searchable PDF - Association for Mexican Cave Studies

PREFACEThe present volume is the second in a series devotedto the cavernicole and endogean fauna ofNorth America, including Mexico. A majority ofthe species described in this volume are from Texas,but new species are also described from other statesin the United States and from Mexico.Texas has long been known for its unusual anddiverse subterranean fauna. The first species to berecovered from the underground habitat was the remarkableblind salamander Typhlomolge rathbuniStejneger from the old U.S. Fisheries artesian wellin San Marcos, Hays County. The description ofthis species was soon followed by descriptions ofnew species of blind crustaceans.Despite the interest generated by these descriptionsonly occasional collections were made in thestate until Dr. Thomas C. Barr, Jr., then of TexasTechnological College in Lubbock, visited severalcaves and encouraged visits by other speleologists.These discoveries in the late 1950s further emphasizedthe interest of the Texas cave fauna.Shortly following the organization of the TexasSpeleological Survey in 1961 an active program ofbiological study was initiated by James R. Reddellwith the help of numerous cave explorers in theAustin area. Dr. Robert W. Mitchell of Texas TechUniversity in Lubbock encouraged a number of students,including William R. Elliott, to pursue biologicalstudies in Texas caves as part of their graduateresearch.In recent years the listing of five cave arthropods[the Tooth Cave spider Leptoneta (nowNeoleptoneta) myopica Gertsch, the Tooth Cavepseudoscorpion Microcreagris (now Tartarocreagris)texana Muchmore, the Bee Creek Caveharvestman Texella reddelli Goodnight andGoodnight, the Kretschmarr Cave mold beetleTexamaurops reddelli Barr and Steeves, and theTooth Cave ground beetle Rhadine persephone Barr]on the U.S. Fish & Wildlife Service List ofEndangered Species has resulted in an extensivesampling program for Travis, Williamson, andadjacent parts of Burnet and Hays Counties. Thisstudy revealed the presence of a number of newspecies in this, the best-studied, part of the state, aswell as better defining the ranges of other species.Two other species originally confused with two ofthe endangered species are described as new in thisvolume. Because populations of these species wereincluded in the original endangered species listingthey are also considered endangered by the U.S.Fish & Wildlife Service. Texella n.sp. (see Ubickand Briggs, this volume) was originally placed inTexella reddelli, and Batrisodes (Excavodes) n.sp.(see Chandler, this volume) was placed inTexamaurops reddelli.Despite extensive collections throughout the stateby a large number of students of cave biology andan eager cadre of cave explorers the fauna of thestate remains poorly known. The continuing discoveryof new species in the best-studied parts of thestate indicates strongly that we have much to learnabout the cave fauna of this remarkable region. Fewdistributions are adequately known and doubtlessnumerous new species await discovery in areas stillbarely touched.The present volume includes taxonomic works onseven orders of arthropod. Of special significanceare revisions of two of the more important groupsoccurring in Texas caves, the subgenus Cicurella ofthe genus Cicurina, and the harvestman genusTexella. Both groups occur outside Texas, but thecenter of diversity as presently understood is thekarst region of central Texas. Also of considerablezoogeographic interest is the discovery of twoMexican genera of aquatic isopod in Texas: the cirolanidgenus Speocirolana and the stenasellid genusMexistenasellus. Holsinger continues his study ofthe amphipod fauna of Texas subterranean waterswith description of significant new taxa from caves,springs, wells, and hyporheic habitats in Texas.Muchmore reviews the cave pseudoscorpions ofTexas and New Mexico with description of additionalnew troglobites and troglophiles. Cokendolpherand Reddell revise the schizomid family Protoschizomidae,which though primarily Mexicanincludes one (undescribed) species in Texas. Rothdescribes the first troglobitic theridiid spider inNorth America. Finally, Chandler reviews the beetlefamily Pselaphidae in Texas caves, with descriptionsof the first eyeless species of the genus Batrisodes.Altogether 110 new species are described in thisv

volume, of which 61 Texas species are believed tobe troglobites and another 16 are Texas troglophiles.An additional nine troglobites (1 each from Alabama,Arizona, and New Mexico and 6 from Mexico)are also described.The present volume includes only a few of thegroups known to contain undescribed species inTexas caves. It is hoped that a future volume can bepublished covering additional families and orders,including millipedes, terrestrial isopods, entotrophs,and cave crickets.Many of the species in this volume occur in areasundergoing rapid urbanization and may be listed asendangered in the future. Of species treated herein,four (two species of Texella, Texamaurops reddelli,and one species of Batrisodes) are considered asendangered by the U.S. Fish & Wildlife Service.Many of the caves from which fauna has beenrecorded have already been destroyed; others,including some of considerable biological importance,are seriously threatened. It is one of the goalsof the present volume to make available to conservationistssufficient taxonomic and distributional informationas to allow a determination to be made onthe necessity of protecting species by their placementon the U.S. Fish & Wildlife Service List ofEndangered Species.My greatest debt of gratitude is to Dr. WilliamR. Elliott for his invaluable contributions to thepreparation of this book for the printer. His encouragementand friendship not only during preparationof this book but for many years have been aninspiration to me in my own work.I wish to express my particular appreciation tothe authors of the papers for their cooperation,especially where newly obtained material requiredrevisions of the original manuscripts. I speciallythank William R. Elliott, Andrew Grubbs, ScottHarden, David McKenzie, Marcelino Reyes,William Russell, and George Veni for theircollecting efforts for many years. Without theircooperation and friendship we would know far lessabout the cave biology of Texas than we do. Dr.Robert W. Mitchell is thanked for his inspiration tome both as a friend and student; he has alwaysprovided whatever support was asked. Dr. Mitchellis also thanked for contributing the photograph onthe cover of this volume.vi

TABLE OF CONTENTSHOLSINGER, JOHN R. Four new species of subterranean amphipod crustaceans(Artesiidae, Hadziidae, Sebidae) from Texas, with comments on theirphylogenetic and biogeographic relationships 1BOWMAN, THOMAS E. Two subterranean aquatic isopod crustaceans newto Texas: Mexistenasellus coahuila (Cole and Minckley), 1972 (Asellota:Stenasellidae) and Speocirolana hardeni, new species (Flabellifera: Cirolanidae) 23COKENDOLPHER, JAMES C., and JAMES R. REDDELL. Revision of theProtoschizomidae (Arachnida: Schizomida) with notes on the phylogeny of the order 31GERTSCH, WILLIS J. Distribution patterns and speciation in North American cavespiders with a list of the troglobites and revision of the cicurinas of the subgenus Cicurella 75ROTH, VINCENT D. A new and first troglobitic spider from Arizona (Thymoites, Theridiidae) 123MUCHMORE, WILLIAM B. Cavernicolous pseudoscorpions from Texas and NewMexico (Arachnida: Pseudoscorpionida) 127UBICK, DARRELL, and THOMAS S. BRIGGS. The harvestman family Phalangodidae.3. Revision of Texella Goodnight and Goodnight 155CHANDLER, DONALD S. The Pselaphidae of Texas caves (Coleoptera) 241INDEX OF NEW TAXA 255Vll

Holsinger, J.R. 1992. Four new species of subterranean amphipod crustaceans (Artesiidae, Hadziidae, Sebidae) from Texas, withcomments on their phylogenetic and biogeographic relationships. Texas Mem. Mus., Speleol. Monogr., 3:1-22.FOUR NEW SPECIES OF SUBTERRANEAN AMPHIPODCRUSTACEANS (ARTESIIDAE, HADZIIDAE, SEBIDAE) FROMTEXAS, WITH COMMENTS ON THEIR PHYLOGENETIC ANDBIOGEOGRAPHIC RELATIONSHIPSJohn R. HolsingerDepartment of Biological SciencesOld Dominion UniversityNorfolk, Virginia 23529ABSTRACTFour new stygobiont amphipod crustaceans are describedfrom a variety of groundwater habitats (e.g., caves, springs,interstitial media) in south-central and western Texas. The newspecies are Artesia welbourni (Artesiidae), Holsingeriussmaragdinus (Hadziidae), Mexiweckelia hardeni (Hadziidae),and Seborgia hershleri (Sebidae). The new locality records forArtesia in Culberson County, and Holsingerius and Seborgia inVal Verde County are the first for these genera outside theartesian well in San Marcos, Texas, whereas the new record forMexiweckelia in Medina County documents a significant rangeextension for this genus from north-central Mexico tosouth-central Texas. Seborgia relicta, previously described fromthe artesian well in San Marcos, is also newly recorded fromMedina County.Both the phylogenetic and biogeographic implications of thesenew taxa are discussed at some length because they providefurther insight into the origin and evolutionary biology ofstygobiont crustaceans in southern Texas and northern Mexico.INTRODUCTIONBy far the greatest taxonomic diversity of subterraneanamphipod crustaceans in North America isfound in Texas, primarily in groundwater habitats ofthe Edwards Aquifer and associated Balcones faultzone in the south-central part of the state. In themost recently published paper on Texas subterraneanamphipods (Holsinger and Longley, 1980), one newfamily, four new genera, and six new species weredescribed from an artesian well in San Marcos.Some of these species were also recorded fromnearby San Marcos Springs and deep artesian wellsnear San Antonio. Collecting of subterraneanamphipods in Texas has continued since this studywas published in 1980, resulting in many new localityrecords for previously described species and thediscovery of a significant number of taxa new to science.In this paper four of these newly discoveredspecies, all with close morphological affinities to thepreviously noted Edwards Aquifer well amphipodfauna, are described from subterranean groundwatersin Medina, Val Verde and Culberson counties.In addition, one species previously known onlyfrom the artesian well in San Marcos is recordedfrom Medina County.Descriptions of the four new species raise the totalnumber of stygobiont amphipods described fromTexas to 21. This remarkably diverse fauna encompassesrepresentatives of five families and nine genera.One of the new species belongs to the genusMexiweckelia, which was previously recorded from1

the state of Coahuila in northern Mexico (see Holsingerand Minckley, 1971) and is herein reportedfrom Texas for the first time. The taxa treated inthis paper are listed under their respective families,which are in turn arranged alphabetically.SYSTEMATICSFAMILY ARTESIIDAE HOLSINGER, 1980Remarks.-Botosaneanu and Stock (1989) do notaccept Artesiidae and Bogidiellidae as separatefamilies and suggest that the two be united in thefamily Bogidiellidae. Their opinion is based in parton the recent description of Aequigidiella from twocaves in Thailand, a genus which possesses certaincharacters that appear to bridge the gap between thetwo family groups.Artesia welboumi, new speciesFigs. 1-3Material examined.-TEXAS: CulbersonCounty: Border Cave, ca. 26 km SW of White City(New Mexico), female holotype and 2 femaleparatypes, W. C. Welbourn, 15 Apr. 1976.The holotype is deposited in the National Museumof Natural History (Smithsonian Institution)(USNM 239480); paratypes are in the author's collection.Diagnosis.-Medium-sized stygobiont speciesdistinguished from A. subterranea, to which it isclosely allied, by more spinose and/or setose appendages,especially gnathopods, pereopods, uropodsand telson; reaching sexual maturity at largersize; and other small differences noted below. Largestfemale, 10.5 mm; male unknown.Female.-Antenna 1: 55-60% length of body, ca.33 % longer than antenna 2; primary flagellum with18 segments, esthetascs on most flagellar segments;accessory flagellum I-segmented. Antenna 2 moresetose than that of A. subterranea, flagellum with 7segments. Mandible: molars conical, feeble, eachwith 1 apical seta; spine row with 7-8 plumosespines; cutting plates rather broad, lacinia mobilis ofleft fmely serrate, that of right 4-dentate; palp segment2 with 2-3 setae on convex inner margin, segment3 subequal in length to segment 2 and bearingmixture of few long and short C(?) and E setae oninner margin toward distal end (apparently rightmandible with more setae than left), outer margin ofsegment 3 with 2 or 3 B setae. Lower lip: innerlobes very broad, broader than outer lobes; lateralprocesses very short. Maxilla 1 very similar to thatof A. subterranea, except apical spines on outerplate are weakly pectinate; palps symmetrical.Maxilla 2 similar in shape to that of A. subterraneabut with more apical setae. Maxilliped: inner andouter plates and segments of palp with few more setaeeach than those of A. subterranea.Gnathopod 1: propod large and prominent, palmelongate, becoming slightly convex distally, bearingrow of about 18 short, peg-like spines on outermargin and row of 8 rather short setae on innermargin, proximal part of palm extending from endof dactyl nail to sharp defining angle nearly straightand bearing row of ca. 6 spines of unequal length;posterior margin very short, without setae. Dactylcurved, not reaching to defining angle. Carpus shortand squat, with posterior lobe bearing setae and pubescens.Coxa very small, shallow, with 2 shortmarginal setae at anterolateral corner. Gnathopod 2:propod as long as, but only ca. 2/3 as broad as thatof gnathopod 1; palm elongate, slightly convex distally,bearing ca. 19 rather short, peg-like spines onouter margin and few short setae on inner margin,part of palm between end of dactyl nail and definingangle with ca. 3 long spines and long setae; posteriormargin of propod relatively short, with 1 set of2 short setae; superior medial setae singly inserted,inferior medial setae in 1 cluster near defining angle.Dactyl curved, not reaching to defining anglewhen closed. Carpus with posterior lobe bearing 2or 3 clusters of setae. Coxa small, shallower thancorresponding body segment, bearing 3 short marginalsetae.Pereopods 3 and 4 subequal, although one or theother sometimes with few more spines and/or setae;coxae about as deep as corresponding body segments,longer than broad, margins with 4 or 5 setaeand lateral faces with few hair-like setae; bases with5-6 short spines on anterior margin, 5-8 setae ofvarying lengths on posterior margin. Pereopod 7 littlelonger than pereopod 6, ca. 55-60% length ofbody, ca. 20% longer than pereopod 5. Coxae ofpereopods 5 and 6 broadly expanded and deeperthan corresponding body segments, that of 5 broaderand more setose than that of 6; ventral marginsbroadly rounded, with 1 to 4 setae, lateral faces withfew hair-like setae; coxa of pereopod 7 shallow andnot expanded. Bases of pereopods 5 and 6 comparativelynarrow, widening distally but lacking distoposteriorlobes; anterior and posterior marginsrather spinose, anterior with short spines, posteriorwith longer spines; basis of pereopod 7 broader thanthose of pereopods 5 and 6, narrowing distally but2

crBDEGFig. I.-Artesia welboumi, new species, paratype female (10.5 mm), Border Cave, Culberson Co., TX: A, B, left & rightmandibles (cutting plates of left enlarged, incisor probably broken); C, lower lip; D, E, maxillae I, 2; F, inner and outer plates ofmaxilliped; G, telson; H, I, J, uropods 1,2,3. (Maxillae and plates of maxilliped to larger scale than other mouthparts; telson anduropods to same scale.)3

Fig. 2.-Anesia welboumi, new species, paratype female (10.5 mm), Border Cave, Culberson Co., TX: A, gnathopod I (lateral menlarged); B, gnathopod 2 (lateral view; palmar spine enlarged); C, pereopod 4; D, antenna 2. (All structures to same scale.)4

also lacking distoposterior lobe. Segments 5 and 6of pereopod 7 with long setae, those of pereopods 5and 6 without long setae. Dactyls of pereopods 5-7relatively long, those of 5 and 6 about 50% lengthof corresponding propods, that of 7 about 38 %length of corresponding propod. Dactyl of pereopod5 with 1 ventral seta, that of pereopod 6 with 1 setof 2 setae and that of pereopod 7 with 2 sets of 2 setaeeach. Coxal gills 2-6 oblong, peduncles short butdistinct. Brood plates narrow, lacking setae inspecimens examined.Pleonal plates with small, distoposterior comersand 1 short seta each; ventral margins lackingspines. Pleopods: inner and outer rami with nearlytwice number of flagellar segments as those of A.subterranea, all setae plumose, none clothespin like;peduncles with 2 coupling spines each. Uronites likethose of A. subterranea, i.e., uronites 1 and 2 eachbearing 2 dorsolateral spines. Uropod 1 very similarto that of A. subterranea, except peduncle usuallywith 5, instead of 4, basofacial spines (becomingprogressively larger toward distal end). Uropod 2rami about like those of A. subterranea, but pedunclemore spinose and bearing up to 9 long spines ondistal half. Uropod 3 approximately 16 % length ofbody: inner and outer rami subequal in length; outerlittle less broad, armed with clusters of spines onouter margin and row of plumose setae on distal halfof inner margin; outer ramus bearing spines andplumose setae on inner and outer margins and spinesonly on apex. Telson little broader than long, withdeep U-shaped cleft; lobes with 9-10 apical spineseach; lateral margins bearing 1 long, threadlike,partly plumose seta each toward distal end.Type-Iocality.-Border Cave, Culberson Co.,Texas, is developed in an outcrop of gypsum bedrockof the Permian-aged Castile formation thatfloors a rather broad gypsum plain just east of theGuadalupe and Delaware mountains (see Fig. 11).The cave is over 305 m long and gives access in twoplaces to lower level lakes of deep phreatic water.Unconfirmed reports by SCUBA divers indicate thatone or both of these lakes may be more than 40 mdeep. The single entrance to Border Cave is situatedat the end of a relatively deep desert arroyo, which,along with the entire cave, apparently floods completelyduring any heavy rainfall.Distribution and ecology.-This species isknown only from its type locality, where it is apparentlyquite rare and is known only from threespecimens that were collected by W. C. Welbourn in1976 from one of two deep lakes situated approximately152 to 213 m from the entrance. I visitedBorder Cave and this lake with colleagues in June1978, but we failed to find additional specimens atthat time. However, more recent visits to this lake in1985 and 1986 by S. J. Harden and associates, resultedin the collection of 37 specimens of an undescribedspecies of the hubbsi group of thesubterranean amphipod genus Stygobromus andspecimens of an undescribed species of asellidisopod. But additional specimens of A. welbourniwere not found. To my knowledge, the second,more remote, deep lake of this cave has not beenbiologically explored. However, according to S. J.Harden (in litt.), SCUBA divers have reported seeingnumerous "crustaceans" [probably both amphipodsand isopods] on dives in both Border Caveand nearby Wiggley Cave, which is located just eastof the former and probably gives access to the samegroundwater aquifer.In response to rumored sightings of amphipods incaves to the north of Border Cave, I visited theParks Ranch-Resurgence caves complex in June1986. These caves are located in New Mexico approximately21 km NE of Border Cave and on thesame gypsum plain. Rather diligent searching inseveral parts of this system revealed only an occasionalspecimen of the widely distributed, epigeanamphipod Hyalella azteca but no other crustaceans.These caves are shallow, apparently flood veryquickly after a rainfall, and do not extend deepenough to intersect phreatic water. Thus, any potentialthey might have as habitats for populations ofstygobiont amphipods is severely limited.Because of their great size and depth (over 300meters), both Carlsbad Caverns and LechuguillaCave, which are located in New Mexico only approximately24 km north-northeast of Border Cave,would appear to have the potential for stygobiontcrustacean faunas. Unlike Border Cave, however,they are excavated in limestone bedrock at higherelevations in the Guadalupe Mountains just west ofthe gypsum plain. To date, despite extensive explorations,neither is known to access phreatic water orcontain any aquatic organisms.Etymology.-It is a pleasure to name this speciesin honor of its collector, W. Calvin Welbourn,Curator of the Acarology Laboratory at Ohio StateUniversity.FAMILY HADZIIDAE S. KARAMAN, 1943Genus Mexiweckelia Holsinger and MinckleyMexiweckelia Holsinger and Minckley, 1971:426 [inpart]; Barnard and Barnard, 1983:644.Revised diagnosis.-Without eyes and pigment,5

\\. \ - -' ..........---~\ I// --.. i/ I//IFig. 3.-Artesia welbourni, new species, paratype female (10.5 nun), Border Cave, Culberson Co., TX: A, B, pereopods 3 & 4 (inpart); C, pereopod 7 (dactyl enlarged); D, pleopod 3. (All structures to same scale.)6

of stygobiont facies. Interantennal lobe small,rounded anteriorly. Antenna 1 longer than antenna2; flagellum with esthetascs present only in male;accessory flagellum vestigial (?) or absent. Peduncularsegment 4 of antenna 2 longer than segment 5.Buccal mass prognathous. Upper lip symmetrical,rounded and unnotched apically. Mandible: molarprominent, triturative, left lacking seta; laciniamobilis present on both right and left mandibles;palp absent. Lower lip: inner lobes vestigial or absent,outer lobes (shoulders) high, lateral processesslender but distinct. Maxilla 1: inner plate notgreatly expanded, with apical setae (sometimes plumose);outer plate with 7 apical serrate spines; palpsusually asymmetrical (i.e., right broader apicallyand more spinose). Maxilla 2: inner plate narrowingtoward apical end, with oblique row of naked facialsetae, inner margin slightly convex. Maxilliped: innerplate rather long, expanded distally but notgreatly so; outer plate extending beyond inner plate,broadening medially or distally, inner margin withshort row of bladespines at or toward apex; palpsegment 2 longest, segment 3 with distomedial lobe,segment 4 (dactyl) not greatly elongate, nail relativelyshort.Gnathopod 1 weakly sexually dimorphic. Propodof female gnathopod 1 shorter than carpus, palmshort, generally transverse, weakly armed with fewsmall spines and setae; dactyl short and thick, nailshort; carpus large, subtriangular, with prominentpubescent posterior lobe and several long setae.Propod of male gnathopod 1 proportionately littlelonger, palm with few more spines. Gnathopod 2strongly sexually dimorphic. Propod of femalegnathopod 2 relatively long and narrow, longer thancarpus, palm short, oblique, bearing 2 rows ofshort, unnotched spines; posterior margin longerthan palm, with several clusters of setae; carpus subtriangular,with pubescent posterior lobe and severalsets of long setae. Propod of gnathopod 2 of malemuch broader distally, palm longer, oblique, bearingdouble row of short, unnotched spines; carpuscomparatively shorter but also with distinct pubescentposterior lobe. Coxae of gnathopods 1 and 2subequal, about as broad as deep; coxae 3 and 4similar; coxa 4 unlobed; coxa 5 lobate. Pereopods 3and 4 subequal. Pereopod 7 at least 50 % length ofbody, little longer than pereopod 6, much longerthan pereopod 5; bases of pereopods 5-7 not muchexpanded, distoposterior lobes usually distinct butnot large; segment 6 moderately spinose and/or setose;dactyls rather short, lacking setules on uppermargins. Coxal gills 2-6 relatively large, typicallyovate and/or ellipsoidal, with distinct peduncles.Brood plates small and narrow, not bearing setae inmaterial examined.Pleonal plates variable, comers not produced,each bearing 1 setule. Pleopods normal, not sexuallydimorphic, peduncles with few (typically 3) couplingspines on inner margins distally. Uronites free(not fused), 1 and 2 each anned with 2 dorsalspines, 3 with 2 or often more. Uropods 1 and 2 notsexually dimorphic; peduncle of 1 bearing 1 or 2basofacial spines. Uropod 3 elongate, magniramous;outer ramus I-segmented, outer margin with strongspines in clusters but lacking setae. Telson relativelyshort,about as long as broad; depth of apical incision(cleft) variable, very shallow or up to ca. 2/3distance to base; lobes with 1 to 3 apical spineseach.Relationship.-Despite some fundamental differencesthat are noted below, Mexiweckelia is apparentlyclosely related to Holsingerius, a genus originallybased on H. samacos (Holsinger) (in Holsingerand Longley, 1980) from the artesian well in SanMarcos and described by Barnard and Karaman(1982). These two genera have a number of importantmorphological characters in common. Mexiweckeliais also closely related to Texiweckelia, anothergenus recorded from the artesian well in SanMarcos, as well as San Marcos Springs (see Holsingerand Longley, 1980).Remarks.-The addition to Mexiweckelia of thenew species described below, re-examination of thetwo species previously assigned to this genus, andre-evaluation of the closely allied genus Holsingerius,necessitates the revised diagnosis given above.Mexiweckelia hardeni, n. sp., is the first recordfor Mexiweckelia outside Mexico and brings the totalnumber of species in the genus to three. Two ofthese species occur in northern Mexico: M. coleifrom groundwaters of the Bolson de CuatroCienegas in Coahuila and M. mitchelli from Cuevade la Siquita in Durango (see Holsinger andMinckley, 1971; Holsinger, 1973).Two discrepancies in the original descriptions ofM. colei (see Holsinger and Minckley, 1971) andM. mitchelli (see Holsinger, 1973) are noted andshould be corrected as follows. Structures on theflagellum of antenna 1 referred to erroneously as"tiny" or "slender" calceoli are esthetascs; basofacialspines were inadvertently omitted from the illustrationsof uropod 1 (1 for M. mitchelli and 2 for M.colei); in M. mitchelli, the right molar has a seta,the left does not.7

