of epiandrous spigots along the male epigas- tric furrow. Data on the spinning organs of Pararchaea have been gleaned from Platnick et al. (1991) and Schiitt (2000). In neither paper were the spinnerets fully illustrated, so some characters, especially for the ALS, cannot be scored. Pararchaea are ecribellate. The spinneret cuticle is squa- mate and there are tartipores. The PMS has two AC spigots and a CY spigot with a long, tapering shaft (Schiitt 2000, fig. 10C). The occurrence of ampullate gland spigots on the ALS and PMS and PI spigot morphology remain unknown. The male palpus lacks tibial processes, and the cymbium lacks trichobothria or chemosensory scopulae but has a paracymbium. The apex of the Pararchaea tegulum has a "complex distal plate" (Forster and Platnick 1984: fig. 237) with two flanges, a hook, and a broad scaly surface, which we code as a C. The entelegyne female genitalia have an epigy- num. Phyxelididae Lehtinen, 1967 Phyxelididae comprise 12 genera and 54 described species (Platnick 2004) that occur in Africa and Madagascar, the southeastern Mediterranean, and south east Asia (Vytfutia). Lehtinen (1967) treated Phyxelidinae as a subfamily of the Amaurobiidae. Griswold (1990) revised this subfamily, adding Vytfutia from the Agelenidae. Griswold Coddington, Platnick and Forster (1999) raised Phyxelididae to family rank and suggested sister group relationship with the Titanoecidae (Fig. 212, "Titaneocoids"). We have observed the behavior of several genera of phyxelidids, including Xevioso in the field and Phyxelida in the field and lab. All build cribellate webs. Xevioso (Figs. 202B, E-F) builds an appressed web radiating out from a retreat, as does Vidole (Fig. 202A; Griswold 1990: fig. la): it walks erect on the web. There may sheet-like cribellate components (Fig. 202F). Phyxelida builds webs with more three-dimensional structure, typically with small sheet-like components beneath which the spider hangs inverted (Figs. 202D). In phyxelidids the cribellate silk carding leg is braced with a mobile leg IV and at least Phyxelida and Ambohima wrap prey with slow alternating movements of legs IV after first biting it. Carlson (in lit.) has studied the fine structure of Phyxelida tanganensis cribellate silk. The cribellate band is entire, cribellar fibrils are cylindrical with nod- ules, and axial fibers and reserve warp are present (Figs. 121A-C). Phyxelidids have eight eyes in two nearly straight rows (Figs. 202A-C), canoe-shaped tapeta, chelicerae with a large boss and teeth and thickened setae along the fang furrow (Fig. 13 IF). The cheliceral gland does not open on a mound (Griswold 1990: fig. 14c, d). The chilum may be entire (Vytfutia) or divided (Phyxelida and Xevioso). Tarsal trichobothria are lacking and there is only a single, subapical trichobothrium on the metatarsus: the base has transverse ridges (Figs. 147A, 155A). The capsulate tarsal organ has a round orifice (Fig. 152J). Most have only plumose setae (Fig. 147A), but at least Malaika has feathery scales as well. The palpal femora of both sexes have probasal thickened setae modified as thorns (Figs. 173D-E). Males of most species have metatar- sus I modified with a posterolateral projection that may be surmounted by a spine and a distal con- cavity (Figs. 173F, 202C). At least in Phyxelida tanganensis the male grasps the female by her sec- ond trochanter with this clasping organ while they hang face to face during copulation. There are three claws but serrate accessory setae, claw tufts and scopulae are absent (Figs. 132C-D). 36 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Preening combs on metatarsi III and IV may be present (Xevioso: Fig. 141G) or absent. The phyx- elidid calamistrum is linear and located in the middle of metatarsus IV (Fig. 143D). Deeleman- Reinhold (1986) reports that Vytfutia has four simple tracheae; Xevioso and Phyxelida each have simple lateral tracheae and median tracheae with few to many branches. Males of at least Xevioso and Phyxelida have epiandrous spigots grouped into two lateral bunches (Figs. 160A-B); epiandrous spigots are absent in Vytfutia (Fig. 161G). All phyxelidids have divided cribella (Fig. 47A) with fields of uniformly distributed strobilate cribellate spigots. Numerous (13-29) PC spigots encircle the anterior margin of the female PMS (Figs. 47C, 49C). The bases are elongate and pressed together and flattened, and each is surmount- ed by a single strobilate shaft (Figs. 46C, 47C). Spigot cuticle is ridged. Phxelidini and Vidoliini have a characteristic, stout, curved seta laterally on the PLS (Fig. 49D), but we couldn't find this seta in Vytfutia. It may have broken off in our specimens. Otherwise phyxelidid spinning organs vary. Griswold (1990) briefly described phyxelidid spigots based on Namaquarachne tropata and Xevioso arnica: here we elaborate on this discussion, redescribing Xevioso and describing Phyxelida and Vytfutia for the first time. The female ALS of Phyxelida tanganensis has a narrow, bare margin (Fig. 49A). There are two MAP at the inner edge and a field of more than 30 PI spig- ots with round base margins and interspersed with tartipores (Fig. 49B). The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. 50B). The female PMS anterior mar- gin is encircled by at least 16 PC, there is a large mAP and a nubbin and tartipore posteriad of these, and posteriorly 8-10 AC spigots (Fig. 49C). We identify only one posterior CY. The male PMS has the PC replaced by an encircling row of nubbins, a large median tartipore and a nubbin that replaces the mAP (Fig. 50C). The female PLS has a domed apical segment and a stout lateral seta. There is an apical MS, a field of about 15 AC and two mesal CY spigots (Fig. 49D). Males lack the CY and have the MS replaced by a large nubbin (Fig. 50D) that resembles a tartipore in being apically dim- pled. Xevioso arnica females (we have not scanned the male) resemble P. tanganensis in most details except that the PMS has a single large nubbin and at least four CY spigots and the PLS has a single PC spigot and a slender nubbin ("cuticular finger") next to the MS spigot (Griswold 1990: figs. 30a-f). Most of the PLS spinning field of our specimen is infolded and obscured: only one CY spigot is visible, but there may be others. The PLS is similar to Namaquarachne, which has an apical MS, slender nubbin, and two CY (Fig. 46B). Vytfutia pollens differs from these in several details. Both females (Fig. 47B) and males (Fig. 48B) have only a single anterior MAP and poste- rior nubbin on the ALS. The PMS of both sexes have a large median mAP but lack nubbins: the female has four CY spigots, three of which are visible in Figure 47C. The PLS appears to lack the stout lateral setae characteristic of other phyxelidids (they may be broken off in our specimens), and the female has an apical MS, one PC, and at least two CY (Fig. 47D). There is a spigot adja- cent to the MS that differs from all others. Unlike the AC it has a slender, cylindrical shaft without longitudinal striations. Although it lacks the encircling ridges characteristic of PC, we code it as a PC (Fig. 47D: PC). It, as well as the MS spigot, is replaced by nubbins on the male PLS (Fig. 48D): this is typical of MS and PC ontogeny. The male palpal tibia of all phyxelidids has a dorsoapical process that may be simple and scle- rotized (e.g., Namaquarachne, Vytfutia), partly sclerotized and partly hyaline (Xevioso and Vidole) or bifid with the larger part rolled (Ambohima, Phyxelida) (Figs. 173A-B). Vytfutia has an addi- tional RTA. The cymbium lacks trichobothria or chemosensory scopulae (Fig. 173A). Male palpal bulbs are diverse (Figs. 170B, 173C, 188C-D). All have at least two processes: C and MA. Conductors are fleshy to sclerotized and may oppose (Vytfutia, Xevioso) or embrace (Phyxelida) the embolus. Xevioso C may be hyaline at the outer edge (Fig. 170B). Both Phyxelida (Figs. 173C, 188C-D) and Vytfutia have flexibly attached, cylindrical MA whereas Xevioso (Fig. 170B) has GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 37 three to five conical processes including an MA and extra tegular processes (e.g, TA1, TA2, TA3, TA3a and TA4: Griswold 1990). We label these all as TA, but assume that one is the MA homo- logue. Epigyna are simple with median and lateral lobes without teeth (Fig. 170E). Psechridae Simon, 1890 Psechridae comprise two genera and 24 species (Platnick 2004). Psechrus and Fecenia occur from southeast Asia to northern Australia. The New Zealand endemics Haurokoa and Poaka, although lacking the psechrid peculiarities of three claws plus claw tufts, grate-shaped tapeta and an elongate, "rectangular" calamistrum, were placed in this family by Forster and Wilton (1973). Recently Raven and Stumkat (2003) moved Poaka to the Amaurobiidae and Haurokoa to the Tengellidae. We discuss Poaka morphology under the Psechridae below and note that its placement in the Amaurobiidae, Psechridae, or elsewhere is still problematic. Our exemplars are Psechrus species from China (Fig. 208A-B, D-E), India, Nepal, Papua New Guinea and Thailand, and Poaka graminicola from New Zealand. All are cribellate. Because of their numerous differences we discuss these genera separately. We have observed Psechrus in the field in Yunnan, China. There they build huge (0.5 to lm across) webs comprising a cribellate sheet (Fig. 208E). At one edge is a circular retreat that leads to a cave, rock crevice, or other cavity (Fig. 208A). The spider hangs beneath the web (Fig. 208B) and the cribellate silk carding leg is braced with a mobile leg IV (Fig. 208D). We never saw them wrap pray. The fine structure of Psechrus cribellate silk has not been studied. The eight eyes are in two nearly straight rows (Fig. 208B) and have a grate-shaped tapetum, the chelicerae have a large boss, teeth on the fang furrow and thickened setae near the fang base, and the chilum is bilateral. The legs have three claws, and remarkably, claw tufts (Figs. 132E-F, 139C), but no serrate accessory setae or preening combs and no scopulae on the posterior legs of females. The legs have only plumose hairs and there are two rows of metatarsal and one of tarsal trichobothria on the legs, but none on the palpal tarsus. The trichobothrial bases have transverse ridges on the hood and the capsulate tarsal organ has a round orifice (Fig. 149L; Griswold 1993: fig. 65). The calamistrum comprises several irregular rows of calamistral setae and is a modified form of the "oval calamistrum" typical of the Lycosoidea and their kin (Figs. 143H, 145F). The posterior respiratory system comprises four simple tracheal tubes (Lamy 1902:168). The male lacks epiandrous spigots. We scanned female (Fig. 109A) and male (Fig. 110A) spinnerets. The cribellum is divided and evenly covered with strobilate cribellar spigots (Figs. 97B, E). Paracribellar spigots are absent from both the PMS and PLS. The female ALS has numerous piriform spigots with round bases; at the mesal margin are two large MAP spigots and a tartipore (Fig. 109B). The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. HOB). The female PMS (Figs. 109C, 111 A) has an anterior large spigot with a squat base, which we interpret as an mAP, and a smaller spigot with a nearby tartipore, which we tentatively identify as a mAP as well because this spigot is replaced by a nubbin in the male (Figs. HOC, 112A-B). Posterior to this group is a field of more than twenty small spigots with slender bases and shafts and at least six larger spigots each with a long, cylindrical base and conical shaft (Figs. 109C, 111 A-B, 112A). Both types of spigots also occur on the female PLS, but only the small type occurs on the male PMS and PLS (Figs. 110C-D). We interpret the smaller spigots as AC, and the larger as CY The female PLS has a broad band of more than twenty AC and about 11 CY spigots in a band along the outer margin of the AC spigot field (Fig. 109D). At the PLS apex the female has a presumed MS spigot (Fig. HID), flanked by two small nubbins in males (Fig. 112C): the MS occurs in a male from Papua New Guinea (Figs. 38 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II HOD, 112D) but is replaced by a nubbin in a male from Thailand (Fig. 56D). The male genitalia of Psechrus are in many ways both typical and atypical for Lycosoidea. There is no apophysis on the palpal tibia, but the cymbium has a well developed dorsal chemosen- sory scopula (Fig. 167A). The tegulum and subtegulum have interlocking lobes, but the C is fleshy rather than hyaline and there is no MA (Figs. 167B-C). Psechrids are typical lycosoids, though: the related genus Fecenia has an RTA, hyaline C and a MA (Levi 1982). The entelegyne female geni- tala have an epigynum that lacks lateral teeth (Figs. 164C-D). Poaka graminicola was placed "with considerable doubt" (Forster and Wilton 1973:297) in the Psechridae. Raven and Stumkat (2003:106) moved Poaka to the Amaurobiidae because of its sim- ilarities to Manjala, an Australian spider described in the Amaurobiidae by Davies (1990). As our analysis suggests, the definition of Amaurobiidae is still unsettled, and the ultimate placement of Poaka remains an open question. This monotypic genus is endemic to New Zealand, where the spi- der is common in grasslands and lives by hunting without constructing a snare. Poaka has eight eyes in two straight rows, and, unlike typical psechrids, the tapeta are canoe- shaped. We were unable to discern a chilum. The chelicerae have a large basal boss, teeth on the fang furrow, and the characteristic stout seta at the retromargin of the fang. The legs are spinose and sexually dimorphic: female tibiae and metatarsi I and II have several pairs of ventral spines, though there are few spines on these segments in males. All leg setae are plumose and the calamistrum is linear and extends for the basal half of the fourth metatarsus (Fig. 142D). The leg metatarsi have two irregular rows of trichobothria and the tarsi have a single row of three to five trichobothria that increase in length distally. The trichobothrial base has a smooth hood (Fig. 150D), and the capsulate tarsal organ has a round (Fig. 150D) to keyhole-shaped (Fig. 150E) open- ing. There are three claws but unlike typical psechrids claw tufts are absent (Fig. 142A). Preening combs, serrate accessory setae, and scopulae are also absent. The male lacks epiandrous spigots (Fig. 157F). The tracheal system comprises four simple tubes. We report on Poaka spinning organs for the first time. We scanned both female and male spin- nerets (Figs. 107A-D, 108A-D). The cribellum is divided (Fig. 157B) and covered with evenly spaced strobilate spigots and the spinneret cuticle texture is ridged (Fig. 157D). The female ALS has a narrow spinning field margin and a single MAP flanked by a posterior nubbin at the median edge (Fig. 107B). There are about fifteen PI gland spigots with rounded base margins. The male ALS is similar, with the PI field interspersed with tartipores (Fig. 108B). The female PMS has a large anterior spigot with squat base and tapering shaft that we code as mAP (Fig. 107C). The cen- tral part of the PMS spinning field has several AC spigots and several PC spigots with a single stro- bilate shaft emerging from each base. The PC extend from the anterior margin to the middle of the spinning field. We code as CY two large spigots at the posterior margin of the spinneret with shafts much stouter than the mAP The male PMS retains only the anterior mAP and four AC spigots (Fig. 108C). Surprisingly, there are no nubbins representing vestiges of the PC of the female. The domed apical segment of the PLS of our female is slightly collapsed making spigot interpretation difficult (Fig. 107D). There are several AC spigots but we cannot confirm the presence or absence of CY or PC spigots. One apical spigot has a shaft much stouter than the others: we provisionally code this as an MS. The male retains only several AC and has several nubbins (Fig. 108D). The male palpal tibia has four retrolateral processes: an apical blade, two posteromedian teeth and a median stout cone (Fig. 184F). We code this as a "complex RTA." The cymbium lacks processes or chemosensory scopulae but has at least one trichobothrium. The complex palpal bulb has a tegulum with a prolateral straight, slender embolus. The base of the embolus extends forward in a sclerotized projection that parallels the embolus and embraces its apex (arrows in Figs. 184 C-E). There are two flexibly-attached tegular apophyses, one arising near the embolic base that we GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 39 code as the C, and the other near the retroapex of the bulb that we code as a MA (Figs. 184A-C, E). The entelegyne female genitalia have a simple vulva and an epigynum with median and later- al lobes that lack teeth. Segestriidae Simon, 1893 Segestriidae comprise three genera and 106 species (Platnick 2004). The genera Ariadna and Segestria are cosmopolitan, and Gippsicola, known from only one species, is known only from Australia. Segestriids occupy tubular cavities, which they line with silk. The web comprises silken "trip lines" that radiate out from the retreat (Fig. 202G). They are members of Dysderoidea, a homogeneous group of haplogyne families including dysderids, oonopids, and orsolobids (Forster and Platnick 1985). Dysderoids have a peculiar tracheal system (described, among others, by Lamy 1902, Purcell 1910, and Forster and Platnick 1985). The lateral tracheae are advanced just behind the epigastric furrow, bearing two separate spiracles similar to the booklung spiracles (Fig. 162A). Each spiracle leads to a broad, anteriorly directed tracheal trunk that splits in many thin tracheoles, serving both the abdomen and the carapace (Fig. 162B-F). The internal female genitalia of dys- deroids are also distinctive (described, among others, by De la Serna de Esteban 1976, Uhl 2000, and Burger et al. 2003). The posterior wall of the atrial cavity connects to a large posterior recep- tacle (Fig. 162B), while the anterior part leads to an anterior, often bilobate receptacle. Our representatives are Ariadna boesenbergi from Argentina and Ariadna maxima from Chile. There are six eyes with primitive tapeta, the anterior medians having been lost. The carapace is elongate, and the thoracic fovea is absent. There are no sternal or labial sigillae. The chelicera has three teeth on its anterior margin, and a small apical retromarginal tooth (Fig. 130F). The chelicer- al lamella found in other haplogynes is absent. A cheliceral boss is absent, but in its place there is a transversly elevated ridge. The third leg is characteristically oriented forwards, an apparent adaptation to the tubular silk retreat (Fig. 202G). The tarsal organ is exposed (Fig. 15IB; Forster and Platnick 1985: fig. 862). Tarsal trichobothria are lacking (Fig. 138B) and there is only a single subdistal trichobothrium on the metatarsus. The trichobothrial bases have a transverse ridge (Fig. 151A). The tarsal cuticle is smooth. There are three tarsal claws (Fig. 138A) and no accessory claws, claw tufts, or scopulae. The setae are of the plumose type and there are no scales of any kind, including feathery scales. There is a preening comb on the retrolateral apical margin of metatarsus IV. The male metatarsus I is cylindrical, although other Ariadna species bear clasping processes. Male epiandrous spigots are absent. We have scanned the spinnerets of male and female Ariadna boesenbergi (Figs. 18, 19). Our observations concur with those compiled in Platnick et al. (1991) for Ariadna and the other two segestriid genera. The spinnerets are not sexually dimorphic. The colulus is well defined and pilose. The ALS has three segments, with the basal segment crossed by a diagonal membranous area, giv- ing the impression of a four-segmented spinneret (Figs. 18A, 19B). There is only one MAP spigot without accompanying nubbins or tartipores, and about 11-12 PI (Figs. 18C, 19B). The PMS have one mAP and one AC (Figs. 18E, 19C), and the PLS have 4 AC (Figs. 18F, 19D). The ALS PI and the PLS AC have elongated, curved shafts and flattened bases. We have not found tartipores, except for a scar corresponding to one of the male PLS AC; because this scar occurred on only one side, we scored tartipores absent, which concurs with previous findings (Platnick et al. 1991). The male palp has a simple, cylindrical tibia without processes, and a very short tarsus, where the piriform copulatory bulb arises. There is no hematodocha other than the articulation of the bulb with the tarsus, and no bulb sclerites or processes other than the simple embolus. 40 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Stiphidiidae Dalmas, 1917 Stiphidiidae comprise 13 genera and 94 species occurring primarily in Australia and New Zealand (Platnick 2004). Records of Ischalea from Madagascar and Mauritius are dubious until modern material is available for study. Stiphidiids are both cribellate and ecribellate. As exemplars we chose Stiphidionfacetum, which we observed in New Zealand and Australia, and Pillara griswoldi (recently described by Gray & Smith 2004), observed in the Barrington Tops area of New South Wales, Australia. Pillara was identifed as "Baiami" in Griswold et al (1999). Stiphidion makes a characteristic web, dubbed the "sombrero web," which has a central conical retreat encircled by a loosely-hanging curtain of cribellate silk (Figs. 204D-F). The spider hangs beneath the web. Pillara makes a flat sheet of cribellate silk beneath which the spider hangs: there is a retreat at one corner of the web. Cribellate silk carding has not been observed in either genus, and we never observed Stiphidion to wrap prey that it had caught. The fine structure of Stiphidion cribellate silk was studied by Eberhard and Pereira (1993), who found the cribellate mass entire, cribellar fibrils cylindrical with nodules, and axial fibers and reserve warp in the cribellate silk. Our stiphidiid exemplars have two rows of eyes with grate-shaped tapeta. Homann (1971) depicts a weakly undulate grate-shaped tapetum in Stiphidion; Pillara has the tapetum making a few broad loops (Griswold, pers. obs). We code the tapeta of each genus as grate-shaped. The PER of Pillara is straight, but that of Stiphidion is recurved. The chelicerae have teeth, thickened setae near the base of the fang furrow, and a large boss, and the chilum is median. Pillara has deeply notched trochanters, whereas those of Stiphidion are entire. Feathery scales are present on the legs in both genera (Fig. 1471). There are three claws but serrate accessory setae, preening combs, claw tufts and scopulae are lacking. Tarsal organs are capsulate (Figs. 153C) and there is a single row of tarsal trichbothria with smooth bases (Figs. 1471, 155F); the tarsal trichobothria increase in length distally. The calamistrum is linear. Male epiandrous spigots are absent. The posterior respiratory system consists of four simple tubes (Griswold pers obs., Pillara; Forster and Wilton 1973, Stiphidion). We examined the spinning organs of both sexes of Pillara and Stiphidion. The cribellum (Figs. 69A, 71A) is divided into two fields of strobilate spigots (Fig. 72A), spinneret cuticle is ridged (Fig. 71C) and the ALS has a wide bare margin surrounding the spinning field (Figs. 69B, 71B). The female ALS of Stiphidion has two MAP clustered at the mesal margin: the anterior is much larger than the posterior. The PI field comprises nearly 40 spigots with rounded base margins inter- spersed with tartipores (Fig. 69B). The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. 