Mexiweckelia hardeni, new speciesFigs. 4-6Material examined.-TEXAS: Medina County:Hondo Creek hyporheic, 6.5 km E of Hondo, femaleholotype and 4 female paratypes, S. J. Harden,18 Mar. 1988; additional paratypes collected by S.J. Harden include 1 female, 20 Jan. 1986 and 1male, 23 Feb. 1988.The holotype is deposited in the National Museumof Natural History (USNM 239477); paratypesare .in the author's collection.Diagnosis.-A small, interstitial species closelyallied to M. colei from northern Mexico, but easilydistinguished from that species and also from M.mitchelli by the telson, which has a very shallowapical notch, and peduncle of uropod 2, which isheavily spined toward distal end along dorsomedialmargin. Largest male, 3.2 mm; largest female, 4.5mm.Female.-Antenna 1 about as long as body, ca.50 % longer than antenna 2; primary flagellum with35 segments, esthetascs absent. Antenna 2 with 14flagellar segments. Mandible very similar to that ofM. colei: incisor 7-dentate; right mandible with3-dentate lacinia mobilis, 2 accessory spines in spinerow and molar seta; left mandible with 4-dentatelacinia mobilis, 3 accessory spines and molar setaabsent. Lower lip like that of M. colei. Maxilla 1:inner plate with 11-12 apical, plumose setae; outerplate with 7 apical, serrate spines; palps asymmetrical,right distally expanded and bearing about 7short, "thick" spines; left not expanded and bearingonly about 4 generally weaker spines. Maxilla 2:inner plate narrowing distally, with oblique row of15-16 facial setae. Maxilliped: inner plate with 3bladespines and setae on apex and 2 medial setaesubapically; outer plate broadening medially ordistally, with short row of bladespines apically andsubapically on inner margin; palp segment 2longest, segment 3 narrower and with smalldistomedial lobe.Gnathopod 1 very similar to those of M. coleiand M. mitchelli; propod about nearly as long ascarpus, palm very short, scarcely armed, bluntlyrounded just below tip of closed dactyl, posteriormargin rather long, without setae; dactyl short andthick, nail short and indistinct; carpus subrectangular,widening distally into broad pubescent lobe,lobe broadest at distal end, bearing several long setae;coxa about 4/5 as deep as broad, margin with 2setae. Propod of gnathopod 2 narrow, elongate,palm short and oblique, with double row of 3 or 4small spines, defining angle with 1 spine and 2 longsetae; posterior margin long, with 3 sets of setae;superior medial setae singly inserted in row near anteriormargin. Dactyl of gnathopod 2 bearing smallbladespines on inner margin, nail rather short; carpussubtriangular, posterior margin broadly lobiformand pubescent toward distal end, lobebroadest at distal end, with 4 or 5 clusters of longsetae. Coxa of gnathopod 2 subequal in size to thatof gnathopod 1. Pereopods 3 and 4 subequal, coxaelittle broader than deep, each with 2 marginal setae,4 not excavate or lobate. Pereopod 7 about 66 %length of body, little longer than pereopod 6, about33 % longer than pereopod 5. Bases of pereopods5-7 relatively narrow, distoposterior lobes small,scarcely produced; dactyls 30% to 40% length ofcorresponding propods. Coxal gills 2-6 large,prominent, ellipsoidal and/or subovate, with distinctpeduncles. Brood plates small, narrow, nonsetose inmaterial examined.Pleonal plates: posterior margins slightly convexor nearly straight, comers rounded, each bearing 1seta; plate 3 with 1 ventral margin spine. Pleopodpeduncles with 3 coupling spines on inner margins.Uronites 1 and 2 each with 2 small, dorsodistalspines, 3 with 6 such spines in groups of 3. Uropod1: inner ramus little longer than outer, shorter thanpeduncle, armed with about 5 apical spines; outerramus with about 4 apical spines and 2 short spineson outer margin; peduncle with 10 spines, 2 ofwhich are basofacial. Uropod 2: inner ramus longerthan outer, shorter than peduncle, with about 5 apicalspines; outer ramus bearing about 5 spines;peduncle rather spinose, bearing 3 or 4 spines onouter anterior margin distally and 9 spines (in row)on distal 3/5 of dorsomedial margin. Uropod 3 relativelylong, ca. 22 % length of body, rami subequalin length; outer ramus slightly narrower, outermargin armed with sets of spines in clusters of 3 butlacking setae, medial margin with singly insertedspines and plumose setae, apex with cluster of ca. 6spines; inner ramus with singly inserted spines andplumose setae on both margins, apex with cluster ofabout 5 spines. Telson about as broad at base aslong, narrowing distally, apical margin with veryshallow notch, apical lobes with 3 spines each.Male.-Differing principally from female inhaving esthetascs on antenna 1 and in structure ofgnathopods as follows. Propod of gnathopod 1 littlelonger, palm with few more small spines, 1 or 2relatively long inferior medial setae; dactyl littlelonger. Propod of gnathopod 2 proportionatelylarger, palm rather long, oblique, armed withdouble row of about 7 small, unnotched spine teeth,posterior margin about equal to palm in length, with8

BcEoFLK3JFig. 4.-Mexiweckelia hardeni, new species, paratypes, Hondo Creek hyporheic, Medina Co., TX, female (4.5 mm): A,maxilliped; B, C, left & right mandibles; 0, right maxilla I; E, apex ofpalp of left maxilla I; F, lower lip; G, H, uropods 2, 2. Female(3.7 mm): I, palp of left maxilla I; J, coxal gill; K, pleonal plates; L, coxae 1-4. (Lower lip to smaller scale than other mouthparts;coxae and pleonal plates to same scale, coxal gill to smaller scale.)9

2 sets of 1 or 2 setae each; dactyl rather long andslightly curved, closing beyond palm spines, nailrather short.Type-Iocality.-Interstitial medium of shallowgravel banks of Hondo Creek, just south of the Balconesescarpment, in Medina Co., Texas (see Fig.12). The bedrock in this area is composed of shalesand sandstones of Late Cretaceous age.Distribution and ecology.-This species is presentlyknown only from the type locality, where ithas been collected on three occasions from shallowpits dug into gravel banks of Hondo Creek or frombaited jars placed in these pits. The amphipods apparentlyinhabit an interstitial habitat in the hyporheic(and/or parafluvial) zone of this stream. Arather detailed investigation of the hyporheic zone ofHondo Creek by S. J. Harden (in lilt.) over a periodof several years has resulted in the collection offlatworms, oligochaetes, hydrobiid snails, ostracodes,stenasellid isopods (Mexistenasellus coahuilaCole and Minckley - see Bowman, 1992, in thisvoL), and amphipods. In addition to M. hardeni, theBFig. 5.-Mexiweckelia hardeni, new species, paratypes, Hondo Creek hyporheic, Medina Co., TX, female (4.5 mm): A, B,gnathopods 1,2 (palms and dactyls enlarged). Male (3.2 mm): C, gnathopod 2 (spines enlarged). (All gnathopods to same scale).10

plate; and e) coxa 1 larger than coxa 2. Despitethese important differences, this genus and Me.xiweckeliaare apparently closely related, as alreadyindicated. This relationship is strongly supported byseveral mouthpart characters of the new species describedbelow that appear to be intermediate betweenthe two genera. Holsingerius is also closely relatedto Te.xiweckelia, as evidenced by their number ofshared similarities, which include especially thestructure of the gnathopods, coxae, and uropods (seeHolsinger and Longley, 1980).Holsingerius smaragdinus, new speciesFigs. 7-9Material examined.-TEXAS: Val VerdeCounty: Emerald Sink (Cave), 3.2 km N of Langtry,female holotype and 3 female paratypes, D. Cannyand S. J. Harden, 25 May 1985; 2 female paratypes,R. M. Waters, 31 Mar. 1984.The holotype is deposited in the National Museumof Natural History (USNM 239478); paratypesare deposited in the author's collection.Diagnosis.-A medium-sized cavemicolous speciesintermediate in some mouthpart characters betweenMe.xiweckelia and Holsingerius but havingmore characters of the latter and herein provisionallyassigned to this genus. Distinguished from Holsingeriussamacos, the only other species in this genus,by presence of a vestigial lacinia mobilis onright mandible; smaller inner plate of maxilla 1 withapproximately 1/2 as many medial setae; shorter innerplate of maxilla 2, which is not sub-rectangularin shape and has significantly fewer facial setae;shorter row of plumose setae on inner margin of innerplate of maxilliped; absence of setules from uppermargins of dactyls of pereopods 5-7; and presenceof long, plumose setae on inner or outer ramiof uropods 1 and 2. Largest female, 8.5 mm; maleunknown.Female.-Antenna 1 85-90% length of body, ca.65 % longer than antenna 2, primary flagellum with39-42 segments, lacking esthetascs; accessory flagellumabsent. Antenna 2 with 13 flagellar segments.Mandible: molars prominent; right mandible withspine apparently representing vestigial lacinia mobilis,2 long plumose accessory spines and molar seta;left with well developed, 4-dentate lacinia mobilis, 2accessory spines, but lacking molar seta. Inner lobesof lower lip vestigial or absent. Maxilla 1: innerplate with 19-20 medial, plumose setae; outer platewith 7 apical, serrate spines; palps symmetrical,broadest distally, bearing 4 spines at or near apexand 4-5 setae subapically (cf., Holsingerius samacos).Maxilla 2: inner plate narrowing distally, withoblique, submarginal row of ca. 29 naked, facial setae.Maxilliped: inner plate rather broad, bearing 4bladespines apically and setae apically/subapically,and row of plumose setae on inner margin; outerplate broadest medially, bearing short row of bladespineson inner margin near apex; palp segment 3with inner distal lobe.Gnathopod I: propod proportionately small, palmshort, with few tiny spines and several setae, medialfacial setae few, posterior margin longer than palm,with setae; carpus longer than propod, producedposteriorly into prominent, pubescent lobe which isbroadest proximal to distal end (as in H. samacos),bearing several groups of long setae; posterior marginof basis setose; coxa rather deep and broadly expanded,with 2 short, marginal setae. Gnathopod 2:propod relatively narrow, elongate, narrowingslightly and unevenly distally, palm oblique, short,armed with double row of ca. 6 small spine teeth,defining angle with several tiny spines and 2 longsetae, medial (inner facial) setae absent, but both anteriorand posterior margins bearing row of long setae;dactyl short, rather stout, nail short; carpussubtriangular, posterior margin lobiform and pubescent,lobe broadest proximal to distal end (as ingnathopod 1 and in H. samacos), bearing 5 or 6clusters of long setae; coxa smaller than that ofgnathopod 1, little deeper than broad. Pereopods 3and 4 subequal, bases rather broad and bearing shortspines on anterior margin and longer (slender)spines on posterior margin; coxa about as deep asbroad, with 2 marginal setules; coxa 4 not lobate.Pereopod 7 65-70% length of body, little longerthan pereopod 6, ca. 25 % longer than pereopod 5.Bases of pereopods 5-7 moderately broad, posteriormargins broadly convex, distoposterior lobes welldeveloped; dactyl of pereopod 5 about 50% lengthof corresponding propod, dactyls of pereopods 6 and7 ca. 35-40 % length of corresponding propods; dactylswithout setules on upper margins. Coxal gillslarge, prominent, usually ellipsoidal, sometimessubovate, with distinct peduncles. Brood platessmall, narrow, and nonsetose in material examined.Pleonal plates similar to those of H. samacos,comers produced (especially plates 2 and 3), bearingI setule each. Pleopod peduncles with 6-7 couplingspines each on inner margins. Uronites 1 and 2 with2 small dorsodistal spines each, uronite 3 with 6-8such spines. Uropod I: inner ramus subequal inlength to outer, shorter than peduncle, with about 5apical and 2 lateral spines; outer ramus with apicaland lateral spines and row of 5 rather long, plumose12

AFig. 7.-Holsingerius smaragdinus, new species, paratype female (8.0 mm), Emerald Sink, Val Verde Co., TX: A, B, left & rightmandibles; C, left maxilla I; D, palp of right maxilla I; E, lower lip; F, maxilla 2; G, maxilliped; H, I, antennae I, 2. (Lower lip tosmaller scale than other mouthparts; antennae to same scale).13

setae on upper margin; peduncle bearing about 12spines, 3 of which are relatively large basofacial (cf.H. samacos). Uropod 2: inner ramus subequal inlength to outer, with apical spines and row of ratherlong plumose setae on lower margin; outer ramuswith about 5 apical spines; peduncle subequal torami in length, anned with about 3 spines. Uropod 3relatively long, about 22 % length of body; rami ofequal length but outer bearing spines only (in clustersof 2 or 3) on outer margin; inner margin ofouter ramus and both margins of inner ramus withspines and plumose setae. Telson rather long, about25% longer than broad; apical margin with deep,V-shaped cleft extending ca. 75 % distance to base;apical lobes bearing 4 or 5 apical spines each; lateralmargins lacking spines.Type-Iocality.-Emerald Sink (or EmeraldCave), located on the Stockton Plateau in Val VerdeCo., Texas (see Fig. 12), extends to a depth of approximately91 m below the entrance and contains adeep lake of phreatic water on the lower level. Thecave is excavated in limestones of the Buda andDevils River formations of Cretaceous age; a descriptionof the cave was published by Kunath andSmith (1968).Relationship.-Although this species does notFig. 8.-Holsingerius smaragdinus, new species, paratype female (8.0 mm), Emerald Sink, Val Verde Co., TX: A, B, gnathopodsI, 2 (palms enlarged), C, D, uropods I, 2; E, telson. (Gnathopods to smaller scale than uropods and telson.)14

have as many setae on the inner plates of maxiIlae 1and 2 as H. samacos, the number is significantlygreater than those observed in species of Mexiweckelia.Moreover, a close alliance between H. smaragdinusand H. samacos is evidenced by occurrence ofthe following character states in the former: elongate1st antenna; vestigial lacinia mobilis of rightmandible; symmetrical palps of maxilla 1; long,narrow propod of female gnathopod 2, whichnarrows unevenly toward distal end; less than distalposition of posterior lobes of the carpus of bothgnathopods; coxa 1 larger than 2; broad bases ofpereopods with prominent distoposterior lobes;produced comers of pleonal plates; 6 or 7 couplingA-- .-/--- ./" --- /B\,\\II -, ' \\/,"-~ I/\ \/DFig. 9.-Holsingerius smaragdinus, new species, paratype female (8.0 mm), Emerald Sink, Val Verde Co., TX: A, B, coxae andupper part of bases of pereopods 5 & 6; C, pereopod 7; D, coxa, gill and brood plate of pereopod 3; E, pereopod 4 (in part); F,uropod 3. (pereopods 5,6 & 7 to smaller scale than pereopods 3 & 4, uropod 3 to larger scale.)15

spines on pleopod peduncles; 3 large basofacialspines on peduncle of uropod 1; and deep, relativelywide cleft of telson. Although some of thesesimilarities are probably more variable than others,this high number of shared characters, many ofwhich are synapomorphies, clearly indicates a muchstronger phylogenetic relationship of H.smaragdinus with H. samacos than with species ofM exiweckelia.Distribution and ecology.-This species isknown only from its type locality, where it has beencollected twice from a phreatic lake, probably exceedinga depth of 10.6 m. The lake is also inhabitedby the stygobiont cirolanid isopod Speocirolanahardeni, described by Bowman (1992, in this vol.).Etymology.-The epithet smaragdinus is fromLatin, meaning "emerald-green," and is used in referenceto the type locality which is sometimes saidto contain "emerald-green water. "FAMILY SEBIDAE WALKER, 1908Genus Seborgia BousfieldSeborgia Bousfield, 1970:164; Karaman, 1982:86-87.Relictoseborgia Karaman, 1982:91-92.Remarks on taxonomic status.-Seborgia waserected by Bousfield (1970) for a single species, S.minima, which he described from an oligohalinebrackishwater lake on Rennell Island in the BritishSolomon Islands in the South Pacific. A secondspecies, S. relicta, was described from the artesianwell in San Marcos, Texas, and assigned to this genusby Holsinger and Longley (1980).On the basis of small differences in morphologybetween these species, Karaman (1982) proposed thenew genus Relictoseborgia for S. relicta, while retainingSeborgia for S. minima. In support of hisproposal, Karaman (1982) pointed out that the Texasartesian well species differed primarily by having ashallow lateral cephalic (interantennal) lobe combinedwith an obsolete anteroventral (inferior antennal)sinus, "non-telescopically-shaped" antenna I,proportionately longer mandibular palp segment I,larger inner lobes of lower lip, unequal-sizedgnathopods, and peduncle of uropod 3 shorter thanunisegmented ramus.In my opinion most of these differences are minoror variable and all are within the accepted range ofmorphological parameters generally expected betweenspecies in the same genus (see also Holsingerand Longley, 1980; Holsinger, 1986a). However, incombination with each other and with the extremegeographic separation and significant habitat differenceof these species, I agree with Karaman thatthese differences probably support some kind of recognitionabove the species level for these taxa.However, considering the lack of strong morphologicaldifferences and the fact that only two (nowfour) species are involved, full generic status maynot be warranted. Instead, I propose dividing Seborgiainto subgenera along the lines indicated below.Subgenus Seborgia Bousfield, NEW RANKSeborgia Bousfield, 1970: 164; Karaman, 1982:86-87.Diagnosis.-Corresponding to the diagnosis forgenus Seborgia given by Karaman (1982:86), exceptthat eyes may be present and the inferior antennal(anteroventral) sinus is variable in depth.Remarks.-This subgenus contains two species:S. minima Bousfield (1970), already mentionedabove; and S. schieckei Ruffo (1983) from a brackishmesopsammic habitat on the coast of South AndamanIsland in the Indian Ocean. Although S.schieckei is more closely allied morphologicallywith S. minima than S. relicta, its slightly less producedinterantennal lobe and weak inferior antennalsinus are apparently intermediate between the twoextremes noted in the other two species. To date, S.schieckei is the only eyed member of the genus.Subgenus Relictoseborgia Karaman, NEW RANKRelietoseborgia Karaman, 1982:91-92.Diagnosis.-Corresponding to the diagnosisgiven for genus Relictoseborgia by Karaman (1982).Remarks.-This subgenus contains two species:S. relicta (Holsinger, in Holsinger and Longley,1980) from the artesian well in San Marcos, HaysCo., Texas and a new locality from Medina Co.,Texas, noted below; and S. hershleri, new species,from Val Verde Co., Texas, described below.Seborgin (Relictoseborgin) hersh/eri, new speciesFig. 10Material examined.-TEXAS: Val VerdeCounty: unnamed spring on east side of DevilsRiver, ca. 32 km N of Del Rio, female holotype and3 female paratypes, R. Hershler, 8 Sept. 1986;16

Antenna 1 about 33 % length of body, about 25 %longer than antenna 2; primary flagellum with esthetascs;accessory flagellum 2-segmented (terminalsegment rudimentary). Mouthparts generally likethose of S. relicta, except for differences inmandible (but note that left and right mandibles arereversed in original description of S. relicta inHolsinger and Longley, 1980, fig. 23(e and f).Mandible: molar apparently lacking setule, incisor5-dentate, lacinia mobilis of left 5-dentate; palpsegment 1 relatively long, 2/3 to 3/4 as long assegment 2; palp segment 2 apparently little broaderthan that of S. relicta, with pilose inner margin and2 setae distally; segment 3 lacking all but ca. 3 apical(E) setae. Lower lip with relatively large innerlobes. Maxilla 1: inner plate tapering distally, lackingapical setae; outer plate with 7 very weakly serratespines; palps symmetrical, with few setae apicallyand subapically. Maxilla 2 reduced to singleplate with broad base and 4-5 apical setae. Maxilliped:inner plate sublinear, with few apical setae;outer plate much broader, rounded apically, withfine setae on apex and few longer setae on innermargin distally.Gnathopods subchelate. Gnathopod 1: propodprominent, almost twice size of propod ofgnathopod 2, widest distally, palm transverse,bearing row of few setules on inside, defining angleproduced into large spine-like process, posteriormargin long, oblique, and piliferous in part; dactyllong, closing against inside of defining angle boss,nail indistinct; carpus short, squat, posterior lobebearing 2 thick setae; basis long and slender, with 2long setae on posterior margin near distal end; coxadeep, about 2 times deeper than broad, ventral marginwith 2 relatively long setae. Gnathopod 2: propodlittle wider distally than proximally, not muchexpanded, palm short, transverse, bearing fewsetules on inside, derming angle produced into smallspine-like process, posterior margin long and partlypiliferous , with I or 2 short setae; dactyl closingagainst inside of defining angle, nail indistinct; carpusabout 50% length of propod, slightly lobed posteriormargin with 2 short setae; basis long andslender, bearing 2 long setae on outer margin neardistal end; coxa deep and rather narrow, bearing 1or 2 long setae toward distal margin. Pereopods 3and 4 subequal except coxae differ; coxa 3 verydeep, subrectangular, extending perhaps 90% lengthof corresponding basis, distal margin with 2 shortsetae; coxa 4 also very deep, but much broader, extendingto end of corresponding basis, distal marginwith 2 short setae; dactyls long, at least 50% lengthof corresponding propods. Pereopod 7 subequal inlength to pereopod 6, longer than pereopod 5, ca.50 % length of body. Bases of pereopods 5-7 withbroad distoposterior lobes, posterior margins convexand partly serrate; dactyls rather long, ca. 50%length of corresponding propods. Coxal gills andbrood plates like those of S. relicta.Posterior comers of pleonal plates weakly produced,acuminate, without setules. Pleopods normalfor genus. Uronites without dorsal spines. Uropod1: inner ramus little longer than outer ramus andpeduncle, bearing 2 short spines near distal end,outer ramus with 1 or 2 short spines near distal end;peduncle with 2 spines distally. Uropod 2: innerramus short, only about 1/2 length of outer ramusand peduncle; armed with 2 short spines near distalend; outer ramus with 1 or 2 short spines near distalend; peduncle with 2 spines distally. Uropod 3uniramus; peduncle at least 2/3 length of ramus; ramuswithout spines. Telson little longer than broad,gently tapering distally, lacking spines or setae; apicalmargin entire (not cleft).Type-Iocality.-Unnamed spring on east side ofDevils River in a canyon just downstream fromSlaughter Bend, Val Verde Co., Texas (see Fig.11). This spring is also the type locality for a stygobionthydrobiid snail, Phreatodrobia coronaeHershler, and is described in some detail byHershler and Longley (1987).Distribution and ecology.- This species isknown only from its type locality, where it has beenfound in association with two other stygobiont amphipods:Stygobromus (possible new species of thejlagellatus group) and an undescribed new genus ofthe family Hadziidae; the latter being much morecommon than the former. Seborgia hershleri wasalso collected with planarians, the hydrobiid snailsmentioned above, copepods, and cirolanid isopods(Cirolanides texensis Benedict - see Bowman,1992, this voL). Of a total of 9 specimens collectedto date (all during September), 8 were females and 5of them were ovigerous. The females measured 1.4to 1.6 mm in length; 4 were carrying 1 egg each,whereas the other had 2 eggs.Devils River, which drains south to the RioGrande, lies on the southwest comer of the EdwardsPlateau and just west of the area included within theconfines of the Edwards Aquifer (see Holsinger andLongley, 1980; Longley, 1981). The springs alongthe east side of Devils River emerge from subterraneanaquifers developed in the Georgetown limestoneof Cretaceous age. Much of the stygobiontfauna from these springs is closely allied taxonomicallywith that found farther east in the San Antonioand San Marcos pools of the Edwards Aquifer.18

of subterranean amphipod faunas of Texas andMexico.n has already been hypothesized that most of thenon-crangonyctid subterranean amphipods of Texasand Mexico (e.g., artesiids, bogidiellids, hadziids,sebids) as well as other groups of stygobiont crustaceans(such as cirolanid and stenasellid isopods, andthermosbaenaceans), originated from marine ancestorsthrough stranding in Late Cretaceous and/orearly Cenozoic times (Holsinger and Longley, 1980;Bowman, 1982; Holsinger, 1986b). The inland,freshwater stygobiont members of these groups occurat present in areas that were once covered byshallow marine embayments. The distribution patternsof the new taxa fall well within areas previouslysubjected to Cretaceous embayments, andtherefore offer further corroboration of the strandingtheory.The new locality for Artesia from Border Cave inCulberson County on the western fringe of the GreatPlains, approximately 650 km west of the only otherknown locality for this genus in San Marcos (Fig.11), is a significant range extension for this poorlyunderstood stygobiont group. However, consideringthe great distance and potential dispersal barriersbetween these widely separate localities, themorphological similarity of A. subterranea from theartesian well in San Marcos and A. welbourni fromBorder Cave is surprising.Border Cave is of further interest biogeographicallyfor subterranean amphipods because it isinhabited by an undescribed species of the hubbsigroup of Stygobromus. With one exception, thisgroup is restricted to far western United States andsouthwestern Canada (Holsinger and Shaw, 1986).In the southwestern United States only one otherspecies is recorded from southeastern Arizona, approximately500 km to the west of Border Cave.rII-_IP~COsTEXASEDWARDSPLATEAUBALCONES -----ESCARPMENT • AUSTIN" 1...• 0 SAN ANTONIO1GULFOFMEXICOoKM200 400Fig. 12.-Geographic distribution of HolsingerillS and Mexiweckelia in Texas and Mexico. Localities for Holsingerills indicated byopen circles: I, H. samacos (artesian well in San Marcos, Hays Co.); 2, H. smaragdimlS (Emerald Sink, Val Verde Co.). Localities forMexiweckelia indicated by closed circles: I, M. hardeni (Hondo Creek hyporheic, Medina Co.); 2, M. colei (Bolson de CuatroCienegas, Coahuila); 3, M. milchelli (Cueva de la Siquita, Durango).20

Of biogeographic interest for the hadziid amphipodsis the newly extended range of Mexiweckeliafrom Cuatro Cienegas, Mexico northeastwardacross the Sierra Madre Oriental for approximately400 km to Medina Co., Texas. The present distributiontrack of this genus (see Fig. 12) extends fromnortheastern Durango, Mexico to a point justsouthwest of San Antonio in Medina Co., Texas.This track overlaps the northern part of the distributiontrack of the stygobiont stenasellid isopod Mexistenasellus(see Bowman, 1982; 1992, this vol.). Thecongruence of these tracks can be interpreted to suggestthat the evolutionary history of these two crustaceangenera was influenced by the same factors.The high number of characters shared by M. coleiin Mexico and M. hardeni in Texas suggest thatthese taxa are probably sister-species with an immediatecommon ancestor. Paralleling the closephylogenetic relationship of the amphipods is thefmding by Bowman (1992, in this vol.) that populationsof the stygobiont isopod genus Mexistenasellusfrom the same localities as the amphipods are sosimilar that they apparently represent the samespecies. There is also significant overlap of thisnorthern Mexiweckelia-Mexistenasellus track withthe northern part of the range of the stygobiontcirolanid isopod Speocirolana (see also Bowman,1992, this volume), offering further evidence for animportant generalized crustacean track fromnorth-central Mexico north and northeast intosouth-eentral Texas.Also of interest for hadziids is Holsingeriussmaragdinus, the second described species for thisgenus and the first record outside the artesian well inSan Marcos. The new locality extends the range ofthe genus westward for approximately 350 km, fromthe Edwards Plateau across the Pecos Valley to theStockton Plateau (Fig. 12). With respect to some ofits mouthparts, H. smaragdinus appears to be intermediatebetween Holsingerius and Mexiweckelia,suggesting that these two genera are closely alliedphylogenetically. This close relationship is stronglysupported by a cladistic analysis (Holsinger, inprep.), which shows that Mexiweckelia and the severalhadziid genera from the Edwards aquifer, includingboth Holsingerius and Texiweckelia,compose a nested subset of genera with a relativelyrecent common ancestry.Finally, two new localities are noted for the rare,"dwarf" genus Seborgia, marking both significantrange extensions and additional habitats for thisgroup. The new locality for S. relicta in the hyporheiczone of Hondo Creek in Medina County isthe first for this species outside the artesian well inSan Marcos and extends the range of this speciessouthwest for approximately 120 km (Fig. 11). Italso documents the occurrence of this species from ashallow groundwater (hyporheic) habitat outside thedeep phreatic waters of the Edwards Aquifer, per se.Signifying an even greater range extension forthis genus is the discovery of the closely similarspecies, Seborgia hershleri, from a spring in theDevils River gorge, approximately 250 km west ofSan Marcos (Fig. II) and just west of the Uvaldepool of the Edwards Aquifer. The morphologicalsimilarities of S. relicta from Hays and Medinacounties and S. hershleri from Val Verde County arestriking, and the few differences noted between thetwo are conceivably within the limits of geographicvariation. However, because nothing is presentlyknown about variation in the few rare species of thisgenus, and because of the apparent geographic isolationof the populations, I have elected to treat theVal Verde populations as a separate species.ACKNOWLEDGMENTSI am deeply grateful to Scott J. Harden, RobertHershler, W. Calvin Weibourn and others for collectingthe material utilized in this study. Theseworkers and James R. Reddell also provided mewith critical information on localities and habitats.A final draft of the manuscript was improved byhelpful comments from Thomas E. Bowman and JanH. Stock. I also thank Jun Zhang for inking thedrawings and helping with preparation of the figures,and the Publications and Graphics ServicesOffice at Old Dominion University for preparationof the distribution map.LITERATURE CITEDBarnard, LL., and C.M. Barnard. 1983. Freshwater Amphipodaof the World (parts 1 & II). MI. Vernon, Virginia: HayfieldAssociates. 830 pp.Barnard, LL., and G.S. Karaman. 1982. Classificatory revisionsin Gammaridean Amphipods (Crustacea), part 2. Proc. BioI.Soc. Washington, 95(1):167-187.Botosaneanu, L., and J.H. Stock. 1989. A remarkable newgenus of cavernicolous Bogidiellidae (Crustacea, Amphipoda)from Thailand. Studies in honour of Dr. PieterWagenaar Hummelinck. Foundation for Scientific Researchin Surinam and the Netherlands Antilles, Amsterdam,123: 171-184.Bousfield, E.L. 1970. Terrestrial and aquatic amphipod Crustaceafrom Rennell Island. The Natural History of RennellIsland British Solomon Islands, 6: 155-168.Bowman, T.E. 1982. Three new stenasellid isopods fromMexico (Crustacea: Asellota). Assoc. Mexican Cave Stud.Bull., 8:25-38rrexas Mem. Mus. Bull., 28:25-38.21

Bowman, T.E. 1992. Two subterranean aquatic isopod crustaceansnew to Texas: Mexistenasellus coahuila (Cole andMinckley, 1972) (Asellota: Stenasellidae), and Speocirolanahardeni, new species (Flabellifera: Cirolanidae). TexasMem. Mus., Speleol. Monogr., 3:21-Hershler, R., and G. Longley. 1987. Phreatodrobia coronae, anew species of cavesnail from southwestern Texas. TheNautilus, 101 (3): 133-139.Holsinger, J .R. 1973. Two new species of the subterraneanamphipod genus Mexiweckelia (Gammaridae) from Mexicoand Texas, with notes on the origin and distribution of thegenus. Assoc. Mexican Cave Stud. Bull., 5:1-12.Holsinger, J.R. 1986a. Amphipoda: Sebidae. pp. 568-569 in: L.Botosaneanu, ed., Stygofauna Mundi. Leiden: Brill.Holsinger, J.R. 1986b. Zoogeographic patterns of North Americasubterranean crustaceans. pp. 85-106 in: R.D. Gore andK.L. Heck, eds., Crustacean Issues 4, Crustacean Biogeography.Rotterdam: A.A. Balkema.Holsinger, J.R., and W.L. Minckley. 1971. A new genus andtwo new species of subterranean amphipod crustaceans(Gammaridae) from northern Mexico. Proc. BioI. Soc.Washington, 83(37):425-444.Holsinger, J.R., and G. Longley. 1980. The subterranean amphipodcrustacean fauna of an artesian well in Texas. SmithsonianContr. Zool., 308:1-62.Holsinger, J.R., and D.P. Shaw. 1986. A new stygobiont amphipodcrustacean (Crangonyctidae, Stygobromus) from glaciatedkarst on Vancouver Island, Canada. Comm. 9thInternatl. Congr. Speleol., Barcelona, 2:98-101.Karaman, G.S. 1982. Contribution to the knowledge of theAmphipoda 127. New freshwater subterranean genus Relictoseborgia,n. gen. with remarks to genus Seborgia Bousfield(Fam. Sebidae). Studia Marina, 11-12:85-94.Kunath, C.E., and A.R. Smith, eds. 1968. The caves of theStockton Plateau Texas. Texas Speleol. Surv., 3(2):1-111.Longley, G. 1981. The Edwards Aquifer: Earth's most diversegroundwater ecosystem. Internatl. J. Speleol., 2(1-2):123-128.Ruffo, S. 1983. Nuovi Anfipodi mesopsammici delle !soleAndamane (Crust. Amphipoda) (Studi sui Crostacei Anfipodi,XCVIII). Boll. Mus. Civ. St. Nat. Verona,10:485-509.This is publication No. N.S.-54 of the Texas Memorial Museum.22

Bowman, T .E. 1992. Two subterranean aquatic isopod crustaceans new to Texas: Mexistenasellus coahuila (Cole and Minckley, 1972)(Asellida, Stenasellidae) and Speocirolana hardeni, new species (Flabellifera: Cirolanidae). Texas Mem. Mus., Speleolol. Monogr.,3:23-30TWO SUBTERRANEAN AQUATIC ISOPOD CRUSTACEANS NEW TO TEXAS:MEXISTENASELLUS COAHUILA (COLE AND MINCKLEY, 1972) (ASELLOTA:STENASELLIDAE) AND SPEOCIROLANA HARDENI, NEW SPECIES(FLABELLIFERA: CIROLANIDAE)Thomas E. BowmanDivision of CrustaceaSmithsonian Institution NHB-163Washington, DC 20560ABSTRACTMexistenasellus coahuila is reported from wells and hyporheicwaters in Bexar and Medina Counties. lllustrations aregiven of selected features and differences from type materialfrom Coahuila, Mexico, are noted. Speocirolana hardeni, newspecies, is described from Bexar and Val Verde Counties; it issimilar to S. thennydronis, but has longer and more slenderpereopods and uropods.In this paper I add 2 species, one new, to theknown isopod fauna of Texas, and bring up to datethe records of Cirolanides texensis, which nownumber 24.STENASELLIDAE DUDICH, 1924Mexistenasellus Cole & Minckley, 1972Until now North American isopods of the familyStenasellidae have been reported only from Mexico,where 6 species of Mexistenasellus and 2 species ofEtlastenasellus are known to occur (summarized inBowman, 1982b). The northernmost of thesestenasellids has been found recently in southernTexas, as documented below.Mexistenasellus coahuila Cole and Minckley, 1972Figs. 1, 5aMexistenasellus coahuila Cole and Minckley, 1972: 315-320,figs. 1-31; Cole, 1984:8-9.Mexistenasellus sp.: Henry, Lewis, and Magniez, 1986:460.Material.-TEXAS: Bexar County: spring on Ebank of San Antonio River, 23 Oct. 1985 (Scott J.Harden), 2 males, 4 females (USNM 242471); 26Oct. 1985 (Scott J. Harden, D. Kern, C. Lindblom),1 male, 1 female (USNM 242472); Leon Creekpowerplant well no. 1, 30 July 1979 (R. Rutland), 1male (USNM 250357); well in Brackenridge Zoo,San Antonio, 3 Sept. 1976 (H. Karnei), 1 female(USNM 250358); 22 May 1981 (M. Brzozowski), 1female (USNM 250359); 10 June 1981, 1 male(USNM 250360). Medina County: hyporheic zone,Hondo Creek: 4 mi. E Hondo, spring 1986 (Scott J.Harden), 3 males, 6 females (USNM 250187); 12May 1987 (Scott J. Harden), 5 males, 8 females(USNM 250188); S of Highway 90, spring 1986(Scott J. Harden), 1 male, 8 females (USNM250189).Remarks.-Mexistenasellus coahuila was describedfrom thermal springs (> 30°C) of the Cuatro23

Cienegas basin, central Coahuila, Mexico. Untilnow it has been known only from the type-locality.The Texas specimens fit Cole and Minckley's descriptionexcept for some minor details as follows:The uropods are slightly longer than the pleotelson.In M. coahuila they are distinctly shorter.The pleotelson has a few dorsal setae not shownby Cole and Minckley.On the exopod of the male pleopod 1 the distomedialsetae are much longer than the distolateralsetae. In Cole and Minckley's Fig. 24 they are onlyslightly longer. But in one of their paratypesmounted on a slide the medial setae are muchlonger.The arrangement of setae on the exopod of pieopod3 is slightly different.The exopod of pleopod 5 is unarmed in theMexican specimens; in the Texas specimens a seta ispresent on the suture between segments 1 and 2.The Mexican specimens were "bright red in life. "gf I •'I \III j r \~ 1LI.. ~I • I J L "'I /,Fig. l.-Mexislenasellus coahuila, a-b female, b-i male: a, antenna I, distal segments of flagellum; b, pereopod I; c, pleopod 2,anterior; d, pleopod 2, posterior; e, pleopod 1; f, pleopod 3; g, pleopod 4 exopod; h, pleopod 5 exopod.24

A note by Scott Harden that accompanied the 23 Octoberspecimens reads, "Bright, +/- 'crimson' redpigment on head, telson, pereopods, and posteriormargin of pereonites. Darker ('brighter') in somespecimens than others. "The penes, previously undescribed in M. coahuila,are shown in Fig. Sa. They are widely separatedand about half the length of pereonite 7. Theybroaden gradually in the proximal 2/3, then tum lateradand narrow to a rounded apex. The paired vasdeferens, packed with sperm in the specimen illustrated,are parallel and separated by less than theirwidth in pereonite 7. Near the distal end of thepereonite they turn laterad at nearly right angles andat the bases of the penes bend abruptly posteriad.CIROLANIDAE Dana, 1853Speocirolana Bolivar y Pieltain, 1950The genus Speocirolana comprises 6 known species(summarized in Bowman, 1982a), all fromcaves in Mexico. The new species described hereinis the first found outside of Mexico. It is mostsimilar to its geographically nearest congener, S.thermydronis Cole and Minckley, 1966, but aspointed out below, the 2 species are easily distinguishable.Speocirolana hardeni, new speciesFigs. 2-4MateriaI.-TEXAS: Bexar County: ArtesianWell no. 4, 1978-1979 (Henry Kamei, GlennLongley), 14 specimens; CPS Leon Creek Well no.1, 27 Aug. 1979 (Rick Rutland), 1 specimen; VerstraetenWell, 1977-1978 (Henry Kamei), 46 specimens.Val Verde County: Emerald Sink, 2 miles Nof Langtry, 31 March 1984 (Randy M. Waters),male holotype, 17.9 mm (USNM 250182);paratypes, male 18.8 nun, 2 females 17.0, 19.0 mm(USNM 250183); 26 May 1985 (D. Canny, ScottHarden), 2 paratypes, male 17.1 mrn, female 19.8mm (USNM 250184). Slaughter Bend Springs(29°39'N, lOo o 55'W), 29 September 1984 (RobertHershler), 1 juvenile (non-type) 4.7 mm (USNM250185).Description.-Length of largest male 22.4 mrn,of largest female 27.0 mm. Body about 3.7X as longas wide, widest at pereonite 6. Anterior margin ofhead with slight median concavity, without rostrum.Frontal lamina conical, narrowly rounded anteriorly,not projecting ventrally. Measured in dorsalmidline, pereonites 1-4 subequal in length,pereonites 5-7 subequal in length, each about 1/3longer than pereonites 1-4. Pereonites 4-7 withtransverse furrows, that of pereonite 7 almostreaching lateral margins. Coxae 2-3 reaching posteriormargins of pereonites, rounded posteriorly;coxae 4-7 produced beyond posterior margins ofpereonites into pointed processes, coxa 7 reaching orslightly exceeding posterior margin of pleonite 1.Epimera of pleonites with posteroventral comers angularbut scarcely produced. Pleotelson about aslong as basal width, about 1.4X length of pleon; lateralmargins sparsely armed with minute setae,slightly convex; apex pointed, slightly obtuse.Antenna 1 reaching beyond midlength ofpereonite 1, peduncle segment 3 about 1.4X lengthof segment 2; flagellum with up to 18 segments,each segment with esthete except segments 1, 14,and 18. Antenna 2 reaching posterior margin ofpereonite 6, peduncle segment 5 nearly 2X length ofsegment 4, flagellum with up to 40 segments.Mandible incisors 3-cuspid, cusps more deeplyseparated in right incisor, which is covered externallyby that of left; left lacinia with 13 marginalspines, molar with 27 marginal spines; palp segment2 about 1.5X length of segment 1, with row of setaeon distal 3/5 of lateral margin. Maxilla 1 exopodwith 12 terminal spines; endopod with 3 long plumosespines and 2 shorter setae. Maxilla 2 with 5and 7 pectinate spines on palp and exopod respectively;endopod with 13 setae. Maxilliped enditewith 13 marginal setae and 2 retinacula; palp segmentsdensely setose on medial margins, with 1 setaon lateral margin of segment 3, 2 on segment 4.Pereopod 1 propus about 2.3X as long as wide;posterior margin (palm) laterally with 2 shallowlobes each bearing a stout spine, medially evenlyrounded, hirsute. Pereopod 2 propus more slender,1.7X as long as wide, palm weakly and evenly convex,bearing 3 spines more slender than those inpereopod 1. Pereopod 3 like pereopod 2, but segmentsslightly more slender and elongate.Pereopods 5-7 slender, progressively longer posteriorly.Anterior and posterior margins wtihsetules; basis with 1 or 2 retrorsoplumose setae onproximal part of posterior margin; merus, carpus,and propus each with short spine near midlength ofanterior margin and clusters of long spines atantero- and posterodistal comers.Exopods of pleopods 1-4 with plumose setae onposterior and much of lateral margins; pleopod 5exopod with setae limited to distolateral corner;exopod of pleopod 1 undivided; exopods of pleopods2 and 3 with partial sutures; exopods of pleopods4 and 5 completely divided. Endopods25

Fig. 2.-Speocirolana hardeni, a-b, d-j female, c, k male: a, habitus, dorsal; b, head and pereon, lateral; c, pleon, left side,ventral; d, antenna 1; e, antenna 1, distal segments of flagellum; f, antenna 2, peduncle, dorsal; g, buccal area, ventral; h, leftmandible; i, left maxilla I; j, left maxilla 2, k, penes.26

~~..~ ~~ I))\\ "-,'.=bFig. 3.-Speocirolana hardeni, a-b female, c-i male: a, left maxilliped; b, endite of same; c, pereopod 1; d, pereopod 2; e,pereopod 4; f, pereopod 5; g, pereopod 6; h, pleopod 2; i, pereopod 5.27

':'\]' , '. '. '. '.'. '.'.'... '~),\'v~ , .,. I,:", ••,~ l···~. ', .\\'II~\---/~j(~f/~",..... .....7rFig. 4.-Speocirolana hardeni, a-d male, e female: a, pereopod 7; b, pleopod I; c, pleopod 3; d, pleopod 4; e, uropod.Speocirolana thennydronis male: f, uropod.28

unarmed, except in pleopod 2, which has 2 short setaeon distal margin. Appendix masculina insertednear base of pleopod 2 endopod, slightly taperingdistally to round tip reaching apex of endopod.Uropod protopod about 3.6X as long as width atmidlength, about 1.3X as long as endopod; distomedialcorner with 2 apical and 2 subapicalspines. Endopod subtriangular, about 3X as long aswide, apex with concavity bearing 5 long setae, 2additional long setae inserted subapically on dorsalsurface. Exopod about 0.8X length and 0.6X widthof endopod, apex oblique, produced medially, withcluster of about 5 long setae arising from shallowconcavity and directed posterolaterally.Etymology.-Named for Scott Harden, whoseenergetic collecting has led to great improvements inour knowledge of the aquatic isopods of Texas.Comparisons.-Speocirolana hardeni and S.thermydronis differ from the other 5 species ofSpeocirolana in having the apex of the pleotelsonpointed; the apex is rounded or truncate in the otherspecies. Speocirolana hardeni and S. thermydronisalso differ in lacking spines on the margins of bothuropod rami; the other species have marginal spineson the endopod or on both rami. The absence (ornear absence) of marginal setae on the endopods ofpleopods 1 and 2 also separates S. hardeni and S.thermydronis from the remammg species, all ofwhich have marginal setae on these endopods.Speocirolana hardeni and S. thermydronis differmost obviously in the shape and armament of theuropods. The rami of S. thermydronis are shorterand more rounded, and the armament of the apicesof the rami and the medial prolongation of the protopodare quite different (compare Fig. 4e and Fig.4f). the pereopods of S. hardeni are more slenderand elongate than those of S. thermydronis (compareFig. 4a herein with Fig. 16 of Cole and Minckley,1966).Cirolanides texensis Benedict, 1896Fig.5b,cIn my redescription of Cirolanides texensis in1964, I listed the distributional records known atthat time. I take this opportunity to bring this list upto date. Records are listed by county in alphabeticalorder.Records.-TEXAS: Bexar County: Artesian Wellno. 4; Isopit Cave; Leon Creek Powerplant Well no.1; O.R. Mitchell Well; Twin Pits Cave, 10 km N ofSan Antonio; Verstraeten Well. Burnet County:Longhorn Caverns. Comal County: Honey CreekCave; Python Pit; LCRA Well, New Braunfels.Fig. 5.-Mexislenase/lus coahuila: a, male pereonite 7, ventral, showing vas deferens and penes. Cirolanides lexensis, fromVerstraeten Well, right pereopod 1, lateral: b, 5.7 mmjuvenile; c, 13 mm male.29