70B). There is a conspicuous tartipore near the MAP nubbin. The female PMS (Fig. 69C) has nearly thirty AC spigots, one anterior mAP, and a posterior and lateral CY. The PC comprise two huge bases on the anterior margin of the spinning field with six to eight strobi- late shafts emerging from each base (Fig. 68B). The male PMS lacks the CY, and the PC are replaced by two or three large anterior nubbins (Fig. 70C). The conical apical segment of the female PLS (Fig. 69D) has two basomedian CY (only one is visible in Fig. 69D), more than 40 AC, and an apical MS flanked by three PC (Fig. 68D). The male PLS lacks the CY and the apex has the MS and PC replaced by four nubbins (Fig. 70D). Pillara spinnerets resemble those of Stiphidion (Figs. 71A-D, 72B-D). The female ALS has two MAP clustered at the mesal margin and a PI field of about 30 spigots with rounded base margins interspersed with tartipores (Fig. 7IB); the male retains only the anterior MAP with the posterior replaced by a nubbin (Fig. 72B). The female PMS (Fig. 71C) has one large anterior mAP, one posterior CY and only five AC spigots. Two huge PC bases on the anterior margin of the spinneret give rise to 15 and 7 strobilate spigot shafts. The lat- eral PC base has a single PC base and shaft emerging from its side. The male retains the mAP and GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 41 seven AC and the PC are replaced by two anterior nubbins (Fig. 72C). The female PLS (Fig. 7ID) has a basal and a median CY, about 15 AC, and an apical MS flanked by three PC (Fig. 68C). The MS and PC are replaced by nubbins in the male (Fig. 72D). Male pedipalpal tibiae (Figs. 179F, 189A) have a VTA and an RTA that may be simple (Stiphidion) or complex (Pillara). The cymbium lacks chemosensory scopula but has trichoboth- ria. The palpal bulb lacks a MA: in our analysis this optimizes as a synapomorphy for Stiphidiidae, but this may be artifactual, as at least some ecribellate stiphidiid genera (e.g., Cambridgea, Ischalea) may have MA (Forster and Wilton 1973). The C is sclerotized and embraces the embo- lus. We score this as a"desid" type C, but these are heterogeneous in stiphidiids. The C of Stiphidion is broadly T-shaped, with two points extending retrolaterad and embracing the embolus and the other simple one pointing prolaterad (Fig. 179E). Pillara has a C that embraces only the embolus tip. Female genitalia are entelegyne, the epigynum lacks lateral teeth and the vulva is sim- ple. Tengellidae Dahl, 1908 Tengellidae comprise eight genera and 32 described species (Platnick 2004). Most occur in the Americas. Tengella is the type genus. Our exemplars are Tengella radiata from Costa Rica (Fig. 206C, E-F) and an unidentified Tengella species from Mexico. We have observed Tengella radiata in the field in Costa Rica and in the lab. They make a large sheet of cribellate silk with a funnel-shaped retreat at one side, and run on top of the sheet (Figs. 206C, E). The cribellate silk carding leg is braced with a mobile leg IV. We never saw them wrap prey. The fine structure of Tengella cribellate silk was studied by Eberhard and Pereira (1993) who found that the uniform cribellate silk band has both axial fibers and reserve warp, and that the cylindrical cribellar fibrils have nodules. Tengella has eight eyes in two rows (Fig. 206F), a canoe-shaped tapetum, chelicerae with a large boss and teeth and thickened setae near the fang furrow. The chilum is bilateral. There are multiple rows of trichobothria on the tarsi; trichobothrial bases have transverse ridges (Fig. 149K). The tarsal organ is capsulate with a round orifice (Fig. 149F; Griswold 1993: fig. 64). There are three claws and scopulae but claw tufts are lacking (Fig. 139B). The calamistrum is oval and females have dense scopulae beneath the posterior tarsi. There are four simple tracheae. We report on the spinning organs of a male and two female Tengella radiata from Costa Rica (Figs. 98A, 99A). The cribellum has a divided spinning field with strobilate spigots uniformly dis- tributed. The ALS has a narrow, bare margin. The female has two MAP (Figs. 100C) at the mesal margin with a nearby tartipore, and a field of numerous PI spigots interspersed with tartipores (Figs. 98B). The male is similar except that the posterior MAP is replaced by a nubbin (Fig. 99B). The female PMS lacks PC spigots, has two mAP with squat bases and long, conical shafts, the external one accompanied by a tartipore, and two posterior CY spigots interspersed with about 40 AC spigots (Fig. 98C). The male is similar, retaining the two mAP spigots but lacking the CY (Fig. 99C). The female PLS has more than 30 AC spigots and two basal and one median CY spigots (Figs. 98D, 100A-B). There is an apical MS on the anterior margin, flanked by two presumably AC spigots (Fig. 98D inset). The group is replaced by nubbins in the male (Figs. 99D, F). Both PMS and PLS have some marginal spigots slightly larger than the AC, which we tentatively iden- tify as a different class of AC (marked with arrows on plates). The males lack epiandrous spigots. The male palpal tibia has a simple retrolateral process. The cymbium lacks a chemosensory scopula. Tengella palpal bulbs are typical of primitive Lycosoidea and their kin (Griswold 1993; Wolff 1977: fig. 3). They have interlocking lobes on the tegulum and subtegulum, and two process- es in addition to the embolus on the tegulum: an apical hyaline C and a median, convex, flexibly 42 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II attached MA. The epigynum has an enlarged median lobe; the lateral lobes lack teeth (Wolff 1977: fig. 5). Titanoecidae Lehtinen, 1967 Titanoecidae comprise five genera and 46 described species (Platnick 2004). Most occur in Eurasia and the Americas though a few species occur in Africa, Madagascar, and New Guinea. All are cribellate. Our exemplars are Goeldia species from South America (Figs. 203F-H) and Titanoeca species (Fig. 203E) from the USA. Titanoecids make cribellate webs that radiate from a retreat (Figs. 203F, G). These webs may be appressed to the substrate and may include sheet-like components. Szlep (1966) has described the web and spinning behavior of Titanoeca albomaculata from Israel. She states that the spider emerges from the retreat at night and "sits on the catching web." We have observed the behavior of Titanoeca nigrella in captivity. They moved on top of the web and were never observed to wrap prey after biting it. Carlson (in lit.) has studied the fine structure of cribellate silk of Titanoeca nigrella. The cribellate silk has both axial fibers and reserve warp (Fig. 121D). Titanoecids have eight eyes in two nearly straight rows (Fig. 203H), canoe-shaped tapeta, che- licerae with a large boss and teeth and thickened setae along the fang furrow. The chilum is entire. Tarsal trichobothria are lacking and there is only a single, subapical trichobothrium on the metatar- sus: the base has transverse ridges (Figs. 154H, I). The capsulate tarsal organ has a round to oval orifice (Figs. 1521, L). There are three claws, but claw tufts, serrate accessory setae, preening combs and scopulae are lacking. The linear calamistrum begins near the base and extends for near- ly the entire length of metatarsus IV (Fig. 143E). The posterior respiratory system comprises four simple tracheae. Males of Titanoeca americana have numerous epiandrous spigots scattered ante- riad of the epigastric furrow (Figs. 161B, D). We were unable to observe this region in our Goeldia male specimens and code their epiandrous spigots as unknown. We illustrate titanoecid spinning organs for the first time. All titanoecids have divided cribel- la (Fig. 51 A) with fields of uniformly distributed strobilate cribellate spigots. Paracribellar spigots are absent from the PMS but present on the PLS. The female ALS of a Goeldia from Chile has the bare margin narrow surrounding the spinning field at least anteriorly (Fig. 53B). The are two MAP with a nearby tartipore and a field of about 30 PI spigots with rounded bases and interspersed with tartipores. The MAP and tartipore are removed from the mesal margin of the spinneret and are par- tially surrounded by the PI spinning field. The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. 54A). The female PMS (Fig. 53C) has a large anterior mAP with a squat base and a single AC mesad and two AC laterad of the mAP. Posteriorly there are two coni- cal CY with stout, conical shafts. The male PMS has the mAP replaced by a huge dimpled nubbin, and retains only the three AC spigots (Figs. 54C-D). The female PLS has a domed apical segment with a terminal field of about nine AC and a single PC (Fig. 53D). There is a lateral CY and near the base are at least two PC spigots (Fig. 55E). There is no sign of an MS near the apical PC (Fig. 55F). Males have only the AC spigots and a large nubbin (Fig. 54B), presumably of the PC. The spinnerets of Titanoeca are similar but differ in several features. Like Goeldia the MAP are removed from the mesal margin of the spinneret and partially surrounded by the PI field (Figs. 51B, 52B, 55A) and the female PMS has a large mAP and two posterior CY (Fig. 51C) with the mAP replaced by a huge nubbin in the male (Fig. 52C). The PMS has more AC spigots than Goeldia: five (Fig. 52C) or six (Fig. 51C). The Titanoeca americana female PLS is also like that of Goeldia in lacking an MS and having an apical and three basal PC spigots (Fig. 5ID). A peculiar, elongate nubbin appears to replace the apical PC on the male PLS (Fig. 52D) and other nubbins replace the proximal PC. Titanoeca nigrella is identical to T. americana (Figs. 52A, 55A, C). The nubbins that GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 43 replace the mAP and PC are peculiar in titanoecids, resembling tartipores in having folds or dim- ples (Figs. 52C, 54C). Coddington (1990b: 70) scored Titanoeca silvicola as having a MS on the female PLS, but we are unable to find such a spigot in the other titanoecid species. Because Coddington's T. silvicola female was damaged, we consider the presence of a MS ambiguous, and code Titanoeca as lacking this spigot. The male palpal tibia of all titanoecids has a complexly folded dorsoapical process (Figs. 174A, C, E, 188A-B). Male palpal bulbs have at least two processes in addition to the embolus. A convex process is flexibly attached to the center of the tegulum: we code this as an MA because of its similarity in form and position to the unambigouous MA of other families (Figs. 174B, 188A-B). There is a triangular process near the embolic base, which we code as a conical extra tegular process (Figs. 174D, 188B). There is no obvious C, but the embolus inserts in a unique groove along the outer margin of the tegulum (Fig. 174D, 188B). The female genitalia are enteleg- yne and the epigyna are simple with median and lateral lobes without teeth. Uloboridae Thorell, 1869 Uloboridae is a worldwide family comprising 18 genera and 248 species (Platnick 2004, Grismado 2004). All are cribellate. Our exemplars are Octonoba octonaria from the USA, several species of Uloboms, and Hyptiotes from California. Uloborids are the only spiders that build typical orb webs with cribellate capture lines (Figs. 201A-C, 206C), although some genera make webs reduced to a single sticky line (Miagrammopes, Lubin 1986). They exhibit the whole suite of behaviors that go into constructing the orb web. Uniquely, tertiary radii are doubled during orb construction (Eberhard 1982, Coddington 1986a). They hang beneath the orb, card cribellate silk with the carding leg braced by the other, mobile leg IV, and wrap their prey before biting. The fine structure of uloborid cribellate silk has been stud- ied extensively (Friedrich and Langer 1969; Opell 1989a, b, 1995, 2001; Peters 1983, 1984, 1987). The cylindrical cribellar fibrils have nodules. The cribellate band is puffed (Figs. 119A-B) and axial fibers are present but the reserve warp is absent (R. Carlson, in lit). The cribellate threads of Hyptiotes are typical (Fig. 120D). Uloborids are small to medium-sized spiders. There are eight eyes that lack tapeta. The large chelicerae have teeth on the fang furrow but lack stout setae near the fang base. A basal boss may be small (Octonoba) or absent (Uloborus). The chilum is not visible. Uloborids, together with the Mesothele (Haupt 2003), are unique among spiders in lacking venom glands. The legs have feath- ery scales (Figs. 147H, 148B), the tarsi lack trichobothria and there is but a single subapical tri- chobothrium on the metatarsi, but a row of trichobothria occurs on femur IV. Trichobothrial bases are smooth (Figs. 147H, 154G) and the minute tarsal organ is capsulate (Figs. 152E-F). There are three claws with associated serrate accessory claw setae (Figs. 137D, 140A) and the hind tarsus has a series of stout setae that we code as the "deinopoid tarsal comb" (Fig. 140A). Claw tufts, scopu- lae and preening combs are absent. The linear calamistrum begins basally and extends for most of metatarsus IV length (Figs. 145A-C). The posterior respiratory system comprises two thick medi- an trunks with lateral branches (Lamy 1902, Opell 1979, 1987) and simple lateral tracheae. Male epiandrous spigots are scattered along the epigastric furrow. The spinning organs of Uloboridae have been extensively studied (Octonoba octonarius: Coddington 1989: figs. 6-9; Uloborus: Kovoor 1978, Peters 1983, 1984, Peters and Kovoor 1980, 1989). We present some details of Octonoba (Figs. 44A-B) and Uloborus (Figs. 45A-C). The cribellum is entire (Fig. 44A) and the spigots have ridged texture. Unlike deinopids but like arane- oids, the ALS of Uloborus has a single MAP spigot accompanied by a nubbin (Fig. 45B). Like 44 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Deinopidae (Fig. 43B) and Austrochilidae (Fig. 11B) there is a wide, bare region between the MAP and the PI field (Fig. 45B). The PI spigots of the ALS have sharp apical margins. The PMS has an apical mAP with a squat base and short shaft, several median AC and several posterior CY (Fig. 45C). There appears to be a tartipore next to the mAP in our Uloborus individual illustrated (Fig. 45C) but there is a nubbin in that place in Octonoba (Coddington 1989), Waitkera (Platnick et al. 1991), and Conifaber (Grismado 2004). The AC and CY are similar in structure, with long cylin- drical bases, sharp base margins, and short conical shafts. In another Uloborus that we examined both a nubbin and tartipore accompany the mAP. The CY are recognizable by their larger size and their absence in males. Bunched along the anterior margin of the PMS are more than 20 PC spig- ots that have closely-spaced annulations on the shafts (Fig. 45C). These have previously been clas- sified as being of the "deinopoid" type (Griswold et al. 1999: character 82, state 0). The PLS of Octonoba has CY and AC spigots with the same morphology as in Uloborus and an apical MS set well apart from the spinning field (Fig. 44B). The male palpal tibia of uloborids lacks processes, and the cymbium lacks chemosensory scopulae, trichobothria or a paracymbium. The bulb has a C and MA. The C is large, sclerotized, and flexibly attached. We code this as the "uloborid" type, similar only to that of araneids. The entelegyne female genitalia have an epigynum. Teeth are lacking, although some Uloborus have paired projections anteriad of the epigynal median sector. Zorocratidae Dahl, 1913 Zorocratidae comprise five genera and 21 described species from Mesoamerica, Africa, Madagascar, and Sri Lanka (Platnick 2004). Most are cribellate but at least Campostichomma man- icatum and some Uduba are ecribellate. The family name was first used by Dahl (1913). The gen- era Campostichomma, Raecius, Uduba, Zorocrates and Zorodictyna were associated as a clade by Griswold (1993) in a study of lycosoid spiders and their relatives (Fig. 213), and Dahl's family name was applied to this clade by Griswold et al. (1999, Fig. 212, "Zorocratidae"). In their analy- sis the zorocratid exemplars Raecius and Uduba were joined by the male tibial crack (Fig. 141F), clumped cribellar spigots (Figs. 97C-D, F), and a male palpal tibial ventroapical process (Figs. 185B, 194C). Zorocratidae are nevertheless heterogeneous (e.g., all known Zorocrates lack the male tibial crack and clumped cribellar spigots). Subsequent work by Silva (2003) associated Zorocrates and Tengella, apart from other zorocratids (Fig. 214). Raven and Stumkat (2005) placed zorocratids as a subfamily of Zoropsidae (Fig. 215, "Zorocratinae"). Our exemplars are Zorocrates from North America (Fig. 207B), Raecius from Africa (Figs. 207G, H), and Uduba from Madagascar (Figs. 207A, C-F). We have observed Raecius in the field in Cameroon and Tanzania, Uduba in the field in Madagascar, Zorocrates in the field in Arizona and Mexico, and all three genera in captivity. Zorocratids may be running hunters or fossorial pred- ators. There may be a collar (Figs. 207F, G) or funnel (Fig. 207E) of cribellate silk extending out from the retreat. Leg autospasy readily occurs at the tibial crack of Uduba, in contrast to Zoropsis (see below). Zorocrates lacks a tibial crack. Webs comprise irregular cribellate lines or sheets that radiate out from a burrow or retreat, although some Uduba make no webs and wander in search of prey. The cribellate silk carding leg is braced with a mobile leg IV; we never observed prey wrap- ping after the initial bite. Carlson (in lit.) has studied the fine structure of the cribellate silk from a juvenile Raecius scharffi from Tanzania. The cribellate silk has both axial fibers and reserve warp (Figs. 124D, 125C). Zorocratids have eight eyes in two nearly straight rows (Figs. 207A-B, D, H) and canoe- shaped tapeta (Homann 1971: fig. 27B), chelicerae with a large boss and teeth and thickened setae GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 45 near the fang furrow (Fig. 1301). The chilum is bilateral. There are multiple rows of trichobothria on the tarsi (Fig. 141D); trichobothrial bases have transverse ridges (Figs. 151E, 156G). The tarsal organ is capsulate with a round (Figs. 151F, 153L) to multilobed orifice: in Campostichomma and some Uduba the orifice may make an asterisk-like or stellate pattern (Fig. 1530). There are three claws and dense scopulae that may obscure the ITC, but claw tufts are absent (Fig. 146A-C). At least Raecius has feathery scales (Fig. 147J), but Zorocrates has only cylindrical scales (Fig. 146D). The calamistrum is oval (Fig. 143C, 146E-G) and females have dense scopulae beneath the posterior tarsi (Fig. 141D); males of most species (but not in Zorocrates) have a tibial crack (Fig. 141F). There are four simple tracheae. We report on the spinning organs of a female Raecius asper from Cameroon, a male and female of Raecius jocquei from Cote d'lvoire (Figs. 105A-D, 106A-D), both sexes of two species of Uduba from Madagascar (Figs. 104A-D) and of two Zorocrates species from North America (Figs. 101A-D, 102 A-F, 103A-G). The spinning organs of Raecius have recently been discussed (Griswold 2002), but some of our interpretations here differ from those previously published. In all species the cribellum has linear groups of strobilate spigots (Fig. 97D); PC spigots are lacking from the PMS and PLS; three or more CY spigots occur on the PMS and PLS, and male epiandrous spig- ots are absent (Fig. 161H). Female Raecius have an entire, wide and short cribellum (Fig. 97D). The ALS has a narrow, bare margin surrounding the spinning field, two mesal MAP, a large tarti- pore mesad of these, and a field of more than 30 PI spigots with rounded bases and interspersed with tartipores (Fig. 105B). The male ALS is similar except that the posterior MAP is replaced by a nubbin (Fig. 106B). The PMS of R. jocquei has a large anterior mAP with a broad, squat base and slender shaft, several AC spigots (one anterior to the mAP and several behind) and there appears to be a second, slender mAP next to a large tartipore (Fig. 105C). There are three CY spigots. Males lack the CY spigots but retain the anterior mAP, AC and tartipore, and have a large nubbin next to the tartipore (Fig. 106C). We interpret this as a mAP nubbin, and conclude that at least R. jocquei has two PMS mAP (this differs from the conclusions in Griswold [2002], which considered the sec- ond mAP as a possible CY). The female PLS has a domed apical segment with more than 20 AC spigots and three large, conical CY spigots (Fig. 105D). We identify an MS spigot at the apex, and the male has an apical nubbin that may be an MS vestige (Fig. 106D). The female of Raecius asper resembles R. jocquei except that there are four CY on the PMS and five on the PLS (Griswold, 2002). The two exemplar species of Uduba are alike in having a divided cribellum (Figs. 97C, 104A). The ALS has two mesal MAP and large tartipore mesad of these in females, an MAP and nubbin in males, and a field of more than fifty PI with rounded bases and interspersed with tarti- pores (Fig. 104B). The female PMS has two large anterior mAP, a possible second small mAP near it, 8-12 AC, and 5-10 CY spigots on the posterior surface (Fig. 104C). Males retain only the mAP and a few AC spigots, and have a large nubbin that is probably the vestige of the second mAP. The female PLS has a domed apical segment with more than 20 AC spigots with long, slender shafts, and six large, conical CY spigots (Fig. 104D). A spigot with a stouter shaft closely flanked by two smaller spigots is identified as a MS spigot. Males have only the AC spigots. We present scans of the spinning organs of a female of Zorocrates cf. mistus and a male of Zorocrates sp. (Figs. 101-103). The cribellum is wide, bipartite, and the cribellar spigots are strobilate, uniformly dis- tributed (Figs. 101A-B). The male has only a few vestigial cribellar spigots (Fig. 103G). The female ALS has two MAP and a tartipore (Fig. 102B), while in the male ALS the posterior MAP is replaced by a nubbin (Fig. 103B). The female PMS has two mAP spigots and a tartipore, sever- al small AC spigots, and 4 or 5 large CY spigots on its posterior side (Fig. 102C-D). The male has only one mAP and a tartipore and the AC spigots (Fig. 103C-D). The female PLS has two large basal CY spigots, many AC spigots with interspersed setae, and one external MS spigot (Fig. 46 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 102E). There is some asymmetry in the PLS of this female: in the right PLS the MS is flanked by one vestigial shaft sharing the base with the MS, and one nubbin (Figs. 101C, 102E), while in the left side the MS has the fused shafts and accompanying spigots well developed. The male PLS has only AC and the three nubbins of the MS and its accompanying spigots (Fig. 103E-F). The zorocratid male palpal tibia has only a retrolateral process in Zorocrates (Fig. 186D-E), and this plus a ventral process in Uduba and Raecius (Figs. 185B, 194B-C). In most species there is a cymbial scopula of chemosensory setae, though some species of Raecius and Zorocrates lack this. Raecius palpal bulbs are typical of primitive Lycosoidea and their kin (Griswold 1993). They have interlocking lobes on the T and ST (Figs. 186F, 194C), and three processes in addition to the embolus on the tegulum: an apical hyaline C, a median, flexibly attached MA, and a large sclero- tized process (a TA that we code as "sclerotized tegular process" or STP) (Figs. 185A, 194C). In Zorocrates the embolus is articulated, and the tegular locking lobe corresponds with the embolar base (Fig. 186A, 194A); there is no subtegular locking lobe. The palpal bulb of Uduba is highly modified (Figs. 185C, 194B). There is no obvious vestige of interlocking lobes on the tegulum and subtegulum. The elongate embolus is flexibly attached and passes behind the bifid, deeply concave tegulum. There is a flexibly attached MA and a hyaline ridge on the retrolateral lobe of the tegu- lum that is probably the hyaline C. Another lobe of the tegulum is a TA. The epigynum has a medi- an and lateral lobes. Teeth are lacking in Uduba and Zorocrates but may be present or absent in Raecius: the female exemplar of Raecius congoensis that was coded for the previous study (Griswold et al. 1999) lacked them. Zoropsidae Bertkau, 1882 Platnick (2004) records five genera and 22 species in the Zoropsidae. Traditionally containing only Zoropsis, the family was enlarged by Lehtinen (1967), who proposed Takeoa for the East Asian Z nishimurai, and by Bosselaers (2002) who proposed Akamasia for Z cyprogenia, from Cyprus. Raven and Stumkat (2003, 2005) have recently studied zoropsid relationships in depth and greatly enlarged the family. They added the the Australasian genera Uliodon and Huntia, described four new Australasian genera (Birrana, Kilyana, Krukt and Megateg) and moved the subfamilies Zorocratinae (formerly Zorocratidae) and Griswoldiinae (formerly Miturgidae, comprising Devendra, Griswoldia and Phanotea) into the Zoropsidae (Fig. 215). Zorocratine exemplars are discussed above under Zorocratidae. Our zoropsid exemplar summarizes the data from two species of Zoropsis. Zoropsis is widespread in Eurasia, especially the circum-Mediterranean region. Zoropsis spinimana has recently been intoduced into North America (Griswold and Ubick 2001). All are cribellate. We observed captive Zoropsis spinimana from California and France. Zoropsis are running hunters that make no apparent use of silk for prey capture but spin a curtain of cribellate silk to sur- round a retreat where they make their eggsac (Fig. 208F). The cribellate silk carding leg is braced with a mobile leg IV. Carlson {in lit.) has examined the fine structure of Zoropsis cribellate silk. The cribellate mass is irregular and contains both axial fibers and reserve warp (Fig. 125A-B). We never observed prey wrapping. Male Zoropsis have a crack near the base of leg tibiae. In contrast to Uduba (see above), leg autospasy at the tibial crack did not occur in leathered individuals. There are eight eyes with the posterior row strongly recurved (Fig. 208G) and the tapetum grate-shaped. The chelicerae have a large boss, teeth on the fang furrow and thickened setae near the fang base, and the chilum is bipartite. The legs are very spinose (more than 7 pairs of ventral spines on the first tibia), the capsulate tarsal organ has an oval orifice (Fig. 153N) and the tri- chobothrial bases have transverse ridges (Fig. 1561). Two to three dorsal rows of trichobothria GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 47 occur on the tarsi. The short, basal calamistrum is oval (Figs. 143G, 145G, H). The ITC is reduced and there are well developed claw tufts (Figs. 139D, 141E), and at least the posterior tarsi of females have scopulae. The respiratory system consists of 4 simple tracheal tubes. Males lack epiandrous spigots. We describe the spinning organs of a female of Zoropsis spinimana (Fig. 113 A) and a male of Zoropsis rufipes (Fig. 114A). The cribellum is broad and short, with strobilate spigots distributed in transversal bands across a divided spinning field (Figs. 113C-E; Silva Davila 2003: figs. 32A-B). The ALS has a narrow, bare margin, a mesal pair of large MAP and a tartipore, and numer- ous (>60) PI spigots interspersed with tartipores (Fig. 113B, 114B). The bases of the PI spigots have a flat margin around the origin of the shaft. The PMS (Fig. 113F) lacks a paracribellum and has two large anterior spigots with short stout bases and tapering shafts: we identify them as mAR The male has the shaft broken, but one of them remains and the mAP can be identified by their bases (Fig. 114C). A large tartipore occurs anteriad of the mAP (Fig. 100E). Posteriad of these are two or three small AC spigots with cylindrical bases and slender shafts, interspersed posteriorly with at least 10 large CY spigots with conical bases and long, conical shafts (Fig. 100F). The domed apical segment of the PLS has five to six AC and a small distal spigot that we presume to be the MS (Figs. 100D, G). Laterally and basally on the PLS are 10 large CY spigots with conical bases and long, conical shafts (Figs. 100D, 113G). The male lacks the CY, and has the MS and the two accompanying spigots reduced to nubbins (Fig. 114D). The male pedipalpus has a simple, triangular apical RTA (Fig. 185D) and the cymbium has a dorsal chemosensory scopula (Fig. 185F). The bulb (Fig. 185E) has retrolateral locking lobes on the tegulum and subtegulum and three processes in addition to the embolus: an apical, hyaline C that opposes the tip of the embolus, a flexibly attached, concave MA, and a fleshy to membrane- ous TA arising near the embolic base, which we code as an extra tegular process ("MTP" = mem- braneous tegular process of Griswold 1993). The epigynum has lateral lobes without teeth and a scape-like median lobe, and the vulva is simple. CHARACTERS All character scorings are listed in Appendix 2. Comments for specific cells are listed in Appendix 3. LEGS.? These characters were scored by our observations of voucher specimens. Trichobothria were scored present if observed with light microscopy but are recorded as absent from tibiae, metatarsi and tarsi only if absence was confirmed by scanning electron microscopy. 1. Femoral trichobothria: (0) absent; (1) present. In our dataset femoral trichobothria are unique to uloborids. Although we didn't scan the femo- ra of all exemplars, the femoral trichobothria are typically long and conspicuous, and their pres- ence and absence are adequately observed with light microscopy. 2. Tarsal organ: (0) exposed; (1) capsulate. Exposed tarsal organs (state 0) have the nerve endings visible on the cuticle (Figs. 150A, 152A). The capsulate form (state 1) has the nerve endings invaginated into a pocket and accessible to the outside through a hole. This hole may be simple and round to oval (Figs. 152B-L, 153A-N) or have elaborately scalloped, stellate edges (Fig. 1530) 3. Tarsal trichobothria: (0) absent; (1) present. Tarsal trichobothria are absent in hypochilids, austrochilids, filistatids, eresids, oecobiids, Orbiculariae, nicodamids, palpimanoids, phyxelidids and titanoecids, and are present (Fig. 147E) in most dictynids (but absent in Dictyna and Nigma), neolanids, stiphidiids, agelenids, amphinec- tids, desids, amaurobiids and the Lycosoidea and their kin. 48 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 4. Tarsal trichobothrial rows: (0) one; (1) two or more. We code a single line of trichobothria as one row (Fig. 147E). An arrangement of trichoboth- ria that is staggered or forming two or more rows (Fig. 141D) we code as two or more. 5. Metatarsal trichobothria: (0) one or two; (1) three or more. A single, subapical to apical metatarsal trichobothrium occurs in hypochilids, austrochilids, eresids, oecobiids, Orbiculariae, palpimanoids, nicodamids, phyxelidids and titanoecids. A row of three or more occurs in filistatids, neolanids, stiphidiids, agelenids, amphinectids, desids, amauro- biids and the Lycosoidea and their kin. Most dictynids have only one or two but Tricholathys has a row of several trichobothria. 