Crockett County: 0-9 Well. Edwards County:Devil's Sinkhole, Rocksprings. Hays County:Ezell's Cave, San Marcos; Marcia's Well, San Marcos,cave at 85 ft. depth; artesian well at fish hatchery,San Marcos. Kerr County: Stowers Cave, 13mi. NW of Hunt. Medina County: Valdina FarmsSinkhole. Real County: Bonner Fallout ShelterCave, 5 mi. N of Leakey. Schleicher County: CaveY, 5 mi. N of Eldorado. Uvalde County: IndianCreek Cave, 22 mi. NW of Uvalde; McNair Cave;Rambie's Cave, 2 mi. N of Uvalde. Val VerdeCounty: Diablo Cave, 15 mi. NW of Del Rio, calyxhole entrance; Four-Mile Cave, 4 mi. N of Del Rio;H.T. Miers Cave, 24 mi. N of Del Rio; Unnamedspring, ca. 20 mi. N of Del Rio, just E of Devil'sRiver; Little Diablo Cave, 1 mi. from calyx hole entranceto Diablo Cave; Slaughter Bend Springs(29°39'N, lOo o 55'W).Gnathopod development.-The gnathopod(pereopod 1) of Cirolanides has a precocious development.Fig. 5 shows, to the same scale,gnathopods from a 5.7 mm juvenile and a 13 mmadult male from the same sample. Although thejuvenile body length is only 44 % of that of theadult, the length of its propus is 82 % of that of theadult; thus the juvenile propus has grown allometricallyat nearly twice the rate of the body. In addition,the palm of the juvenile propus has 3 conspicuousnarrow teeth that are reduced to lowrounded lobes in the adult. The juvenile dactyl isslightly longer and much more slender than that ofthe adult.Precocious gnathopod development does not occurin Speocirolana. Its presence in Cirolanidessuggests a need for the juvenile to handle objectsthat a gnathopod developing isometrically could notcope with.DISCUSSIONThe occurrence of(Mexistenasellus coahila)the same speciesand a species pair(Speocirolana thermydronis and S. hardeni) in localitiesin northern Mexico and southern Texas separatedby about 300 (Speocirolana) or 400(Mexistenasellus) kID suggests the earlier existenceof continuous ranges from which fragmentation hasled to the present isolated populations. A similarhistory for the subterranean amphipod genus Mexiweckeliais discussed by Holsinger (1973).LITERATURE CITEDBenedict, J.E. 1896. Preliminary descriptions of a new genusand three new species of crustaceans from an artesian well atSan Marcos, Texas. Proc. United States Nat!. Mus.,18(1087): 615-617.BolIvar y Pieltain, C. 1950. Estudio de una Ciro/ana cavernlcolanueva de la region de Valles, San Luis POtOSI, Mexico.Ciencia, Mexico, 10:211-218.Bowman, T.E. 1964. Antra/ana lira, a new genus and species oftroglobitic cirolanid isopod from Madison Cave, Virginia.Internat!. J. Speleol., 1(1+2):231-236, pIs. 50-57.Bowman, T.E. 1982a. Speociro/ana pubens and S. endeca, newtroglobitic isopod crustaceans from Mexico (Flabellifera:Cirolanidae). Assoc. Mexican Cave Stud. Bull., 8:13-23.Bowman, T.E. 1982b. Three new stenasellid isopods fromMexico (Crustacea: Asellota). Assoc. Mexican Cave Stud.Bull., 8:25-38.Cole, G.A. 1984. Crustacea from the Bolson of CuatroCienegas, Coahuila, Mexico. J. Arizona-Nevada Acad. Sci.,19:3-12.Cole, G.A., and W.L. Minckley. 1972. Stenasellid isopod crustaceansin the Western Hemisphere -- a new genus and speciesfrom Mexico -- with a review of other North Americanfreshwater isopod genera. Proc. Bio!' Soc. Washington,84(39) :313-326.Dana, J.D. 1853. Crustacea, part II. United States ExploringExpedition, 14:689-1618.Dudich, E. 1924. Uber Prate/sonia hungarica Mehely. Zoo!'Anz.,60:151-155.Henry, J.-P., J.J. Lewis, and G. Magniez. 1986. Isopoda:Asellota: Aselloidea, Gnathostenetroidoidea, Stenetrioidea.pp. 434-464 in: L. Botosonaneau, ed., Stygofauna Mundi.Leiden: E.J. Brill, Dr. W. Backhuys. vi + 740 pp.Holsinger, J.R. 1973. Two new species of the subterraneanamphipod genus Mexiweckelia (Gammaridae) from Mexicoand Texas, with notes on the origin and distribution of thegenus. Assoc. Mexican Cave Stud .Bull, 5:1-12.This is publication No. N.S.-55 of the Texas Memorial Museum.30

Cokendolpher, LC., and l.R. Reddell. 1992. Revision of the Protoschizomidae (Arachnida: Schizomida) with notes on the phylogenyof the order. Texas Mem. Mus., Speleol. Monogr., 3:31-74REVISION OF THE PROTOSCHIZOMIDAE (ARACHNIDA: SCHIZOMIDA)WITH NOTES ON THE PHYLOGENY OF THE ORDERJames C. Cokendolpher2007 29th St.Lubbock, TX 79411andJames R. ReddellTexas Memorial MuseumThe University of Texas at Austin2400 TrinityAustin, TX 78705ABSTRACTThe predominantly Mexican family Protoschizomidae isrevised to include 11 species in two genera. ProroschizomusRowland is recognized for seven species, five of which areplaced in two species groups: the pachypalpus group includes P.pachypalpus (Rowland) (Tamaulipas), P. rowlandi n. sp. (SanLuis Potosi), and P. occidenralis Rowland (Colima); the sprouseigroup includes P. sprousei n. sp. and P. purificacion n. sp. (bothfrom Tamaulipas). Two species known only from females, P.treacyae n. sp. and P. gertschi n. sp. (both from Tamaulipas),are not placed in species groups. Agastoschizomus Rowlandincludes A. lucifer Rowland (San Luis Potosi), A. huitzmolotitlensisRowland (San Luis Potosi), A. stygius n. sp. (Hidalgo),and A. patei (Tamaulipas). Several undetermined immaturespecimens, including the first species of the family from Texasand the U.S.A., are also recorded. The female genitalia aredescribed for the first time for members of the Protoschizomidae.A new system for naming articles in the female flagellum and arevised method for numbering flagellar setae are provided. Thephylogeny of the order Schizomida is discussed, and a cladogramof the families and protoschizomid genera and species isprovided. The available data on habitats and cave life arerecorded. Agasroschizomus patei may be paedomorphic, the firstfor the order.INTRODUCTIONThe family Protoschizomidae is a small distinctivegroup of the order Schizomida known only fromsouthwestern Texas (U.S.A.) and Mexico. The firstmember of the family Protoschizomidae was describedby Rowland (1971) as Agastoschizomus luciferand placed in the Hubbardiidae (=Schizomidae).The subfamily Megaschizominae was named byRowland (1973b) to include Agastoschizomus andthe African genus Megaschizomus Lawrence. Rowland(1975b) later recognized the distinctness of theAmerican forms and proposed the family Protoschizomidaeto include A. lucifer, A. huitzmolotitlensisRowland, Protoschizomus pachypalpus Rowland,and P. occidentalis Rowland. Megaschizomus wasretained in the monotypic subfamily Megaschizominaein the Hubbardiidae (=Schizomidae).In the present paper we redescribe the knownspecies, describe five new species of Protoschizomus31

and two new species of Agastoschizomus and describefor the first time the female genitalia of theProtoschizomidae. A phylogeny for the two familiesof the order and members of Protoschizomidae isprovided.METHODSThe methods and terminology essentially followthose of Reddell and Cokendolpher (1985), exceptfor some anatomical terms which follow van derHammen (1986). We have not followed all of vander'Hammen's scheme as some doubt of its validityexists (see Shear et aI., 1987; Shultz, 1989). Thefemale genital sternites were examined in lactophenol.Routine examinations were of specimens placedin alcohol in small dishes. Positioning of specimenswas best maintained by anchoring them in fine whitesand. Black sand was sometimes used to view lightcolored structures. Examinations using compoundmicroscopes were made while the specimen or dissectedpart(s) were placed on a microscope slide(generally a depression slide) in 100% glycerol. Asmall tuft of cotton placed in the glycerol was usedto anchor the specimens. Specimens were nevermounted with Balsam or similar media, a practicethat should be discouraged since detailed studies athigher magnifications are severely hampered whenspecimens cannot be rotated.We, as well as other earlier authors, have usedmisleading terminology regarding the "spines" ofthe pedipalps. According to "The Torre-BuenoGlossary of Entomology," a spine is not separatedfrom the cuticle by a joint. This is not the case withthe "spines" on the schizomid pedipalp. These structuresare more correctly referred to as spinose-setae.Likewise, the use of spur(s) has sometimes beenmisused. A spur is a thick cuticular appendage connectedto the body wall by a joint. The use of spurfor the process on the mesal margin of the pedipalptrochanter appears correct, whereas such a referenceto structures on the pedipalp basitarsus-tarsus is incorrect.Although these pedipalpal structures aremore correctly termed spines, these structures havebeen commonly referred to as spurs for many years.For that reason we shall continue to refer to thebasitarsal-tarsal spurs. Our use of the term plumosein earlier publications is somewhat misleading. Asused by Harvey (in press), pilose is a better term todescribe this condition.The propeltidium was measured from the posteriormostseta on the anterior process to the posteriormargin of the propeltidium. Leg and pedipalp segmentswere measured from dorsolateral joint to dor-solateral joint. The numbering of the setae on thechelicerae follows Lawrence (1969) (see Fig. la).The setae on the protoschizomid flagellum are essentiallydivided into two groups: larger setae andmicrosetae. The microsetae are relatively uniform insize and shape and are found only on the distal tip ofthe flagellum and on the ventral surface of the maleflagellum either on or posterior to the retractablea6. '-'IbFig. I.-la, Chelicera of male Agastoschizomlls hllitzmolotitlensis,Setal gn;up numbering follows that of Lawrence(1969). Type 5 setae lacking in Protoschizomidae, S = serrula.lb, Pro-, meso-, and metapeltidium of female Protoschizomusrowlandi, S = sclerite between propeltidium and splitmesopeltidium; A = anterior process with row of two setae,followed by pair of setae at base. Ie, Dorsal view of femaleflagellum of Megaschizomlls mossambiclis (redrawn fromLawrence, 1958), Numerals refer to segment/article positions;see methods for further details, Scale lines = 0.2 mm.la32

lobes. The larger setae vary greatly in size and occuras simple pointed structures, or somewhat flattenedwith pilose borders, or as round pointed structureswith pilose distal ends. Harvey (in press) noted therecurring pattern of setae and introduced the firstnotation for labeling the larger setae on the flagellumof schizomids (Hubbardiidae of Australia).While protoschizomids have all of the setae noted byHarvey, several other setae can be found on protoschizomidflagella, thus necessitating a renumberingof the setae. We have attempted to correlate ournumbers with those of Harvey. Whereas Harveyonly numbered larger setae, we have attempted tonumber all setae (except a tenninal microseta, whichis difficult to observe). We have done this becausesome setae will be microsetae on some species andlarger setae on other species.Female flagellum: Dorsal surface bearing up tofive medial setae [dml, dm2, dm3 (pair of setae),dm4] and up to four pairs of lateral setae (dll, d12,d13, dI4). Ventral surface bearing up to eight medialsetae [vml, vm2 (pair of setae), vm3 (submedialpair of setae), vm4 (pair of setae), vmS] and twopairs of ventrolateral setae (vII, vI2). From therelative positions on the adult female flagellum, thefollowing correlations are suggested [Harvey (inpress) = present system]: dml=dml, not present=dm2,not present=dm3, dm2=dm4, dll=dll,not present=dI2, dI2=dI3, not present=dI4,vml =vml, vm2=vrn2, not present=vm3,vm3=vm4, vm4=vmS, vll =vll, vI2=vI2. Thelateral setae are variable as to position and often appearto switch positions from dorsolateral to ventrolateral.An alternate hypothesis is that the said setaeare not homologous and represent losses and gains,respectively. Since this would involve two steps(loss and gain) we feel it is more parsimonious toassume a shift in position. Several setae are shortand sometimes thinner in the Protoschizomidae:dm3, dll, d14. In some specimens dl4 appear to bemicrosetae. A single microseta (larger in A. pateifemales), which we have not numbered, is oftenpresent on the most terminal area of the flagellum.Male flagellum: The setation is similar, but neverexactly the same as in females. In A patei, severalpairs of setae have shifted to a ventral position. Thisshift is not noted in females or other males. Possiblythe presence of ventrolateral lobes in the otherspecies precludes the varying of setae. Numberingof setae is greatly aided by noting that vm3 areusually off-eentered and that dm3, dll, and dl4 aregenerally small in size.There are five positions on the female flagellumof schizomids where annuli can occur. These areasmay be evident as thinnings of the cuticle(sometimes only on ventral surface) or as cleanbreaks in the cuticle. We have numbered these areas,based on the genus with the greatest number of annuli(Megaschizomus), to standardize our descriptionsand discussion (Fig. Ic). The positions of theannuli are always in the same place relative to thelarger setae: segment I contains no setae; segment IIwith dml, dm2, dll, vml-vm3; segment III withdm3, vm4; segment IV with dm4, d12, vll, andsegment V with d13, d14, vmS, v12. Position 2 annulifound in Hubbardiidae do not occur in protoschizomids.The distinction between segments andarticles is not entirely clear. When examined underlOOx the divisions between segments appear asbreaks in the cuticle. The lines are generally thinand sometimes difficult to see. No musculature isevident and it is unlikely that these segments can bemoved. Attempts to bend the flagellum at a segmentaljunction in a preserved specimen result in a cleanbreak at that border. Subdivisions of the segmentsare evident by areas of thinned cuticle. These zonesare sometimes quite wide and appear lighter in colorthan the surrounding cuticle. Segments do not appearto break at these subdivision zones and in preservedspecimens this zone will withstand somebending. When the flagellum is divided by breaks inthe cuticle we refer to the resulting subdivisions assegments. When a segment or entire flagellum isdivided by only thinnings in the cuticle we refer tothe resulting units as articles. Annuli are alsopresent in juveniles but not in the flagellum of maturemales.The following acronyms have been used to designatemuseums in which the specimens are deposited:AMNH - American Museum of Natural History,New YorkCAS - California Academy of Sciences, SanFranciscoTMM - Texas Memorial Museum, Austin (allspecimens formerly reported in the collection ofTexas Tech University have been transferred toTMM).Character relationships were first examined usingMcClade (Maddison and Maddison, 1987). An exactanalysis was performed using Hennig86 (Farris,1988) which is "guaranteed" to find all of the mostparsimonious cladograms. All multistate characterswe treated as non-additive (unordered). To attemptto reduce the ambiguity of the strict consensus,successive approximations character weighting wasapplied. Because none of the characters conflicted,this produced identical results. Autapomorphiccharacters were not coded for the matrix. By doing33

so we reduced the data matrix from 92 to 44characters.In determining the polarity of character states wehave generally followed the five "rules of thumb"used by Shear and Gruber (1983): "I) occurrence inoutgroups speaks for plesiomorphy, 2) less differentiated,more homonomously patterned meristic charactersare plesiomorphic, 3) states resembling thosein juveniles are plesiomorphic, 4) characters consistentlycorrelated with others known to be apomorphicare likely themselves to be apomorphic, and (5)correlations between morphological and chronologicalorecological positions (as in the foregoing) areto be used with caution. "PHYLOGENY OF THE PROTOSCHIZOMIDAEThe Uropygi and Schizomida (superorder Camarostomata)differ from other members of the Arachnidea(Tetrapulmonata, Megoperculata) by numeroussynapomorphies. These include: fusion of thepedipalp coxae, posterior narrowing of the intercoxalregion, reduction or loss of a postcerebralpharynx, presence of 2-3 transverse bridges with 1-2fenestra on the endosternite, great elongation of thepatellae of leg J, presence of two well-developedtrichobothria distodorsally on leg tibiae I, singletrichobothrium on tibiae II-IV each, presence of pretarsaldepressor muscles originating from the posteriorwalls of patellae II-IV, absence of postgenitaleversible appendages, presence of a pygidial exocrinegland, presence of a multi-jointed pygidial flagellum,absence of a distinct middle piece on thespermatozoan axonome, female grasping of maleabdomen during mating, and the presence of aprenymph and four nymphal instars (Weygoldt andPaulus, 1979; Shear et aI., 1987; van der Hammen,1989; Shultz, 1990; Shultz, pers. comm.). Severalother characters are known to separate Uropygi andSchizomida from other Arachnidea, but their polaritieshave not been established. These characters arefound in spermatogenesis and spermatozoa morphology(Alberti and Palacios-Vargas, 1987).In order to better understand the relationshipswithin the Protoschizomidae, we first investigatedTable I.-Characters and presumed polarities for cladisticanalyses of selected taxa of the superorder Camarostomata. 0 =plesiomorphic; 1,2,3 = apomorphic.O. Heart segments V and VI well developed (0), prosomal heartsegements absent (I).I. Prodorsum not divided into pro-, meso-, and metapeltidia(0), divided (I).2. Anterior process absent (0), present (I).3. Anterior prodorsum without seta (0), with setae (I).4. Median eyes present (0), absent (I).5. No setal pairs on prodorsum (0), with pairs of setae (I).6. Two pairs stigmata in segments VID and IX (0), one pair inVIII (I).7. Eight pairs intestinal diverticula in the abdomen (0), with sixpairs (I).8. Anterior abdominal neuromeres absent (0), eight abdominalneuromeres present (I).9. Propeltidium and tergites without distinct setal pattern (0),with (I).10. Tergites without microsetae (0), tergites I-II with pairs ofmicrosetae (I).II. Pointed process on coxa 2 absent (0), present (i).i2. Unable to jump, femur IV not enlarged (0); able to jump,femur IV and its associated muscles enlarged (i).i3. Male flagellum divided into many segments (0), undivided(I).14. Cheliceral fixed digit with two teeth (0), with three teeth (I),with more than three teeth (2).15. Cheliceral sermla with rounded knobs (0), with hyaline teeth(I).16. Cheliceral brush absent (0), present (I).17. Row of two setae on anterior process (0), one seta (I), pairof setae followed by single seta on anterior process (2).18. No seta at base of anterior process (0), pair of setae (I).19. More than two pairs dorsal propeltidial setae (0), twoanteriorly placed setal pairs present (i), two widely spacedsetal pairs present (2).20. Pedipalps sexually dimorphic (0), not dimorphic (i).2!. Male and female pedipalps of approximaly equal length ascompared to the body length (0), male pedipalps longer(I).22. Pedipalpal trochanter not produced (0), slightly produced(I).23. Basitarsal-tarsal spurs symmetrical (0), slightly asymmetrical(I), asymmetrical (2).24. Femur IV less than 4.8 times longer than deep (0), more (I).25. Trochanter IV about 1/3 length of femur IV (0), about 1/2length (1,2).26. Tergite III with two setae (0), with four setae (I).27. Male sternites with scattered or irregular rows of setae (0),with two distinct rows of setae (I).28. Sternite VI long (0), short (I).29. Eight dorsoventral muscles (0), seven (i).30. Female flagellum with segments and articles (0), withsegments only (I), without segments or annuli (2).31. Female flagellum with position 2 annulus (0), annulus absent(I).32. Flagellar setal patterns same in both sexes (0), patternsdifferent (I).33. Dm2 seta not present on female flagellum (0), present (1,2).34. Dm2 seta not present on male flagellum (0), present (I).35. Vm4 setae present on male flagellum (0), absent (I).36. Dm4 seta present on female flagellum (0), absent (I).37. Male flagellum not expanded distally (0), expanded (1,2).38. Male flagellum with distinct stalk (0), without stalk (I).39. Male flagellum over 3x long as wide (0), less than 3x (1,2).40. Male flagellum distally rounded (0), laterally compresseddistodorsally (I).41. Microsetae at base of flagellum lobes (0), microsetae onlobes (I), microsetae absent (2).42. Spermathecae with one pair lobes with pits (0), one pairlobes without pits (I), two or more pairs lobes without pits(2).43. Receptaculum margins smooth without pits (0), smooth withpits (I), lobed with pits (2), saw-toothed with pits (3).34

the relationships of the order and then those formembers of the family. The character numbers inthe following discussion are the same as those appearingin Tables 1-2 and Fig. 2.Some authors consider the Schizomida to be asuborder of Uropygi, whereas others consider it anorder. We, like some previous authors, prefer torecognize the Schizomida and Uropygi as separateorders within the Camarostomata. Most authorscombining the two orders emphasize the existence ofplesiomorphies or characters of undetermined polarity.Some of the differences between Schizomida andUropygi are as great as the differences among theUropygi, Amblypygi, and Araneae. Shear et al.(1987) suggested that the pygidial exocrine glandwas an autapomorph for the Uropygi. This is not thecase, as such glands are also present in the Schizomida(Brignoli, 1973).Synapomorphies for the Uropygi are: pedipalpschelate, coxal gland orifices associated with leg IIIabsent, lateral tergocoxal muscle inserted on thepleural membrane adjacent to the coxa, presence ofommatoids on dorsal surface of anal somite, flagellumwith clear regions ventrally which are sensitiveto light, presence of five posterior abdominalneuromeres, and modification of cheliceral cleaningorgan.The Schizomida can be separated from theUropygi as well as other Arachnidea by the followingsynapomorphies: prosomal heart segments absent(character 0); prosoma divided into pro-,meso-, and metapeltidia (character 1); modificationof the anterior portion of the prosoma into a pointedprocess (character 2); presence of setae on and basalto the anterior process of the propeltidium (character3); median eyes absent (character 4); presence ofpairs of dorsal setae on the propeltidium (character5); stigmata lacking on segment IX (character 6);with six pairs of intestinal diverticula in the abdomen(character 7); with eight anterior abdominalneuromeres (character 8); male flagellum not dividedinto numerous segments (character 13); propeltidiumand tergites with distinct setal pattern(character 9); tergites I-II with pairs of microsetae(character 10); anterior distal tip of coxa II prolongedas a sharp process (character 11); ability tojump, femur IV and its associated muscles beingenlarged (character 12). Three additional charactersare known which appear to be synapomorphies forthe Schizomida, but because they have only beendetermined (examined) in a single species of Hubbardiinaewe do not code them for the data matrix:absence of a posteriorly directed dorsal endosternitesuspensor muscle; absence of anterior transpatellarmuscle; and absence of the posterior patellotibialmuscle. Additional characters are known (Millot,1942) for which we have not determined the polarities.Specifically, uropygids have four pairs of prosomaticdiverticula, schizomids have one palr;Table 2.-Character states in selected members of thesuperorder Camarostomata (0 codes plesiomorphic condition;1,2,3 codes apomorphic conditions; ? codes unknown conditionor condition not relevant to region of cladogram being resolved).1, Uropygi; 2, Protoschizomus pachypalpus; 3, P. rowlandi; 4,P. occidentalis; 5, P. sprousei; 6, P. genschi (male unknown);7, P. purificacion (male unknown); 8, P. treacyae (maleunknown); 9, Agastoschizomus lucifer; 10, A. huitzmolotitlensis(female unknown); 11, A. srygius (male unknown); 12, A. patei;13, Megaschizominae; 14, Hubbardiinae.TaxaI 2 3 4 5 6 7 8 9 10 11 12 13 14Character0.0 1 I I 1 1 1 1 I I 1 1I. 0 1 I I I 1 I I I I I I2.0 I I I 1 I 1 1 I I I I3.0 I I 1 1 1 1 I 1 1 1 14.0 I I 1 1 I I 1 1 1 1 15.0 1 1 I I 1 1 1 1 I 1 16.0 1 1 1 1 1 1 1 I 1 I 17. 0 I I 1 1 I 1 I I I I I8.0 1 1 1 1 1 I 1 1 1 1 19.0 1 I 1 1 1 I I I 1 1 I10.0 1 1 I 1 I 1 1 1 I 1 111. 0 I I I I 1 1 I I I I I12.0 1 I 1 I 1 1 1 1 I 1 I13.0 1 I 1 I ? ? ? I I ? I I 114.0 0 0 0 0 0 0 0 0 0 0 0 1 2IS.? 0 0 0 0 0 0 0 0 0 0 0 1 116.0 0 0 0 0 0 0 0 0 0 0 0 1 117.? 0 0 0 0 0 0 0 1 1 1 1 2 0,318.0 1 1 1 1 1 1 1 1 1 1 1 0 019.? 1 I 1 0 0 0 0 0 0 0 0 0 0,220.0 1 I II? ? ? 1 ? ? 1 0 021.? 1 1 1 0 0 0 0 0 0 0 0 ? ?22.? 0 0 0 I 0 1 0 0 0 0 0 ? ?23.? 0 0 0 0 0 0 0 0 0 0 0 1 224.? 0 0 0 0 0 0 0 I I I I 0 025.0 I I I 1 1 I I 0 0 0 0 2 026.? I 1 1 0 0 0 0 0 0 0 0 ? ?27.? 0 0 0 0 ? ? ? II? I 0 028.? I I I 1 I 1 1 0 0 0 0 0 029.0 0 0 0 0 0 0 0 0 0 0 0 1 130.? 0 0 ? 0 0 0 0 0 ? 0 2 0 I31.? I I ? I 1 I II? 1 I 0 0,232.? I I ? I ?? 1 1 0 133.? 0 0 ? 0 0 0 0 ? 2 0 0 034.? 0 0 0 0 O? 1 I ? 1 0 035.? I I 0 0 0 0 0 0 0 0 0 ? ?36.? 0 0 ? I 0 I 0 0 ? 0 0 0 037.? 1 I 1 I ? ? ? 0 0 ? 0 0 0,238.? 1 I 1 I ? ? ? II? 1 0 039.? 0 0 0 0 ? ? ? II? 1 0 0,240.? I I 0 0 ? ? ? 0 0 ? 0 ? ?41.? 0 0 0 1 ? ? ? 0 2 ? 0 ? ?42.? 0 0 0 0 0 0 0 0 ? 0 0 I 243.? 1 1 1 0 0 0 0 0 ? 3 2 ? ?35

uropygids have no frontal gland, whereas one frontalgland is present in schizomids; uropygids havepaired ovaries, but ovaries are unpaired in schizomids;uropygids have 12 esophageal neuromeres,schizomids nine.The Schizomida are currently divided into threefamilies, one of which is extinct. Because the specimensof the extinct family Calcitronidae (from Mioceneor Pliocene deposits in Arizona, U.S.A.) aregenerally in bad condition, character states are hardif not impossible to determine. Rowland (1975a)reviewed the characters and accepted the leg tarsi ashaving 7:5:4:4 segments. The leg tarsi in theUropygi, Hubbardiidae, and Protoschizomidae are7:3:3:3. Because the few specimens of Calcitronidaeknown do not show characters needed to verify theirplacement in the Schizomida (synapomorphies listedabove), an alternate hypothesis can be produced inwhich the. tarsal count 7:3:3:3 is derived in theSchizomida (minus Calcitronidae) and Uropygi. Theplesiomorphic condition is found in the Calcitronidae.In such a scheme the Calcitronidae wouldbecome the sister group and represent an unnamedorder. While Amblypygi have numerous tarsal Isegments (many more than 7), their legs II-IVcounts are like the Schizomida and Uropygi (Shultz,1990). The genus Calcoschizomus (currently placedin Hubbardiidae) is likewise extinct and like themembers of the Calcitronidae the only knownspecimens are not preserved sufficiently well to determinethe character pOlarities. Members of theCalcitronidae and Calcoschizomus will not be mentionedin the following discussion, because of theuncertainty of the characters.Rowland (1975b) in erecting the family Protoschizomidaedid not discuss the phylogenetic relationshipsof the family but from the name of thefamily he clearly indicated he felt it to be moreprimitive than the Hubbardiidae (=Schizomidae). Inhis unpublished dissertation, Rowland (1975a)elaborated on his reasons for considering the Protoschizomidaeto be primitive. He presented severalproposed transformation sequences in which thestructure of the chelicerae and female flagellum wereused to support his hypothesis.The family Protoschizomidae was defined on thebasis of the presence of eight pairs of dorsoventralabdominal muscles, the female flagellum with distinctsegments, the absence of a serrula and brush onthe chelicera, only two teeth on the fixed digit of thechelicera, the presence of "true spines" (see materialsand methods on terminology) on the pedipalp,the symmetrical placement of the spurs of the pedipalp,the ratio of claw and spur lengths to the dorsallength of the pedipalp basitarsus-tarsus, thelength/depth ratio of trochanter IV and femur IV,and the degree of separation of the mesopeltidialplates. Rowland (1975b) did not compare the Protoschizomidaewith the Uropygi, the generally recognizedsister group of the Schizomida.Rowland (1975b) and Rowland and Reddell(1979a) distinguished the genera Agastoschizomusand Protoschizomus on the basis of several ratios:gap between mesopeltidial plates/width of plate,length/width ratio of metapeltidial plates,width/length ratio of stemites IV-VII, length of clawand spurs/dorsal length of pedipalp basitarsus-tarsus.These ratios appeared quite distinctive atthat time since only two small epigean(Protoschizomus) and two large cavernicole(Agastoschizomus) species were known. With thediscovery of several additional cavernicoles(described herein) exhibiting varying degrees of adaptationto the subterranean habitat, the ratios usedto separate the two genera appear to have limitedsignificance at the generic level. We retain the twogenera with greatly different diagnoses based oncharacters apparently unrelated to the degree ofspecialization for cave existence. The following is adiscussion of characters used earlier for separationof the genera and characters which we consider to bemore phylogenetically significant.Cheliceral teeth (character 14): The fixed digit ofthe chelicerae of the Protoschizomidae all have twoteeth with essentially the same shape as in theUropygi. The apomorphic conditions of three andmany more than three teeth are found in the Hubbardiidae.Megaschizominae have three teethwhereas Hubbardiinae have many more than threeteeth.Cheliceral serrula (character 15): The Hubbardiidaepossess a distinct row of hyaline teeth (serrula)on the mesal margin of the movable jaw of thechelicerae which probably function as a cleaning organ.The family Protoschizomidae lacks a true serrula,but does have in the same place a series of smallrounded to sharp knobs or teeth similar in positionand shape to those in the Amblypygi. Uropygidslack teeth. We have recorded the number of teeth inthe serrula, but intraspecific variation in this charactermakes it of limited value even for species recognition.Accessory teeth: Harvey (in press) first noted theimportance of small rounded teeth on the lateralmargins of the movable jaw of the chelicerae andnamed them accessory teeth. Megaschizominae haveseven teeth, whereas Hubbardiinae have 0-3 teeth.Because these teeth are also missing in all protoschi-36

zomids their absence is considered to be plesiomorphic.Cheliceral brush (character 16): The Hubbardiidaehave a brush of setae (type 5) on the base of thefixed cheliceral digit. The plesiomorphic conditionof no brush is known in the Protoschizomidae andUropygi.Mesal setation of the chelicerae: We have recordedthe number of setae for each species of Protoschizomidae,but intraspecific variation in thischaracter makes it of limited value.The anterior process of the propeltidium is downturnedin most Hubbardiidae and Protoschizomidae.The straight forward-pointing process in A. patei isconsidered to be an autapomorphy.Within the order, several patterns have been observedin the setation on the anterior process of thepropeltidium. As already noted the presence of thesesetae is a synapomorphy for the order. In the Protoschizomidaethere is either one or a row of two setaeon the process. Protoschizomus (Fig. Ib) sharesa row of two setae with all New World Hubbardiidae,with the exception of species of Hubbardia. InHubbardia, Megaschizomus, Trithyreus, and mostOld World species assigned (most incorrectly) toSchizomus the anterior process bears a pair of setaefollowed by a single seta. The presence of a row oftwo setae in Protoschizomus and Hubbardiidae impliesthat this is the plesiomorphic condition in theprotoschizomids. Agastoschizomus lost one seta(character 17).Pair of setae at base of process (character 18): Inthe Protoschizomidae there is a pair of setae at thebase of the anterior process of the propeltidium.This pair is absent from the Hubbardiidae andUropygi.Setae on anterolateral margin of propeltidium:Megaschizominae have six or seven setae, in additionto the setae on and at the base of the anteriorprocess, on the anterior margin of the propeltidium.All other Schizomida and Uropygi lack setae in thisregion.The number of dorsal setae and placement on thepropeltidium varies somewhat intraspecifically, butspecies of the P. pachypalpus group all have threeor four pair, whereas other protoschizomid specieshave one or two pair. Members of Hubbardiidaehave two to five pair. In the Protoschizomidaewhere the lower number of pairs occurs it is the posteriormostpairs that are absent, whereas in speciesof Hubbardiidae with two pair, the middle pairs areabsent. Because different setal pairs appear to havebeen lost, we suggest the plesiomorphic conditionfor the order is more than two pairs (character 19).Further support for this polarity assignment is thewidespread occurrence of this condition in the order.Eyes: The Protoschizomidae and Megaschizominaelack eyespots. The Hubbardiidae usually havedistinct eyespots, but these are sometimes absent,most frequently in cavernicolous species. A fewspecies of Old World Hubbardiidae have distinctfaceted lateral eyes. Lateral eyes are present in otherArachnidea indicating this is a plesiomorphic condition.However, the widespread loss of eyes in theSchizomida suggest that the presence of lateral eyespotsand eye facets is a reversal or a new structure.The apomorphic interpretation of the eyespots iscorrelated with other characters in the Hubbardiidaeknown to be apomorphic.Posterodorsal abdominal process: Some speciesof Hubbardiinae have a distinctly developed processat the posterior margin of abdominal segment XII.The absence of this process in the Protoschizomidae,Megaschizominae, and many species of Hubbardiidaeimplies that absence is the plesiomorphic condition.Pedipalps: The shape of the pedipalp is useful inspecies recognition. In the Hubbardiidae the pedipalpmay be highly modified both in shape and armatureand appears of use in constructing phylogenies.Unlike many species of Hubbardiidae andUropygi, the male and female pedipalps in the Protoschizomidaeare essentially the same shape andpossess the same armature (character 20). However,the male pedipalp is distinctly longer than the femalepedipalp in relation to body length in the P. pachypalpusgroup. This condition is considered to be areversal in the Protoschizomidae (character 21). Theincrease in size of the spurs and claw in the Protoschizomidaeis generally correlated with bodysize. The claw is greatest in A. patei and almost aslarge in other species of Agastoschizomus. These arealso the most troglomorphic species and this isprobably an adaptation to cave life and possibly correlatedwith a more active hunting strategy of thecavernicole in a food poor environment. Large clawsalso occur in Megaschizominae and Trithyreus, bothwith large species.Pedipalp trochanter produced distally: The distalmargin of the trochanter of the pedipalp in mostspecies of Protoschizomidae is not produced; it isslightly produced in P. sprousei and P. purificacion.It is distinctly produced in Megaschizominae. In theHubbardiinae it may not be produced or may behighly produced and bear spurs or spinose-setae onthe distal margin. The pedipalps ofjuveniles are notproduced, implying that this is the plesiomorphiccondition. The slightly produced trochanter in some37

Protoschizomus is considered a derived condition inthe Protoschizomidae (character 22). The trochanterin the Uropygi can be slightly produced or not.When slightly produced, it is widened as well andnot laterally compressed as in schizomids. We believethe slightly "produced" state in the Uropygiarose independently and is not homologous with theproduced state in the Schizomida.The pedipalpal trochanter has a spur on the mesalside in most Hubbardiidae. However, the spur is absentin some apparently unrelated genera in theHubbardiinae. It is absent from all Protoschizomidaeand Uropygi and its presence in the Hubbardiidae istentatively considered to be a synapomorphy. Apparently,the spur has been secondarily lost (or derived)on several different occasions in the Hubbardiinae.Further studies, with emphasis on the occurrenceof the mesal spur, are needed to confirmthe polarity of this character.Basitarsal-tarsal spurs on pedipalp (character 23):The spurs on the protoschizomid pedipalp are symmetricallyplaced with respect to the claw. InMegaschizominae they are slightly asymmetrical,whereas in the Hubbardiinae they are distinctlyasymmetrical.Spinose-setae on pedipalps: One of the charactersused by Rowland (1975b) in defining the Protoschizomidaewas the presence of thick "spines" withsocketed bases on the pedipalp (see materials andmethods on terminology). This character is, however,also shared by Megaschizomus, Trithyreus,and Schizomus ashmolei Reddell and Cokendolpher.There is a complete gradation between hair-like setaeand spinose setae and it is obvious that spinosesetae are only modified setae. The absence of rigidspinose setae on most species of Hubbardiinae,however, implies that the character is of some phylogeneticsignificance. All species of Protoschizomusand Agastoschizomus possess spinose setae onthe tibia and some species also have spinose setae onthe patella. Protoschizomus gertschi also has spinosesetae on the femur. Megaschizomus mossambicus,Trithyreus, and Schizomus ashmolei have spinosesetae on all segments. Most protoschizomids (exceptA. patei, P. pachypalpus, P. rowlandi, P. sprousei,P. purificacion) have spinose setae on the patella.Because the resulting matrix does not follow anypattern observed with other characters, we assumethe difference between spinose setae and setae isminor.Anterior sternum: The number of setae is givenfor the anterior sternum, but this character is oflimited value even in species recognition. The presenceof only one sternapophysial seta occurs in A.patei. A single male of P. sprousei and the onlyknown specimen of P. treacyae also have a singlesternapophysial seta. All other species of Schizomidapossess two sternapophysial setae. The presenceof only one seta is therefore of considerable interestbut its phylogenetic significance is unknown since itoccurs in species apparently not closely related.Legs: The general increase in length and slendernessof the legs is almost certainly a troglomorphicadaptation. The epigean species (P. pachypalpusgroup) have legs I and IV shorter than the body,whereas all of the cave species have these legslonger than the body.Tarsus I: Agastoschizomus patei has the firstsegment of tarsus I longest, whereas it is approximatelyequal to or shorter than the last segment inall other Schizomida.Anterodorsal margin of femur IV: The anterodorsalmargin of femur IV slopes posteriorly in all Protoschizomidae,Megaschizominae, Trithyreus, andSchizomus ashmolei. In the remaining Hubbardiinaethe margin forms about a 90° angle.Length of femur IV (character 24): Femur IV isless than 4.8 times longer than deep in Protoschizomusand Hubbardiidae, but more slender in Agastoschizomus.This is probably related to the greaterspecialization to cavernicolous life by the latter species.The effects of the change in femur thicknessand associated musculature on the ability to jumphave not been investigated.Length of trochanter IV (character 25): TrochanterIV is about 1/2 the length of femur IV inMegaschizominae and Protoschizomus, and about1/3 the length in Agastoschizomus, Hubbardiinae,and Uropygi. We presume this condition arose separatelyin the two groups and therefore we have assignedseparate apomorphic states to those formswith longer trochanters.Mesopeltidial plates: The smaller species of Protoschizomidaeall have smaller plates, with the gapbetween them being greater. The increase in size ofthe plates and subsequent decrease in the gap appearsdirectly correlated with size and probablyprovides greater support to the larger species. It is,therefore, of limited value phylogenetically.The condition of the metapeltidium was originallyused to separate genera in the Hubbardiidae,with species having an entire plate being placed inSchizomus (and later Megaschizomus was added) andthose with a split plate being placed in Trithyreus.The value of this character for generic identificationwas questioned by Hansen and Sorensen (1905), althoughthey provisionally utilized it in separatingsubgenera. They noted that the metapeltidium in38

some species could be either split, entire, or onlypartially separated. Most later authors have rejectedit as a character of generic value. It is, however, ofvalue when used in conjunction with other charactersin delineating phylogenetic lineages. The stateof the metapeltidium appears consistent within atleast some groups of species. In the Protoschizomidaethe metapeltidium is divided in Protoschizomusand A. lucifer; it is entire in A. huitzmolotitlensis, A.srygius, and A. patei. The divided metapeltidium ofA. lucifer is considered to have been derived inAgastoschizomus.Setation of the abdominal tergites: This characteris fairly consistent in the family Hubbardiidae and isof some use in determining relationships. Megaschizominaehave a submarginal row of four anterior setaewhich is not found in other Hubbardiidae. Protoschizomidaefrequently have extra anterior setaebut these do not form a distinct row and are presumablynot homologous to the setae of Megaschizominae.The Hubbardiidae most frequently have two largeposterior setae on segments I-VII, with four setae onsegments VIII-IX. In some species of Hubbardiinaethe lateral setae on segment VIII are missing. Inother species there are multiple setae on one or moretergites. There is some variation in number of setae,but usually the extra setae are either unpaired orminute; where there is a question it is assumed thatthe typical number is the normal state. As alreadymentioned, the presence of microsetae on tergitesI-II is a synapomorphy for the order. Some"Schizomus" have lost some or all of the microsetae.When microsetae are present on tergite I, there arethree closely spaced (in a row) on each side of tergiteII. A derived condition occurs in A. patei wherethere are still three microsetae per side but they occuras a triad (one centered in front of pairs of microsetae).Tergite I contains two posterior setae inall protoschizomid species except A. srygius inwhich the setae are missing. Tergite II always hastwo setae in the Protoschizomidae as do some Hubbardiidae.Tergite III has four setae in the P. pachypalpusgroup, but two setae in all other protoschizomidspecies (character 26). This character state isunresolved on the ordinal level because members ofthe Hubbardiidae have two, four, and more thanfour setae. Tergites IV-VII have four setae in the P.pachypalpus group but vary in other groups andgenera, with both character states being present insome specIes.Setation of the abdominal sternites: In the familyProtoschizomidae there are always two rows ofsubmarginal setae on sternites IV-VIII and sometimeson IX in the females. In the males of the P.pachypalpus group the setae are scattered or at mostform two close-set highly irregular rows near theposterior margin of the sternites. In the only speciesof the P. sprousei group for which males are known,the anterior row of setae is situated near the middleof the sternite rather than near the anterior margin.Most species of Hubbardiinae have two irregularrows in both sexes. The apomorphic condition(character 27) is found in Agastoschizomus males,which have two distinct rows of setae.Width/length ratio of sternites IV-VII: There is ageneral decrease in the width/length ratio with increasedbody size. Even so, this character appears tobe useful in separating Protoschizomus from Agastoschizomus.The width/length ratio of sternite VI(character 28) is below 2.3 for Protoschizomus. Thisremains true even for the species of Agastoschizomusthat are smaller than the largest Protoschizomus.The ratio is greater than 2.3 in Agastoschizomusand the Hubbardiidae.Abdomen: In some species of Hubbardiinae(Hubbardia Cook, Stenochrus Chamberlin, and"Schizomus"), some segments of the abdomen maybe extremely elongate. This elongation does not occurin the Megaschizominae or the Protoschizomidae.Abdominal muscles (character 29): The Protoschizomidaeshare with the Uropygi and Amblypygithe presence of eight dorsoventral muscles.The Hubbardiidae have lost the posteriormost pair.All protoschizomids, except P. rowlandi, have4-6 ventral setae on segment X of the abdomen. Thederived condition, in P. rowlandi, is the occurrenceof two setae.Abdominal segment XII of protoschizomids generallyhas two dorsal setae. That of A. srygius arequite heavy and more spinose. The setae on male A.huitzmolotitlensis are very long and are extendedover the flagellum.The flagellum of some male Hubbardiidae(Hubbardia, "Schizomus", and Megaschizominae) isapparently (observation of preserved material only)carried with the posterior end up or arched forwardtowards the anterior end of the animal. Most Hubbardiinaeand protoschizomids apparently do nothold the flagellum up. The IX stemite and tergite ofsome Hubbardiidae are modified so that the flagellumcan be held up. These modifications include ashorter tergite IX than sternite IX length and archedand reduced ventral stemite.Female flagellum: The flagellum is divided byannuli or rings in most schizomids. The annuli canoccur as either thinnings in the cuticle or as breaks39