6. Tarsal trichobothria: (0) normal; (1) longer distally. A row of tarsal trichobothria that increase in length distally is a classic key character (e.g.; Kaston 1972: couplet 38a, fig. 107). We code as "normal" a line or lines of trichobothria of equal or of irregular, differing lengths. 7. Palpal tarsal trichobothria: (0) absent; (1) present. We have examined male cymbia for palpal tarsal trichobothria. 8. Trichobothrial base hood texture: (0) smooth; (1) with transverse ridges. We code as smooth all trichobothrial hoods that are all smooth (Figs. 154A-C). We also code as smooth those with fine wrinkles or striations that don't differ from the surrounding cuticle and that may be transverse (Figs. 135B) or longitudinal (Figs. 154G, 155F, G). We code as having trans- verse ridges those hoods with deep, broad transverse grooves that are larger than surrounding cutic- ular ridges (Figs. 147B, 149H, 154D, H, I) 9. Trichobothrial base: (0) simple; (1) notched. A trichobothrial base is simple if it has an evenly curved, entire, distal margin (Fig. 156E). Forster et al. (1987: figs. 103-105) observed a small notch on the distal margin of the base in aus- trochilids. 10. Leg cuticle texture: (0) fingerprint; (1) squamate; (2) smooth. We scored most of our terminals from scans of the tarsal cuticle. Most representatives have cuticular sculpturing forming a fingerprint pattern (Figs. 148A, F). Araneus (Fig. 149A), Mimetus (Fig. 149B), Nicodamus (Fig. 149E), and to a lesser degree Archaea (Fig. 149H), Huttonia (Fig. 1491), andDesis (on tibiae, Fig. 151J, but not on tarsi, Fig. 1511) all have squamate cuticle. Smooth cuticles are less common, e.g., in Megadictyna (Fig. 154F) and Matachia (Fig. 156B). A few rep- resentatives are scored polymorphic (0, 2) because they only have faint fingerprint marks (Nigma, Retiro) or differ among tarsal and tibial cuticle (Desis). This classification of cuticle texture was proposed by Lehtinen (1996). 11 . Deep trochanteral notch: (0) absent; (1) present. We code as notched only those trochanters that have a deep, semicircular notch (e.g., Griswold 1993: fig. 2). 12 . Tarsal claws: (0) three; (1) two. Three claws occur in most of our exemplar taxa (Figs. 132B-C, F, 133A, 137C). Loss of the ITC leaving only two claws occurs only in ctenids (Fig. 139E) and zoropsids (Fig. 141E). 13. Claw tufts: (0) absent; (1) present. Tufts are dense groups of setae in the pretarsal region, set on a separate plate beside the claws, and obscuring the pretarsus and ITC (if present). These are absent in most of our exemplar taxa (Figs. 132B-C, 133A, 137C). Claws tufts are present in spiders with three (Figs. 132E-F, 139C) or two (Figs. 139E, 141E) claws. Extended scopulae that obscure the ITC may be confused with claw tufts, e.g., as in tengellids (Fig. 139B) and zorocratids (Fig. 141D) that have dense scopula but lack tufts. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 49 14. Serrate accessory claw setae: (0) absent; (1) present. Many three clawed spiders that hang beneath webs have modified setae beneath the STC and near the ITC. These setae are serrate along the lower margin, and may be flattened. They may func- tion to grasp and release the silken line, and for this reason are sometimes called "false claws." Such modified setae occur in austrochilids (Figs 133A-D), Orbiculariae (Figs. 135C-E, 137C-D), in the nicodamids (Figs. 137A-B, 139A) and in the "palpimanoids" Mimetus (Fig. 142B), Pararchaea, and Archaea (Figs. 134E-F). Serrate accessory claw setae are absent from other spi- ders in our dataset, including some that hang beneath webs, e.g. Hypochilus (Fig. 132A), Neolana (Fig. 132B) and Phyxelida (Figs. 132C-D). 15. Male palpal femoral thorns: (0) absent; (1) present. Setae that are conspicuously thickened, or which at least have thickened bases, we refer to as "thorns." These occur on the inner surface of the male palpal femur of some austrochiloids, some amaurobiids (e.g., Pimus: Fig. 147C) and all phyxelidids (Figs. 173D-E). Most of our exemplar taxa lack thorns. 16. Female palpal femoral thorns: (0) absent; (1) present. Palpal femoral thorns similar to those in males occur in females of all phyxelidids, in Thaida, Huttonia, and at least Retiro in the Amaurobiidae. 17. Hair type. (0) plumose; (1) serrate. Plumose hairs (sensu Lehtinen 1975) have many fine, short to long barbs arranged in spiral whorls around the shaft (Figs. 147B, 154B). Serrate hairs, typical of the Araneoidea, have flattened, ovelapping scales along the shaft (Fig. 147G). These also occur in Mimetus and Pararchaea. Pseudoserrate plumose hairs (Fig. 147F) were reported to be present in deinopids and uloborids (Green 1970:8; Coddington 1986a: 327), but after a wider taxon sampling was examined, the dis- tinction between plumose and pseudoserrate became unclear (compare Figs. 148A-D). 18. Feathery scales: (0) absent, (1) present. Hill (1979) studied the diversity of shapes in scales, a specific type of seta. Scales are thinner than hairs, are bent in an angle just above the socket (Fig. 146D), and lack innervation. Scales may have lateral prolongations, named setules by Hill. Flattened scales with long setules are tradition- ally named "feathery hairs" (Figs. 135A, 147D, H-J, 148A-B). State 0 (absent) applies to those terminals lacking scales of the feathery type, or lacking scales at all. 19. Metatarsal preening combs: (0) absent; (1) present. We define preening combs as setae that form an even series at the apex of a leg segment, and that differ conspicuously from surrounding setae (Fig. 141G). We don't know if these combs are actually used in preening. Such combs occur in the phyxelidid Xevioso, the agelenid Neoramia, the desid Matachia (but not other desids), in amphinectids, in the amaurobiids Amaurobius and Callobius, and in Huttonia. 20. Tibial ventral spine number: (0) fewer than seven pairs; (1) seven or more pairs. The anterior tibia of Zoropsis and Acanthoctenus is very spinose, with more than seven pairs of ventral spines. Females, but not males, of Poaka also have very spinose first tibiae and metatar- si. No other exemplar taxon has more than five pairs of ventral spines on the anterior tibiae. 21. Reduced leg spination: (0) no; (1) yes. Most of our exemplar taxa have extensive leg spination: e.g., dorsal spines on the leg femora, pairs of ventral spines on tibiae and metatarsi I and II, and spines on most surfaces of tibiae and metatarsi III and IV. Those taxa in which spines are absent or reduced to a few scattered examples, i.e., Hypochilus, Archaea, Huttonia, and Pararchaea, eresids, and dictynids (including Aebutina), are coded as "reduced". 22. Male metatarsus I: (0) unmodified; (1) modified. 50 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Among the Phyxelididae the male first metatarsus (and second as well in Ambohima) is mod- ified with a deep median concavity and, in most species, a clasping spine (Figs. 173F, 202C). At least in Phyxelida tanganensis the male grasps the female by her second trochanter with this clasp- ing organ while they hang face to face during copulation. 23. Male tibial crack: (0) absent; (1) present. The male tibial crack is a conspicuous suture line visible through the cuticle at the base of the leg tibiae of most male Zorocratidae and in Zoropsis just distal to the basal pair of ventral spines; it is visible on the surface as a shallow, depressed ring, and autospasy may occur at this point (Fig. 141F; Griswold 1993: figs. 3-4). Autospasy through the base of the tibia apparently does not occur easily in Zoropsis but occurs readily in Uduba (Griswold, pers. obs.). The tibial crack differs from patella-tibia autospasy, which is found in Thaida, where it occurs through the base of the tibia, and in Filistatidae, where it occurs at the patella-tibia joint. 24. Female posterior leg scopula: (0) absent; (1) present. Females of tengellids, zorocratids, and lycosoids (expect Psechrus) have dense scopulae of tenent setae beneath the hind tarsi and metatarsi (Fig. 141D). Other spiders may have scopulae, at least in males, but extensive development in females is unique to the exemplars listed above. 25. Deinopoid tarsal comb: (0) absent; (1) present. The thick and blunt tarsus IV macrosetae (Figs. 140A, 141B-C) that form a clear line in deinopids and uloborids have previously been suggested as a synapomorphy for the Deinopoidea (Coddington 1986a, 1990a, 1990b; Griswold et al. 1998). This result is corroborated by our analy- sis (although these setae are absent in the deinopid Menneus, Fig. 1408). Austrochiloids and basal Entelegynae typically have tarsal macrosetae. Megadictyna has thick setae uniformly distributed (Fig. 141A), and Menneus, Oecobius, and Nicodamus (compare Fig. 140A with B-D), as well as Araneus and Mimetus, all have tarsal macrosetae, but not forming a clear line. We code the deinopid tarsal comb present only if the macrosetae form a clear line, a coding more restrictive than in previous studies (e.g., Griswold et al. 1999). The sustentaculum, a tarsal macroseta found in araneids, may be a vestige of this comb. 26. Calamistral rows: (0) two; (1) one; (2) three. Most of our cribellate exemplar taxa have the calamistral setae in a single row (state 1; Figs. 143D-F). In Hypochilus the calamistral setae occur in two parallel rows (state 0; Fig. 143A). Kukulcania and Filistata have the calamistral setae in three staggered rows (state 2, Fig. 143B), but the more basal calamistral setae still show the triseriate arrangment found in the basal filistatine Sahastata and the Prithinae. We do not score this character for those taxa with an oval patch of calamistral setae (see character 14 below) except for Eresidae, which have a single linear row as well as an oval patch of calamistral setae (Fig. 144D). 27. Calamistrum: (0) linear, with a straight row or rows of calamistral setae; (1) oval to rec- tangular. Most of our cribellate exemplar taxa have linear calamistra, with a single (Figs. 143D-F) or double (i.e., Hypochilus: Fig. 143A) row of calamistral setae (state 0). We code the filistatids, which have staggered rows, as state 0 also. An oval to rectangular patch, with the bases of calamis- tral setae not forming a line, occurs in the Lycosoidea and their kin (state 1; Figs. 143C, G, H, 145F-J). Eresids have a linear calamistrum and a dorsal patch of smaller calamistral setae (i.e., with lines of teeth, Fig. 144D-F). In some specimens of Eresus the line of larger setae is not clear- ly distinguishable from the dorsal patch. We did not score the eresids differently (e.g., as a poly- morphism with both states) because other terminals (e.g., Dictyna, Fig. 145D-E) have a similar patch, but there the calamistral setae lack the lines of teeth, thus are not easily distinguishable from the normal setae. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 51 28. Calamistrum origin: (0) basal to subbasal; (1) median. Calamistrum origins were classified based on the following formula: length from the metatar- sus base to calamistrum origin divided by the metatarsus length. A ratio of less than 0.30 was con- sidered basal to subbasal (Figs. 143E, 145A). A ratio of greater than 0.30 was considered median origin (Figs. 143D, 144A). CARAPACE.? Most characters were scored by examination of exemplars. Character data taken from the literature include the suite of classical characters from spider internal anatomy (characters 51-56; Platnick 1977, ex Millot 1931a, b, 1933a-c, 1936, Marples 1968, 1983, Homann 1971). Some past studies have assumed the states for many unexamined taxa (e.g., Platnick et al. 1991; Griswold et al. 1998; Griswold et al. 1999), but here we have coded only those for which there are observations. We have been conservative in the scoring of these characters, only using data from congeneric representatives, thus many of the cells are empty. Whenever possible tapetum form was scored from exemplars, but classical data from Homann (1971) and from Levi's translation of Homann (in lit.) were also included. The "primitive" tapetum fills the eyecup with shiny reflective surface. If a tapetum was oval and bisected by a straight dark line it was considered canoe-shaped (Homann 1971: figs. 10A, 27B, 32A). The grate-shaped tapetum is a complex structure in which the rhabdoms are bilaterally arranged in a folded row that penetrates the tapetum (Homann 1971), giving the shiny tapetum an appearance like that of a fireplace grate or barbecue grill. It is this com- plexly folded structure that causes the "sparkling" eyeshine of lycosids, pisaurids, and other Lycosoidea. We have not examined the morphology of the retina: if the tapetum had the appear- ance of a folded grate, resembling that figured by Homann (1971: figs. 28a-d, 32d, e), it was scored as grate-shaped. 29. Carapace shape: (0) oval; (1) square; (2) round. In dorsal view the carapaces of most taxa studied are oval to pear-shaped, with the cephalic region narrower than the thoracic (state 0; Figs. 196F, 202B, 203H, 208G). Eresids have the cara- pace anteriorly truncate, giving it a square or rectangular shape (state 1, Fig. 199B). Oecobiids have the carapace heart-shaped to nearly round in outline (state 2, Figs. 199E-F). 30. Clypeal hood: (0) absent; (1) present. A clypeal hood is a median extension of the clypeal margin over the base of the chelicerae (Fig. 129A). It is typical of austrochiloids and eresids. Absent refers to a clypeal margin that is straight or evenly curved in the center (Figs. 128C, 129B-C). 31. Pars cephalica shape: (0) not elevated; (1) elevated. Uniquely among our exemplars, in Archaea and Pararchaea the entire pars cephalica is ele- vated from the fovea to the clypeus (Figs. 127A, 195D). 32. Carapace anterior shape: (0) normal; (1) prolonged around chelicerae. In Archaea and Pararchaea the elevated pars cephalica is prolonged around the chelicerae and the bases of the elongated chelicerae are completely surrounded by sclerotized cuticle. In all other exemplars soft cuticle surrounds the chelicerae ventrally. 33. Cheliceral diastema: (0) absent; (1) present. In most representatives the bases of the chelicerae are close to the endites. Mimetus, Huttonia, Archaea and Pararchaea have an evident diastema (Fig. 128A), i. e., a wide space between the che- licerae and endites. Schiitt (2002) first noted the importance of this character. 34. Thickened setae at fang base: (0) absent; (1) present. There may be a series of thick, plumose setae present along the margins of the fang furrow. One posterior seta may be longer than the others and curves gracefully along the posterior side of the fang (Fig. 1301, 13 IF). We code this character as present if this one long, curved seta arising 52 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II on the retromargin near the fang base is present. Taxa lacking this characteristic seta are scored as absent (state 0) (Figs. 126C-D, 130A-C, 13 IB, D). 35. Male chelicerae: (0) normal; (1) bowed. The median margins of the chelicerae of most male spiders in our exemplar set are straight or slightly diverging distally (Fig. 129C). In Dictyna and Nigma the chelicerae are deeply excavate medially and bowed out laterally (Fig. 129D). 36. Chelicerae: (0) normal; (1) small. The chelicerae of most of our exemplar spiders are fairly robust (state 0 "normal") (Fig. 129C). The small chelicerae of oecobiids are weak, short and thin (state 1 "small") (Fig. 129B). 37. Medial cheliceral concavity: (0) present; (1) absent. Hypochilids have distinct concavities on the inner subbasal face of the paturon into which the fang tips fit when closed (Fig. 130B); a similar depression occurs in Desis (Fig. 130H). Such con- cavities are lacking in all other taxa in our exemplar set (Figs. 126C, 130C). 38. Chelicerae: (0) free; (1) fused at base. The chelicerae of most of our exemplar spiders are separate all the way to the clypeal margin (state 0, "free": Figs. 129B-C), and the articulation between them is membranous and flexible. Filistatid chelicerae have a stiff, sclerotized articulation near the cheliceral margin, and the patur- on bases are separated anteriorly and posteriorly by strips of pale, fleshy tissue (state 1: Fig. 126C). Mimetus also has the chelicerae with a slightly fused, stiff basal articulation (Fig. 128C). 39. Cheliceral teeth: (0) present; (1) absent. Teeth, which are non-articulate cuticular projections, are present near the fang furrow in most of our exemplar taxa (state 0). These teeth may be large (Figs. 130A-B, 131C, E-F) or small and difficult to detect, as in the eresids (Figs. 131A-B, D). Filistatids (Fig. 126C), oecobiids (Fig. 130C), Huttonia and Pararchaea lack cheliceral teeth. 40. Cheliceral chela: (0) absent; (1) present. In filistatids the apex of the paturon is extended to meet the tip of the fang when flexed, form- ing a chela (Figs. 126B-C). Our other exemplar taxa lack this modification (Figs. 126D, 130A-B). Eresidae have the apex of the paturon extended toward the tip of the fang but this is not so pro- nounced as the chela in filistatids (see Figs. 13IB, D). 41. Cheliceral peg teeth: (0) absent; (1) present. The palpimanoids, Archaea, Huttonia, Mimetus and Pararchaea, have the chelicerae armed with a series of stout, socketed setae that Forster and Platnick (1984) referred to as "peg teeth" (Figs. 127B, 128D). True teeth, i.e., conical projections of the cheliceral cuticle, may or may not be present in palpimanoids. No other taxa in our exemplar set have peg teeth. 42. Cheliceral gland mound: (0) absent; (1) present. The cheliceral gland opens through a set of pores on the paturon near the tip of the fang when at rest. The glands of the palpimanoids Archaea, Huttonia, and Pararchaea open on an area of raised cuticle or mound (Fig. 127C). Although mimetids have previously been coded as having a gland mound (Platnick et al. 1991), Schiitt (2002) suggests that mimetids lack clearly developed gland mounds. Forster and Platnick (1984: figs. 379, 381) refer to a mimetid mound, but this is a barely raised group of pores. We code mimetids as lacking gland mounds. All others in our exem- plar set have the cheliceral gland pores opening through flat cuticle. 43. Cheliceral boss: (0) absent; (1) present. The boss is a retrobasal swelling on the paturon that varies in size and shape (Figs. 129A, C-D) from a low mound to a prominent, teardrop-shaped protuberance. The boss is absent in many lower araneomorphs (Figs. 126A, 128A), even those with large chelicerae such as hypochilids, aus- trochilids, and deinopids. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 53 44. Cheliceral boss: (0) small; (1) large. We differentiate between the low mound found in some eresids and araneoids (state 0, "small": Fig. 129A), and the prominent, teardrop-shaped protuberance typical of higher entelegynes (state 1 "large": Figs. 129C-D). 45. Cheliceral stridulatory striae: (0) absent; (1) present. Some spiders may have transverse ridges on the outer margin of the paturon that presumably function as stridulatory striae. These are prominant in Archaea (Fig. 127D). Our Huttonia species have a small area with stridulatory ridges, similar to that in Archaea. Presence or absence of striae is variable in Pararchaea (Forster and Platnick 1984), so we code this genus as polymorphic. Austrochiloids also have striae. Thaida males have striae, but females have only aligned nodules. 46. Chilum: (0) absent; (1) median; (2) bilateral. The chilum is a sclerite located in the fleshy tissue at the base of the chelicerae beneath the clypeal margin. Due to the retraction of the cheliceral bases beneath the clypeus in some specimens it was not possible to score this character for all taxa. Nor was scoring possible in most cases for those taxa with clypeal hoods, i.e., austrochilids and eresids. For those taxa in which the preche- liceral region was visible we noted that there was no sclerite (state 0: "chilum absent"), a single sclerite in the middle (state 1: "chilum median") or two sclerites separated in the center by soft cuti- cle (state 2: "chilum bilateral"). 47. Tapetum: (0) primitive; (1) canoe-shaped; (2) grate-shaped; (3) absent. A tapetum, when present, occurs only in the indirect eyes: the laterals and posterior medians. Homann (1971) referred to a primitive tapetum ("primitivem Typus") in which the eye has a com- plete, even layer of light-reflecting crystals (state 0: primitive). This tapetum type is readily seen in hypochilids, austrochilids and filistatids. We have also scored the tapeta of oecobiids as primi- tive. Homann referred to Oecobius as having the primitive type but noted that the tapetum of Uroctea combined primitive and canoe-shaped features. Homann's canoe-shaped tapetum ("kahn- formigem Tapetum") has two shiny oval parts bisected by a longitudinal dark line (state 1; Homann 1971: figs. 10A, 27B, 32A). It is typical of most entelegyne spiders. The grate-shaped tapetum (state 2) has each half weakly (e.g., Stiphidion: Homann 1971: fig. 32D) to strongly (e.g., Homann 1971: figs. 12C, 32E) folded so that it resembles a fireplace grate or barbecue grill (Homann's "ros- tformigem Tapetum"). This occurs in psechrids (Homann 1971:224, 261; Levi 1982: figs. 73, 74, 88), ctenids (Homann 1971:224, 261; pers. obs.), zoropsids (Homann 1971:261; pers. obs.), and stiphidiids (Homann 1971: fig. 32D; pers. obs.). Tapeta are absent in eresids, deinopids and ulo- borids (Homann 1971:242-243). 48. Posterior eye row: (0) straight; (1) recurved. When viewed from above the posterior eye rows of Hypochilus (Fig. 195C), eresids (Fig. 199B), deinopids, uloborids, Stiphidion, Zoropsis (Figs. 208F, G) and Acanthoctenus (Fig. 208C) are recurved. All other taxa in our exemplar set have straight posterior eye rows (e.g., Figs. 203A, H, 205A, 207A). 49. Serrula tooth rows: (0) multiple; (1) single. The serrula is a group of toothlike structures situated at the ventral tip of the anterior surface of the pedipalpal coxal endites. Marples (1968) noted that there are two kinds. In hypochilids the serrula consists of a plate bearing several rows of teeth (state 0; Platnick 1977: figs. 14, 15). In all other taxa the serrula is a single row of closely spaced teeth (state 1; Platnick 1977: figs. 11, 12). We have scored this for all taxa by observation. 50. Sternal sigilla: (0) present; (1) absent. Sigilla are depressions or dimples in the cuticle of the sternum or labium that correspond to the attachments of ventral extensions of the endostemite. Marples (1968:20) reported labial and 54 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II sternal sigilla similar to those of mygalomorphs and mesotheles in the hypochilid Ectatosticta, but only a pair of labials in Hypochilus. Lehtinen (1967:300) has reported sternal sigilla in filistatids which, in Filistata and Kukulcania, are depressed, whitish areas in front of coxae III and IV. All other taxa in our exemplar set lack sigilla. 51. Venom gland: (0) present; (1) absent. Uloborids are the only spiders among our exemplars that lack venom glands (Millot 1931a). Additional scorings were taken from Millot (1931b, 1933a-c), who made carapace sections to study the midgut diverticula of spiders, and from Forster (1955) and Forster and Platnick (1984). 52. Venom gland: (0) endocheliceral; (1) extends into carapace. Petrunkevitch (1933) discovered that the venom glands of Hypochilus are confined to the che- licerae and Millot (1933b) found the same configuration in Ectatosticta. All examined non- hypochilid araneomorphs with venom glands, including the austrochilids Hickmania (Marples 1968) and Thaida (Zapfe 1955), have these glands extending into the carapace. The sources of our data are the same as for the previous character. 53. Coxal gland duct: (0) convoluted; (1) simple. The coxal glands of hypochilids have highly convoluted ducts (Marples 1968) similar to those of mygalomorphs (Buxton 1913). Other araneomorphs, including austrochiloids, have simple, inverted, U-shaped ducts. Only a few of our exemplar families have been studied, i.e., hypochilids and austrochiloids (Marples 1968), and filistatids, araneids, dictynids, agelenids, and ctenids (Buxton 1913). We scored Araneus and Dictyna after an unspecified dictynid, both from Buxton (1913), but the ctenids and agelenids that he studied may not be closely related to the terminals in this analysis and we therefore did not code our ctenid and agelenid exemplars. 54. Midgut diverticula in chelicerae: (0) present; (1) absent. Hypochilids have diverticula of the proximal portion of the midgut (the thoracenteron) that extend anteriorly into the base of the chelicerae (Millot, in Bristowe 1933; Marples 1968). Austrochiloids (Marples 1968) and the other araneomorphs studied by Millot (1931a) lack these anterior extensions. Platnick (1977) argued that the extension of the thoracenteron into the chelicer- ae was a synapomorphy for the hypochilids with acquisition of a parallel condition in liphistiids (Millot, in Bristowe 1933). 55. Pharynx dorsal Ml muscle origin: (0) carapace; (1) rostrum. The dorsal dilator muscle Ml of the pharynx of most spiders originates dorsally on the cara- pace (state 0) (Marples 1968, 1983). In hypochilids the pharyngial dilators originate on an apodeme of the rostrum (state 1), similar to Mesothelae. Our scorings are from Marples (1968, 1983). Deinopis and Filistata are scored inapplicable because the Ml muscle is absent (the anterior mus- cle M2 seems to act as dilator). ABDOMEN.? Most characters were scored by examination of exemplars. Character data taken from the literature include the suite of classical characters from abdominal internal anatomy (char- acters 57-58: Platnick 1977, ex Millot 1931a, 1933a-d, 1936; Marples 1968). As with the carapace characters, we only used data from congeneric representatives, thus many data are missing. Respiratory system data came from dissections by Lamy (1902), Forster (1970), Forster and Wilton (1973), Ramirez (2000) or Griswold and Ramirez (pers. obs.). Male epiandrous spigots were exam- ined with SEM and the condition checked in additional specimens with light microscopy. 