in the cuticle. As already mentioned in the methodssection, we consider segments to be subdivisionsseparated by breaks in the cuticle and subdivisionsseparated by a thinning of the cuticle as articles.This results in the use of opposite terms used bymost previous authors. They apparently did not examinethese annuli with sufficient magnification.Segments occur in all major groups of schizomids,whereas articles are apparently found only inMegaschizominae and Protoschizomidae. The lossof articles in the Hubbardiinae is considered a synapomorphy(character 30). The complete absence ofall annuli in A. patei is an autapomorphy. While thereverse (development of articles is a synapomorphyfor Megaschizominae and Protoschizomidae) is moreparsimonious, not requiring a reversal, it is notconsistent with other characters considered to beapomorphic. Treating the absence of articles asapomorphic resulted in a longer tree length (55rather than 54) when examined using Hennig86.The maximal number of annuli (five) occurs inthe Megaschizominae. Lawrence (1958) reportedthat the female flagellum of Megaschizomus consistedof three segments, but his illustration(redrawn in Fig. lc) showed annuli at five positions.From his description it appears he considered annuliat positions 4 and 5 to be true segmental divisions.Our examination of a female Megaschizomus suggestsjust the opposite; positions 1-3 are segmentaldivisions and positions 4 and 5 are only zones ofthinned cuticle. The flagellum we examined was incomplete,missing the latter two positions. When wetried to position the specimen for examination, thebasal segments broke apart cleanly, indicating thatthey were indeed segmental junctions. As alreadynoted, no true articles are recorded for the Hubbardiinae.The number of subdivisions in the Protoschizomidaeranges from 0-5.An annulus (segmental break) in position I isconsidered plesiomorphic as it occurs in both theHubbardiidae and Protoschizomidae. The loss ofthis annulus in A. srygius is an autapomorphy. Anannulus (segmental break) occurs at position 2 inonly members of the Hubbardiidae (both Megaschizominaeand Hubbardiinae). This annulus was apparentlysecondarily lost in some Hubbardiinae(character 31). An annulus at position 3 (segmentalbreak) is known from all family level groups ofschizomids. As it is also missing from numerous unrelatedtaxa in the Hubbardiinae and Protoschizomidaepresence/absence of an annulus at position 3 willlikely be useful only in separating congeneric species.An annulus at position 3 is found in A. lucifer,P. pachypalpus, P. purificacion, P. rowlandi, andP. sprousei. An annulus (thinning in cuticle inMegaschizominae and Protoschizomidae) is presentat position 4 in both families of schizomids and itsabsence in some Hubbardiinae is considered to be areversal. An annulus (thinning of the cuticle) ispresent at position 5 in the Megaschizominae andProtoschizomidae. It is apparently absent from theHubbardiinae and P. gertschi and is therefore consideredto have been lost on at least two occasions.As noted by Harvey (in press), the flagellar setationof the Hubbardiidae appears to be unique.While the same setae are present in the Hubbardiinaeas are present in the Megaschizominae and Protoschizomidae,the absence of setae resulting in thepattern dml, dm4, dB, vml, vrn2, vm4, vIl, vmS,vl2 is a synapomorphy for the Hubbardiinae.Harvey also noted that male and female flagella havethe same number of setae and that the setae occur atapproximately the same positions in the Hubbardiinae.This is not the case in the Protoschizomidaeand is considered to be a derived state (character32). Among female protoschizomids, drn2 is lackingfrom all species except P. gertschi and A. srygius(character 33). Because these species are apparentlyunrelated (based on other characters) and becausedrn2 do not occur in the Hubbardiidae, we suggestthese setae developed independently. Seta drn2 isalso lacking in penultimate males of P. sprousei andP. purificacion as well as adult males of P.sprousei, P. pachypalpus, P. rowlandi, and P. occidentalis(character 34). Several other setae areknown to be absent from male protoschizomid flagella:vm4 from P. pachypalpus and P. rowlandi(character 35); drn3 from A. patei; vmS from A.huitzmolotitlensis [note that vmS may also be missingfrom male A. patei - we have not labeled it inour drawing but suggest it is one of the off- centeredpairs numbered dl2, the second dl2 seta being absent(aberrant individual). Discovery of additional maleA. patei should resolve this matter.]. Only femalesof P. sprousei and P. purificacion lack dm4(character 36).Male flagellum: The male flagellum of protoschizomidsprovides the best characters for distinguishingspecies. It also appears to be useful in delineatinggeneric limits. In Protoschizomus the flagellumis distinctly enlarged distally (character 37), whereasin Agastoschizomus it does not increase in width distally.Since most species of Hubbardiidae also haveapically enlarged flagella it is likely that this is theplesiomorphic state.The male flagellum of protoschizomids (exceptA. patei) and Megaschizominae has soft, sometimeseversible areas. The males with these soft areas also40

have small pores over the surface of the flagellum(especially distolaterally). Because the flagellum isgrasped by the female during mating we suggestthese soft areas might be squeezed by the female,possibly causing an exocrine secretion to be emittedthrough the pores. Such a system is known in Opiliones(cheliceral glands in the Ischyropsalidoidea),where the male produces a material that is eaten bythe female during courtship and mating. The apparentabsence of flagellar glands or sacs in the Hubbardiidaesuggest this is a synapomorphy for theother schizomids. Because this is inconsistent withother known apomorphies for the order, we suggestthese glands are retained in some Hubbardiidae.Rowland and Lawrence (both keen observers) studiedprotoschizomids and Megaschizominae but failedto note the soft regions or pores. With this in mindit is easy to understand why these structures couldbe missed on the much smaller Hubbardiinae. Thepublished drawings of many Hubbardiinae revealnumerous lobes and surface depressions which couldbe the soft regions of the flagellum. It is for thisreason we are considering the presence of thesecharacters to be plesiomorphic. The loss of soft regionsand surface pores in A. patei and some Hubbardiinaeare considered derived.The male flagellum of protoschizomids(excluding A. patei) and Megaschizominae havenumerous "spicules." Although it is suggestive ofsome form of stridulation we are unable to locateopposing sound producing structures. Because thesurface of the female flagellum is imbricate, it ispossible these spicules are the remnants of the tipsof the imbricate "scales." The male flagellum of A.patei is like that of juveniles and females. It lackssegmentation, eversible lobes, and pores, and thesurface integument is imbricate. Because all otherschizomids show marked adult and sexual dimorphismin the flagellum, we suggest A. patei is paedomorphic.The absence of ventrolateral lobes also serves toseparate A. patei from Protoschizomus and otherspecies of Agastoschizomus. Because these lobes arenot present in the Hubbardiidae, we assume theywere secondarily lost in A. patei. Among the specieswith ventral lobes, all (except P. sprousei) have thetips of the lobes curved inward. The tips of P.sprousei curve outward.The flagellum of the Protoschizomidae is withouta distinct ventrally inserted stalk (character 38),whereas in the Hubbardiidae the stalk is distinct andespecially so in Megaschizominae where it gives theappearance of comprising a separate segment. Themale stalk is homologous to the first three segments(two in protoschizomids) in the female and juvenileflagella, as determined by setation. The maleflagellum of Protoschizomus and most Hubbardiidaeis less than three times as long as wide (character39), whereas it exceeds three times inAgastoschizomus.The distodorsal end of the male flagellum islaterally compressed in P. pachypalpus and P.rowlandi. This area is rounded in otherprotoschizomids (character 40).Microsetae are often present on the ventralsurface of male flagella which also have ventrallobes. Such setae are missing from A.huitzmolotitlensis, placed on the ventral lobes in P.sprousei, or placed near the base of the lobes in A.lucifer, P. pachypalpus, P. occidentalis, and P.rowlandi. This latter condition is consideredplesiomorphic (character 41). Because no specieswithout lobes have microsetae, we conclude themicrosetae were independently lost In A.huitzmolotitlensis.In P. sprousei males, the apex of the flagellum istriangular in shape. The apex in otherprotoschizomids is rounded.In P. pachypalpus, the basal half of the maleflagellum is strongly constricted and this character isconsidered an autapomorphy. Other protoschizomidshave the base gradually or not constricted. The basalstalk of the flagellum in male P. sprousei is verybroad, whereas the stalk is narrow in all otherprotoschizomids.The male flagellum of P. occidentalis has twopairs of slit-sensilla lateral to the dm1 seta. Thiscondition is considered an autapomorphy.Some of the setae (dm3, dm4, d12, d13, v12) onthe male flagellum of A. patei are enlarged and havepilose borders on the distal half. This is consideredan autapomorphy.Female genitalia: Brignoli (1973) first noted thevalue of the sperrnathecae to the taxonomy of theHubbardiinae. Rowland (1973c) indicated that thesperrnathecae were of limited value in distinguishingspecies due to interspecific variation. Rowland andReddell (1979a, 1979b, 1980, 1981) described andillustrated the spermathecae of all New Worldspecies of the subfamily Hubbardiinae for whichfemales were available. Reddell and Cokendolpher(1985) illustrated the spermathecae of the Africansubfamily Megaschizominae and the eastern AsiaTrithyreus and Cokendolpher (1988) the species ofHubbardiinae from Japan and Taiwan. Reddell andCokendolpher (1991) illustrated the only truemember of Schizomus and Harvey (in press)illustrated the spermathecae of five new genera of41

other Hubbardiinae. The spermathecae of no speciesof the Protoschizomidae have been up to nowdescribed or illustrated.The spermathecae of the subfamily Hubbardiinaeconsist of one to multiple pairs of lobes, some ofwhich are simple and not sclerotized, whereas othersterminate in enlarged sclerotized bulbs (receptacula).The spermathecae of the Megaschizominae consistof a single pair of sclerotized lobes with the basaltwo-thirds being somewhat rugose and thespermathecae of Trithyreus have one pair of doubledlobes without pits (Reddell and Cokendolpher,1985). The spermathecae of the Protoschizomidaecomprise a single pair of lobes which are oftenshallowly pitted over most of their surface (character42). The spermathecae of the Uropygi (Weygoldt,1971, and pers. obs.) are quite different from thoseof the Schizomida and are not easily compared. Thespermathecae of many primitive spiders andAmblypygi consist of two lobes which are pitted andappear very similar to those of theProtoschizomidae. While of some value in speciesrecognition, the spermathecae do not appear to behelpful in distinguishing protoschizomid genera. Insome species the spermathecae are distinctlyenlarged distally. The spermathecae of the onlyknown female of P. treacyae differ; one lobe isenlarged, the other simple.The margins of the receptaculum are relativelysmooth, without numerous pits in A. lucifer, P.sprousei, P. gertschi, P. purificacion, and P.treacyae. The margins are saw-toothed with manypits in A. stygius; lobed with numerous pits in A.patei, and smooth with numerous pits in P.pachypalpus group members (character 43).The spermathecae are widened basally or withenlarged receptaculum in all protoschizomids exceptA. stygius. The gradually narrowed (towards thereceptaculum) spermathecae are considered anautapomorphy for A. stygius.A gonopod has been discovered in someHubbardiinae. It is apparently absent from theremaining Hubbardiidae and Protoschizomidae.CladogramAt first glance, one might question why we havepresented a c1adogram. While it is true many of therelationships and characters are unresolved, we feltour data could serve as a starting point. This is thefirst attempt to apply cladistics to any taxa withinthe order. It is hoped our efforts will stimulateothers to examine characters more closely. It isimmediately evident that the unresolved areascontain species in which one sex is unknown.Unlike many other groups of animals, one cannotsimply re-collect at the type locality to obtain theneeded material. In almost all cases the typelocalities are caves (some very deep andinaccessible) and specimens are scarce. Additionalmaterial, if obtained, may help resolve thec1adogram further.The analysis resulted in 117 equally parsimonioussolutions, each 54 steps long and with no characterconflicts (i.e., the consistency and retention indicesare both 1.00). The strict consensus of those 117c1adograms is presented in Fig. 2.Based on our cladistic analysis, the orderSchizomida is easily divided into two families:Protoschizomidae and Hubbardiidae. TheProtoschizomidae are likewise split into two clades:Protoschizomus and Agastoschizomus. We can easilydistinguish four species in Agastoschizomus: Alucifer, A. huitzmolotitlensis, A. stygius, and A.patei.The genus Protoschizomus can be separated intothe P. pachypalpus group, the P. sprousei group,and two unplaced species. As with Agastoschizomus,the missing data from species known from singlesexes is the major problem.DISTRIBUTIONThe family Protoschizomidae is known only fromthe eastern part of the Sierra Madre Oriental in thestates of Hidalgo, San Luis PotOSI, and Tamaulipasand the state of Colima in Mexico; and the southernedge of the Edwards Plateau in Texas, U.S.A. (Fig.3). The genus Protoschizomus includes threeepigean species, one each in Colima, San LuisPotosI, and Tamaulipas and four cavernicole speciesin the state of Tamaulipas. Agastoschizomusincludes four cavernicole species in the states ofHidalgo, San Luis PotOSI, and Tamaulipas. A singleimmature specimen not placed in a genus isrestricted to a single cave in Val Verde County,Texas. Each species is extremely limited in itsdistribution, with most known only from onelocality. Protoschizomus pachypalpus has beencollected only within a few kilometers of G6mezFarias, Tamaulipas. Agastoschizomus lucifer occursin three caves within 15 kilometers of each other.Agastoschizomus patei is definitely known (fromadult specimens) only from one cave, but may occurin two other caves within a few kilometers of thetype-locality.The three epigean species of Protoschizomus arewidely separated. Protoschizomus pachypalpus42

inhabits tropical deciduous forest along the easternslopes of the Sierra de Guatemala in Tamaulipas; P.rowlandi occurs in the coastal plain east of CiudadValles in San Luis Potosi; P. occidentalis is the onlyspecies of Protoschizomidae known fromsouthwestern Mexico and the Pacific drainage.The cavernicole species of Protoschizomus appearto be less troglomorphic than the species ofAgastoschizomus, but are known only from caves.Protoschizomus gertschi inhabits a single cave nearMiquihuana in the western Sierra Madre Oriental. Itis separated from the species of the Purificaci6nregion by the Rio Guayalejo. The occurrence of fourspecies of Protoschizomidae in the Purificaci6nregion is remarkable. Agastoschizomus patd is ahighly troglomorphic species and is probablyrepresentative of a very early invasion of the cavesof this region. Agastoschizomus sp. cf. patei (knownonly from immature specimens) is sympatric inCueva del Tecolote with P. sprousei. The threespecies of Protoschizomus in the Purificaci6n regionare each known only from a single cave, but shouldbe found in other nearby caves with additionalcollecting. Protoschizomus sprousei occurs only inCueva del Tecolote near the village of Los SanPedro at an elevation of 1,450 m. The remainingtwo species, P. purijicacion and P. treacyae occurin Cueva X at an elevation of 1,950 m and in Cuevadel Borrego at an elevation of 1,980 m,respectively; both of these caves are located near thevillage of Conrado Castillo. The occurrence of threespecies of the same genus in the caves of this area isparalleled by the carabid beetle genusMexaphaenops Bolfvar y Pieltain with four speciesin the same area. Mexaphaenops jebriculosus Barr isfound at lower elevations only a few kilometersfrom Cueva del Tecolote. Three species, M.jamesoni Barr, M. mackenziei Barr and M.sulcifrons Barr inhabit caves in the Conrado Castilloarea. As with Mexaphaenops, the presence of threespecies of Protoschizomus in the same area mayrepresent different times of invasion of thecavernicole habitat. Protoschizomus purijicacionexhibits a lesser degree of troglobitic modification\j~19, 35,40, 43P. pnc/'ypnlplls group19,21,26,43vP. sprnlfse; group22,36Agnsrosc/'izofllfls17,24,27,34,39PrOlOschizomrlS25,2R,37Ilubbardiidae14-16,23,29,42Pro(oschizomidae1R, 20,31,32,38'--__---,,--Schizolllida0-13CaillarostomataFig. 2.-Cladogram of selected members of the superorder Camarostomata. See text for explanation and Tables 1 and 2 forcharacter list (autapomorphies have been excluded) and presumed polarities.43

than P. treacyae and may be a more recent caveimmigrant.The four species of Agastoschizomus are isolatedfrom each other. Agastoschizomus lucifer inhabitsthe Sierra de EI Abra, a low mountain range alongthe eastern slopes of the Sierra Madre Oriental. TheRio Tamuin cuts through the cavernous limestoneand thus isolates this species from A.huitzmolotitlensis which inhabits a higher elevationin the Sierra Madre Oriental near Xilitla. To thesouth of this region the Rio Moctezuma cuts throughthe cavernous limestone and thus isolates A.lOS'105'102'99'9S'30'30'27'27'24'24'İ:)•21'105'102'99'Fig. 3.-Distribution of Protoschizomidae. I, Genus and species undetermined (immature only); 2, PrOioschizomus purijicacion, P.rreacyae; 3, Agasroschizomus parei, P. sprousei; 4, P. genschi; 5, P. pachypalpus; 6, A. lucifer; 7, P. rowlandi; 8, A. huirzmolorirlensis;9, A. srygius; 10, P. occidenralis. Heavy outlines are rivers. G = RIO Guayalejo; M = RIO Moctezuma; T = RIO Tamuln.44

huitvnolotitlensis from A. stygius to the south inHidalgo. Agastoschizomus patei is isolated from A.lucifer by the Rio Guayalejo.The undescribed species from Texas is a tropicalrelict now isolated in the cavernicole environmentby the surrounding semi-arid surface. Other tropicalrelicts in Central Texas caves include among othersthe pseudoscorpion Leucohya texana Muchmore,two species of the opilionid genus HoplobunusBanks, the amblypygid Phrynus n. sp., and thenicoletiid silverfish Texoreddellia texe/lSis (Ulrich).FAMILY PROTOSCHIZOMIDAE ROWLANDMegaschizominae: Rowland, 1973c: 136 (part).Protoschizomidae Rowland, 1975b:I-2; Rowland,1975a:25, 27,339; Rowland and Reddell, 1979a:161, 162, 166, 174; Rowland and Reddell,1980:1; Cokendolpher, 1981:6; Reddell,1981:65, 124, 125; Rowland and Reddell,1981:19; Levi, 1982:76; Elliott, 1984a:15;E[lliott], 1984b:8; Reddell and Cokendolpher,1984:5; Reddell and Cokendolpher, 1985:48;Elliott and Reddell, 1985:214.Diagnosis.-Medium to large, 4.0 to 12.4 mmtotal length excluding flagellum. Anterior process ofpropeltidium generally downturned, with one or twosetae and with pair of setae at base of process.Mesopeltidia large, gap between plates 0.1 to 0.6anterior width of one plate. Eight pairs ofdorsoventral abdominal muscles. Female flagellumwith or without segments and articles. Sternite VI2.0 to 4.8 times wider than long. Male flagellumwith or without retractable lateral lobes. Femalespermathecae consisting of one pair of single lobeswith shallow pits over most of their surfaces.Cheliceral serrula not composed of hyaline teeth, butrepresented by a row of blunt, nearly hemisphericalknobs or stout teeth; brush (type 5 setae) at base offixed digit absent; fixed digit with two teeth; ratioof pedipalpal claw length to dorsal length ofbasitarsus-tarsus 0.6: 1 to 1.3: 1; basitarsal spurssymmetrical, long, about 0.3 to 0.7 dorsal length ofbasitarsus-tarsus. Femur IV 3.0 to 8.25 times aslong as deep.Type-genus.-Protoschizomus Rowland, 1975b(by original designation).Distribution.-MEXICO: Tamaulipas, San LuisPotosi, Hidalgo, Colima. U.S.A.: Texas.Included genera.-Agastoschizomus Rowland,1971; Protoschizomus Rowland, 1975b.Protoschizomus RowlandProtoschizomus Rowland, 1975b:2; Rowland,1975a: 25, 27, 334, 337, 339, 343-346, 378,figs. 286-287 (manuscript name); Rowland andReddell, 1979a:166; Levi, 1982:76; Elliott andReddell, 1985:214.Type-species.-Agastoschizomus pachypalpusRowland, 1973a (by original designation).Diagnosis.-Epigean and cavernicole species.Medium to large species, 4.0 to 8.3 rom total lengthexcluding flagellum. Propeltidium with two setae onanterior process. Anterior process down-turnedapically. Anterior sternum with one or twosternapophysial setae. Gap between mesopeltidialplates 0.3 to 0.6 anterior width of one plate.Sternites V-VIII of male with two rows of setae orwith scattered setae. Sternite VI 3.1 to 4.8 timeswider than long; width/length ratio versus bodylength, 0.8 to 2.3. Female flagellum with segmentsand articles. Male flagellum enlarged distally, withretractable ventrolateral lobes. Female pedipalp inproportion to body length 0.7 to 1.0 pedipalp lengthof male. Pedipalpal spur about 0.3, claw about 0.6to 0.8 dorsal length of basitarsus-tarsus. Femur IV3.0 to 4.5 times longer than deep.Distribution.-MEXICO: Tamaulipas, San LuisPotosi, Colima.Included species.-P. pachypalpus (Rowland),P. rowlandi n. sp., P. occidentalis Rowland, P.sprousei n. sp., P. gertschi n. sp., P. purificacionn. sp., P. treacyae n. sp.pachypalpus groupDiagnosis.-Epigean. Medium sized species, 4.0to 4.6 mm total length. Propeltidium with three orfour pairs dorsal setae. Anterior sternum with twosternapophysial setae. Males with scattered or twovery irregular rows of setae on sternites IV-IX.Pedipalp claw about 0.6 to 0.8 dorsal length ofbasitarsus-tarsus. Pedipalps of females in proportionto body length 0.6 to 0.7 pedipalp length of male.Legs I and IV shorter than body length.Distribution.-MEXICO: Tamaulipas, San LuisPotosi, Colima.Included species.-P. pachypalpus (Rowland),P. rowlandi n.sp., P. occidentalis Rowland.Protoschizomus pachypalpus (Rowland)Figs. 2-5, 22-23, 38-45Agastoschizomus pachypalpus Rowland,1973a:6,45

Key to Species and Genera of Protoschizomidaela. Male flagellum without ventrolateral lobes (Fig. 97); female flagellumwithout segments or articles (Fig. 102) (Tamaulipas)A. pateilb. Male flagellum with ventrolateral lobes; female flagellum segmented 22a. Anterior process of propeltidium with one seta; male flagellum notdistally enlarged Agastoschizomus 32b. Anterior process of propeltidium with two setae (Fig. Ib); maleflagellum distally enlarged Protoschizomus 53a. Metapeltidium divided (San Luis PotosI) A. lucifer3b. Metapeltidium undivided 44a. Femur IV 4.8 times longer than wide; abdominal tergite I with onepairs oflarge posterior setae (San Luis Potosl}4b. Femur IV 6.0 times longer than wide; abdominal tergite I withoutlarge posterior setae (Hidalgo)A. huitzmolotitlensisA. stygius5a. Propeltidium with three or four pairs dorsal setae (Fig. lb); malestemites IV-VIII with scattered or two irregular rows of setae P. pachypalpus group 65b. Propeltidium with two pairs dorsal setae; male sternites IV-VIII withtwo regular rows ofsetae 86a. Male flagellum broadly joined at base (Fig. 52), vm4 seta present(Fig. 27); sperrnathecae robust (Fig. 57) (Colima)P. occidentalis6b. Male flagellum less broadly joined at base, vm4 seta missing;sperrnathecae long and slender. 77a. Propeltidium with three pairs setae; male flagellum 1.7 times as wideas long, strongly constricted basally (Fig. 39) (Tamaulipas)7b. Propeltidium with four pairs setae; male flagellum 2.0 times as wideas long, not strongly constricted basally (Fig. 46) (San Luis Potosl}P. pachypalpusP. rowlandi8a. Female flagellum with dm4 seta missing; pedipalp trochanter slightlyproduced P. sprousei group 98b. Female flagellum with dm4 seta present; pedipalp trochanter notproduced 109a. Female flagellum with segment/article V present (Fig. 9); sperrnathecaelong, slender, not enlarged apically (Fig. 64) (Cueva del Tecolote, Tamaulipas)...... ..P. sprouse;9b. Female flagellum with segment/article V absent (Fig. 73); sperrnathecaeshort, enlarged apically (Fig. 76) (Cueva X, Tamaulipas)P. purificacionlOa. Female flagellum with all setae present (Figs. 10-11), segment/articleV missing (Fig. 10); femur IV 4.2 times as long as deep (San Luis Potosl}lOb. Female flagellum with drn2 seta missing (Fig. 14); segment/article Vpresent (Fig. 14); femur IV 3.7 times as long as deep (Cuevadel Borrego, Tamaulipas)P. gertschiP. treacyae46