56. Fifth median abdominal endosternite: (0) present; (1) absent. The internal anatomy of the abdomen of spiders was studied by Millot (1933a-d, 1936) and Marples (1968), who indicated that the abdomen of most araneomorphs has four ventral median endosternites (non-cuticular, internal tendonous pieces where muscles insert). Liphistius has nine GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 55 median abdominal endosternites, including the four from the anterior abdominal segments and five additional representing the primitive abdominal segmentation. Hypochilids retain the first of these additional five (state 0), the fifth abdominal endosternite (Marples 1968), where the muscle dilator of the anus inserts. The fifth endosternite is lost in the remaining araneomorphs (state 1). 57. Intestine: (0) M-shaped; (1) straight or only curved. Millot (1933b: figs. 1, 2) discovered that the opisthosomal portion of the midgut is M-shaped in Ectatosticta, Liphistius and mygalomorphs (state 0), as opposed to straight in other araneo- morphs (state 1). He did not specify the other taxa studied. He also noted (Millot 1931a:740) that the study of the abdominal intestine is extremely difficult. Millot reported (1933b:228) that the Haplogynae with globose abdomens (Scytodes, Physocyclus) have very curved, but not M-shaped intestines. Marples (1968) reported that Ectatosticta (and to a lesser degree Hypochilus) and the austrochiloids also have M-shaped midguts. Lucrecia Nieto (in lit.) reported an M-shaped intestine in the filistatid Kukulcania. This is a new coding for filistatids, differing from that in previous stud- ies (e.g., Platnick et al. 1991; Griswold et al. 1999). 58. Heart ostia number: (0) four; (1) three or two. Hypochilids (Petrunkevitch 1933) and austrochiloids (Zapfe 1955; Forster 1955; Marples 1968) have four pairs of heart ostia (state 0), whereas other araneomorphs (the Araneoclada of Platnick 1977) have three or two pairs (state 1). Petrunkevitch (1933) reported the number of heart ostia for filistatids, eresids, oecobiids, deinopids, uloborids, araneids, dictynids, agelenids, desids, amaurobiids, psechrids, zoropsids and ctenids. We scored our exemplars following Petrunkevitch (1933) and Millot (1936) when they studied congeneric representatives. Lucrecia Nieto (in lit.) confirmed the presence of three pairs of ostia in Kukulcania. 59. Third dorsoventral abdominal muscles: (0) present; (1) absent. These muscles run from sigillae on the dorsal surface of the abdomen to the ventral median endosternites corresponding to the tracheal (or posterior booklung) segment (IX segment in Millot 1933b, 1936). Many spiders have two pairs of large sclerotized sigillae on the abdominal dorsum, marking the insertions of the dorsoventral muscles corresponding to the genital and tracheal seg- ments. Scorings for this dataset were primarily taken from Millot (1933b, 1936), Crome (1955), Marples (1968), Ramirez and Grismado (1997) and Ramirez (2000). The presence of muscles was in some cases indirectly inferred from the corresponding dorsal abdominal sigillae. 60. Posterior spiracles: (0) two widely separated; (1) one narrow median. Four lunged spiders such as Hypochilus, Gradungula, and Hickmania have a pair of posterior spiracles united by a furrow (state 0), whereas those with posterior tracheae have but a single spir- acle (state 1). Thaida, Kukulcania and Filistata have similar posterior respiratory systems connect- ing to a wide furrow (and filistatines have posterior booklungs in the hatching stage, Ramirez pers. obs.) We also code these as state (0). Archaea has two separate, round openings (Fig. 127F) that we code as state (0). The general condition in Entelegynae is a narrow spiracle connecting to the tracheae. 61. Posterior booklungs and modifications: (0) booklungs; (1) reduced booklungs; (2) lateral tracheae. Hypochilus, Gradungula, and Hickmania retain typical posterior booklungs. The peculiar mor- phology in Thaida (Forster et al. 1987; Ramirez 2000) is ambiguous, with intermediate morpholo- gy between reduced booklungs and lateral tracheae, and is scored here as polymorphic (1, 2). Filistatines have reduced booklungs, with a few leaves in the hatching stage, of which only one is retained later in development (state 1), while other spiders develop the tracheae without passing through cuticular booklung leaves (Ramirez 1995, Anyphaenidae, and pers. obs., Loxosceles). In nearly all of our other exemplar taxa the posterior respiratory system comprises median tracheae 56 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II derived from the elongated entapophyses of the third abdominal segment and lateral tracheae derived from modification of the posterior booklungs (state 2) (Purcell 1909). We code this char- acter as unknown for those taxa that have lost their lateral tracheae, i.e., Archaea, Nicodamus and the dictynids Dictyna, Nigma, Lathys and Tricholathys. 62. Lateral tracheae: (0) simple; (1) branched; (2) absent (lost). Most taxa, even those with branched median tracheae, have the lateral tracheae forming a sim- ple, unbranched tube. Desids (Matachia, Desis, Phryganoporus and Badumna) are alike in having the lateral tracheae highly branched. The dictynids Dictyna, Lathys and Tricholathys have a thick median tracheal trunk with many fine lateral branches, but lateral tracheae are lacking. Lateral tra- cheae are also lacking in Nicodamus and in Archaea. Taxa with posterior booklungs are scored inapplicable. 63. Median tracheae or 3rd entapophyses: (0) muscle apodemes; (1) median tracheae. The 3rd abdominal entapophyses are cuticular muscle apodemes that occur between the pos- terior booklungs in primitive spiders with four booklungs, and between the lateral tracheae in some haplogyne spiders (Ramirez 2000). Purcell (1909) demonstrated that these apodemes develop in elongate structures with thin walls, to give the median tracheae, which in some cases still retain their connection with the abdominal muscles (Lamy 1902). Ramirez (2000) found that the median tracheae are a synapomorphy of Entelegynae (convergently with Austrochilinae). Hypochilus, Gradungula, Hickmania and the filistatids have normal muscle apodemes, and Ariadna has short muscle apodemes. All other terminals have median tracheae. 64. Median tracheae: (0) simple; (1) branched. Simple tracheae comprise unbranched tubes. We code as branched those median tracheae that have few to many branches, and both those that branch from a basal rosette and those with thick trunks giving rise to lateral branches. Huttonia has a median, unbranched tube (Forster and Platnick 1984). Aebutina has median tracheae that are flat and wide, until a point where they are abruptly truncated and then extend into a thin tube. The truncate border has the ragged texture typical of muscle attachments. We code them as simple. 65. Anal tubercle: (0) small; (1) large. The anal tubercle of most araneomorphs is little longer than the PMS (state 0: Figs. 47A, 59A). Oecobiids have a large anal tubercle, nearly as long as the elongate PLS and much larger than the PMS (Figs. 27A, 30A). Deinopids have an anal tubercle that extends beyond the PMS (Fig. 44A), but is not nearly as large as in oecobiids. 66. Epiandrous spigots: (0) present; (1) absent. Epiandrous spigots, which presumably serve in making the sperm web, occur in a variety of taxa (Figs. 160A-B, 161A-E). Males of many taxa lack them (Figs. 160C-D, 161F-H): how or if they construct sperm webs is unknown. 67. Epiandrous spigot distribution: (0) dispersed; (1) in two bunches. Most phyxelidid males have the epiandrous spigots grouped into two separate bunches (Figs. 160A-B), as do Thaida (Fig. 158A) and Hickmania, Callobius and Amaurobius, and Pararchaea. In other males with epiandrous spigots these are scattered along the margin of the epigastric fur- row (Figs. 161A-C, E). SPINNERET MORPHOLOGY.? Spigots were classified according to the criteria in Coddington (1989). Spigots are named for the glands that they presumably serve. Glands and spigots have been associated in several key taxa. Although the glands of most of our exemplar taxa have not been studied, we think that reasonable inferences about their gland types can be made from the struc- ture, number, distribution and ontogeny of spigots. Piriform gland spigots (PI) occur as multiples GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 57 on the ALS of both males and females. Ampullate gland spigots (MAP and mAP) are typically present as singles or as a few on the ALS and PMS in both males and females, and if one is miss- ing in a male it is replaced by a nubbin. Paracribellar spigots (PC) may occur on the PMS, PLS, or both, and have a characteristic annulate shaft similar to that of cribellar spigots. These may be pres- ent in both sexes of "lower" araneomorphs, but are in most cases replaced by nubbins in the males of higher araneomorphs. The PLS modified spigot (MS) (e.g., "pseudoflagelliform" gland spigot of Kovoor [1977b]) may occur on the PLS of both sexes, or if absent in the male, is replaced by a nubbin. Aciniform gland spigots (AC) occur as multiples on the PMS and PLS of both sexes. Cylindrical gland spigots (CY) occur only in adult females, where they may be found on the PMS and PLS, and are not represented by nubbins in males. We also suggest that shaft morphology is a more reliable indication of spigot type than base morphology. We follow Tillinghast and Townley (1994) and Townley and Tillinghast (2003) in defining nubbins and tartipores. All specimens were critical point dried before scanning electron microscope (SEM) examination of spinning organs. 68. Ampullate spigot shaft texture: (0) smooth or longitudinally striated; (1) with papillae; (2) concentrically striated. Most of our representatives have smooth or longitudinally striated MAP and mAP shafts (Figs. 65B-C, 71C). Eresids are unique in that their ampullate shafts have small papillae or imbricate pro- trusions (state 1, Figs. 33C, E, 37C). Filistatids are also unique in having all spigot shafts with faint concentric ridges (state 2, Figs. 3E-F, 5F). 69. Spigot base texture: (0) all squamate; (1) only PMS and PLS squamate; (2) all smooth or with "fingerprint" pattern. Most of our representatives have smooth or finely striated spigot base cuticle (state 2, Figs. 65B-C, 71C). Hypochilus has scale-like annuli encircling all the spigot bases (state 0, Figs. 1A-D, 2C-D). Deinopids and Nicodamus have some spigot bases with similar squamate-annulate texture, but only on the posterior spinnerets (state 1, Figs. 41C-D, 43C-D, 44E). 70. Tartipores: (0) absent; (1) present. Tartipores were first described by Kovoor (1986). We follow the terminology of Townley and Tillingast (2003) for nubbins and tartipores. They define a tartipore as "a cuticular scar, morpho- logically singular or multiple, that results, after ecdysis, from a collared opening forming in the exoskeleton during proecdysis; the opening accommodates a silk gland duct, allowing it to remain attached to a spigot on the old exoskeleton during proecdysis." This mechanism allows spiders to use silk during the proecdysis, through gland ducts that pierce the forming cuticle to remain attached to the shedding spigots (Townley et al. 1993, Tillinghast and Townley 1994). A direct pre- diction is that hypochilids and filistatids (as well as most haplogynes), which lack tartipores (Figs. IB, 5F), are unable to produce silk in the days before moulting. Similarly, spiders should be able to use, during proecdysis, only the basic spigot types with tartipores (namely MAP, mAP, PI and AC, but not MS, PC, or cribellum). Tartipores typically have a dimple, scar, or crease (Figs. 9D, 10D). Tartipores of the MAP and mAP are larger than PI or AC tartipores. The ALS of higher entel- egynes have a basic complement of two MAP (which may be replaced by a nubbin in the adult), accompanied by one MAP tartipore (Figs. 13C, 86B). The same pattern generally occurs with the mAP in the PMS, but we found more variation there. 71. Cribellum: (0) present; (1) absent. Most of our exemplars are cribellate (Figs. 2A, 47A) but Palpimanoidea are ecribellate (Fig. 24A). We include a few other ecribellate taxa, i.e., Gradungula, Araneus, Nicodamus, and Uroctea (Fig. 30A) that may exemplify character combinations more representative of their higher taxa, and which offer clearer evidence for taxonomic placement. The cribellar spigots are lost in males when they reach maturity. 58 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 72. Cribellum: (0) entire; (1) divided. Hypochilus, austrochilids, Orbiculariae, most dictynids and several other taxa have the cribel- late spinning field entire with the spigots spread evenly across (Figs. 2A, 10A, 38D, 43A). Divided cribella have a bare line down the center, dividing the spinning field into two lateral parts (Figs. 5A, 97B-C). 73. Cribellate spigots: (0) uniform; (1) clumped. In most taxa, whether the cribellum is entire or divided, the cribellar spigots are uniformly spread across the spinning field (Figs. 34C, 97E). In Acanthoctenus and some zorocratids these spigots are clumped into tightly packed groups surrounded by bare cuticle (e.g., Acanthoctenus: Figs. 97A, G; Uduba: Figs. 97C, F), which may form short, longitudinal segments (Acanthoctenus, Raecius). Zoropsis has two transversal bands of spigots isolated by bare cuticle (Figs. 113C-F), in similar disposition as the aligned clumps of Acanthoctenus; we code Zoropsis ambiguous for this character. 74. Cribellate spigots: (0) strobilate; (1) claviform. Strobilate cribellar spigots are cylindrical to tapering at the tip and ringed by evenly spaced, raised annular ridges (Figs. 34C, 96B, 97E). Claviform cribellar spigots, typical of the Filistatidae, are club-shaped, blunt at the tip, and have fine transverse ridging (Figs. 5B-C, 14E). 75. ALS segment number: (0) three; (1) two. Primitive Araneomorphae retain the intermediate ALS segment as an incomplete lateral ring (Figs. 2A, 32A). Other Araneomorphae have lost this segment (Fig. 76A). 76. ALS MAP: (0) clustered; (1) dispersed. One or a few ampullate glands serve the ALS and PMS. The large spigots served by these glands are called "major ampullate" (MAP) on the ALS and "minor ampullate" (mAP) on the PMS. They can be distinguished from other spigots on these spinnerets by their large size, usually squat base, and small number. If present in females but absent in males they are typically replaced by a nubbin. In most taxa the MAP are clustered along the inner margin of the spinning field ("clus- tered": Figs. IB, 43A, 113B), but in filistatids, eresids and oecobiids some MAP spigots are inter- spersed among the piriform spigots ("dispersed": Figs. 6B, 29A, 35B). We score this for taxa with a single MAP if it is accompanied by a nubbin, but code it "?" for those have have only one unac- companied MAP, i. e., Oecobius and Pararchaea. 77. ALS MAP field separated by deep furrow: (0) absent; (1) present. In most Araneomorphae the mesal MAP field is separated from the PI field by more or less flat or weakly folded cuticle (Figs. 67B, 104B). In Araneoidea, Mimetus and Archaea, the MAP field is separated by a deep furrow (Figs. 20E-F, 21A, 25B). Huttonia has only an indentation between both fields (Fig. 24B). 78. ALS MAP number in female: (0) more than three; (1) three; (2) two, or one plus a nubbin. Hypochilus, Gradungula, eresids and deinopids have four or more MAP (state 0, Fig. 43B); filistatids have three (state 1, Fig. 6B); most of our exemplars have a consistent pattern (state 2) of two marginal MAP (Figs. 7B, 49B, 53B) or only one, in that case accompanied by a posterior nub- bin (e.g., Uloborus, Fig. 45B; dictynids, Figs. 59B, 61B, 63B, 65B; Retiro, Fig. 92B). Except in deinopids and Uroctea, the other taxa with more than one MAP have two mesal, anterior MAP larg- er than the rest (Fig. 2C). These are presumably homologues of the two MAP that remain in high- er entelegynes. Ariadna, Oecobius, Desis and (presumably) Pararchaea are the only representa- tives with only one MAP and no nubbin (Fig. 27B), so we code them as nonapplicable for this char- acter. 79. ALS female posterior MAP: (0) functional; (1) reduced to nonfunctional nubbin. There has been extended confusion between nubbins and tartipores, but Townley and Tillingast GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 59 (2003) recently presented a coherent terminology. Nubbins are nonfunctional, vestigial spigots, usually in the form of protuberances, in the place where a spigot occurred in the previous instar. They lack the characteristic cuticular scar found in tartipores. Nubbins of MAP only occur in high- er entelegynes that have only two MAP We can identify that it is the posterior MAP that is consis- tently reduced in the adults. The posterior MAP is the secondary one, that is, the one that remains functional during proecdysis, and produces the tartipore (Townley et al. 1993). The posterior MAP in adult Archaea is smaller than the anterior one (although still has a shaft, Figs. 21B, 22B), but both MAP are well developed in the immature (Fig. 20B), thus we interpreted the vestigial adult posterior MAP as a nubbin. We scored this character as inapplicable in Oecobius and Pararchaea, which seem to have only one MAP. 80. ALS piriform margin: (0) rounded; (1) flat to sharp. Piriform glands serve multiple spigots on the ALS. These spigots extrude the glue that attach- es silken lines. The junction of the base and shaft may be gradual and rounded (Fig. 83B) or the junction may be abrupt, with the edges of the base at the junction forming a shelf or ledge ("flat to sharp": Figs. 58B, 84B). We score this as nonapplicable in Mimetus and Pararchaea, which have short, reduced bases (Fig. 25B). 81. PMS mAP: (0) absent; (1) present. Like the MAP on the ALS, minor ampullate spigots (mAP) occur as one, two, or a few on the PMS (Figs. 40C, 92C). They tend to have squat bases and slender, tapering shafts (e.g., Fig. Ill A). Unlike cylindrical gland spigots (CY), these large spigots occur in both females and males or, if absent in males, are replaced by nubbins. Among our exemplars only Hypochilus (Fig. 1C), Gradungula, Huttonia and Uroctea (Fig. 29B) lack mAP 82. PMS mAP number in female: (0) more than two; (1) two; (2) one plus nubbin; (3) one. The definition of nubbins and tartipores is the same as for the MAP. Many of our exemplar taxa have a single mAP on the PMS (Figs. 40C, 92C). The lycosoids and their kin have two (Figs. 98C, 104B), as do the austrochilids, Deinopis, Pimm, and the enigmatic Aebutina. Araneoids, ulo- borids, Mimetus (Fig. 25C) and some phyxelidids (Fig. 49C) have one mAP plus a nubbin. According to Kovoor and Lopez (1979) eresids have numerous ampullate gland spigots opening on the PMS (Fig. 34E). 83. PMS mAP position: (0) median to anterior; (1) posterior. The mAP is located in the median to anterior part of the PMS spinning field in most of our exemplars (Figs. 59C, 92C). A posterior mAP is found in the Araneoidea, and in Mimetus, Megadictyna (Fig. 40C) and some dictynids (Fig. 61C). 84. PMS aciniform number: (0) more than three; (1) one to three. Most of our exemplar taxa have numerous AC spigots on the PMS (Figs. 49C, 60C) but in a few species these are reduced in number to three or fewer (Figs. 64C, 74C, 85C, 113F). Despite the suggestions of Kovoor and Lopez (1979), ontogeny compels us to code aciniform gland spigots present in eresids. 85. PMS-PLS AC shaft size: (0) uniform; (1) two size classes. This character accounts for the finding in Gradungula, Hickmania, Uroctea, Nicodamus, and Tengella of an uncertain type of spigot on male and female PMS and PLS, which we tentatively describe as a second class of AC (Figs. 17E-F, 30C). These larger spigots generally occur on the margins of the spinning fields, but in Hickmania there is also a median line of large spigots on the PLS (Figs. 9A-D). In Nicodamus the larger spigots form a central group in the PMS (Figs. 41C, 42C). 86. PMS cylindrical gland spigots: (0) absent; (1) present. Cylindrical glands serve the PMS and PLS and produce silk used in forming the eggsac. 60 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Therefore, they occur in adult females but are absent from juveniles and males, even as nubbins. In the absence of gland data we rely on ontogeny to identify CY spigots. If a class of spigots occurs on the PMS and PLS of females but not males, we assume these to be CY. Cylindrical gland spig- ots tend to have slender, tapering bases and shafts, and resemble aciniform gland spigots except in being larger and less numerous (Figs. 49C, 63C, 113F). Cylindrical gland spigots were considered as characteristic of the Entelegynae (Platnick et al. 1991), but we report them from the aus- trochiloids Thaida and Hickmania. They are absent in hypochilids (Figs. 1C-D) and Haplogynae including filistatids (Fig. 6C), but putative homologs of CY occur in telemids and leptonetids as well (Platnick et al. 1991). Some entelegyne taxa also lack apparent CY spigots, e.g., Matachia (Figs. 83C-D). We are unable to determine if CY spigots occur in Megadictyna, and therefore code this and other CY characters as unknown. 87. PMS cylindrical gland spigot number: (0) one or two; (1) many. Most of our exemplar taxa with CY spigots have one or two on the PMS (Figs. 53C, 61C, 88C). We code three or more as "many". The phyxelidids Vytfutia and Xevioso have three (Fig. 47C). According to Kovoor and Lopez (1979), eresids have numerous CY glands serving the PMS. Thaida (Fig. 11C) and Hickmania (Fig. 7C) have several PMS CY and deinopoids have several large CY posteriorly on the PMS (Fig. 43C). Zorocratids and Lycosoidea have numerous large CY spigots posteriorly on the PMS: three or four occur in Raecius (Fig. 105C), four in Acanthoctenus (Fig. 115C), and eight or more occur in Psechrus (Fig. 109C), Uduba (Fig. 104C) and Zoropsis (Fig. 113C). 88. PMS paracribellars: (0) absent; (1) present. Spigots with shafts that resemble cribellar spigots may occur on the PMS and/or PLS (Peters and Kovoor 1980). Their function may be to produce fine fibrils that accompany the axial fibrils in the capture threads (Peters 1984; Peters 1987: fig. 65a "Is"). Paracribellars on the PMS occur widely in the Neocribellatae (Figs. 11C, 45C-D, 71C) and are absent in the hypochilids (Fig. 1C) and sporadically absent among the neocribellates (Figs. 27C, 51C, 92C, 113F). They are present in females but in most taxa are replaced by nubbins in males (Figs. 48C, 76C). We code ecribellates as nonapplicable for this character. 89. PMS paracribellar spigots in male: (0) present; (1) absent. The adult males of basal Araneomorphae, i.e, filistatids and austrochilids, retain the PC spig- ots (e. g., Filistata, Fig. 4C, Hickmania, Fig. 8C, Thaida, Fig. 12C), while in our representative entelegynes the PC are reduced to nubbins in the males (Figs. 40C, 50C, 74C). There are many missing entries for this character both because PC are absent in both sexes (i.e., nonapplicable) and because the males of some exemplars have not been scanned (i.e., unknown). The PC shaft found in the male of Neoramia (Fig. 82B) is here regarded as abnormal. Even isolated cribellar shafts are occasionally found in males (Fig. 158C). 90. PMS paracribellar form: (0) strobilate; (1) floppy. In the majority of taxa with paracribellar spigots their shafts resemble those of the cribellar spigots. Most of our exemplars have strobilate PC shafts, with widely spaced, annular ridges (Fig. 82D) (state 0). Previously we recognized a class of paracribellar spigots that have long shafts with many, closely spaced annuli and that are typical of the Deinopoidea (Figs. 45C-D) (Griswold et al. 1999: character 82, "deinopoid"). Similar PC shafts occur in austrochilids, e.g., Thaida (Figs. 11C, 12C) and Hickmania (Fig. 10E, Platnick et al. 1991: fig. 45). We no longer recognize separate stro- bilate and deinopoid states, as some taxa (e.g., Megadictyna, Fig. 38B) have an intermediate mor- phology. The apparent differences in annuli number and position may relate only to the length of the shafts. Filistatid PC morphology is as unique as that of their cribellars (Figs. 3C, F). On the PMS and PLS of filistatids there are peculiar spigots with cylindrical bases and flattened, trans- GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 61 versely ridged, "floppy" shafts. The shafts of these spigots may be pointed or claviform (Fig. 4C) (state 1). 91. PMS paracribellar distribution: (0) at anterior margin of spinning field; (1) mid-field; (2) at posterior margin of spinning field. Among most taxa with PMS paracribellars these occur along the anterior margin of the PMS spinning field (Figs. 49C, 63C, 83C). In the agelenid Neoramia (Fig. 73C), among the amphinec- tids (Fig. 77C) and in some desids (Fig. 87C) and dictynids the PC spigots occur at mid-field, sur- rounded by AC spigots. Uniquely, the PMS paracribellars of filistatids arise at the posterior mar- gin of the spinning field (Figs. 3C, 4C). 92. PMS paracribellars: (0) bunched; (1) encircling spinneret anteriorly. Austrochiloids, phyxelidids, and some other taxa have paracribellar spigots that may be regu- larly spaced in a row that encircles the PMS anteriorly (Figs. 7C, 11C, 40C, 49C, 60C). In other taxa the anterior PC are irregularly arranged or bunched together (Figs. 43C, 83C). 93. PMS paracribellar shafts: (0) arising from single bases; (1) grouped. The paracribellar shafts of most taxa arise from single bases (Figs. 45D, 47C). In stiphidiids, neolanids, amphinectids, and some desids and dictynids several PC shafts may arise from a single, greatly enlarged base (Figs. 66A, 68B, 73C, 87C). We code this striking morphology as "grouped" or "fused." This character defined in part the "fused paracribellar clade" of Griswold et al. (1999, Fig. 212). In some taxa both single and grouped PC shafts may be found in the same individual, e.g., Dictyna (Fig. 59C). We also code these as grouped. Abnormal, duplicate shafts are occasion- ally found in spiders (Fig. 111A), hence we have coded Callobius PMS PC (Fig. 96C) as arising from a single base. 94. PMS paracribellar bases: (0) cylindrical; (1) long, narrow, flattened. The encircling PC of phyxelidids are so tightly packed that the bases are laterally flattened to fit together (Figs. 46C, 47C). Other taxa with single PC bases, whether bunched or encircling, have the bases well separated and round in cross section (Fig. 63C). This character is nonapplicable in those taxa for which all PC shafts arise from one or a few common bases. 95. PLS aggregate gland (AG) spigot: (0) absent; (1) present. In Araneoidea the paired aggregate gland spigots (Figs. 38C, E) flank the flagelliform gland (FL) spigot and coat the FL fiber with sticky glue as it is spun (Figs. 119C-E). Aggregate glands have been observed in all araneoid families for which gland histology has been studied (Kovoor 1977a), but are absent from other spiders (Figs. 43D, 44B). 96. PLS modified spigot: (0) absent; (1) present. At the apex of the PLS in many spiders are one or more enlarged spigots that differ from the AC, CY and PC. Typically they are larger than the nearby AC spigots and have a thick, cylindrical or even clavate shaft (Figs. 68A, D). They are present in females but often replaced by a nubbin in males (compare Figs. 75D and 76D). Glands serving these spigots were first identified in uloborids by Kovoor (1977b), who called them "pseudoflagelliform," and were subsequently found in amau- robiids, eresids, psechrids and zoropsids, but not in filistatids and dictynids. These glands presum- ably produce the axial fibers in the cribellate band (Peters 1984). Nevertheless, classifying the spig- ot type by presumed gland morphology may present a dilemma. For example, pseudoflagelliform glands have not been found in dictynids, and it is therefore not surprising that their cribellate silk does not have axial fibers. But some female dictynids do have an enlarged spigot at the apex of their PLS (Fig. 59D) that is replaced by a nubbin in males (Fig. 60D). We divorce classification of this spigot from data on gland types and refer to these simply as "modified spigots" (MS). We sug- gest a broad homology between the pseudoflagelliform gland spigot of Deinopoidea, flagelliform gland spigot of Araneoidea, and this peculiar spigot type in other spiders. Typical MS occur wide- 62 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II ly (Figs. Fig. 43D, 46B, 82E, 88D, 113G). Hypochilus has a set of three or four large spigots with cylindrical shafts at the apex of its PLS (Fig. ID). We code these as MS although the morphology is unique. The MS is often flanked by one to three spigots, forming a compact group; in many taxa this group is a triad, made of the MS and two smaller spigots, one at each side. This triad is an addi- tional guide to identify the MS in those terminals. In the taxa that possess PC, the flanking spigots in the triad are PC (Figs. 82C, E); in others, they are similar to AC spigots, but differ in being replaced by nubbins in the male (compare Figs. 33D and 33F; 98D and 99F; 100G and 114D). These accompanying spigots similar to AC seem to belong to a further class of spigots for which glands have not yet been identified, because AC do not degenerate into nubbins. We have not attempted to name these spigots. Interestingly, the AG spigots of araneoids also flank the MS form- ing a triad, suggesting that there may be a broad homology of the otherwise different spigot types accompanying the MS. If that is the case, our character codings involving the PLS-MS triad (PC and MS presence, disposition, and number) only reflect imperfectly these relations. In eresids the MS is anterobasal and separated from the rest of the PLS spinning field (Figs. 3 ID, 33D, J, 37D) but the spigot is otherwise ontogenetically like typical MS. Aebutina has a typical apical MS but also two anterobasal spigots that are replaced by nubbins in the male (Figs. 57D, 58D), which we interpreted as the two flanking spigots of the triad that are displaced to a basal position. Another possible interpretation, which we considered in previous versions of this manuscript, is that all the isolated, basal spigots of eresids and Aebutina are also MS. The MS is absent from filistatids (Figs. 3D, 6D), palpimanoids, oecobiids (Fig. 27D), and some dictynids and titanoecids. Although Coddington (1990a) codes the MS present for Titanoeca, and the male of Titanoeca has what could be a MS nubbin (Fig. 52D), we can't identify an MS on the PLS of any Titanoeca or Goeldia that we have examined. 97. PLS modified spigot position: (0) at margin of spigot field; (1) segregated. The modified spigots (MS) of most taxa are located at the margin of the PLS spinning field (Figs. 46B, 105D). Among the Orbiculariae the MS is laterally separated from the spinning field (Figs. 38E, 43D, 44B), as is the case in Thaida (Fig. I ID). The eresid MS is distant from the rest of the spinning field. Although Aebutina has one MS in the spinning field, the two flanking spig- ots of the triad seem to be segregated, therefore we code Aebutina as uncertain (see below). 98. PLS MS position: (0) at field margin or only slightly segregated; (1) basal anterior. In eresids the MS is well separated from the rest of the PLS spinning field (Figs. 3 ID, 37D). At least the Eresus MS is ontogenetically like that of other spiders in occurring in females but being reduced in males (Fig. 32D). In Stegodyphus the MS produces the same cribellar axial threads as in Deinopis (Peters 1992a), and is accompanied by two small spigots (Fig. 33D, F) that are reduced to nubbins in males (Fig. 33F). Aebutina has the MS in the spinning field, but the pair of accom- panying spigots is apparently segregated (Figs. 56B, 57D). 99. PLS paracribellar: (0) absent; (1) present. Spigots with shafts that resemble cribellar spigots may occur on the PLS. They may function like the PC on the PMS to produce fine fibrils that accompany the axial fibers in the capture threads. Hypochilids (Fig. ID) lack PLS paracribellars and in the Neocribellatae PLS PC are spo- radically present (Fig. 87D) or absent (Fig. 98D). Surprisingly, occurrence of PLS PC does not par- allel that of PMS PC. Some taxa with PMS PC lack these on the PLS (Fig. 44B), and, remarkably, titanoecids lack PMS PC but have them on the PLS (Fig. 53D). They are present in females but are typically replaced by nubbins in males (Fig. 76D). 100. PLS paracribellar number: (0) one; (1) two or three. Megadictyna, Xevioso and dictynids (Fig. 59D) have a single PLS PC, whereas two or more occur in filistatids, titanoecids (Fig. 5 ID), some amaurobiids (Fig. 96D), Neolana, stiphidiids, GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 63 amphinectids and some desids (Figs. 82E, 87B). Thaida has one unambiguous PLS PC and a sec- ond spigot of intermediate morphology (Fig. 13E). The male has two apparent PC nubbins on the PLS (Fig. 12D), so we code Thaida as having two PLS PC. 101. PLS PC distribution: (0) apical only; (1) with an additional basal external group. In most representatives the PLS PC spigots are apical and are situated close to the MS (if this spigot is present), often one at each side (Figs. 68D, 88D, 96D). In titanoecids, besides an apical PC, there is a basal external group of three PC, separated from the rest of the PLS spinning field (Fig. 55D). 102. PLS paracribellar and modified spigot: (0) separate; (1) united. In most taxa that have both a modified spigot (MS) and PC on the PLS the PC flank the MS (Figs. 59D, 96D). In Badumna (Fig. 87D), Phryganoporus (Fig. 87B) and Amaurobius (Fig. 88D) one PC spigot shares a common base with the MS. 103. PLS apical segment: (0) domed to conical; (1) elongate. Most of our exempars have PLS apical segments that are domed (Fig. 4D), conical (Figs. 40D, 44A, 69D, 72D) or flattened (Figs. 7A, D, 11 A). The PLS apical segments of oecobiids are elon- gate (Figs. 27A, D, 30A). This was character 92 in Griswold et al. (1999) except that domed and conical are no longer distinguished. Upon including more taxa we found that we could no longer code domed and conical segments as distinct states. 104. PLS apical enlarged seta: (0) absent; (1) present. Among the phyxelidids the Phyxelidini and Vidoliini have a characteristic, stout seta laterally on the PLS (Fig. 49D). We did not find this seta in Vytfutia, though it may have been broken off in our specimen. Such setae are lacking in our other exemplars (Figs. 64A, 69D). MALE GENITALIA.? Some past phylogenetic studies of spiders, e.g., Coddington (1990b), considered any apophysis on the male palpal tibia to be homologous to the RTA (retrolateral tibial apophysis). Our coding recognizes that a single tibia may have apophyses on as many as four sur- faces (prolateral, dorsal, retrolateral and ventral) and that these may occur in a great variety of com- binations. We therefore divide the male tibial apophyses into four homology hypotheses (pro-, dor- sal, retro- and ventral) and further subdivide some of these in terms of the morphology of the apophysis (e.g., simple or complex) or its origin (e.g., proximal or distal). With regard to apophy- ses on the tegulum of the male palp, we have arbitrarily chosen to allocate homoplasy in our dataset to the median apophysis rather than to both it and the conductor (Coddington 1990a, Griswold et al. 1998). The embolus is readily identifiable. We code three basic homologies on the bulb in addi- tion to the embolus. If a second apophysis is present, we code it as the conductor (C) (see Gradungula, titanoecids and Aebutina for exceptions), if a third, the median apophysis (MA), and if a fourth, we code this as an "extra tegular apophysis" (TA) (see characters 126 and 127). We fur- ther subdivide each of these homologies in terms of shape, texture, and/or position of origin. 105. Retrolateral tibial process (RTA): (0) absent; (1) present. The retrolateral surface of the male palpal tibia may be unarmed (Figs. 166A, 187A) or there may be a spur or projection (RTA: Figs. 175C, 178B, 189B). Although the psechrid Fecenia has an RTA, Psechrus lacks this (Figs. 167A-B) and we code the RTA as absent. 106. Retrolateral tibial process form: (0) simple; (1) complex; (2) desine incised blade. The retrolateral surface of the male palpal tibia may have a simple conical spur or blade (state 0; Figs. 175C, 183D, 189A), or two or more (up to four) spurs may arise from the same region (state 1; Figs. 178B, 189B). Desines have a particular configuration of a relatively simple blade, with a deep incision delimiting a narrow more dorsal process (state 2; Fig. 177E; Forster 1970: fig. 52). 64 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 107. Ventral tibial process (VTA): (0) absent; (1) present. The ventrolateral surface of the male palpal tibia may be unarmed (Figs. 166A, 178C) or there may be a spur or projection (VTA: Figs. 179F, 182A, D, 192A-C). We also consider the large, rounded bumps in Araneus and Mimetus as potential homologues. 108. Dorsal tibial process (OTA): (0) absent; (1) apical; (2) proximal. The dorsolateral surface of the male palpal tibia may be unarmed (Figs. 166A, 177A) or there may be a spur or projection (OTA). That process may be near the apex of the segment ("apical": e.g., Phyxelida, Figs. 173A-B, Titanoeca, 174C) or arise near the base ("proximal," e.g., Megadictyna, Figs. 171A-C, Nicodamus, Figs. 172A-C, Dictyna, Figs. 176D-E, Maniho, Fig. 180C). The proximal process may be simple (Fig. 178E) or surmounted by stout setae (Fig. 176B). 109. Dorsal tibial process form: (0) simple; (1) complex, folded. The dorsal process may be simple (Fig. 178E) or inrolled (Fig. 173B) (state 0: "simple") or be complexely folded (state 1: Figs. 174C, E) as in the Titanoecidae. 110. "Nicodamid" OTA (0) absent; (1) present. The nicodamids Megadictyna (Figs. 171A, C) and Nicodamus (Figs. 172A-C) have a unique type of DTA that arises basally and forms a broad, sweeping curve. 111. Prolateral tibial process (PTA): (0) absent; (1) present. The prolateral surface of the male palpal tibia may be unarmed (Figs. 166A, 172B) or there may be a spur or projection (PTA: Figs. 18IB, 182A, 193A). 112. Paracymbium: (0) absent; (1) present. The paracymbium is a process arising from the cymbial margin. Absent from most entelegy- ne families (Figs. 171B, 175C, 182E), the paracymbium is characteristic of the Araneoidea (Figs. 171E-F). Mimetus (Fig. 169A) and Pararchaea also have paracymbia. We also code a projection from the cymbial margin of Hypochilus as a paracymbium (Fig. 166A). 113. Cymbial dorsal scopula: (0) absent; (1) present. The dorsum of the cymbium may be clothed with ordinary setae with some interspersed chemosensory setae, similar to leg segments (Figs. 173A, 174C), or may have a dense patch of chemosensory setae (Figs. 167A, 185F). 114. Piriform bulb: (0) absent; (1) present. We code as piriform a palpal bulb that has the tegulum and subtegulum fused, lacks process- es other than the embolus, and is spindle-shaped (Figs. 166D, 167D). This type is characteristic of the Filistatidae. Other bulbs have additional processes. Hickmania, although lacking the conductor and median apophysis, does not have a spindle-shaped bulb and retains clear demarcation between tegulum and subtegulum: we do not code it as piriform. 115. Subtegular locking lobe: (0) absent; (1) present. In some taxa the subtegulum has a promarginal lobe on its dorsolateral surface which inter- locks with a corresponding lobe on the tegulum (Figs. 186F, 194C) in the unexpanded bulb. At least some taxa with a tegular lobe have no corresponding subtegular lobe (e.g., Amaurobius). 116. Tegular locking lobes: (0) absent; (1) on embolus base; (2) on tegulum, far from embolus base. In some taxa the embolar base has a promarginal lobe on its dorsolateral surface, which inter- locks with the subtegulum (Fig. 186F) in the unexpanded bulb (state 1). This lobe may interact with a corresponding lobe on the subtegulum (e.g., in Tengella, Raecius, Zoropsis), but there is no cor- responding subtegular lobe in Amaurobius, Macrobunus and Zorocrates. In tengellids, zorocratids, zoropsids and ctenids with embolus fused to the tegulum, this lobe is near the embolic base. In Zorocrates (Fig. 186A), which has the embolus articulated by a hematodocha, the locking lobe is a process at the base of embolus; a similar disposition occurs in Uliodon (Zoropsidae), GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 65 Liocranoides (Tengellidae), and Xenoctenus (Zoridae), among others (Ramirez, pers. obs.)- A sim- ilar lobe, though far from the embolic base (state 2), occurs in Nicodamus (Figs. 172A, D). 117. Tegular groove acting as conductor: (0) absent; (1) present. In titanoecids the embolus rests in a complex groove in the tegulum (Figs. 174D, 188A-B). 118. Conductor (C): (0) present; (1) absent. If a palpal bulb has one apophysis in addition to the embolus we consider it the conductor. If the bulb has two or three apophysis in addition to the embolus we consider the one that is most closely associated with the embolus to be the conductor. A conductor is present in almost all of the taxa treated here (Figs. 166A, 175A, 182B, 189D). Hickmania and filistatids (Figs. 166D, 167D) clearly lack a conductor. We make coding exceptions for the titanoecids, Gradungula, and Aebutina. In titanoecids the embolus rests in a groove in the tegulum (see character 117). Whereas this is not clearly a "process" we code this tegular groove as the conductor for this family. Additional titanoecid palpal bulb processes are coded as a median apophysis and additional tegu- lar process (see below). Forster et al. (1987) code the small process on the Gradungula bulb as a median apophysis, not as a conductor, and we follow this interpretation. The tegulum of Aebutina has only a flexibly attached process that arises far from the embolus and resembles the median apophyses of other exemplars (Figs. 169C-D). In this case we code the conductor of Aebutina as absent. 119. Conductor position: (0) subterminal; (1) terminal; (2) central. Hypochilus (Fig. 166A), Archaea (Fig. 168A, D), Huttonia, Pararchaea, and the eresids (Fig. 170D) have the tegular sclerites, which we code as conductors, at the apex of the bulb (state 1). In most other exemplars that have conductors, these originate laterally, or basally on the bulb (state 0). A few terminals have a simple, median lobe in the center of the bulb, which is coded as "cen- tral" (state 2: Figs. 171B, D, 178D). 120. Conductor and embolus: (0) separate; (1) conductor embraces embolus. Many conductors wrap, cradle or embrace the embolus for part or all of its length (Figs. 166B, 170D, 172 A, 175 A, 179D-E, 189D). 121. Conductor types: (0) sclerotized, relatively simple; (1) dictynid posteriorly directed; (2) desid-amphinectid type, hiding most embolus; (3) complex, large, with many processes; (4) hya- line. We have distinguished a few morphological types of conductors, among the wide variety of known shapes, dispositions and textures of conductors. Sclerotized conductors that are little differ- ent in texture and color from the cuticle of the bulb from which they arise, and are relatively sim- ple, are all lumped in state 0 (e.g., Figs. 166B-C, 170B, 171D-E, 179B, 180A, 187A-B). We have indentified the conductor in Archaea as the sclerotized ridge that spirals around a central pit that contains the embolus and median apophysis (Fig. 168D). Huttonia has an almost piriform bulb. The embolus is a short apical spine accompanied by a small pointed process that we code as a sim- ple conductor (Forster and Platnick 1984: figs. 350-352). The "dictynid" type (state 1) has a long groove that embraces most of the embolus, and the apex is characteristic in turning back proximad, in most species extending proximad of the base of the bulb (Figs. 175A-B, D); the conductor apex may be straight, curved, or form a spiral. We call the "desid-amphinectid" type of conductor (state 2) one that is hypertrophied and accompanies the embolus from its origin, is highly convoluted, and with a membranous furrow that hides the embolus for much of its trajectory (Figs. 178A, C); extreme developments of this kind of conductor occur in Desis and Metaltella (see character 122). The conductors of araneids (Fig. 171E), mimetids, Pararchaea, uloborids, and oecobiids are large, complex articulate structures with many processes (state 3). Oecobiid palpal bulbs have 3^1 processes in addition to the embolus; by default one of these is coded as conductor (one of the TA's 66 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II in Uroctea, Figs. 170A or 187B, or in Oecobius, Fig. 170C). The tegulum of Mimetus has several sclerotized processes including an anterior ridge with groove for the embolus, a subapical hook, two small median teeth, and two broad retrolateral flanges (Figs. 169A-B); one of these TA is the conductor. The apex of the Pararchaea tegulum has a "complex distal plate" (Forster and Platnick 1984: fig. 237) with two flanges, a hook, and a broad scaly surface. The hyaline conductor (state 4) presents a characteristic, stereotyped morphology. It is membranous in texture, transparent or translucent, strongly contrasts to the texture of the nearby tegular cuticle (Fig. 193A), is fan- shaped, arises from the retroapex of the bulb, and opposes the embolic tip (Fig. 185E). Psechms has a sclerotized conductor but Fecenia, another psechrid, has a hyaline conductor like that of other Lycosoidea. 122. Embolus origin: (0) exposed; (1) internal to complex conductor. In Metaltellinae and in Desis the conductor is greately developed, forming much of the exposed structure of the copulatory bulb. The origin of the long embolus is entirely hidden in intri- cate membranous loops of the conductor, and it is only visible by transparency, or by breaking through the conductor (Figs. 190A-B, 191 A-B). 123. Median apophysis (MA): (0) present; (1) absent. If a palpal bulb has two apophyses in addition to the embolus (i.e., three apophyses) and one fits the criteria listed above for conductor (e.g., embracing embolus, hyaline fan opposing embol- ic apex, etc.), we code the other as a median apophysis. This is typically the farthest apophysis from the embolus origin. Many taxa have a median apophysis (Figs. 166B, 171E, 178A) but this process is typically absent in filistatids (Fig. 166D), eresids (Fig. 170D), deinopids (Fig. 171D), Megadictyna (Fig. 171B), dictynids (Fig. 175A), and our stiphidiid exemplars (Fig. 179E). Although the psechrid Fecenia has a median apophysis, this is lacking in Psechms (Fig. 167C). Forster et al. (1987) code the small process on the Gradungula bulb as a median apophysis. Among the several fixed, sclerotized processes on the tegulum of Mimetus we code one as median apoph- ysis. Archaea has a hooked MA within the crescent-shaped conductor (Fig. 168D). 124. Median apophysis shape: (0) convex; (1) concave. Median apophyses exhibit a great variety of origins and shapes including being swollen or resembling cones or blades (state 0 "convex": Figs. 171E, 188B). We code as a separate state those that have one surface concave or excavate (state 1 "concave"; Griswold 1993: figs. 8,26). 125. Median apophysis attachment: (0) fixed; (1) flexibly attached. The median apophysis may be fixed to the tegulum (state 0: Figs. 166B, 170B, 178D) or have a characteristic, flexible attachment (state 1: Figs. 171E, 187A, 188C, 193A-C). The median apophysis may be moved at this articulation. 126. 'Extra' tegular processes (i.e., in addition to conductor and median apophysis): (0) absent; (1) present. Many taxa have only the standard conductor and median apophysis (Figs. 166B, 173C, 178A, D). If a palpal bulb has a third apophysis in addition to the embolus (i.e., at least four apophyses) we code that bulb as having a conductor, median apophysis and "extra tegular apophysis" (TA: Figs. 180B, 193A, 194B-C). Deciding which apophysis is which may be problematic. In most cases the conductor is associated with the embolus. Median apophysis assignment may be based on similarity in form and position to the median apophyses in taxa without extra apophyses. The remaining apophysis, without special similarity to conductor or median apophysis, is by default the "extra tegular apophysis". There may be more than one extra tegular apophysis. In some taxa, e.g. Xevioso (Fig. 170B) or Mimetus (Fig. 169B) no special similarities enable us to distinguish one of the four tegular apophyses as the median apophysis, yet we code it as present by default. Others, e.g., oecobiids, are even more problematic. These bulbs have several apophyses in addition to the GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 67 embolus (Figs. 170A, C, 187B), but none can be chosen over others as median apophysis or con- ductor, and these, as well as extra tegular apophyses, are coded by default. 127. 'Extra' tegular process form: (0) conical; (1) sclerotized tegular process (STP); (2) mem- branous (MTP). We distinguish extra tegular apophyses based on their shape and texture. A variety of apophy- ses are conical (Figs. 170B-C, 180B), and, although they differ among themselves, we cannot sub- divide this state (state 0). The STP (sclerotized tegular process, state 1) typical of amaurobiids and zorocratids is a sclerotized plate or blade that arises near the embolic base (Figs. 181 A, 182B-C, 185A). Finally, in Zoropsis there is a translucent process (state 2, "membranous") that resembles the STP in origin but differs in texture (Fig. 185E). 128. Palpal tarsus M29 muscle: (0) present; (1) absent. Huber (1994, in lit.) studied the distribution of muscles that attach to the male palpal bulb. The claw flexor, termed "M29" by Ruhland and Rathmeyer (1978), originates at the palpal tibia and attaches to the basal part of the genital bulb (Fig. 167D). Huber noted the presence of the M29 mus- cle in Liphistius, Mygalomorphae, Haplogynae (except Oonops) and Palpimanus, and its absence in all entelegynes. Huber {in lit.) examined many of our exemplar families including filistatids, oecobiids, eresids, deinopids, uloborids, araneids, dictynids, neolanids, stiphidiids, amphinectids, and amaurobiids and we code our exemplars according to his observations, even if he examined genera different than ours. After Huber's examination of Palpimanus, we code the M29 present in Huttonia. We observed cuticular apodemes in the palp of Kukulcania hibernalis that seem to cor- respond to these muscles (Fig. 167D); however Huber {in lit.) made sections of the palp of K. hibernalis and concluded that the M29 is absent. 129. Palpal tarsus M30 muscle: (0) present; (1) absent. Huber (1994, in lit.) studied the distribution the claw extensor, termed "M30" by Ruhland and Rathmeyer (1978), which originates at the cymbium and attaches to the basal part of the genital bulb (Fig. 167D). He noted that the M30 is present in lower araneomorphs and absent from entel- egynes except Uroecobius (Oecobiidae), Tama and Hersilia (Hersiliidae), Palpimanus, Mecysmauchenius and Argyroneta. Because Palpimanus is presumably closely related to Huttonia, we code the M30 present in the latter. Although the M30 is present in Uroecobius, Huber (1994) notes its absence in both Oecobius and Uroctea. FEMALE GENITALIA.? All characters were scored by observation of exemplars. 130. Female genitalia: (0) haplogyne; (1) entelegyne. Hypochilids, austrochiloids, filistatids, huttoniids and archaeids lack a separate fertilization duct connecting the spermathecae with the oviduct (state 0, haplogyne) (Figs. 164B, E); all other exemplar taxa have a fertilization duct (state 1, entelegyne) (Figs. 164D, F). 131. Epigynum: (0) absent; (1) present. We code as an epigynum any sclerotized modification of the cuticle around the female genital region. Epigyna are typical of entelegyne spiders, but we also code the female genital sclerotiza- tion of Thaida as an epigynum. The genital areas in Hickmania and Gradungula are almost iden- tical to Thaida except for the sclerotization, so we also code these as an epigynum. The palpi- manoids Archaea and Huttonia lack epigyna but Mimetus and Pararchaea have epigyna. 132. Epigynum teeth: (0) absent; (1) present. Among taxa with epigyna several have paired conical structures arising from the lateral lobes. These may be short or long (Fig. 180D). We code Raecius as having teeth although this varies with- in the genus: Raecius asper and R. jocquei have teeth but R. congoensis lacks them. Many other entelegynes lack teeth (Figs. 164C, 170E). 68 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 133. Oecobiid spermathecae: (0) absent; (1) present. The entelegyne female genitalia of oecobiids are unique in comprising a copulatory duct that leads to an anterior large, membranous sac, from where another long, sclerotized duct runs to the posterior margin, where the fertilization ducts discharge. In addition Oecobius has a membranous sac at the base of the fertilization duct (Baum 1972). This character is inapplicable to hapologynes. 134. Convoluted vulval ducts of amphinectid type: (0) absent; (1) present. Amphinectid spiders such as Maniho and Metaltella have characteristic, complex vulvae in which the copulatory ducts make at least three anterior-posterior switchbacks (e.g., Fig. 164F; Forster and Wilton 1973: figs. 531, 534). Most entelegyne taxa treated here have simple lobate vul- vae, vulvae with spirals, or with fewer switchbacks (state 0). Matachia also has a complex vulva (Forster 1970: figs. 61-65), which we code like the amphinectid vulva (state 1). 