8-10, figs. 5-7; Rowland, 1973c:136; Rowland,1975a:27, 40; Rowland, 1975b:2.Protoschizomus pachypalpus: Rowland, 1975a:14-15, 27, 40-42, 46-49, 167-168, 211, 213,map 3, figs. 12, 15, 18-19 (part-all recordsexcept 51.5 mi. (82.9 km) E Ciudad Victoria onhighway 70); Rowland, 1975b:2-6, fig. 2 (partallrecords except 51.5 mi. (82.9 km) E CiudadVictoria); Rowland and Reddell, 1979a:162,167-169, 174, figs. 2, 5-7 (part-all recordsexcept 51.5 mi. (82.9 km) E Ciudad Victoria);Reddell, 1981: 124.Material examined.-MEXICO: Tamaulipas:Nacimiento del RIO Frio, 4.8 km S of GomezFarias, 140 m elev., 12 March 1969 (J. Reddell),female holotype and female paratype (AMNH);Arroyo Nacimiento del RIO Frio, 140 m elev., 16Feb. 1970 (R.W. Mitchell), 2 female, 1 immatureparatypes (AMNH); Gomez Farias, 300 m elev., 6Jan. 1964 (J. Reddell, D. McKenzie, L. Manire), 1female paratype (AMNH); 10 km SE of GomezFarias, 4 km W of Highway 85, 21 May 1982 (N.Strenth), 1 male (AMNH), 1 male, 1 female, 1immature (TMM); 21 May 1982 (H.L.McCutchen), 1 female (TMM); 24 May 1982 (H.L.McCutchen), 1 immature (TMM); 96.5 km S ofCiudad Victoria, 17 Nov. 1948 (H.B. Leech), exbromeliad, 2 females, 1 immature (CAS). Vialcontaining immature paratype from type localityactually contains 1 immature and 1 female labeled"paratypes." Vial from Arroyo Nacimiento del RIOFrio contains only 1 female paratype and is labeled"paratype. "Diagnosis.-This species is most closely relatedto P. rowlandi but the male flagellum is more robustwith a more distinct basal stalk. Protoschizomuspachypalpus possesses three pairs of setae on thepropeltidium, whereas P. rowlandi has four pairs.Description.-Male (length from distal margin ofpropeltidium to base of flagellum, 4.26 mm). Lightbrown, cephalothorax and pedipalps slightly darker.4 s 6 7 8o 0 d13dl4dm4" , dl2dm3dlldmlvm2o 0v12vmSvll. vm4vm3, vmldlldl4dl2vm2viivm4vm3vmldl4 0 0 dl3,dl2dlldmldm39101112 13vl2vmSviivm4vm3vm2vml.' 0dm4dl3dl4dl2dm3I..:dtm.: ..n zCJdllv12vmS0 0vllvm3vmlvm4vm20dl3dl4v12vmSvll0 dl2 vm4dm3dllvm30 vm2dmlvmlFigs. 4-13-Setation and annuli placement on flagella of female protoschizomids: Proroschizomus pachypalpus: 4, dorsal aspect; 5,ventral aspect. P. rowlandi: 6, dorsal aspect; 7, ventral aspect. P. sprousei: 8, dorsal aspect; 9, ventral aspect. P. gercschi: 10, dorsalaspect; II, ventral aspect. P. purificacion: 12, dorsal aspect; 13, ventral aspect.47

Cephalothorax: Propeltidium 1.12 mm long, 0.68mm wide; anterior process with row of two setaeand with one pair of setae at base; with two pairsdorsolateral setae (anterionnost slightly moredorsad) about 1/3 from anterior margin and one pairdorsal setae about 1/4 from posterior margin. Gapbetween mesopeltidial plates about 0.6 anteriorwidth of one plate. Metapeltidium separated bydistinct membranous strip. Anterior sternum with 10setae, plus two sternapophysial setae; posteriorsternum with four setae.Abdomen: Tergite I with two pairs anteriormicrosetae (in row) and one pair large posteriorsetae; tergite II with three pairs anterior microsetae14 15 16 17dI2vm5dl3dI4dI2dm4dm3o 0 vm4. dm3dllvm30dII.0vm2dmi.vmlvmI0• vI2vm5• vIIvm4vm3vm218 19 20 21dI40dI3vI2dI3vI20 vm5odI4 vm5dI20vIIdm4 dm4 vm40vII 00dl2dm3'.::0dm3 vm4 dII00dm2o dII vm3 dmi0dmivmIvm2Figs. 14-2J.-Setation and annuli placement on flagella of female protoschizomids: Protoschizomus lreacyae: 14, dorsal aspect; 15,ventral aspect. Agastoschizomus lucifer: 16, dorsal aspect; 17, ventral aspect. A. stygius: 18, dorsal aspect; 19, ventral aspect.Agasloschizomus pale;: 20, dorsal aspect; 21, ventral aspect.48

(in row) and one pair large posterior setae; tergitesIII-VII with one pair dorsal and one pair dorsolateralsetae each; tergite VIII with six setae along posteriormargin and four dorsolateral setae near middle oftergite; tergite IX with two dorsolateral and twolateral setae. Stemite VI about 4.0 times as wide aslong; width/length ratio versus body length, 1.1.Stemites V-VIII with two irregular rows of setaenear posterior margin, IX with scattered setae.Segments X-XI telescoped. Segment X with one pairlateral and four ventral setae. Segment XI with onepair lateral and seven ventral setae. Segment XIIwith one pair dorsal, one pair dorsolateral, one pairlateral, and eight ventral setae. Flagellum (Figs.22-23, 38-39) 0.46 mm long, 0.26 mm wide; withshort base, expanding to club about as wide as longand enlarged distally; with pairs of shortunsclerotized ventral lobes near lateral margins ofclub; dm2 and vm4 setae absent.Pedipalps (Fig. 40): Trochanter not produceddistally. Patella with two spinose setae onventrolateral margin. Tibia with three spinose setaeon ventrolateral margin. Spur about 0.3, claw 0.8dorsal length of basitarsus-tarsus.Chelicerae: Serrula with nine teeth. Setae: 1=3;2=2; 3=10; 4=1; 5=0; 6=1; all strongly piloseexcept seta 6 pointed.Legs: Leg I, including coxa, 4.62 mm long;basitarsal-tarsal proportions: 15:3:4:3:4:3: 13. FemurIV (Fig. 41) about 3.0 times as long as deep.Female holotype (length from distal edge ofpropeltidium to base of flagellum, 4.42 mm).Propeltidium 1.02 mm long, 0.66 mm wide. As inmale except as follows: Stemites V-IX with twodistinct submarginal rows of setae. Stemite VI about3.8 times wider than long; width/length ratio versusbody length, 1.2. Flagellum (Figs. 4-5, 42) 0.42mm long; dm2 seta missing; segments/articles I,III-V present. Spermathecae (Figs. 43-44): one pairlong, slender, distinctly curved lobes only veryslightly increasing in width apically. Pedipalp lengthin proportion to body length about 0.7 as long as2223dl3vI2vmSodl2242526 2728 29dl3vl2o 0d12 o d14°0dm4d130d12vm4vm3vm2vmldmlvm3 0vmlvm2Figs. 22-29.-Setation on flagella of male protoschizomids: Protoschizomus pachypalpus: 22, dorsal aspect; 23, ventral aspect. P.rowlandi: 24, dorsal aspect; 25, ventral aspect. P. occidentalis: 26, dorsal aspect; 27, ventral aspect. P. sprousei: 28, dorsal aspect; 29,ventral aspect.49

male pedipalp; claw about 0.6 dorsal length ofbasitarsus-tarsus. Leg I, including coxa, 3.94 nunlong; basitarsal-tarsal proportions: 9:3:3:3:3:3: 12.Femur IV about 3.4 times as long as deep.Variation: Tergite VIII in one specimen possessedthree anterior and six posterior setae, whereasanother specimen possessed five anterior and sevenposterior setae. Rowland (l973a) stated that theholotype female possessed five pairs of dorsalpropeltidial setae; one of these pairs is doubtless thatat the base of the anterior process, but are-examination of the holotype and other specimensstudied by Rowland reveals only three pairs ofdorsal propeltidial setae. He also stated that theanterior sternum possessed 11 setae. We could findonly ten on all specimens studied, including theholotype. Finally, we have been unable to verify thesetation of the abdominal tergites. The holotype hasapparently cleared somewhat since collected and it ispossible that we have missed some broken setae.Measurements (mm).-Male (female holotype):Pedipalp: trochanter 0.10 (0.10); femur 0.64 (0.48);patella 0.58 (0.42); tibia 0.54 (0.40); basitarsustarsus0.28 (0.22); total 2.14 (1.62). Leg I:trochanter 0.32 (0.28); femur 0.94 (0.82); patella1.02 (0.88); tibia 0.94 (0.78); basitarsus 0.3430 31 32 33ym4dm4dm3dl3• dl2dII dm2 dIIyrn2dmlyl2ym4'o 0ym3yrnl34 35f\.d14dm40 dl3dl2dm3dm2·36 37dm4dl4vl2viI •dm3 0 dl3 dI2. vm4·vi!. vm3. drn2 . vm3'vm4dll vm2 dml vmIdmlvmIdIlvrn2Figs. 30-37.-Selation on flagella of male protoschizomids: Protoschizomus purificacion: 30, dorsal aspect; 31, ventral aspect.Agastoschizomus lucifer: 32, dorsal aspect; 33, ventral aspect. A. huitzmolotitlensis: 34, dorsal aspect; 35, ventral aspect. Agastoschizomuspatei: 36, dorsal aspect; 37, ventral aspect.50

(0.34); tarsus 0.64 (0.58); total 4.20 (3.68). Leg II:trochanter 0.16 (0.18); femur 0.76 (0.72); patella0.44 (0.42); tibia 0.48 (0.44); basitarsus 0.36(0.38); tarsus 0.34 (0.36); total 2.54 (2.50). LegIII: trochanter 0.20 (0.22); femur 0.74 (0.70);patella 0.36 (0.36); tibia 0.40 (0.36); basitarsus0.42 (0.52); tarsus 0.36 (0.36); total 2.48 (2.52).Leg IV: trochanter 0.56 (0.54); femur 0.94 (0.82);patella 0.50(0.48); tibia 0.72 (0.64); basitarsus 0.54(0.52); tarsus 0.42 (0.38); total 3.68 (3.38).Habitat.-This species has been collected fromthe undersides of rocks on steep hillsides and alonga roadcut at the type-locality. It has also beenreported from bromeliads. Vegetation in the GomezFarias region is classified as tropical deciduousforest and is well-described by Martin (1958).Protoschizomus rowlandi new speciesFigs. Ib, 2-3, 6-7, 24-25, 46-51Agastoschizomus pachypalpus Rowland, 1973a:6,8-10, figs. 5-7; Rowland, 1975a:27, 40 (part);Rowland, 1975b:2 (part).Protoschizomus pachypalpus: Rowland, 1975b:2-6,39tt)v ..~~ \\-j'- (~41--:::::::::::=========~444543Figs. 38-45.-Protoschizomus pachypalpus: 38, male flagellum, lateral aspect; 39, male flagellum, dorsal aspect; 40, malepedipalp, lateral aspect; 41, male femur IV, lateral aspect; 42, female flagellum, lateral aspect; 43-44, female spermathecae (43 fromNacimiento del RIO Frio, 44 from 10 km SE Gomez Farias); 45, detail of spermathecal lobe. Scale lines = 0.25 mm for Figs. 38-42,0.1 mm for Figs. 43-44, and 0.01 mrn for Fig. 45.51

fig. 2 (part-51.5 mi. (82.9 km) E CiudadVictoria on highway 70 record only); Rowland,1975a: 27, 40-42, 46-59, 167-168, 211, 213,map 3, fig. 15 (part-51.5 mi. (82.9 km) ECiudad Victoria on highway 70 record only);Rowland and Reddell, 1979a:162, 167-169, 174,figs. 2,7 (part-51.5 mi. (82.9 km) E CiudadVictoria on highway 70 record only).Material examined.-MEXICO: San LuisPotosi: 51.5 mi. (82.9 km) E of Ciudad Valles onHighway 70, 17 Oct. 1972 (B. Firstman, V. Roth),male holotype, female paratype (AMNH).Etymology.- This species is named for J. MarkRowland in recognition of his many contributions tothe study of schizomids.Diagnosis.-This species is distinguished fromP. pachypalpus by the shape of the male flagellum,which is more slender and has a less distinct stalkthan in P. rowlandi. There are four pairs of setae onthe propeltidium in P. rowlandi versus three pairs inP. pachypalpus.Description.-Male holotype (length from distaledge of propeltidium to base of flagellum, 4.64mm); dark orangish red (probably artificiallydarkened due to preservation and storage).Cephalothorax: Propeltidium 1.16 mm long, 0.70mm wide; with row of two setae on anterior processand one off-centered pair setae at base of process;with one pair small widely spaced setae about 1/5from distal edge, two pairs (second pair tiny) setae2/5 from distal edge, and one pair closely spacedsetae about 1/4 from posterior margin. Gap betweenmesopeltidial plates 0.4 anterior width of one plate.Metapeltidium distinctly separated by membranoussuture. Anterior sternum with 10 setae, plus twosternapophysial setae; posterior sternum with foursetae.464851, -!========Figs. 46-51.-Proloschizomus rowlandi: 46, male flagellum, dorsal aspect; 47, male flagellum, lateral aspect; 48, male flagellum,ventral aspect; 49, male pedipalp, lateral aspect; 50, female flagellum, lateral aspect; 51, female spermathecae. Scale lines = 0.25 mmfor Figs. 46-49, 0.1 mm for Figs. 50-51.52

Abdomen: Tergite I with two pairs anteriormicrosetae (in row) and one pair large posteriorsetae; tergite II with three pairs anterior microsetae(in row) and one pair large posterior setae; tergitesIII-VII with one (?) pair dorsal and one (?) pairdorsolateral setae; tergite VIII with two rows of foursetae; tergite IX with one pair dorsolateral and onepair lateral setae. Stemites V-IX with scatteredsetae. Stemite VI about 3.1 times as wide as long;width/length ratio versus body length, 1.5.Segments X-XI telescoped; segments X-XI with onepair dorsolateral, one pair ventrolateral, and onepair ventral setae each; segment XII with one pairdorsal, one pair dorsolateral, one pair lateral, andten (?) ventral setae. Flagellum (Figs. 24-25, 46-48)0.56 mm long, 0.28 mm wide; with graduallywidening stalk ending in prominent bulb; dorsalsurface extending over pair of unsclerotizedventrolateral lobes; setae dm2 and vm4 missing; viisetae are on the inner face of the ventral lobes;unnumbered unpaired setae present laterad to dm4.Pedipalps (Fig. 49): Trochanter not produceddistally. Patella with two ventrolateral spinose setae,distalmost very long. Tibia with four strongventrolateral spinose setae. Spur about 0.3, clawabout 0.7 dorsal length of basitarsus-tarsus.Chelicerae: Serrula with eight teeth, increasing inlength basally. Setae: 1=3; 2=2; 3= 11; 4=2;5=0; 6= 1.Legs: Leg I, including coxa, 4.70 mm long;basitarsal-tarsal proportions: II:3:4:3:3:4: 14.Femur IV about 3.25 times as long as deep.Female paratype (length from distal edge ofpropeltidium to base of flagellum, 4.20 mm); lighttan. Propeltidium 1.06 mm long, 0.58 mm wide.Gap between mesopeltidial plates 0.5 anterior widthof one plate. Tergites as in male except: tergitesIII-VI with one pair dorsal and one pair dorsolateralsetae each; tergite VII with one pair dorsal, one pairdorsolateral, and one pair lateral setae; tergite VIIIwith one pair dorsal, one pair dorsolateral, and onepair lateral setae near posterior margin and two pairsdorsolateral setae near anterior margin. StemitesV-IX with two distinct submarginal rows of setae.Stemite VI about 3.1 times as wide as long;width/length ratio versus body length, 1.4.Flagellum (Figs. 6-7, 50) 0.46 mm long; dm2 setamIssmg; segments/articles I, III-V present.Spermathecae (Fig. 51) of one pair of elongate,curved lobes only slightly increasing in sizeapically. Pedipalp length in proportion to bodylength about 0.7 length of male pedipalp; claw about0.8 dorsal length of basitarsus-tarsus. Leg I,including coxa, 3.96 mm long; basitarsal-tarsalproportions: 9:2:3:3:3:3: 12. Femur IV about 3.0times as long as deep.Measurements (mm).-Male holotype (female):Pedipalp: trochanter 0.14 (0.14); femur 0.74 (0.54);patella 0.86 (0.44); tibia 0.70 (0.44); basitarsustarsus0.32 (0.22); total 2.76 (1.78). Leg I:trochanter 0.32 (0.26); femur 1.02 (0.86); patella1.06 (0.90); tibia 0.94 (0.78); basitarsus 0.26(0.22); tarsus 0.66 (0.54); total 4.26 (3.56). Leg II:trochanter 0.14 (0.16); femur 0.80 (0.72); patella0.48 (0.44); tibia 0.46 (0.44); basitarsus 0.36(0.32); tarsus 0.36 (0.38); total 2.60 (2.46). LegIII: trochanter 0.24 (0.20); femur 0.76 (0.70);patella 0.40 (0.34); tibia 0.40 (0.32); basitarsus0.46 (0.40); tarsus 0.42 (0.40); total 2.68 (2.36).Leg IV: trochanter 0.56 (0.48); femur 1.04 (0.92);patella 0.56 (0.48); tibia 0.74 (0.64); basitarsus0.68 (0.56); tarsus 0.48 (0.44); total 4.06 (3.52).Comments.-Previous reports of this specieshave been given as "E Ciudad Victoria," but thecorrect locality is "E Ciudad Valles."Protoschizomus occidentalis RowlandFigs. 2-3, 26-27, 52-58Protoschizomus occidentalis Rowland, 1975b:3-6,fig. 1; Rowland, 1975a:27, 41-43, 46-59,167-168, map 3, fig. 14; Rowland and Reddell,1979a:162, 167-169, 174, figs. I, 7; Reddell,1981: 124.Material examined.-MEXICO: Colima: 20.9km SW Colima, 16 July 1972 (A. Jung), maleholotype, female paratype (AMNH).Diagnosis.-This species has a wider basal stalkin the male flagellum than its closest relatives, P.pachypaLpus and P. rowLandi. Females may beseparated by the shorter spermathecal lobes in P.occidentaLis than in P. pachypaLpus. Males may alsobe distinguished from both P. pachypaLpus and P.rowlandi by retaining seta vm4, which is lost inboth of those species.Description.- Male holotype (length from distaledge of propeltidium to base of flagellum, 4.00mm); light orangish tan.Cephalothorax: Propeltidium 1.04 mm long, 0.64mm wide; with row of two setae on anterior processand one pair setae at base of process; with one pairof widely spaced dorsal setae about 2/5 from distaledge and one pair closely spaced setae about 1/4from posterior margin; third pairs of tiny setaeslightly posterior to larger anterior pair. Gapbetween mesopeltidial plates 0.5 anterior width ofone plate. Metapeltidium divided by narrow suture.53

Anterior sternum with 11 setae plus twosternapophysial setae; posterior sternum with foursetae.Abdomen: Tergite I with two pairs anteriormicrosetae (in row) and one pair large posteriorsetae; tergite II with three pairs anterior microsetae(in row) and one pair large posterior setae; tergitesIII-VII with one pair dorsal and one pair dorsolateralsetae each; tergite VIII with one pair dorsal, onepair dorsolateral, and one pair lateral setae; tergiteIX with one pair dorsolateral and one pair lateralsetae. Sternites V-VIII with two irregular rows of551565857---Figs. 52-58.-Proloschizomus occidentalis: 52, male flagellum, dorsal aspect; 53, male flagellum, lateral aspect; 54, male flagellum,ventral aspect; 55, male pedipalp, lateral aspect; 56, female pedipalp, lateral aspect; 57, female spermathecae; 58, detail ofspermathecallobe. Scale lines = 0.25 mm for Figs. 52-56, 0.05 mm for Fig. 57, 0.01 mm for Fig. 58.54