135. Gonopore separated from the epigastric fold: (0) absent; (1) present. In the austrochiloids Gradungula, Hickmania and Thaida, the gonopore is externally visible in the projecting genital area (Fig. 163A), while in other spiders the opening is hidden in the epigas- tric fold. In austrochiloids the epigastric fold leads to a blind pocket (the postepigastric fold, Figs. 163A-D), where the strong longitudinal muscles VIII and IX attach (see Millot 1936). Forster et al. (1987) described this posterior structure as a receptacle, but besides having some powerful mus- cles attached, it lacks any feature commonly associated with sperm receptacles (e.g., a constriction followed by a large lumen, or glandular pores). This invagination is quite general for spiders, but similarly developed posterior extensions were found in palpimanid spiders (Platnick et al. 1991) and by Ramirez (pers. obs.) in several other spiders, including liphistiids, gradungulids, mimetids, and corinnids. This has formerly been coded as a "posterior receptaculum" in Platnick et al. 1991. SILK.? Silk ultrastructure data are taken from the discoveries and summary in Eberhard and Pereira (1993), from unpublished observations by Carlson (in lit), and from our own observations. Terminology follows Peters (1987) with modifications by Eberhard and Pereira (1993). 136. Cribellate silk axial lines: (0) present; (1) absent. Axial lines are the straight thicker fibers associated with the cribellum fibrils (Figs. 120A-C, 121C-D, 122A, 123A, 124A). They are present in most of our exemplars for which data are avail- able, being absent only in Matachia and dictynids (Figs. 120E-F). 137. Cribellate silk reserve warp: (0) present; (1) absent. Reserve warp lines are the highly-curled or undulating thicker fibers associated with the cribel- lum fibrils (Figs. 120E-F, 121C-D, 122A, 123A, C, 124A). Reserve warp is present in most of our exemplars for which data are available, being absent only in Hickmania (Figs. 120A-C), uloborids (Fig. 120D), Matachia and some dictynids. 138. Cribellate silk nodules: (0) absent; (1) present. Cribellar fibrils of most Neocribellatae have nodules along their length (Figs. 118B, D). Among taxa studied nodules are lacking only in Hypochilus and filistatids (Fig. 118F). 139. Cribellate silk: (0) uniform; (1) puffed. The cribellate band is made of up foundation lines (e.g., reserve warp, axial lines) and a mass of numerous, fine cribellar fibrils. In most cribellates the lateral margins of this band are uniform or entire (Figs. 120A, 121C, 123A, 125A). Austrochilids (Figs. 118A, C), uloborids (Figs. 119A-B), deinopoids (Fig. 120D), Matachia and dictynids (Fig. 120E) have the edges of the cribel- late band with regular puffs. This character is inapplicable for the cribellate bands of filistatids, which are very differently constructed and heavily folded (Fig. 118E, Eberhard and Pereira 1993). BEHAVIOR.? Behavioral observations were made on living animals in the field or lab. Except where references are cited, these are our personal observations. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 69 140. Web posture: (0) inverted; (1) erect. We code this based upon the movement of the spider on the web. We have observed the scored taxa in the field and/or in captivity. Most spiders with space webs or sheet webs hang beneath them (state 0: "inverted"), although some sheet web builders walk on the web, e.g, Tengella. Tengella, as well as spiders that have their webs appressed to the substrate and walk on the web, are coded as as "erect" (state 1). 141. Combing leg support: (0) fixed leg III; (1) mobile, braced leg IV. Eberhard (1988) identified two ways in which spiders spin cribellate silk. In hypochilids and filistatids the combing leg was supported by the contralateral leg III, which was held immobile or nearly so (state 0: "fixed leg III" or "stereotyped combing type 1"); this combing behavior was recently observed for Filistata insidiatrix by one of us (M. Ramirez) in specimens from Siena, Italy (Figs. 196D-E). Other cribellates support the combing leg with the other leg IV and move both legs synchronously (state 1: "mobile, braced leg IV" or "stereotyped combing type 2") (Figs. 205B, 208D). The combing behavior of Thaida has recently been observed by Lopardo et al. (2004): its combing leg is braced by a mobile leg IV 142. Orb web architecture: (0) absent; (1) present. The orb web is characteristic of Orbiculariae (Figs 201A-C). 143. Deinopid web architecture: (0) absent; (1) present. Deinopids make a highly modified orb web with sticky cribellate capture silk held by the first two pair of legs (Figs. 200C-D). 144. Frame construction: (0) absent; (1) present. Frame construction is one of the components of the behavior sequence in spinning orb webs and is characteristic of Orbiculariae. 145. Radius construction: (0) absent; (1) present. Orb weavers make radii according to stereotyped behaviors, which are absent in non orb weavers. 146. Radius construction behavior: (0) cut and reeled; (1) doubled. Eberhard (1982) described several ways in which orb weavers lay and connect radii; these behaviors are phylogenetically informative (Coddington 1990a, Hormiga et al. 1995). Most orbic- ularians cut and reel radii as they are being laid. The spider moves along a pre-existing radius to a vacant spot on the frame, laying out a new line behind. The spider attaches this radius to the frame, and returns to the hub spinning a new radial line. As the spider returns to the hub on the radial line just laid, this radial line is cut, reeled up, and eaten so that the dragline behind forms the only radi- al line. The result is only one radius for a pass from hub to frame and back (state 1: "cut and reeled"; Eberhard 1982: character Fl). Uloborids cut and reel to make frames, but omit cutting and reeling when spinning radii. However, like orbicularians other than nephiline tetragnathids, they attach radii only once to the frame. The result is that uloborid radii are double, with one line laid on the way out and one on the return (state 2: "doubled"; Eberhard 1982: character F4). 147. Hub construction: (0) absent; (1) present. Hub construction is one of the components of the behavior sequence in spinning orb webs and is characteristic of Orbiculariae. 148. Temporary spiral construction: (0) absent; (1) present. Temporary (non sticky) spiral construction is one of the components of the behavior sequence in spinning orb webs and is characteristic of Orbiculariae. 149. Sticky spiral construction: (0) absent; (1) present. Sticky spiral construction, in which threads of flagelliform gland silk are coated with glue from aggregate glands (Figs. 119C-E), is one of the characteristic components of the behavior sequence 70 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II in spinning orb webs and is characteristic of Orbiculariae. 150. Sticky silk localization: (0) absent; (1) present. Orb weavers use their legs to locate themselves during sticky spiral construction. This behav- ior has not been observed in spiders that do not make orb webs. 151. Sticky silk localization, type: (0) outside leg I; (1) inside leg IV. Eberhard (1982) first pointed out the phylogenetic significance of the different legs used by orb weavers to locate themselves during sticky spiral construction, and systematists have empha- sized it ever since (e.g. Coddington 1986a, 1990a, Hormiga et al. 1995). Araneids and uloborids use the outside first leg (away from the hub) to touch the previous sticky spiral before attaching the current segment (state 0). Deinopis use the inside fourth leg (towards the hub) (state 1). 152. L4 shift switch during sticky silk construction: (0) absent; (1) present. See Coddington (1986a), character 82 in Coddington (1990a) and character 58 in Coddington (1990b). 153. Non-sticky line grip: (0) otherwise; (1) with leg IV See Coddington (1986a), character 81 in Coddington (1990a) and character 57 in Coddington (1990b). 154. Prey wrapping with legs IV: (0) absent; (1) present. Many spiders wrap their prey (Fig. 198F) with alternate movements of legs IV, including Thaida, filistatids, Orbiculariae, Oecobius (Eberhard 1967), Megadictyna and phyxelidids. Other spiders have not been observed to wrap. We code this behavior only for those taxa that we have observed in the field or in captivity catching prey and feeding upon it, or for which the behavior is detailed in the literature. Eberhard (1982) referred to this behavior as "attack wrap without rotation in the web" (his character 13). DISCUSSION Many of the representatives and characters in this study were used in previous analyses (Coddington 1990a, 1990b; Platnick et al. 1991 [Fig. 209]; Griswold 1993 [Fig. 215]; Griswold et al. 1998 [Fig. 211]; Griswold et al. 1999 [Fig. 212]; Schiitt 2000, 2002 [Fig. 210]; Silva Davila 2003 [Fig. 214]; Raven and Stumkat 2005 [Fig. 215]). Whereas common themes emerge, the phy- logenetic hypotheses obtained in those studies are also significantly different from one another, suggesting that we are not yet reaching robust solutions, and that future studies will differ as well (cf. Miller 2003). We choose to emphasize the documentation of observations, fundamentally with images, and the conceptualization of homology hypotheses through a wide range of taxa, rather than in trying to produce highly elaborate but ephemeral phylogenetic hypotheses. Our experience in doing this study is that a collection of high-resolution digital images permits a level of analysis and possibilities for discussions that were impossible without this technology. We make available full-resolution versions of the images contained in this study at . In Figures 218 and 219 we present several indices related with the support of groups. All of them measure different properties of the dataset, and are generally correlated, although imperfect- ly. For example, although Eresoidea is found through the whole space of the analysis parameters explored, its Bremer support and resampling frequencies are very low. More importantly, our results indicate that all groups highly dependent upon specific conditions of the analysis are also weakly supported for the other estimators, thus downplaying the importance of the selection of a specific method of analysis. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 71 Character Systems and Homoplasy Levels The homoplasy levels are homogeneously widespread across all character systems (Fig. 220D), except for behavior and the internal anatomy characters from non-cuticular structures, two character systems difficult to record. Their high congruence indices are seemingly related to the elevated proportion of missing or inapplicable entries (Fig. 220A). We diverged from previous analyses (Platnick et al. 1991; Griswold et al. 1999) in not assuming character distributions for the classic characters of internal anatomy. The high proportion of missing entries illustrates a real sit- uation of a much-neglected field of study. When these characters are more adequately sampled, they may show that they have similar levels of homoplasy as any other character system. Many of our behavioral characters are specific to orb weavers, represented in this dataset by only five representatives in three families. Most of the problematic taxa that interfere with the res- olution of orbweavers in this analysis (fundamentally, Palpimanoidea and the nicodamids) do not spin webs, or their spinning behavior has never been observed in detail. The weak support for orb weavers is not due to the missing data, because the results are insensitive to the replacing of miss- ing entries by zeroes in the characters for orb web details (characters 144-153). However, replac- ing the inapplicable scorings by zeroes in characters 144-153 results in an equal weights tree with Palpimanoidea sister to orbweavers, and Nicodamidae sister to all of them (Fig. 218C), thus approximating the strategy and results of Platnick et al. (1991). We no longer endorse such an approach here. Groups NEOCRIBELLATAE.? The monophyly and synapomorphies of Neocribellatae (all Araneomorphae except Hypochilus, Fig. 212) are not tested here. Compared to Mesothelae, Mygalomorphae and Hypochilus (Haupt and Kovoor 1993; Goloboff 1995), the Neocribellatae evolved a greater diversity of gland spigots on the posterior spinnerets, including paracribellar and minor ampullate gland spigots. ARANEOCLADA.? The Araneoclada were defined by Platnick (1977:8) for a large group of spiders believed to have 3 pairs of heart ostia and a straight midgut, comprising most of the Araneomorphae (Fig. 212). Equal weights (Fig. 216) and implied weights (Fig. 217) both refute the Araneoclada. The haplogyne spiders are placed as the basal Neocribellate group. Lopardo et al. (2004) already discussed challenging evidence favoring a more basal placement of Filistatidae, out- side of Araneoclada (e.g., an M-shaped intestine and traces of posterior booklungs in Filistatidae, Type 2 combing behavior in Austrochilinae). We continued finding evidence in conflict with a monophyletic Araneoclada in this study (e.g., the presence of cylindrical gland spigots in Austrochilidae). Our results are novel in the placement of the representatives of Haplogynae (Segestriidae and Filistatidae) in a more basal position relative to Austrochiloidea. Although we should expect some artificial results from a shallow sampling of such a large and diverse group as Haplogynae, this result clearly indicates that the basal clades of Araneomorphae are far from sat- isfactorily resolved. AUSTROCHILOIDEA.? This taxon was proposed by Forster et al. (1987) and corroborated by the quantitative analyses of Platnick et al. (1991 [Fig. 209]) and Griswold et al. (1999 [Fig. 212]). Both equal weights and implied weights support Austrochiloidea, comprising Austrochilidae and Gradungulidae. Our reinterpretation of the female genitalia of austrochiloids renders additional support to the group (the well exposed female gonopore). However, this reinterpretation under- mined the support for Austrochilidae, because we identified their "posterior receptacle" as a sim- ple cuticular fold generally found in other Araneomorphae. Also novel is the finding of sexually 72 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II dimorphic spigots on the posterior spinnerets of austrochilids. We identify these as cylindrical, previously thought to be restricted to entelegynes. The spinning organs of adult cribellate gradun- gulids were never examined with SEM. The micrographs of immature Macrogradungula in Platnick et al. (1991, figs. 297-304) show a pattern of spigots similar to that found in Gradungula (the ALS), and Hickmania and Austrochilus (PMS and PLS), thus it is unlikely that the examina- tion of the adult spinnerets of cribellate gradungulids will produce any radical change in the reso- lution of Araneoclada or Austrochiloidea. In our analysis Austrochilidae is not monophyletic, implying an unlikely convergence in the acquisition of booklungs in Gradungula and Hickmania. ENTELEGYNAE.? This group is well established (Figs. 209, 212) and is corroborated by our analysis (Figs. 216-217). The entelegyne condition has seemingly reversed to secondary haplogy- ny in several taxa (e.g., some Tetragnathidae and Anapidae, all 'palpimanoids' except Holarchaea, Pararchaea and Mimetus). We know of only one well-documented convergence to the entelegyne condition, in pholcids of the genus Metagonia (Huber 1997). Our analysis under equal weights is consistent with this hypothesis, but under implied weights the palpimanoids Huttonia and Archaea are placed more basally (thus implying primary haplogyny), while Pararchaea and Mimetus remain with araneoids. A broader sampling of palpimanoids should shed more light on this prob- lem ERESOIDEA.? Eresidae and Oecobiidae were first associated by Platnick et al. (1991 [Fig. 209]) and this grouping was corroborated by Griswold et al. (1999 [Fig. 212]). Equal, successive and implied weights all support this clade in our new analysis (Figs. 216-219). A distinctive char- acteristic of Eresoidea is the several major ampullate spigots dispersed among the piriform field. Oecobius, with only one MAP spigot is an exception, probably associated with its very small size. CANOE TAPETUM CLADE.? This group was first suggested in the analysis of Platnick et al. (1991) and named by Griswold et al. (1999 [Fig. 212]). The clade is not recovered in our analysis, due to ambiguous placement of the Eresoidea and Orbiculariae. ORBICULARIAE AND PALPIMANOIDEA.? Two issues about orb weaving phylogeny have been heatedly debated: the origin of orb webs, and the placement of certain non orb weavers within Araneoidea. The first issue had lost its original momentum. After the detailed works of Eberhard (1982) and Coddington (1986a-b), who closely examined the sequences and stereotyped move- ments that orb weavers use to make their webs, there has been little questioning of the common origin of orb webs. It surely influenced this debate that the hypotheses of repeated convergence in complex 'adaptive' traits are not as central to evolutionary debates as was the case years ago. It is nowadays accepted that the cribellate Deinopoidea (Deinopidae and Uloboridae) are the sister group of the ecribellate Araneoidea. As for the second issue (the non orb weaving araneoids), the detailed morphology of the silk spinning organs provided rich support to the affiliation of some non orb weaving families in Araneoidea (Coddington 1989; Griswold et al. 1998 [Fig. 211]), and it seems clear that the orb web was modified or lost several times in the evolution of Orbiculariae (e.g., in Deinopidae, Theridiidae, Linyphiidae, Nesticidae, cyatholipoids and several symphytog- nathoids). Our analysis does not solve the placement of Palpimanoidea, but none of the trees obtained under the several explored weighting schemes implies a convergence in the orb web. Schiitt (2000, 2002, 2003) recently argued that micropholcommatids and textricellids, previ- ously considered palpimanoids, have the spigot characters typical of symphytognathoids. However, the placement of Mimetidae and other "palpimanoids" (e.g., Archaeidae, Pararchaeidae, Malkaridae) has remained controversial. Schiitt (2000, 2002) suggested that Mimetidae, Pararchaeidae and Malkaridae belong with the Araneoidea (Fig. 210). Our analyses concur, also suggesting that at least some "palpimanoids" nest within the Orbiculariae. Implied weights places entelegyne palpimanoids with paracymbia (i.e., Mimetidae and Pararchaeidae) sister to Araneoidea GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 73 (Fig. 217). Equal weights and successive weights place all palpimanoids, including the haplogyne Archaeidae and Huttoniidae, closely related to Araneoidea (Fig. 218, when nicodamids are removed, Fig. 219B). The more basal position of Archaea plus Huttonia suggested by implied weights (Fig. 217) is interesting, because it implies primitive, instead of secondary haplogyny. It is worth noting that Forster (in Forster and Platnick 1984) argued that archaeids are primitively hap- logyne and that their peculiar posterior respiratory system evolved directly from booklungs, a sce- nario plausible given the placement of archeids under implied weights. On the other hand, this placement may be simply an artifact due to shallow taxon sampling. The Palpimanoidea traditionally included only the families Huttoniidae, Palpimanidae and Stenochilidae, all haplogyne families having characteristic brushes of setae on the tarsi and metatarsi of legs I and II. Foster and Platnick (1984) enlarged and radically redefined the Palpimanoidea, adding to the three traditional families the Mimetidae, Micropholcommatidae, Textricellidae, Archaeidae, Mecysmaucheniidae, Pararchaeidae, Holarchaeidae, and later (Platnick and Forster 1987) the Malkaridae. Palpimanoidea has remained one of the most controversial hypotheses in spider classification. This concept of an enlarged, monophyletic Palpimanoidea sur- vived the quantitative test of Platnick et al. (1991), but not those of Schiitt (2002, 2003), and our results are ambiguous. It seems clear that an adequate test of the limits of Araneoidea and Palpimanoidea should include not only numerous representatives of both superfamilies (comprising 32 families in total), but also sufficient outgroup representatives to leave room for the contentious taxa to be placed else- where. The affiliation of palpimanids and archaeids with eresids, and of Nicodamus with mimetids in Schiitt (2002) seems indicative of undersampling of outgroup taxa. Conversely, the alternative placements of Huttoniidae and Archaeidae in our implied and equal weights analyses suggest the effect of undersampling of araneoids and palpimanoids. OUTGROUPS TO ORBICULARIAE.? Identifying the phylogenetic intermediates between orb webs and 'sheet' or 'irregular' webs is the holy grail in understanding the evolution of spider webs, hence the interest in knowing the closer relatives of orb weavers. Some preliminary results (Griswold et al. 1999) suggested that nicodamids are good candidates because of the serrate acces- sory claw setae and entire cribellum found in Megadictyna, resembling those of orb weavers and deinopoids, respectively. In a reexamination of our representatives we found that serrate accesso- ry claw setae are more widely distributed than previously thought, even in spiders with plumose setae (e.g., Austrochilidae, Megadictyna), or in spiders that do not use webs to capture prey (Hersiliidae, not shown). Nicodamus, an ecribellate Nicodamidae, only has slightly serrate acces- sory claw setae, which we coded as uncertain (scoring them absent produces the collapsing of Deinopoidea in the equal weights analysis). Our equal weights analysis concurs with that of Griswold et al. (1999 [Fig. 212]) in placing Orbiculariae (including Palpimanoidea) as sister group of all entelegynes other than Eresoidea (Fig. 218, nicodamids removed), but under implied weights (Fig. 219), eresoids and orb weavers are sister groups feebly supported by having the PLS modi- fied spigots at least slightly segregated from the spinning field. NICODAMIDAE.? Cribellate and ecribellate nicodamids are quite heterogeneous in general morphology. In our dataset they appeared monophyletic only under successive and implied weights (with low support), united by the branched median tracheae and a proximal, curved dorsal process on the male palpal tibia. Harvey (1995) proposed the absence of trichobothria on metatarsus IV as a further synapomorphy of the family, but we have not explored trichobothrial patterns in much detail. Under equal weights, both nicodamid representatives swap around orb weaver groups and their two contiguous branches (Fig. 218). DIVIDED CRIBELLUM CLADE.? This clade, first recognized by Griswold et al. (1999 [Fig. 74 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 212]) includes the derived entelegynes beyond Nicodamidae, and is found here under equal and implied weights (Figs. 216-217), but with low support. The cribellum seems to have transformed from entire to divided at least three times (also in Eresoidea and Filistatidae), and secondarily to entire in some members of the RTA clade. TITANOECOIDS.? Griswold et al. (1999 [Fig. 212]) proposed this clade for the Titanoecidae plus Phyxelididae. Implied and successive weights produces this grouping (Fig. 219), but with very low support and only under two concavities of the weighting function. Under equal weights (Fig. 218) Phyxelididae and Titanoecidae are successive outgroups to the numerous families comprising the RTA clade. RTA CLADE.? Coddington and Levi (1991) proposed this group for those spiders having an apophysis in any position on the male palpal tibia. We have scored separately the ventral, dorsal, prolateral and retrolateral processes. This more restrictive RTA clade, comprising taxa with a retro- lateral tibial apophysis, was found in the analysis of Griswold et al (1999 [Fig. 212]). The RTA clade is supported under equal and implied weights (Figs. 218-219), by the RTA itself, and by the increased number of trichobothria on appendages. AMAUROBIIDAE.? We no longer recover a clade of Amaurobioids (all RTA clade except Dictynidae, Fig. 212). Amaurobiidae itself is problematic under equal weights (because of the mac- robunines Macrobunus and Retiro), though at least a core Amaurobiidae comprising the type genus Amaurobius plus Callobius and Pimm is found (Fig. 216). Under implied and successive weights (Fig. 217) Amaurobiidae is supported by having multiple male palpal tibial processes and an addi- tional tegular sclerite other than conductor and median apophysis. Besides Macrobunus, the repre- sentatives Pimm and Retiro are also listed in Macrobuninae (Lehtinen 1967). However, we expect changes in the composition of Macrobuninae. According to ongoing research by Ramirez and Griswold, the closer cribellate relatives of Macrobunus all have entire cribella (Anisacate, some Macrobunus species, plus some undescribed genera). They are grouped with other ecribellate mac- robunines close to Macrobunus by having a stridulatory area on the male palpal cymbium (Figs. 183A-D). Compagnucci and Ramirez (2000) joined the macrobunines Anisacate, Emmenomma and Naevius by the presence of a gland in the male palpal tibia discharging through a dorsal apoph- ysis. It seems likely that Macrobuninae will end up diagnosed by having an entire cribellum (when present) and a stridulatory area on the male palpal cymbium; it will include a group of genera with a male palpal tibial gland. FUSED PARACRIBELLAR CLADE.? This group (FPC) was proposed by Griswold et al. (1999 [Fig. 212]) for some taxa having the PMS paracribellar spigots with two to several shafts arising from a large, common base, i.e., Stiphidiidae, Neolanidae, Agelenidae, Amphinectidae and Desidae (but not Dictynidae). Successive weights and one of the concavities of implied weights (Fig. 219) excludes the Stiphidiidae but newly groups the Dictynidae, which also have fused paracribellar bases, within FPC, sister to some Desidae (Desis and Matachia). This striking morphology is at least corroborated as a synapomorphy, although for a different collection of taxa. STIPHIDIOIDS.? Griswold et al. (1999) retrieved a clade uniting Neolanidae with Stiphidiidae (Fig. 212), only supported by a reversion to an inverted posture on the web. Our analyses do not support this clade. AGELENOIDS.? A clade uniting the cribellate agelenid Neoramia with desids and amphinec- tids (Fig. 212) is not recovered in our analysis. DESIDAE, AMPHINECTIDAE AND DICTYNIDAE.? Both Desidae and Amphinectidae are para- phyletic in our analysis. Under all concavities other than 6 the amphinectid Metaltella and the desid Desis are sister groups, joined by the internal origin of the embolus. Dictynidae is supported under equal, successive and implied weights, but we do not reproduce the basal position of Dictynidae in GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 75 the RTA clade, as found in Griswold et al. (1999 [Fig. 212]). Instead, in our current results Dictynidae is sister to some Desidae (Figs. 216-217). OVAL CALAMISTRUM CLADE.? Griswold (1993 [Fig. 213]) proposed a clade for those spiders with an oval to rectangular calamistrum, grouping the lycosoids together with zorocratids and ten- gellids. This clade passed the quantitative test of Griswold et al. (1999 [Fig. 212]). We obtained an oval calamistrum (OC) clade under equal and implied weights (Figs. 216-217). Under implied weights the group is supported additionally by the tegular and subtegular locking lobes. LYCOSOIDEA.? The Lycosoidea were defined by having a grate-shaped tapetum in the indi- rect eyes (Homann 1971). This idea was corroborated by the quantitative test of Griswold (1993 [Fig. 213]) and partially corroborated by Griswold et al. (1999 [Fig. 212]), who grouped the Psechridae, Zoropsidae and Ctenidae but excluded the Stiphidiidae. We no longer obtain a mono- phyletic Lycosoidea (represented by the cribellate Psechridae, Zoropsidae and Ctenidae), although Zoropsis and Acanthoctenus appear as sister groups (Figs. 216-217). Our results differ from those of Griswold et al. (1999) and Griswold (1993), and should be considered in the light of the recent analyses made by Silva Davila (2003) and Raven and Stumkat (2005). Silva Davila's (2003) dataset included a dense sampling of Lycosoids and their kin and a fairly broad selection of outgroups, including dionychans and amaurobiids, and rooted the analy- sis with Megadictyna. The dataset comprised the seven OC clade representatives included here, most of the representatives considered in Griswold (1993), a denser sampling within Tengellidae and Ctenidae, and many representatives of ecribellate families allegedly linked to lycosoids (Cycloctenidae, Zoridae, dionychans). She also reviewed and used most of the characters of the previous cladistic analyses, including Griswold et al. (1999). Silva Davila recovered a clade of spi- ders with grate-shaped tapetum (GST clade), including lycosoids, but also miturgids, zorids, and the OC clade nested within (Fig. 214). She did not recover Tengellidae and Zorocratidae as mono- phyletic groups. It is interesting that her analysis suggested that Tengella and Zorocrates are sister groups (Fig. 214), a result mirrored in our analysis for all but equal weights (Fig. 219). Her dataset implies tiny support values for the relations among the higher groups, and the results differ signif- icantly between equal and implied weights; this, together with the differences obtained from a wider taxon sampling, is indicative of high instability in the relations of lycosoids and their kin. It is notable that the grate-shaped tapetum appears to have little phylogenetic value. Our cladograms (Figs. 216-217) and those of Silva Davila (Fig. 214) and Raven and Stumkat (Fig. 215) all imply considerable homoplasy in this feature. ZOROCRATIDAE.? Our implied and successive weights analyses concur with Silva Davila's (2003 [Fig. 214]) in joining Zorocrates with Tengella, although the support is weak (Figs. 218-219). Raven and Stumkat (2005) used the densest sampling of lycosoids and their kin of any study to date, though their dataset was not as broad as Silva Davila's or ours. They enlarged the Zoropsidae to include taxa formerly included in the Zorocratidae, which they considered as a sub- family of Zoropsidae (Fig. 215). Their analysis was rooted with Tengella, so does not test the pos- sible relationship between Tengella and Zorocrates. Our analysis did not recover a clade of Zorocratidae, and in the light of the current evidence, it is unlikely that the family is monophylet- ic. None of the characters proposed to define Zorocratidae by Griswold et al. (1999), i.e., clumped cribellar spigots, male tibial crack, or tibial ventroapical process, occur in Zorocrates. AEBUTINA AND POAKA.? The placement of the mysterious Aebutina at the base of lycosoids and allies is very unconvincing. Excluding Aebutina from the equal weights analysis is of no con- sequence, but under implied weights causes Macrobunus to join at the base of the OC clade, as occurs in the equal weight analysis. The relationships of Poaka remain an open question, only resolved in the implied weights tree, but involving groupings of very low support. 76 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II CONCLUSIONS We are making progress. Whereas some results conflict with previous studies and even with- in this study, at least some results are robust. The Austrochiloidea, Entelegynae, Eresoidea, and Divided Cribellum, RTA and Oval Calamistrum clades all survived the tests, and may represent true evolutionary groups. The unique origin of the orb web seems assured, though the composition of the Orbiculariae remains controversial. On the other hand, the weak support for other groups, poor phylogenetic performance of many classic characters, sensitivity of results to taxon sampling and generally ephemeral nature of many phylogenetic hypotheses suggest that we still have a long way to go to understand the big picture of spider evolution. What do we need to make more progress? We probably have adequate methods of analysis. Our results suggest that some groups are robust and some not, regardless of estimator, which down- plays the importance of the selection of a specific method of analysis. This in turn suggests that, given adequate taxon sampling and careful character definition and coding, the currently available analytical methods will give robust and meaningful results. Comparative genomics will undoubt- edly contribute a huge amount to understanding spider evolution. Indeed, current support from the U.S. National Science Foundation to the "Assembling the Tree of Life: Phylogeny of Spiders" proj- ect promises to make DNA data crucial in reconstructing spider phylogeny. But, this should not obscure the continuing importance of "traditional" disciplines. Examination by SEM of cuticular structures has provided hundreds of new and meaningful comparisons. Comparative anatomy of internal soft structures, a field largely neglected since the pioneering work of Millot, Petrunkevitch and Marples in the 1930s through 1960s, deserves rejuvenation. For example, the recent discovery of the primitive M-shaped intestinal configuration in Kukulcania hibernalis, which makes sense given that spider's primitive silk spinning behavior, proves that there are yet valuable insights to be gained by dissecting, sectioning and staining. Conversely, the disappointing behavior of the tapetum (as currently coded) as a phylogenetic character, suggests that reinvestigation of this sys- tem is imperative. Field studies of behavior also have much to offer. The discovery of evolutionar- ily advanced spinning behavior in austrochilines foretold their movement up the spider cladogram. Moreover, behavior is intensely interesting to biologists of all stripes. As we firm up the phyloge- netic tree of spiders, the need for behavioral data to map on this tree will grow. Finally, it is clear that denser taxon sampling is necessary. Much of the disconnect between the various phylogenet- ic studies considered herein may be due to sparse, and different, taxon sampling. This in turn argues that further collecting, especially in inaccessible, undersampled regions, and the continued conser- vation and study of existing collections, is essential. Forty years ago Pekka Lehtinen and Ray Forster started a revolution in spider taxonomy. They brought a worldwide perspective to the subject and focused on the tropics and especially the aus- tral regions. This revolution continues, and the importance of taxa and data from the southern hemi- sphere suggests that a continued focus on the austral regions will be crucial to understanding spi- der evolution. LITERATURE CITED AKERMAN, C. 1926. On the spider, Menneus camelus Pocock, which constructs a moth-catching, expanding snare. Annals of the Natal Museum 5:411^-22. AVILES, L. 1993. Newly-discovered sociality in the Neotropical spider Aebutina binotata Simon (Dictynidae?). The Journal of. Arachnology 21:184-193. BAUM, S. 1972. Zum "Cribellaten-Problem": Die Genitalstrukturen der Oecobiinae und Urocteinae (Arach.: Aran: Oecobiidae). Abhandlugen und Verhandlungen des Naturwissenschaftichen Vereins in Hamburg GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 77 (N.F.) 16:101-153. BOSSELAERS, J. 2002. A cladistic analysis of Zoropsidae (Araneae), with description of a new genus. Belgian Journal of Zoology 132(2): 141-154. BRISTOWE, W. 1933. The liphistiid spiders. With an appendix on their internal anaotomy by J. Millot. Proceedings of the Zoological Society of London, 7952:1015-1057. BURGER, M., W. NENTWIG, AND C. KROPF. 2003. Complex genital structures indicate cryptic female choice in a haplogyne spider (Arachnida, Araneae, Oonopidae, Gamasomorphinae). Journal of Morphology 255:80-93. BUXTON, B.H. 1913. Coxal glands of the arachnids. Zoologische Jahrbucher, II Abteilung fur Systematick 14:231-282. CATLEY, K.M. 1994. Descriptions of new Hypochilus species from New Mexico and California with a cladis- tic analysis of the Hypochilidae (Araneae). American Museum Novitates 3088:1-27. CHAMBERLIN, R.V., AND W. I VIE. 1941. North American Agelenidae of the genera Agelenopsis, Calilena, Ritalena and Tortolena. Annals of the Entomological Society of America 34:585-628. CODDINGTON, J.A. 1983. A temporary slide-mount allowing precise manipulation of small structures. Verhandlungen des Naturwissenschaftlichen Vereins in Hamburg 26:291-292. CODDINGTON, J.A. 1986a. The monophyletic origin of the orb web. Pages 319-363 in W. A. Shear, editor, Spiders: Webs, Behavior, and Evolution. Stanford University Press, Stanford, California, USA. CODDINGTON, J.A. 1986b. Orb webs in 'non-orb weaving' ogre faced spiders (Araneae: Deinopidae): a ques- tion of genealogy. Cladistics 2(1): 53-67. CODDINGTON, J.A. 1989. Spinneret silk spigot morphology: Evidence for the monophyly of orb-weaving spi- ders, Cyrtophorinae (Araneidae), and the group Theridiidae plus Nesticidae. The Journal of Arachnology 17:71-95. CODDINGTON, J.A. 1990a. Ontogeny and homology in the male palpus of orb-weaving spiders and their rela- tives, with comments on phylogeny (Araneoclada: Araneoidea, Deinopoidea). Smithsonian Contributions to Zoology 496:1-52. CODDINGTON, J.A. 1990b. Cladistics and spider classification: araneomorph phylogeny and the monophyly of orbweavers (Araneae: Araneomorphae; Orbiculariae). Ada Zoologica Fennica 190:75-87. CODDINGTON, J.A., AND H.W. LEVI. 1991. Systematics and evolution of spiders (Araneae). Annual Revew of Ecology and Systematics 22:565-592. CODDINGTON, J.A., AND N. SCHARFF. 1994. Problems with zero-length branches. Cladistics 10:415-423. COMPAGNUCCI, L.A., AND M.J. RAMIREZ. 2000. A new species of the spider genus Naevius Roth from Argentina (Araneae, Amaurobiidae, Macrobuninae). Studies on Neotropical Fauna and Environment 35:203-207. CROME, W. 1955. Die Beziehungen zwischen dem dorsalen Zeichnungsmuster und der Metamerie des Spinnenabdomens, II. Zoologische Jarhbiicher, Systematik 83:541-638. DAHL, F. 1913. Vergleichende Physiologic und Morphologic der Spinnentiere unter besonderer Berucksichtigung der Lebensweise. 1. Die Beziehungen des Korperbaues und der Farben zur Umgebung, Jena, pp. 1-112. DAVIES, V.T. 1998. A revision of the Australian metaltellines (Araneae: Amaurobioidea: Amphinectidae: Meteltellinae). Invertebrate Taxonomy 12:211-243. DAVIES, V.T. 1990. Two new spider genera (Araneae: Amaurobiidae) from rainforests of Australia. Acta Zoologica Fennica 190:95-102. DEELEMAN-REINHOLD, C. 1986. A new cribellate amaurobioid spider from Sumatra (Araneae: Agelenidae). Bulletin of the British Arachnological Society 7:34-36. DE LA SERNA OF ESTEBAN, C.J. 1976. Algunas observaciones anatomo-histologicas sobre el aparato reproduc- er de la hembra de Ariadna mollis (Holmberg, 1876) (Araneae, Labidognatha, Haplogynae). Phycis C 35:139-146. EBERHARD, W.G. 1967. Attack behavior of diguetid spiders and the origin of prey wrapping in spiders. Psyche 74:173-181. EBERHARD, W.G. 1982. Behavioral characters for the higher classification of orb-weaving spiders. Evolution 36:1067-1095. 78 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II EBERHARD, W.G. 1988. Combing and sticky silk attachment behavior by cribellate spiders and its taxonomic implications. Bulletin of the British Arachnological Society 7:247-251. EBERHARD, W.G., AND F. PEREIRA. 1993. Ultras tructure of cribellate silk of nine species in eight families and possible taxonomic implications (Araneae: Amaurobiidae, Deinopidae, Desidae, Dictynidae, Filistatidae, Hypochilidae, Stiphidiidae, Tengellidae). The Journal of Arachnology 21:161-174. FARRIS, J.S. 1969. A successive approximation approach to character weighting. Systematic Zoology 18:374-385. FORSTER, R.R. 1955. Spiders of the family Archaeidae from Australia and New Zealand. Transactions of the Royal Society of New Zealand 83:391-403. FORSTER, R.R. 1970. The Spiders of New Zealand, III. Otago Museum Bulletin 3:1-184. FORSTER, R.R., AND N.I. PLATNICK. 1984. A review of the archaeid spiders and their relatives, with notes on the limits of the superfamily Palpimanoidea (Arachnida, Araneae). Bulletin of the American Museum of Natural History 178:1-106. FORSTER, R.R., AND N.I. PLATNICK. 1985. A review of the austral spider family Orsolobidae (Arachnida, Araneae), with notes on the superfamily Dysderoidea. Bulletin of the American Museum of Natural History 181:1-230. FORSTER, R.R., N.I. PLATNICK, AND M.R. GRAY. 1987. A review of the spider superfamilies Hypochiloidea and Austrochiloidea (Araneae, Araneomorphae). Bulletin of the American Museum of Natural History 185(1): 1-116. FORSTER, R. R., AND C.L. WILTON. 1973. The Spiders of New Zealand, Part IV. Otago Museum Bulletin 4:1-309. FRIEDRICH, V, AND R. LANGER. 1969. Fine structure of cribellate spider silk. American Zoologist 9:97-101. GIRIBET, G 2003. Stability in phylogenetic formulations and its relationship to nodal support. Systematic Biology 52:554-564. GOLOBOFF, PA. 1993a. Estimating character weights during tree search. Cladistics 9:83-91. GOLOBOFF, PA. 1993b. Nona 2.0. Computer program available at . GOLOBOFF, P.A. 1995. A revision of the South American spiders of the family Nemesiidae (Araneae, Mygalomorphae). Part 1: Species from Peru, Chile, Argentina, and Uruguay. Bulletin of the American Museum of Natural History 224:1-189. GOLOBOFF, PA., J.S. Farris, M. Kallersjo, B. Oxelman, M.J. Ramirez, and CA. Szumik. 2003. Improvements to resampling measures of group support. Cladistics 19:324-332. GOLOBOFF, PA., J.S. FARRIS, AND K. NIXON. 2003-2004. TNT: Tree analysis using new technology. Version 1.0. Program and documentation, available from the authors, and at . GRAY, M.R. 1983. The taxonomy of the semi-communal spiders commonly referred to the species Ixeuticus candidus (L. Koch) with notes on the genera Phryganoporus, Ixeuticus and Badumna (Araneae, Amaurobioidea). Proceedings of the Linnean Society of New South Wales 106:247-261. GRAY, M.R., AND H.M. SMITH. 2004. The "striped" group of stiphidiid spiders. Two new genera from north- eastern New South Wales, Australia (Araneae: Amaurobioidea: Stiphidiidae). Records of the Australian Museum 56:123-138. GREEN, LA. 1970. Setal structure and spider phytogeny. M.S. thesis. Southern Illinois University, Carbon- dale, Illinois, USA. GRISMADO, C.J. 2004. Two new species of the spider genus Conifaber Opell 1982 from Argentina and Paraguay, with notes on their relationships (Araneae, Uloboridae). Revista Iberica Aracnologia 9:291-306. GRISWOLD, C.E. 1990. A revision and phylogenetic analysis of the spider subfamily Phyxelidinae (Araneae, Amaurobiidae). Bulletin of the American Museum of Natural History 196:1-206. GRISWOLD, C.E. 1993. Investigations into the phylogeny of the lycosoid spiders and their kin (Arachnida, Araneae, Lycosoidea). Smithsonian Contributions to Zoology 539:1-39. GRISWOLD, C.E. 2002. A revision of the African spider genus Raecius Simon, 1892 (Araneae, Zorocratidae). Proceedings of the California Academy of Sciences 53 (10):117-149. GRISWOLD, C.E., J.A. CODDINGTON, G HORMIGA, AND N. SCHARFF. 1998. Phylogeny of the orb-web building spiders (Araneae, Orbiculariae: Deinopoidea, Araneoidea). Zoological Journal of the Linnean Society GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 79 123:1-99. GRISWOLD, C.E., J.A. CODDINGTON, N.I. PLATNICK, AND R.R. FORSTER. 1999. Towards a phytogeny of entel- egyne spiders (Araneae, Araneomorphae, Entelegynae). The Journal of Arachnology 27:53-63. GRISWOLD, C.E., AND D. UBICK. 2001. Zoropsidae: a spider family newly introduced to the USA (Araneae, Entelegynae, Lycosoidea). The Journal of Arachnology 29:111-113. GRISWOLD, C.E., AND X.P. WANG. 2001. Character Data to accompany "'Towards a phytogeny of entelegyne spiders (Araneae, Entelegynae)' by Griswold, C, J. Coddington, N. Platnick, and R. Forster. 1999. J. Arachnol. 27:53-63." HARVEY, M.S. 1995. The Systematics of the Spider Family Nicodamidae (Araneae: Amaurobioidea). Invertebrate Taxonomy 9:279-386. HAUPT, J. 2003. The Mesothele ? a monograph of an exceptional group of spiders (Araneae: Mesothele). Originalabhandlungen aus dem Gesamtgebiet der Zoologie, Stuttgart 154:1-102. HAUPT, J., AND J. KOVOOR. 1993. Silk-gland system and silk production in Mesothelae (Araneae). Annales des Sciences Naturelles, Zoologie, Paris 14:35^-8. HILL, D.E. 1979. The scales of salticid spiders. Zoological Journal of the Linnean Society (London) 65:193-218. HOMANN, H. 1971. Die Augen der Araneae: Anatomie, Ontogenese und Bedeutung fur die Systematik (Chelicerata, Arachnida). Zeitschrift fur Morphologie der Tiere 69:201-272. HORMIGA, G, W.G. EBERHARD, AND J.A. CODDINGTON. 1995. Web-construction behaviour in Australian Phonognatha and the phylogeny of nephiline and tetragnathid spiders (Araneae: Tetragnathidae). Australian Journal of Zoology 43:313-364. HUBER, B.A. 1994. Genital bulb muscles in entelegyne spiders. The Journal of Arachnology 22:75-76. HUBER, B.A. 1997. On American 'Micromerys' and Metagonia (Araneae, Pholcidae), with notes on natural history and genital mechanics. Zoologica Scripta 25:341-363. KASTON, B.J. 1972. How to know the spiders. 2nd Ed. William C. Brown Co., Dubuque, 289 pp. KOVOOR, J. 1977a. La sole et les glandes sericigenes des Arachnides. L'annee biologique 16:97-141. KOVOOR, J. 1977b. L'appareil sericigene dans le genre Uloborus Latr. (Araneae: Uloboridae). I. Anatomie. Revue Arachnologique 1:89-102. KOVOOR, J. 1978. L'appareil sericigene dans le genre Uloborus (Araneae, Uloboridae). II. Donnees histochim- iques. Annales des Sciences Naturelles Zoologie et Biologie Animate 20:3-25. KOVOOR, J. 1980. Les glandes sericigenes d' Urocteai durandi (Latreille) (Araneae: Oecobiidae). Revision, histochimie,affinites. Annales des Sciences Naturelles, Zoologie, Paris 1(3): 187-203. KOVOOR, J. 1986. L'appareil sericigene dans les genres Nephila Leach et Nephilengys Koch: anatomie micro- scopique, histochemie, affinites, avec d'autres Araneidae. Revue Arachnologique 7(1): 15-34. KOVOOR, J., AND A. LOPEZ. 1979. Presence de glandes a sole melanisees chez des Eresidae solitaires (Araneae). Revue Arachnologique 2 (3):89-102. KULLMAN, E. 1975. Die Produktion und Funktion von Spinnenfaden und Spinnengeweben. Pages 318-378 in Netze in Natur und Technik (Inst. Leichte Flachentragwerke ed.). Stuttgart-Vaihingen. LAMORAL, B.H. 1968. On the ecology and habitat adaptations of two intertidal spiders, Desis formidabilis (O.P. Cambridge) and Amaurobioides africanus Hewitt, at the Island (Kommetjie, Cape Peninsula) with notes on the occurrence of two other spiders. Annals of the Natal Museum 20(1): 151-193. LAMY, E. 1902. Recherches anatomiques sur les trachees des araignees. Annales des Sciences Naturelles, Paris (Zoologie 8} vol. 15:149-280. LEHMENSICK, R., AND E. KULLMAN. 1956. Uber den Feinbau der faden einiger Spinnen (Vergleich des Aufbaues der Fangfaden cribellater und ecribellater Spinnen). Zoologischer Anzeiger, Supplement.: 123-129. LEHTINEN, P.T. 1967. Classification of the cribellate spiders and some allied families. Annales Zoologici Fennici 4:199-468. LEHTINEN, P.T. 1975. Notes on the phylogenetic classification of Araneae. Proceedings of the 6th International Congress of Arachnology, Amsterdam 1974:26-29. LEHTINEN, P.T. 1996. The ultrastructure of leg skin in the phylogeny of spiders. Revue Suisse de Zoologie, Geneve hors. Serie 1975:399-421. 80 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II LEVI, H. W. 1982. The spider genera Psechrus and Fecenia (Araneae: Psechridae). Pacific Insects 24:114-138. LOPARDO, L., MJ. RAMIREZ, CJ. GRISMADO, AND L.A. COMPAGNUCCI. 2004. Web building behavior and the phylogeny of austrochiline spiders. The Journal of Arachnology 32:42-54. LUBIN, Y.D. 1986. Web building and prey capture in the Uloboridae. Pages 132-171 in W.A. Shear, ed., Spiders: Webs, Behavior, and Evolution. Stanford University Press, Stanford, California, USA. MARPLES, B.J. 1968. The hypochilomorph spiders. Proceedings of the Linnean Society London 179:11-31. MARPLES, B.J. 1983. Observations on the structure of the fore-gut of spiders. Bulletin of the British Arachnological Society 6:46-52. MCKEOWN, K.C. 1936. Spider Wonders of Australia. Angus & Robertson, Sydney, Australia, xii+252 pp. MILLER, J.A. 2003. Assessing progress in systematics with continuous jackknife function analysis. Systematic Biology 52:55-65. MILLOT, J. 1931a. Les diverticules intestinaux du cephalothorax chez les Araignees vrais. Zeitschrift Morphologie und Okologie der Tiere 21:740-764. MILLOT, J. 1931b. Les glandes venimeuses des Araneides. Annales des sciences naturelles, Zoologie, Paris 14:113-147. MILLOT, J. 1931c. Anatomie comparee de l'intestin moyen cephalo-thoracique chez les Araignees vraies. Compte rendu des VAcademie des sciences, Paris 192:375-377. MILLOT, J. 1933a. Le genre Aebutina (Araneides). Bulletin de la Societe Entomologique de France, Paris 58:92-95. MILLOT, J. 1933b. Notes complementaires sur 1'anatomie des Liphistiides et des Hypochilides, a propos d'un travail recent de A. Petrunkevitch. Bulletin de la Societe Entomologique de France, Paris 58:217-235. MILLOT, J. 1933c. L'anatomie interne des Dinopides. Bulletin de la Societe Entomologique de France, Paris 57:537-542. MILLOT, J. 1933d. L'anatomie interne des Liphistiides. In Bristowe, W.S. The Liphistiid Spiders. Proceedings of the Zoological Society, London. 1933:1045-1055. MILLOT, J. 1936. Metamerisation et musculature abdominale chez les Araneomorphes. Bulletin de la Societe Entomologique de France, Paris 61:181-204. MILLOT, J. 1938. L'Appareil sericegene d'Oecobius cellariorum Duges, suivi de quelques considerations generates sur les glandes secretrices de sole des Araneides. Travaux de la Station zoologique de Wimereux 13:479-487. MILLOT, J. 1949. Classe des Arachnides (Arachnida). I. Morphologie generale et anatomie interne. In P. Grasse. Traite de Zoologie. Paris 6:263-319. MORRISON, R., AND M. MORRISON. 1990. Australia: the Four Billion Year Journey of a Continent. Facts on File, New York, New York, USA. 334 pp. NIXON, K. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15:407-414. OPELL, B. 1979. Revision of the genera and tropical American species of the spider family Uloboridae. Bulletin of the Museum of Comparative Zoology 148:443-549. OPELL, B. 1987. The influence of web monitoring tactics on the tracheal systems of spiders in the family Uloboridae (Arachnida, Araneida). Zoomorphology 107:255-259. OPELL, B. 1989a. Functional associations between the cribellum spinning plate and capture threads of Miagrammopes animotus (Araneida, Uloboridae). Zoomorphology 108:263-267. OPELL, B. 1989b. Measuring the stickiness of spider prey capture threads. The Journal of Arachnology 17:112-114. OPELL, B. 1995. Ontogenetic changes in cribellum spigot number and cribellar prey capture thread stickiness in the spider family Uloboridae. Journal of Morphology 224:47-56. OPELL, B. 2001. Cribellum and calamistrum ontogeny in the spider family uloboridae: linking functionally related but separate silk spinning features. The Journal of Arachnology 29:220-226. PETERS, H.M. 1983. Struktur und Herstellung der Fangfaden cribellater Spinnen (Arachnida: Araneae). Verhandlungen des Naturwissenschaftlichen Vereins in Hamburg 26:241-253. PETERS, H.M. 1984. The spinning apparatus of Uloboridae in relation to the structure and construction of cap- ture threads (Arachnida, Araneida). Zoomorphology 104:96-104. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 81 PETERS, H.M. 1987. Fine structure and function of capture threads, In W. Nentwig ed., Ecophysiology of Spiders, Pp. 187-202, Springer, New York. PETERS, H.M. 1992a. On the spinning apparatus and the structure of the capture threads of Deinopis subrufus (Araneae, Deinopidae). Zoomorphology 112:27-37. PETERS, H.M. 1992b. Uber Struktur und Herstellung von Fangfaden cribellater Spinnen der Familie Eresidae (Arachnida, Araneae). Verhandlungen des Naturwissenschaftlichen Vereins in Hamburg 33:213-227. PETERS, H.M., AND J. KOVOOR. 1980. Un complement a l'appareil sericigene des Uloboridae (Araneae): le paracribellum et ses glandes. Zoomorphology 96:91-102. PETERS, H.M., AND J. KOVOOR. 1989. Die Herstellung der Eierkokons bei der Spinne Polenecia producta (Simon, 1873) in Beziehung zu den Leistungen des Spinnapparates. Zoologische Jahrbiicher, Abteilung Allgemiene Zoologie und Physiologie der Tiere 93:125-144. PETRUNKEVITCH, A. 1928. Systema Aranearum. Transactions of the Connecticut Academy of Arts and Sciences 28:1-270. PETRUNKEVITCH, A. 1933. An inquiry into the natural classification of spiders, based on the study of their inter- nal anatomy. Transactions of the Connecticut Academy of Arts and Sciences 31:299-389. PETRUNKEVITCH, A. 1939. The Status of the Family Archaeidae and the Genus Landana. Annals of the Entomological Society of America 32:479-501. PLATNICK, N.I. 1977. The hypochiloid spiders: A cladistic analysis, with notes on the Atypoidea (Arachnida, Araneae). American Museum Novitates 2627:1-23. PLATNICK, N.I. 2004. The world spider catalog, version 5.0. American Museum of Natural History, online at http://research.amnh.org/entomology/spiders/catalog/INTROLhtml. PLATNICK, N.I., J.A. CODDINGTON, R.F. FORSTER, AND C.E. GRISWOLD. 1991. Spinneret evidence and the high- er classification of the haplogyne spiders (Araneae, Araneomorphae). American Museum Novitates 3016:1-73. PLATNICK, N.I., AND R.R. FORSTER. 1987. On the first American spiders of the subfamily Sternodinae (Araneae, Malkaridae). American Museum Novitates 2894:1-12. PLATNICK, N.I., AND W.J. GERTSCH. 1976. The suborders of spiders: a cladistic analysis. American Museum Novitates 2607:1-15. PURCELL, W.F. 1909. Development and origin of respiratory organs in Araneae. Quarterly Journal of the Microscopical Society (N.S.) 54:1-25. PURCELL, W.F. 1910. The phylogeny of the Tracheae in Araneae. Quarterly Journal of the Microscopical Society. (N.S.) 54:519-564. RAMIREZ, M.J. 1995. A phylogenetic analysis of the subfamilies of Anyphaenidae (Arachnida, Araneae). Entomologica Scandinavica 26:361-384. RAMIREZ, M.J. 2000. Respiratory system morphology and the phylogeny of haplogyne spiders (Araneae, Araneomorphae). The Journal of Arachnology 28:149-157. RAMIREZ, M.J. 2003. A cladistic generic revision of the spider subfamily Amaurobioidinae (Araneae, Anyphaenidae). Bulletin of the American Museum of Natural History 277:1-262. RAMIREZ, M.J., AND C.J. GRISMADO. 1997. A review of the spider family Filistatidae in Argentina (Arachnida, Ataneae), with a cladistic reanalysis of filistatid genera. Entomologica Scandinavica 28:319-349. RAMIREZ, M.J., C. GRISMADO, AND T BLICK. 2004. Notes on the family Agelenidae in Southern South America (Arachnida: Araneae). Revista Iberica Aracnologia 9:179-182. RAVEN, R.J., AND K. STUMKAT. 2003. Problem solving in the spider families Miturgidae, Ctenidae and Psechridae (Araneae) in Australia and New Zealand. The Journal of Arachnology 31 (1): 105?121. RAVEN, R.J., AND K. STUMKAT. 2005. Revisions of Australian ground-hunting spiders: II: Zoropsidae (Lycosoidea: Aranaea). Memoirs of the Queensland Museum 50(2):347^-23. ROBSON, C.H. 1878. Notes on a marine spider found at Cape Campbell. Transactions of the New Zealand Institute 10:299-300. RUHLAND, M., AND W. RATHMEYER. 1978. Die Beinmuskulatur und ihre Innervation bei der Vogelspinnen Dugesiella hentzi (Ch.) (Araneida, Avaiculariidae). Zoomorphologie 89:33-46. SCHUTT, K. 2000. The limits of the Araneoidea (Arachnida: Araneae). Australian Journal of Zoology 48:135-153. 82 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II SCHUTT, K. 2002. The limits and phylogeny of the Araneoidea (Arachnida, Araneae). Thesis submitted to Mathematisch-Naturwissenschaftlichen Fakultat I der Humboldt-Universitat zu Berlin. 153 pp. SCHUTT, K. 2003. Phylogeny of Symphytognathidae s.l. (Araneae, Araneoidea). Zoologica Scripta 32:129-151. SHEAR, W.A. 1969. Observations on the predatory behavior of the spider Hypochilus gertschi Hoffman (Hypochilidae). Psyche 76:407-417. SHEAR, W.A. 1970. The spider family Oecobiidae in North America, Mexico, and the West Indies. Bulletin of the Museum of Comparative Zoology 140:129-164. SIERWALD, P. 1990. Morphology and ontogeny of female copulatory organs in American Pisauridae, with spe- cial reference to homologous features (Arachnida: Araneae). Smithsonian Contributions to Zoology (484): 1-24. SILVA DAVILA, D. 2003. Higher-level relationships of the spider family Ctenidae (Araneae: Ctenoidea). Bulletin of the American Museum of Natural History 274:1-86. SIMON, E. 1892. Histoire naturelle des Araignees, vol. 1, pt. 1. Encyclopedic Robet, Paris, France. 256 pp. SZLEP, R. 1966. Evolution of the web spinning activities: the web spinning in Titanoeca albomaculata Luc. (Araneae, Amaurobidae). Israel Journal of Zoology 15:83-88. TILLINGHAST, E.K., AND MA. TOWNLEY. 1994. Silk glands of araneid spiders: Selected morphological and physiological aspects. Pages 29^4 in D. Kaplan, W.W. Adams, B. Farmer, and C. Viney, eds., Silk Polymers: Materials Science and Biotechnology. American Chemical Society Symposium, Series 544. American Chemical Society, Washington, D.C., USA. TOWNLEY, M.A., E.K. TILLINGHAST, AND N.A. CHERIM. 1993. Moult-related changes in ampullate silk gland morphology and usage in the araneid spider Araneus cavaticus. Philosophical Transactions of the Royal Society of London, B 340:25-38. TOWNLEY, M.A., AND E.K. TILLINGHAST. 2003. On the use of ampullate gland silks by wolf spiders (Araneae, Lycosidae) for attaching the egg sac to the spinnerets and a proposal for defining nubbins and tartipores. The Journal of Arachnology 31(2): 209-245. TULLGREN, A. 1910. Araneae. In Wissenschaftliche Ergebnisse der Schwedischen Zoologischen Expedition nach dem Kilimandjaro, dem Meru und den umgebenden Massaisteppen Deutsch-Ostafrikas 1905-1906 unter Leitung von Prof. Dr Yngve Sjostedt, Stockholm, 20(6):85-172. UHL, G. 2000. Two distinctly different sperm storage organs in female Dysdera erythrina (Araneae: Dysderidae). Arthropod Structure and Development 29:163-169. WANG, X.P 2000. A revision of the genus Tamagrina (Araneae: Amaurobiidae), with notes on amaurobiid spinnerets, tracheae and trichobothria. Invertebrate Taxonomy 14:449^-64. WOLFF, R.J. 1977. The cribellate genus Tengella (Araneae: Tengellidae?). The Journal of Arachnology 5(2): 139-144. ZAPFE, H. 1955. Filogenia y funcion en Austrochilus manni Gertsch y Zapfe (Araneae-Hypochilidae). Trabajos del Laboratorio de Zoologia Facultad de Filosofia Y Education Universidad de Chile 2:1-53. ZIMMERMAN, W 1975. Biologische und rasterelektronenmikroskopische Festellungen an Oecobiinae, Uroecobiinae und Urocteinae (Araneae: Oecobiidae) als Beitrag zum "Cribellaten-Ecribellaten Problem." Thesis. Universitat Bonn, Bonn, Germany. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 83 APPENDICES 84 GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 85 Appendix 1. Taxa Examined to Provide Exemplar Data Exemplar specimens have labels reading "Entelegyne phylogeny atlas exemplar" Taxa marked with an asterisk * are not coded in our data matrix, but reference SEM preparations, photographs, or other observations. AGELENIDAE Neoramia sana Forster: d 9 from Saddle Hill, Dunedin, New Zealand, 29 October 1992, R. Forster, CAS; 9 from Saddle Hill, 5 km W Dunedin, elev. 280-380 m, ca. 45?56'S; 170?20'E, native bush, 14 April 1995, C. Griswold and T. Meikle, CAS (Meikle photo voucher). AMAUROBIIDAE Amaurobius fenestralis (Stroem): d 9 from Tisvilde, Denmark, 19 May 1991, C. Griswold, CAS and USNM. Amaurobius sp.: 9 from Page Mill Boulevard, 0.3 mi. from Skyline Drive, 1600 ft. elev., Santa Clara Co., California, USA, 5 July 1996, R. Carlson and C. Griswold, CAS (R. Carlson silk study voucher). Callobius sp.: 9 from redwood forest, 0.7 mi NE Fort Ross, Sonoma Co., California, USA, 21 June 1996, C. E. Griswold, R. Carlson, CAS (R. Carlson silk study voucher). Callobius bennetti (Blackwall): d 9 from Piscataquis Co., Maine, USA, June 1978, 0. Jennings, CAS; d from Soubunge Mountain, Maine, USA, 8 June 1978, D. Jennings, CAS; 9 from Hampshire Co., West Virginia, USA, 9 June 1985, USNM. Callobius gauchama Leech: d 9 from Seven Oaks, elev. 5600ft., San Bernardino Mts., San Bernardino Co., California, USA, 17-19 May 1996, R. Vetter, CAS (R. Carlson silk study voucher). Callobius nevadensis (Simon): 9 from Norden, California, USA, T. Davies, CAS (Davies photo voucher). Callobius pictus (Simon): 9 from Olympia, Washington, USA, 16 March 1931, H. Exline, CAS; 9 from Elum, Washington, USA, 20 May 1934, H. Exline, CAS. Macrobunus cf. multidentatus (Tullgren): 2c? 1 9 from Arroyo Cole Cole, 25 km N Cucao, Chiloe, Chile, 8-11 February 1991, M. Ramirez, MACN, (MACN-Ar, SEM preparations MJR 958-962, temporary mount MJR 973). Pimus pitus Chamberlin: 9 from Kyburz, California, USA, 15 September 1959, W. Gertsch and V. Roth, AMNH; c? from Yosemite, California, USA, 14 September 1959, W. Gertsch and V. Roth, AMNH. Pimus napa Leech: 4 9 paratypes from Napa Co., 3 mi W Oakville, California, USA, 15 February 1954, Roth and Schuster, AMNH (SEM preparation MJR 761-763); d paratype from 2 mi W Oakville, Napa Co., California, USA, 31 December 1953, AMNH (SEM preparation MJR 764). Pimus spp.: d 9 from Mendocino Co., California, USA, 15 September 1990, 0. Ubick, CAS; 9 from Eel River, Mendocino Co., California, USA, 20 September 1990, D. Ubick, CAS; 9 from Page Mill Road, 0.3 mi. from Skyline Boulevard, Santa Clara Co., California, USA, 1600 ft. elev., 5 July 1996, R. Carlson and C. Griswold, CAS (R. Carlson silk study voucher). Retiro sp.: d 9 from Lima, Peru, H. Exline, CAS. AMPHINECTIDAE Maniho ngaitahu Forster and Wilton: d 9 from Kaituna valley, South Island, New Zealand, 13 April 1964, R. Forster, CAS. Maniho pumilio Forster and Wilton: d 9 from Butterfly Creek, Eastbourne, New Zealand, Nothofagus/tree fern forest, 18 April 1996, L.J. Boutin, CAS (R. Carlson silk study voucher). Maniho tigris Marples: 9 from Butterfly Creek, Eastbourne, New Zealand, Nothofagus/tree fern forest, 18 April 1996, L.J. Boutin, CAS (R. Carlson silk study voucher). *Metaltella rorulenta (Nicolet): 9 from Malalcahuello, Chile, 25 January 1985, N. Platnick, CAS; d from Nahuelbuta, Chile, M. Irwin and E. Schlinger, CAS. Metaltella simoni (Keyserling): d 9 from Riverside, California, USA, 7 July 1996, R. Vetter, CAS (R. Carlson silk study voucher); 3d 11 9 from St. Tammany Co., Pearl River, Lousiana, USA, AMNH; 2d 2 9 from same locality, 1965, L. Roddy, AMNH; d from Villa Madero, Buenos Aires, Argentina, August 1998, C. Scioscia, MACN (MACN-Ar, preparation MJR 975); d from El Palmar, Entre Rios, Argentina, November 86 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II 1988, M. E. Galiano, MACN (MACN-Ar, preparation MJR 976); d from Puerto Obligado, Buenos Aires, Argentina, March 1983, E. Maury and P. Goloboff, MACN (MACN-Ar, preparation MJR 977); juv. from Whittier, California, USA, March 2005, L. Vincent, CAS (Vincent photo voucher). *Metaltellinae sp.: 29 from Llanquihue, Concordia, Fundo Pedernales, Chile, 4 February 1986, T. Cekalovic, AMNH. ANAPIDAE *Crassanapis chilensis Platnick and Forster: 2d" 4 9 from Parque Nacional Puyehue, Aguas Calientes, Osorno, Chile, Moczarski-Tullgren extractor, 13-17 December 1998, M. Ramirez, L. Compagnucci, C. Grismado and L. Lopardo, MACN (MACN-Ar, 1 d and 1 9 mounted for SEM, preparation MJR 677). ARANEIDAE Araneus diadematus Clerck: d from Seattle, Washington, USA, 3 September 1933, H. Exline-Frizzell, CAS; ld49 from Swansea, W of High Park, Ontario, Canada, 43?41'N, 79?28'W, 1 September 1945, W. Ivie and T. Kurata, AMNH (SEM preparations MJR 819-822). *Argiope argentata (Fabricius): 1 9 from 1 mi. S. Millers Landing, Baja California, Mexico, in coastal sand dunes, 6 July 1973, S. Williams and K. Blair, CAS. *Metepeira atascadero Piel: d 9 from Guanajuato, Mexico, September 1976, C. Griswold and R. Jackson, CAS. ARCHAEIDAE Archaea workmani (O. P-Cambridge): d 9 from Vohiparara, Pare National de Ranomafana, ca. 21?14'S, 47?24'E, elev. 1100 m., Fianarantsoa Prov., Madagascar, April 1998, C. Griswold, D. Kavanaugh, M.J. Raherilalao and D. Ubick, CAS; 3d 3 9 and immatures from Pare Nationale Ranomafana, Talatakely, 21?14.9'S, 47?25.6'E, Fianarantsoa Prov., Madagascar, 5-18 April 1998, C. Griswold, D. Kavanaugh, N. Penny, M. Raherilalao, J. Ranoriaranarisoa, J. Schweikert, D. Ubick, CAS (SEM preparations MJR 791-797). (Coville photo voucher) AUSTROCHILIDAE *Austrochilus forsteri Grismado, Lopardo & Platnick: 9 from Contulmo, Chile. February 1992, M. Ramirez, MACN (Ramirez photo voucher). *Austrochilus melon Platnick: juvenile from Cuesta Pucalan, Chile, 19 September 1966, E.I. Schlinger, USNM. Hickmania troglodytes (Higgins and Petterd): d 9 from cave at Mole Creek, Tasmania, Australia, 3 June 1996, J. Boutin, CAS; silk from Newdegate Cave, Tasmania, Australia, 11 October 1998, J. Boutin, CAS (R. Carlson silk study voucher); silk samples from Tasmania, L. Trimmer Cave, 15 July 2000, N. Dioran, MACN (MACN-Ar, SEM preparations MJR 74, 77); Id 1 9 from W of Deloraine, Mole Creek Cave, in small Cavern, Tasmania, Australia, 3 June 1996, L.J. Boutin, CAS (SEM preparations MJR 787, 788). Thaida peculiaris Karsch: d 9 from Los Lagos, Valdivia, Neltumo, Chile, 23 November 1988, V and B. Roth, CAS; d 9 from Region Pucon by Lago Villarica, Chile, 14 December 1988, V and B. Roth, CAS; imma- ture from Osorno, Chile, 12 February 1985, USNM; silk samples from Puerto Blest, Neuquen, Argentina, 7-20 January 2000, L. Lopardo and A. Quaglino, MACN (SEM preparations MJR 72, 78); 1 9 from Puerto Blest, Parque Nacional Nahuel Huapi, Argentina, 7-20 January 2000, L. Lopardo and A. Quaglino, MACN (MACN9976, SEM preparations MJR 675, 676, 839); 1 d from Bellavista, N shore Lago Villarrica, elev. 310 m, site 655, window trap, valdivian rainforest, Cautm, Chile, 15-30 December 1982, A. Newton and M. Thayer, AMNH (SEM preparation MJR 765); 9 from Puyehue, Chile, 15 December 1998, M. Ramirez, MACN (Ramirez photo voucher). CTENIDAE Acanthoctenus spiniger Keyserling: d from Ecuador, collected on bananas in New York, USNM; 9 from Changuinola, Panama, 1965, J. Harrison, CAS; 9 from Tegucigalpa, Honduras, 1953, Gilbert, USNM. Acanthoctenus cf. spinipes Keyserling.- 2d 1 9 from Rio Samiria, Loreto, Peru, fogging, May 1990, T. Erwin et al., MUSM. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 87 Acanthoctenus sp.: 9 from Rio Samiria, Peru, 15 May 1990, D. Silva, MUSM. Acanthoctenus sp.: 9 from 25km N Formosa, 25?59'S, 58?12'W, Estancia Guaycolec, Argentina, elev. 185m, 26 February to 10 March 1999, S. Hey don and J. Ledford, CAS (Griswold photo voucher). DEINOPIDAE Deinopis spinosus Marx: d 9 from Gainesville, Florida, USA, 1 July 1994, C. Griswold, CAS; 9 from Alachua Co., Florida, USA, J. Coddington, 3 August 1985, USNM; 9 from Finca La Selva, Heredia Province, Costa Rica. October 1981, C. Griswold and R. Coville, CAS (Coville photo voucher); 9 from same locality, 13 August 1985, J. Coddington, USNM (Coddington photo voucher). Menneus camelus Pocock: 9 from Twin Streams near Mtunzini, Natal, Zululand, South Africa, 19-20 January 1984, T. Meikle and C. Griswold, NMSA (Meikle photo voucher); 9 from Mpumalanga Pro v., Songimvelo Nature Reserve, Kromdraai, 26?2'33"S, 31?0'5"E, 800m elev., South Africa, 16-23 March 2001, D. and S. Ubick, CAS; 9 from Kaibos, Kenya, 23 May 1980, B. Lamoral, USNM. Menneus sp.: 1 9 from Tembe Elephant Park, elev. 115 m, 27?2'32.8"S, 32?25'24.4"E, KwaZulu-Natal, South Africa, 9-12 April 2001, M. Ramirez, MACN (SEM preparations MJR 623). DESIDAE Badumna longinqua (L. Koch): d 9 from Montevideo, Uruguay, 5 November 1961, R. Capocasale, CAS; d 9 from San Francisco, California, USA, July 1996, R. Carlson, CAS (R. Carlson silk study voucher); 1 9 from Maui, Hawaii, USA, Lennox, MM85-19, USNM; Id 1 immature from Ship Cr., Haast, New Zealand, 8 December 1977, E. I. Schlinger, CAS; 2919 subadults from Pacifica, San Mateo Co., California, USA, 13 May 1995, K. Ribardo, CAS. *Badumna sp.: 9 from from Waitomo, near caves, ca. 38?14'S, 175?08'E, North Island, New Zealand, 3 April 1995, C. Griswold and T. Meikle, CAS (Meikle photo voucher). Desisformidabilis (O. P.-Cambridge): ld3 9 from "The Island", Kommetje, Cape Peninsula, Western Cape, South Africa, May 1966, B. Lamoral, CAS (identified by B. Lamoral, SEM preparation MJR 274, tempo- rary preparation MJR 974); d 9 from same locality, 34?9'S, 18?20'E, 30 air km S Cape Town, intertidal zone, under rocks, 13 March 2001, K. Muller, S. Prinsloo, L. Prendini, D. and S. Ubick, CAS (SEM prepa- rations MJR 275-280); d 9 from Luderitz, Namibia, among intertidal rocks, October 1985, C. Griswold and T Meikle, NMSA (Meikle photo voucher). Matachia australis Forster.- 9 from Saddle Hill, Dunedin, New Zealand, 29 October 1992, R. Forster, CAS. Matachia marplesi Forster: d from Helena Bay, North Island, New Zealand, 3 February 1994, E. Schlinger, CAS. Matachia spp.: 9 from Parakaunui Falls, Catlins Coastal Rainforest Park, 10.8km 201?S Owaka, 846.51592?, E169.55887?, elev. 20m, Nothofagus/podocarp forest, Otago Pro v., New Zealand, 16 February 2005, C. Griswold, D. Silva and H. Wood, OMD (Wood photo voucher); 9 from Banks Peninsula, Hay Scenic Reserve nr. Pigeon Bay, elev. 50 m, ca. 43?42'S, 172?54'E, native forest, 10 April 1995, C. Griswold and T. Meikle, CAS (Meikle photo voucher). Phryganoporus candidus (L. Koch): d d 9 9 from Black Mountain, Canberra, Australia, 7 August 1990, C. Griswold and T Meikle, CAS and USNM. DICTYNIDAE Aebutina binotata Simon: d 9 from Rio Cuyabeno, near via Atipishea, Sucumbos, Ecuador, August 1995, G Cahas, CAS; d from Aguas Negras, near Tarapuy, Napo, Ecuador, 1984, L. Aviles, CAS; 1 d and several 9, from Divisoria, 1700 m, Huanuco, Peru, 23 September -3 October 1946, F. Woytkowski, AMNH. Dictyna arundinacea (Linnaeus): d 9 from Tuva, Russia, 9 June 1995, Y. Marusik, CAS; d 9 from Lyngby, Denmark, 26 May 1991, C. Griswold and N. Scharff, USNM; 9 9 from Skibo Castle, Dornoch, Sutherland, Scotland, August 1935, R. Miller, AMNH; 2d 19 from Helsingfors, Haga, Finland, 3 June 1951, W. Hackman, AMNH (SEM preparation MJR 818). Dictyna bostoniensis Emerton: d 9 from Minnesota, USA, 27 June 1936, H. Exline-Frizzell, CAS. Dictyna sp.: 9 from Whittier, California, USA, March 2005, L. Vincent, CAS (Vincent photo voucher). Lathys delicatula (Gertsch and Mulaik): 9 from Southwestern Research Station, Portal, Arizona, USA, 19 September 1972, D. Ubick, CAS. 88 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Volume 56, Supplement II Lathys humilis (Blackwall): d 9 from Eastling, Kent, United Kingdom, 16 May 1993, A. Russell-Smith, CAS. Lathys immaculata (Chamberlin and I vie): d from Bradley, Arkansas, USA, 2 February 1964, CAS. *Mallos sp.: d 9 from Chiricahua Mountains, Arizona, USA, 26 May 1975, D. Ubick, CAS. *Mallos sp.: 9 from Mt. Lemon, Arizona, USA, D. Ubick, CAS (R. Carlson silk study voucher). Nigma linsdalei (Chamberlin and Gertsch): d 9 from San Francisco, Calfornia, USA, June 1994, D. Ubick, CAS. Tricholathys spiralis Chamberlin and I vie: d 9 from Lenore Lake, Washington, USA, 8 May 1938, Hatch, CAS. Tricholathys sp.: 9 from San Francisco, California, USA, 15 April 1992, K. Dabney, CAS. ERESIDAE *Dresserus subarmatus Tullgren: d 9 from Nairobi, Kenya, AMNH. *Dresserus sp.: d 9 from Mazumbai, Muheza District, Tanzania, elev. 1600-1800 m, 4?49'S, 38?30'E, 11-20 November 1995, C. Griswold, D. Ubick, and N. Scharff, CAS. *Dresserus sp.: immature from Tembe Elephant Park, elev. 115m, 27?2'32.8"S, 32?25'24.4"E, KwaZulu- Natal, South Africa, 9-12 April 2001, M. Ramirez, MACN. Eresus cinnaberinus (Olivier): d 9 from Fiesch Wallis, Switzerland, Schenkel, AMNH. Eresus cf. cinnaberinus (Olivier): Id from Peloponesus, Mistras, Greece, 19 June 1982, B. and H. Malkin, AMNH (SEM preparations MJR 809, 810); 1 9 from Igrherm, Anti Atlas, elev. 1600-1700 m, Morocco, 23-29 May 1974, B. Malkin, AMNH (SEM preparations MJR 811, 831). Eresus sandaliatus (Martini & Goeze): 9 d from road between Rye to Gl. Sal ten, 9?35'E, 56?05'N, SE of Silkeborg, Denmark, 25 November 1994, P. d. place Bj0rn, CAS. Stegodyphus dumicola Pocock: d 9 from Spieonkop Dam, south shore, 30 km SW Ladysmith, elev. 900m, 28?41'S, 29?28'E, mixed grassland and dry bushveld, KwaZulu-Natal, South Africa, December 1985, T. Meikle and C. Griswold, CAS; d 9, same data, 9 January 1986, CAS. Stegodyphus mimosarum Pavesi: d 9 from Spieonkop Dam, south shore, 30 km SW Ladysmith, elev. 900m, 28?41'S, 29?28'E, mixed grassland and dry bushveld, KwaZulu-Natal, South Africa, December 1985, T Meikle and C. Griswold, CAS NMSA (Meikle photo voucher); d 9 from 31 mi. SE Ft. Hill, elev. 1600m, Malawi, 20 February 1958, E. Ross, CAS; 1 juv. from Pare National de Ranomafana, Vohiparara, ca. 21?14'S, 47?24'E, elev. 1100 m., Fianarantsoa Pro v., Madagascar, 5-7 November 1993, C. Griswold, CAS; 1 9 from Phinda Resource Reserve, elev. 38 m, S 27?50'43", E 32?18'49.1", KwaZulu-Natal, South Africa, 13-15 April 2001, M. Ramirez, MACN (SEM preparations MJR 767, 768). Stegodyphus sp.: 9 from ShweSettaw Wildlife Reservation, Magway Division, Myanmar, N20?4'7.4", E94?35'2.8", elev. 137m, deciduous forest, at night, 29 September 2003, C.Griswold, P.Sierwald, DUbick, Aye Aye Cho and Tin Mya Soe, CAS (Dong Lin photo voucher). FILISTATIDAE Filistata insidiatrix (Forskal): d 9 from rock wall in park, Barcelona, Spain, 13 October 1986, J. Coddington, USNM; several females and eggsacs with spiderlings from Siena, 4 km S San Giminiano, Fattoria Voltrona, Reg. Toscania, Italy, 12 July 2001, M. Ramirez, MACN (SEM preparations MJR 798-803, 835). *Filistatinella sp.: 9 from Arroyo Seco, Monterey Co., California, USA, 7 May 1995, 0. Ubick, CAS. Kukulcania hibernalis (Hentz).- d 9 from Archbold Biological Station, 8 mi. S. Lake Placid, Highlands Co., Florida, USA, 26 June 1978, C. Griswold, CAS; d 9 from Clearwater, Pinella Co., Florida, USA, December 1962 - February 1963, O. Paulus, CAS; 1 9 from Alachua Co., Florida, USA, 8 August 1985, USNM; 9 and spiderlings from eggsac reared in lab, from Las Gamas, 20km W Vera, Santa Fe, Argentina, 27-30 October 1994, M. Ramirez and J. Faivovich, MACN (SEM preparations MJR 33-37); silk samples from Buenos Aires, Argentina, 11 January 2001, L. Lopardo, MACN (SEM preparation MJR 73); d 9 from Buenos Aires, Argentina, 1 January 2005, M. Ramirez, MACN (Ramirez photo voucher); Id from Savanna, Georgia, USA, 18 August 2001, T. Sullivan, AMNH (SEM preparation MJR 838). Kukulcania sp.: d from Kings Co., 25th Ave. near Parejo Hill, 35?37'14"N, 119?54'52"W, California, USA, 18 May 1997, D. Ubick and W. Savary, CAS (penultimate observed carding cribellate silk); silk samples, 9 from North Carolina, USA, unspecified locality, alive in AMNH. GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 89 *Misionella mendensis (Mello-Leitao): 9 from Misiones, Argentina, M. Ramirez, MACN (Ramirez photo voucher). *Pritha nana (Simon): 9 from Bolzano, Italy, M. Ramirez, MACN (Ramirez photo voucher). GRADUNGULIDAE Gradungula sorenseni Forster: c? 9 from Saltwater Forest, west coast, South Island, New Zealand, 2 December 1991, P. Walsh, CAS; 7c? 3 9 from South Island, west coast, Saltwater Forest, pit trap in rimu forest, New Zealand, 1 December 1991, P. Walsh, CAS (SEM preparations MJR 789, 790). HUTTONIIDAE Huttonia palpimanoides O. R-Cambridge: c? 9 from dead fern fronds in Leith Saddle forest, Dunedin, South Island, New Zealand, 3 January 1975, R. Forster, CAS. Huttonia sp.: 1 9 from Orongorongo Res. Project, Wellington, New Zealand, 1 June 1992, OMD (SEM prepa- rations MJR 827-829); 1 c? with same data, OMD; 1 9 from Kapiti Island, off SW Coast of North Island, New Zealand, 40?52'S, 174?55'E, ex pitfall trap, 1996, J. Mclartney, CAS (SEM preparation MJR 830); 1 immature from Otago, Trotters Gorge, New Zealand, from ferns, 6 February 1979, R. R. Forster, USNM (tracheae examined). HYPOCHILIDAE Hypochilus sheari Platnick: c? from Yancey Co., Crabtree Meadows, 25 mi. N Marion, North Carolina, USA, 4 September 1976, C. Griswold and R. Jackson, CAS. Hypochilus pococki Platnick: c? 9 from Ramsey Cascade, Great Smokey Mts. N.P, elev. 2080 ft., Sevier Co., Tennessee, USA, 10 August 1995, F. Coyle, CAS; 9 from Haywood Co., North Carolina, USA, 3 October 1960, USNM; many c? and 9 from above Crabtree to Betsey's Gap, 3956 ft. elev., Haywood Co., North Carolina, USA, 3 October 1960, W. Gertsch, W. Ivie, AMNH (SEM preparations MJR 735, 836, 837, 863). Hypochilus kastoni Platnick: 9 from Mount Shasta, California, USA, J. Ledford, CAS (Ledford photo vouch- er). *Ectatosticta sp.: 19 from Taibai Shan S flanks, above Houshenzi, mixed coniferous/Rhododendron forest, elev. 3050 m, Shaanxi Pro v., China, 12-13 June 1997, P. Jager and B. Martens, JGU (SEM preparation MJR 755). MIMETIDAE Mimetus hesperus Chamberlin: c? from Baboquivari Canyon, Baboquivari Mts., Pima Co. Arizona, USA, 21 July 1952, H. B. Leech and J. W. Green, CAS; 9 from Tampico, Tamulipas, Mexico, summer 1966, CAS; 2 9 from Kingston Camp, 30 mi S Austin, Toiabe Range, elev. 3700 ft, Lander Co., Nevada, USA, 12 August 1966, P.P. and M. Rindge, AMNH (SEM preparations MJR 823, 824); 1 9 from Valles, San Luis Potosi, Mexico, 19 July 1956, W. Gertsch, V. Roth, AMNH; lc? from Brown Canyon, Baboquivari Mts., Arizona, USA, 9 July 1952, M. Cazier and W Gertsch, AMNH (SEM preparation MJR 825). NEOLANIDAE Neolana dalmasi (Marples): <3 9 from Lake Okataina, New Zealand, 20 October 1984, 0. Court, OMD; co cn o O H H H 0 0 H H H H 0 0 H 0 0 0 0 0 0 0 0 0 0 0 0 H 0 o o H H H H H ^ H H H H H ^ ^ H H H H ^ H H H H H ^ ^ H H H H H H CM ? ? ? H ? ? ? ? ? CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM H H H H H H H H H H H H H H H H H 1-1 H H H H H H H H H H H H p. 0 o. 0 0 0 0 0 C o. o. o. C 0 c. 0 C 0 0 C 0 0 0 C 0 o c. H H H H H H H H H H H H ^ H H H H ^ H H H H H H H H H H H H H H H H H H H H H H H H H H H H H ,_, H H H H H H H H H H H H O O ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? O O o o 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? o c o o 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o ? H ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? o o o o 0 0 0 0 0 0 0 0 0 0 ? H 0 0 0 0 H H H 0 H 0 0 0 0 0 H H OS o o ? H " ^ ? ii ? ? ? ? ? ? ? ? ? ? ii ? ii ? CM c. ? ii ii ? o o o o 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o US I 01 I OS as c ID 01 a, 01 01 W US OS U as ra ID ID c? c^ 4 01 0 as I OS as ?a ID 01 c^ ID 4 -H 3 L P ?? ? 01 a H ? J3 S c^ 4 0 X| s^ I T3 ra h , -?; ?a I as J3 ID ID 5 W U b H 01 H 0 q 4 s^ a! H. 0 ^ as *- y. a M ID ID ^ M q o US ti- 01 01 b. 0) > >. c 01 M T3 0 _B U H nt '<- OS \ 01 ^ i i S 1 li ID ^ ID as 01 5 ID 9 3 o t, -- 12! Q "3 k i4 k a > % o- 12! w ^ 12! s ? 0 S DH m 9 m ? ^ Cj GRISWOLD, RAMIREZ, CODDINGTON, & PLATNICK: ENTELEGYNE SPIDER DATA 95 -a o ?a a> 3 8 ??a a o u % 3 z td a. a. < oa-ioooooo I>COCNCNCNOJOJCN oi>oooooo OUJOOOOOO r-^eoooooo o o o o o o y^LTlOOOOOO kn^oooooo o o o o o o (N CN (N Ol Ol Ol (N O II <_> 1) r c E n ? IN m -n ^moooooo ^CCNOOOOOO ^JOOOOOOO ^fOOOOOOO o o o o o o o o o o o o o o o o o o o o o o o o rOiHOOOOOO mooooooo o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o OlOOOOOOO CTi o o o o o o 00 O O O O O O o o o o o o o l?> O O O O O O in o o o o o o ^f o o o o o o CO O O O O O H CN O O O O O O o o o o o o o CTi o o o o o o [> o o o o o o ^o o o o o o o o o o o o o to 0) Q) 0 ' CJi o Cl in EH N 0 ?) 0 " CN CN CN CN CN o o o o o o o o o o o o o o o o o o o O O O O O O O o o o o o o o o o o o o o o O O O O O O H O O O O O H H o o o o o o o o o o o o o o o o o o o o o H O O O O O O o o o o o o o o o o o o o o o o o o o o o O O O O O O o O O O O O O O o o o o o o o H O O O O O O O o o o o o o O O O O O O O o o o o o o o O O O O O O O o o o o o o o to -q ra H IN o 1 o u % 3 z u a. a. < a II H CN ai ^. 0 0 ^. ^. 0 rt o Z] g 2 ? ? ' ' o H O o |> o o o o o o i?> o o o o o o Lfl o o o o o o ?^ O O O O O O m o o o o o o CN O O O O O O O O O O O O O ocooooooo 0[>000000 omoooooo o^oooooo omoooooo o o o o o o o o CN CN CO ^ o o o o coooooooo o o o o o o o o o o o o o o o o I O O I o. o o o O. I O I oooooooooooo I ooooooooooo I OOOOOOOOOOO I ooooooooooo ooooooooooo oommoommmoo OOOOOOOOOOO I OOOOCNCNOOOCNO HOOOOOOOOOOO oooooooooooo OOOOOOOOOOOO] oooooooooooo oooooooooooo oooooooooooo oooooooooooo oooooooooooo OOOOOOOOOOCNCN oooooooooooo oooooooooooo oooooooooooo oooooooooooo O O O O O I o I o o o o o o oooooooooooo I O O O O I O I I O O O O I O I I O O O O I O I I O O O O I O I ooooooon-o r-\r-\r-\r-\r-\r-\r-\C-. r-\ ooooooooo ooooooooooo OO I oooooooo o o o o o as II -c : : II O -c c ? 2 o X ft Q m ;.;; 73 00 ? i> 3 = 8 o 3 z td a. a. < o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o I o o o o o o o o o o o o o o o o o H O O v-\ v-\ o o o o o o o o o o o o o o o o o o ro ro ro ro (N (N o o o o o o ? tS 3 0 J3 Si 5 % % p.. o o- o- o o o o r^ oooooooooo ooooon-oooo oooooooooooo i-Hi-HOOOOOOOOOO oooooooooooo MMMMOOOMOOOM oooooooooooo HOOOOOOOOOOO MMMMOOOOOOOO oooooooooooo oooooooooooo I ooooooooooo OOOOMOOMOMMM OOOOOOOOOMOM oooooooooooo oooooooooooo oooooooooooo OOOOOMMMMOOO O I O IOOOOO I O I o o o o o H O H H O O v-i o o o o o o o oooooooooooo oooooooooooo o o o o o o o O O O O O T-\ o oooooooooooo H I MMOOOOOOHH OOOOOOOOOHOO OOMOOOOOOOOO oooooooooooo romrorororoHroroHiHiH oooooooo to 0 bi 0 in CO in 0 0) fO XJ fd fd