setae near posterior margin. stemite IX with oneirregular row of setae. Stemite VI about 4.8 times aswide as long; width/length ratio versus body length.0.8. Segments X-XI telescoped; segments X-XI withone pair dorsolateral. one pair ventrolateral. andfive ventral setae each; segment XII with one pairdorsal. one pair dorsolateral. one pair lateral. andnine ventral setae. Flagellum (Figs. 26-27. 52-54)0.52 mm long, 0.24 mm wide; gradually expandeddistally with triangular apex projecting over pair ofunsclerotized ventrolateral lobes; drn2 seta missing.Pedipalps (Fig. 55): Trochanter not produceddistally. Patella with two short and one long spinosesetae. Tibia with four ventrolateral spinose setae.Spur about 0.3. claw about 0.6 dorsal length ofbasitarsus-tarsus.Chelicerae: Serrula with seven long pointedteeth. Setae: 1=3; 2=6; 3=12; 4=3; 5=0: 6=1.all but seta 6 pilose.Legs: Leg I, including coxa, 4.64 mm long;basitarsal-tarsal proportions: 9:3:4:3:3:3: 13. FemurIV about 3.0 times as long as deep.Female (length from distal edge of propeltidiumto base of flagellum. 4.04 mm). Propeltidium 1.00mm long, 0.60 mm wide. Tergite setation as inmale, except one additional lateral pair each on VIand VII. Sternites V-IX with two distinct rows ofsubmarginal setae. Sternite VI about 4.8 times aswide as long; width/length ratio versus body length.0.8. Flagellum missing. Spermathecae (Figs. 57-58)of one pair of short curved lobes not increasing insize apically. Pedipalp (Fig. 56) length inproportion to body length about 0.7 male pedipalplength. Leg I, including coxa, 3.84 mm long;basitarsal-tarsal proportions: 12:3:4:3:3:3: 10.Femur IV about 2.7 times as long as deep.Measurements (mm).-Male holotype (femaleparatype): Pedipalp: trochanter 0.14 (0.14); femur0.78 (0.50); patella 0.82 (0.46); tibia 0.72 (0.42);basitarsus-tarsus 0.32 (0.22); total 2.78 (1. 74). LegI: trochanter 0.34 (0.28); femur 1.00 (0.78); patella1.06 (0.82); tibia 0.98 (0.78); basitarsus 0.22(0.28); tarsus 0.62 (0.52); total 4.22 (3.46). Leg II:trochanter 0.14 (0.16); femur 0.82 (0.66); patella0.44 (0.38); tibia 0.46 (0.38); basitarsus 0.36(0.28); tarsus 0.34 (0.34); total 2.56 (2.20). LegIII: trochanter 0.20 (0.20); femur 0.74 (0.62);patella 0.38 (0.34); tibia 0.36 (0.32); basitarsus0.42 (0.36); tarsus 0.36 (0.32); total 2.46 (2.16).Leg IV: trochanter 0.54 (0.46); femur 1.04 (0.82);patella 0.54 (0.46); tibia 0.70 (0.60); basitarsus (0.48); tarsus - (0.40); total - (3.22).Comment.-Rowland and Reddell (I979a)reported that the paratype female was immature;examination of the genitalia has shown it to bemature.sprousei groupDiagnosis.-Cavernicoles. Medium to largespecies, 5.2 to 8.2 mm total length not includingflagellum. Propeltidium with two pairs dorsal setae.Anterior sternum with one or two sternapophysialsetae. Sternites V-VIII of male with one distinct rowof setae near middle of sternite and one row nearposterior margin. Pedipalp claw about 0.9 to 1.0dorsal length of basitarsus-tarsus. Female pedipalpsequal in length to males. Legs I and IV longer thantotal body length.Distribution.-MEXICO: Tamaulipas.Included species.-P. sprousei n.sp.. Ppurificacion n.sp.Protoschizomus sprousei new speciesFigs. 2-3, 8-9, 28-29, 59-65New species of schizomid: P. Sprouse. 1985:39.Material examined.-MEXICO: Tamaulipas:Cueva del Tecolote. Los San Pedro, 1,450 m elev.•18 Nov. 1984 (P. Sprouse), male holotype(AMNH), female paratype (AMNH); 21 Nov. 1984(D. Pate), 1 immature (TMM); 21 Nov. 1984 (P.Sprouse), female paratype. 2 immature females(TMM); 24 Nov. 1985 (P. Sprouse), 2 male, 1female paratypes. 5 immatures (TMM); 26 Nov.1985 (P. Sprouse), 3 immatures (TMM); 22-28Nov. 1986 (P. Sprouse), 1 immature (TMM);March 1990 (P. Sprouse), 3 immatures (TMM).Etymology.-This species is named for Peter S.Sprouse in recognition of his outstanding work instudying the caves and cave fauna of Mexico.Diagnosis.-Proroschizomus sprousei hassegments/articles I and III-V in the femaleflagellum. The broad basal stalk, more lateralposition of the retractable lobes which curveoutward, and distinctly triangular apex of the maleflagellum separates P. sprousei from other species ofProtoschizomus for which males are known.Description.-Male holotype (length from distaledge of propeltidium to base of flagellum, 6.12mm). Body, pedipalps. and chelicerae orangishbrown; legs light tan.Cephalothorax: Propeltidium 1.80 mm long. 0.94mm wide; with two setae in a row on anteriorprocess, one pair of closely set setae at base ofanterior process, one pair of widely spaced longsetae about 4/9 from distal margin, and one pair55

closely set short setae about 1/3 from posteriormargin. Gap between mesopeltidial plates 0.3anterior length of one plate. Metapeltidiumdistinctly divided. Anterior sternum with six setaeplus two sternapophysial setae; posterior sternumwith four setae.Abdomen: Tergite I well-sclerotized, with twopairs anterior microsetae (in row) setae and one pairlarge posterior setae; tergite II with three pairsanterior microsetae (in row) and one pair largeposterior setae; tergite III with one pair dorsal setae;tergites IV-VII with one pair large dorsal setae andone pair small dorsolateral setae each; tergite VIIIwith one pair large dorsal setae, one smalldorsolateral seta on left side, and one pair lateralsetae; tergite IX with one pair dorsolateral and onepair lateral setae. Segment X with four ventral andtwo ventrolateral setae; segment XI with five ventraland two ventrolateral setae; segment XII with twovery strong dorsal setae, two smaller dorsolateral60596164-------------------Figs. 59-65.-ProlOschizomlls sprollsei: 59, male flagellum, dorsal aspect; 60, male flagellum, lateral aspect; 61, male flagellum,ventral aspect; 62, male pedipalp, lateral aspect; 63, female flagellum, lateral aspect; 64, female spermathecae; 65, detail ofspermathecallobe. Scale lines = 0.25 mm for Figs. 59-63, 0.1 mm for Fig. 64, 0.01 mm for Fig. 65.56

setae, and ten ventral setae. Stemites V-VIII withtwo distinct rows of setae, one row along posteriormargins and one row near middle of stemites, IXwith anterior row irregular. Stemite VI about 3.6times as wide as long; width/length ratio versusbody length, \.7. Flagellum (Figs. 28-29, 59-61)0.62 rnm long, 0.34 mm wide; widest near middlewith pointed apex and pair of lightly sclerotizedventrolateral lobes projecting posterolaterally fromnear apex; dm2 seta missing.Pedipalps (Fig. 62): Trochanter slightly produceddistally. Patella with three spinose setae onventrolateral surface. Tibia with five spinose setaeon right pedipalp and four spinose setae on leftpedipalp (which increase in length and widthdistally) on ventrolateral margin. Spur about 0.5 andclaw about equal to dorsal length ofbasitarsus-tarsus.Chelicerae: Serrula with eight rounded teeth.Setae: 1=3; 2=3; 3=10; 4=2; 5=0; 6=1. Setae1-3 pilose; 4, 6 pointed.Legs: Leg I, including coxa, about 10.30 mmlong; basitarsal-tarsal segment proportions:17:6:6:7:7:6:22. Femur IV about 4.5 times as longas deep.Female paratype: As in male except as follows:Length from distal margin of propeltidium to baseof flagellum, 8.26 mm. Color reddish orange.Propeltidium 2.26 mm long, 1.10 mm wide. TergiteIV with one pair dorsal setae; tergite VIII with onepair dorsal, one pair dorsolateral, and one pairlateral setae near posterior margin and one pairdorsolateral setae near middle. Setation of segmentsX-XII also varies slightly. Stemites V-IX with twodistinct submarginal rows of setae. Stemite VI about3.6 times as wide as long; length/width ratio versusbody length, 2.3. Flagellum (Figs. 8-9, 63) 0.88mm long; dm2 and dm4 setae IDJssmg;segments/articles I, III-V present. Spermathecae(Figs. 64-65) of one pair of curved elongate lobesonly slightly enlarged apically. Chelicerae serrulawith eight rounded teeth; setae: 1=3; 2=7; 3 = 16;4 = 2; 5 = 0; 6 = 1. All setae except 6 long andpilose. Pedipalp equal in length to male pedipalp.Leg I, including coxa, 10.98 mm long;basitarsal-tarsal proportions: 20:7:7:7:8:7:22.Femur IV about 3.8 times as long as deep.Variation: An examination of four additionaladults revealed the following differences from theholotype: two specimens had only one dorsolateralseta on tergite IV; two specimens had one pairdorsal and one pair lateral setae; one specimen hadone pair dorsal, one pair dorsolateral, and one pairlateral setae on tergite VIII; one specimen had onepair dorsal, one pair dorsolateral, and one pairlateral setae on tergite IX. One specimen had sevensetae and one had five setae on the anterior sternum.One specimen had six and one had four spinose setaeon the pedipalp tibia. The flagellum of two juvenilesconsisted of two segments and was enlarged basally(annuli at positions 4 and 5). Because of thereduction in segmentation and enlargement, wepresume them to be penultimate males. Earlier instarjuveniles (presumably of both sexes) have nosegments in the flagellum. The setation of thejuvenile males is the same as the female except dm4is present.Measurements (mm).-Male holotype (femaleparatype): Pedipalp trochanter 0.22 (0.36); femur1.00 (1.30); patella 0.88 (1.18); tibia 0.88 (1.20);basitarsus-tarsus 0.40 (0.56); total 3.38 (4.60). LegI: trochanter 0.74 (0.76); femur 2.24 (2.56); patella2.90 (2.88); tibia 2.20 (2.24); basitarsus 0.38(0.40); tarsus 1.10 (1.20); total 9.56 (10.04). LegII: trochanter 0.54 (0.44); femur 1.70 (2.04); patella0.88 (\.12); tibia 1.20 (\.38); basitarsus 0.84(0.98); tarsus 0.78 (0.82); total 5.94 (6.78). LegIII: trochanter 0.44 (0.40); femur 1.64 (1.98);patella 0.86 (1.04); tibia 0.98 (1.20); basitarsus0.96 (1.12); tarsus 0.82 (0.88); total 5.70 (6.62).Leg IV: trochanter 1.00 (1.10); femur 2.26 (2.30);patella \.06 (1.26); tibia \.48 (1.90); basitarsus1.28 (1.46); tarsus 0.82 (0.98); total 7.90 (9.00).Habitat.-This species is relatively abundant inCueva del Tecolote. The cave is, with the exceptionof the Sistema Purificaci6n, the most extensive inthe Purificaci6n region. More than 11 kilometers ofpassage have been surveyed in the cave and it is inexcess of 230 m deep. During the rainy season thecave receives an enormous amount of floodwaterand is heavily polluted by sewage from the town ofLos San Pedro (Sprouse et aI., 1987).Protoschizomus purificacion new speciesFigs. 2-3, 12-13,30-31,72-77Material examined.-MEXICO: Tamaulipas:Cueva X, Conrado Castillo, 1,950 m elev., 27 Dec.1986 (P. Sprouse), female holotype (AMNH); 15April 1980 (D. Pate), immature male paratype(TMM).Etymology.-The species name is based on thePurificaci6n region in which area Cueva X islocated; it is used as a noun in apposition.Diagnosis.-Protoschizomus purificacion is distinguishedfrom all other species of Protoschizomusby the segmentation of the female flagellum, wheresegments/articles I and III-IV are present.57

one pair dorsolateral, and one pair lateral setae;tergite IX with one pair dorsolateral and one pairlateral setae. Stemites V-IX with two distinctsubmarginal rows of setae. Stemite VI about 3.8times wider than long; width/length ratio versusbody length, 1.6. Segments X-XI telescoped;segment X with six ventral and ventrolateral setae;segment XI with seven ventral and ventrolateralsetae; segment XII with two dorsal, twodorsolateral, and 10 ventral setae. Flagellum (Figs.10-11, 66) 0.56 mm long; with complete set ofsetae; segments/articles I and IV present.Fig. 71.-Protoschizomus gertschi: a, female spermathecallobe; b-c, details of lobe microtubules; d, detail of microtubulebase.Spermathecae (Figs. 69-71) of one pair lobes, withslender stalks and rounded enlarged apical bulbs.Pedipalps (Fig. 67): Trochanter not produceddistally. Femur with six or seven strong spinosesetae decreasing in size distally. Patella with fourstrong spinose setae along ventrolateral margin andthree or four long pilose strong spinose setae onmesal margin. Tibia with four strong spinose setaeand two long setae along ventrolateral margin. Spurabout 0.5, claw about 0.9 dorsal length ofbasitarsus-tarsus.Chelicerae: Serrula with nine teeth. Setae: 1=3;2=4; 3= 11; 4=2; 5=0; 6= 1. All denticlesrounded and many setae broken, suggesting this isan older adult.Legs: Leg I, including coxa, about 7.90 mmlong; basitarsal-tarsal segment proportions:14-3-5-5-4-4-17. Femur IV (Fig. 68) about 4.2times as long as deep.Male unknown.Immature paratype: As in holotype except thereis a pair of dorsolateral setae slightly anterior tomiddle of the propeltidium; dorsolateral seta onright side of abdominal tergite IV missing; only onepair of dorsal setae on tergite V; and only one pairof dorsal and one pair lateral setae on tergite VIII.Measurements (mm).-Female holotype:Pedipalp: trochanter 0.44; femur 1.80; patella 0.80;tibia 0.74; basitarsus-tarsus 0.38; total 4.16. Leg I:trochanter 0.50; femur 1.88; patella 1.92; tibia1.56; basitarsus 0.38; tarsus 1.02; total 7.26. LegII: trochanter 0.36; femur 1.52; patella 0.88; tibia0.94; basitarsus 0.74; tarsus 0.68; total 5.12. LegIII: trochanter 0.44; femur 1.50; patella 0.78; tibia0.86; basitarsus 0.92; tarsus 0.68; total 5.18. LegIV: trochanter O. 84; femur 1.78; patella O. 86; tibia1.38; basitarsus 1.16; tarsus 0.82; total 6.84.Habitat.-Sotano de Riachuelo is a 60 m deepcave which receives some floodwater (Mothes,1982). The cave is also home to two remarkabletroglobitic carabid beetles, Miquihuana rhadiniformisBarr and Paratrechus laticeps Barr.Comments.-This species, although smaller andless-attenuated than A. lucifer, A. stygius, and A.huitzmolotitlensis, is probably restricted to cave life.The absence of the left anterior seta on thepropeltidium is probably an abnormal condition,since this seta is present in the paratype.Protoschizomus treacyae new speciesFigs. 2-3, 14-15, 78-81Material examined.-MEXICO: Tamaulipas:Cueva del Borrego, 0.5 km S of Conrado Castillo,60

1,980 m elev., 26 Dec. 1986 (T. Treacy Sprouse),female holotype (AMNH).Etymology.-The species is named in honor ofTerri Treacy, who collected this and many otherinteresting cave animals.Diagnosis.-The species is readily identified bythe segmentation of the female flagellum wheresegments/articles I, IV, and V are present.Description.-Female holotype (length fromdistal edge of propeltidium to base of flagellum 6.36mm). Propeltidium, pedipalps, chelicerae, and legs Iorangish brown; abdomen and legs II-IV light tan.Cephalothorax: Propeltidium 1.72 mm long, 0.90mm wide; with two setae in a row on anteriorprocess and one pair of setae at base of anteriorprocess; with one pair of widely spaced setae about1/3 from distal margin, one small seta on right sideimmediately behind pairs of setae, and one pairclosely spaced setae about 1/4 from posteriormargin. Gap between mesopeltidial plates 0.57376Figs. 72-77.-Protoschizomus purificacion: 72, female pedipalp, lateral aspect; 73, female flagellum, lateral aspect; 74, penultimatemale flagellum, dorsal aspect; 75, penultimate male flagellum, ventral aspect; 76, female spermathecae; 77, detail of spermathecallobe.61

anterior width of one plate. Metapeltidium distinctlydivided. Anterior sternum with nine setae, plus onesternapophysial seta; posterior sternum with threesetae.Abdomen: Tergite I well-sclerotized, with twopairs anterior microsetae (in row) and one pair largeposterior setae; tergite II with two pairs anteriormicrosetae (in row) and one pair large posteriorsetae; tergites III-V with one pair dorsal setae each;tergite VI with one pair large dorsal setae and onesmall dorsolateral seta on left side; tergite VII withone pair dorsal setae; tergites VIII-IX with one pairdorsal and one pair lateral setae each. Sternites V-IXwith two distinct submarginal rows of setae. SterniteVI 3.6 times as wide as long; width/length ratioversus body length, 1.8. Segments X-XIItelescoped; segment X with two lateral and fourventral setae; segment XI with two lateral and sixventral setae; segment XII with two dorsal, twodorsolateral, two ventrolateral, and eight ventralsetae. Flagellum (Figs. 14-15, 78) 0.74 mm long;dm2 seta missing; segments/articles I, IV, and Vpresent. Spermathecae (Figs. 80-81) of one pairslightly curved lobes; left lobe slightly enlargedapically.Pedipalps (Fig. 79): Trochanter not produceddistally. Patella with three strong spinose setae onventrolateral margin. Tibia with four strong spinosesetae on ventrolateral margin. Spur about 0.5, clawabout equal to dorsal length of basitarsus-tarsus.Chelicerae: Serrula with nine rounded teeth;setae: 1=3; 2=6; 3= 13; 4=3; 5=0; 6= 1; all setaeexcept seta 6 pilose.Legs: Leg I, including coxa, 8.90 mm long;787980--..Figs. 78-81.-Protoschizomus treacyae: 78, female flagellum, lateral aspect; 79, female pedipalp, lateral aspect; 80, femalespermathecae; 81, detail ofspermathecallobe. Scale lines = 0.25 mm for Figs. 78-79, 0.1 for Fig. 80,0.01 mm for Fig. 81.62

asitarsal-tarsal proportions: 15:4:6:5:6:6:20. FemurIV about 3.7 times as long as deep.Male unknown.Measurements (mm).-Female holotype: Pedipalp:trochanter 0.20; femur 1.02; patella 0.92; tibia0.98; basitarsus-tarsus 0.48; total 3.60. Leg I: trochanter0.62; femur 2.30; patella 2.10; tibia 1.80;basitarsus 0.36; tarsus 0.96; total 8.14. Leg II: trochanter0.34; femur 1.64; patella 0.88; tibia 1.00;basitarsus 0.82; tarsus 0.74; total 5.42. Leg III: trochanter0.44; femur 1.56; patella 0.74; tibia 0.90;basitarsus 0.98; tarsus 0.78; total 5.40. Leg IV: trochanter0.84; femur 1.84; patella 0.92; tibia 1.40;basitarsus 1. 32; tarsus O. 86; total 7. 18.Habitat.-Cueva del Borrego is an extensivecave with 1,186 m of surveyed passage and a totaldepth of 58 m.?Protoschizomus sp.Material examined.-MEXICO: Tamaulipas:Cueva de Encino Magico, Los San Pedro, 1,465 melev., 27 Nov. 1986 (D. Pate), 1 immature(TMM).Comment.-This is probably an early instar orvery small species and is only tentatively placed inProtoschizomus.Genus Agastoschizomus RowlandSchizomus: Reddell, 1967:106; Reddell, 1971:28(part).Agastochizomus Rowland, 1971:13 (lapsus calami);Reddell and Mitchell, 1971 b: 1.Agastoschizomus Rowland, 1971: 13-14; Reddell andMitchell, 1971a: 145, 165; Reddell and Mitchell,1971 b: 1; Briggs and Hom, 1972: 1; Dumitresco,1973:282; Reddell, 1973:33, 38; Reddell andElliott, 1973:171; Rowland, 1973a:5-6, 8-10;Rowland, 1973b: 197,200-203; Rowland, 1973c:136; Brignoli, 1974:150; Rowland, 1975a:8-9,14-15, 27-28, 40, 43-50, 167-168, 181, 211,334, 338-339, 343-346, 378; Rowland,1975b: 1-2, 5, 8-10; Rowland and Reddell,1977:80-82, 85, 96; Rowland and Reddell,1979a: 162, 167-170, 174; Rowland and Reddell,1980:21; Anonymous, 1982:961.New genus of schizomid: Mitchell et al., 1977:56.Agostoschizomus: Levi, 1982:76 (lapsus calami).Diagnosis.-Cavernicoles. Large species, 7.00 to12.40 mm total length excluding flagellum. Propeltidiumwith one seta on anterior process. Anteriorprocess downturned or not apically. Anteriorsternum with one or two sternapophysial setae. Gapbetween mesopeltidial plates 0.1 to 0.3 anteriorwidth of one plate. Sternites V-VIII of male withtwo distinct rows of setae. Stemite VI 2.0 to 2.4times as wide as long; width/length ratio versusbody length 3.4 to 5.4. Female flagellum with orwithout segments and articles. Male flagellum notexpanded distally, with or without retractable ventrolaterallobes. Female pedipalp in proportion tobody length 0.9-1.2 pedipalp length of male. Pedipalpalspur about 0.5-0.7, claw about 1.1 to 1.3dorsal length of basitarsus-tarsus. Femur IV 4.8 to8.2 times longer than deep. Legs I and IV longerthan body length.Type-species.-Agastoschizomus lucifer Rowland,1971 (by monotypy and original designation).Agastoschizomus lucifer RowlandFigs. 2-3,16-17,32-33,82-87Agastoschizomus lucifer Rowland, 1971:14-17, figs.1-8; Reddell and Mitchell, 1971a:145, 165, figs.3-4; Dumitresco, 1973:282; Reddell, 1973:33,38; Reddell and Elliott, 1973: 171; Rowland,1973a: 10; Rowland, 1973b: 197, 200-202, figs.2, 4; Rowland, 1973c:136; Brignoli, 1974:150;Rowland, 1975a:8-9, 14-15, 27-28, 43-50,167-168, 181, 211, map 3, figs. 1, 8, 13, 16;Rowland, 1975b:8-1O, fig. 4; Rowland and Reddell,1977:80-82, 85, 96, fig. 1; Rowland andReddell, 1979a:162, 167-170, 174, figs. 3, 7;Rowland and Reddell, 1980:21; Reddell, 1981:16, 36, 65, 125-127, 320, 324, fig. 21; Anonymous,1982:961, unnumbered fig.Agastochizomus lucifer: Rowland, 1971: 13 (lapsuscalami).New genus of schizomid: Mitchell et al., 1977:56.Material examined.-MEXICO: San Luis PotOSI:Sotano de Matapalma, 21 km N of CiudadValles, 242 m elev., 29 May 1969 (R.W. Mitchell,F.E. Abernethy, T. Albert), 1 immature (AMNH);30 Dec. 1972 (R. Fieseler), 1 immature (AMNH);EI Sotano de la Tinaja, 10 Ian NNE of CiudadValles, 165.5 m elev., 9 April 1966 (J. Fish, D.McKenzie), male holotype, female paratype, 1 immature(AMNH); 18 Feb. 1970 (J.A.L. Cooke), 2males, 1 female, 4 immatures (AMNH); 29 May1974 (J. Prentice), 1 female (TMM); S6tano deYerbaniz, 22.5 km N of Ciudad Valles, 241.5 melev., 7 Jan. 1970 (colI. unknown), 1 female, 1 immature(TMM); 7 Jan. 1970 (S. Wiley), 1 female,Simmatures (TMM); 9 Jan. 1970 (W. Elliott, S.63

Wiley), 1 female (TMM); 17 Feb. 1970 (coli. unknown),1 female, 7 immatures (TMM); 28 March1970 (W. Elliott), female allotype, female paratype(AMNH); 8 Jan. 1971 (W. Elliott, J. Shepperd),male paratype (AMNH), 1 immature (TMM).Diagnosis.-This is the only species of Agastoschizomuswith a divided metapeltidium. Malesmay also be separated from males of A. huitzmolotitlensisby the shape of the male flagellum, which ismore slender and more pointed apically in A. huitzmolotitlensisthan in A. lucifer. The spermathecallobes of A. lucifer are long and slender with a slightapical enlargement, whereas in A. stygius the lobesare short and decrease in size apically. Females mayalso be distinguished from A. stygius by the presenceof segments/articles I and III-V in the femaleflagellum. Agastoschizomus stygius has only segments/articlesIV and V present. The presence ofventrolateral lobes on the male flagellum and thepresence of segments/articles on the female flagellumreadily separate this species from A. patei.Description.-Male holotype (length from distaledge of propeltidium to base of flagellum, 8.14mm). Color orangish.Cephalothorax: Propeltidium 2.24 mm long, 1.04mm wide; with one apical seta on anterior process8283848687IFigs. 82-87.-Agastoschizomus lucifer: 82, male flagellum, dorsal aspect; 83, male flagellum, lateral aspect; 84, male flagellum,ventral aspect; 85, male pedipalp, lateral aspeet; 86, female flagellum, lateral aspect; 87, female spermathecae. Scale lines = 0.25 mmfor Figs. 82-86, 0.05 for Fig. 87.64

and one pair setae at base of process; with one pairwidely spaced setae 5/11 from anterior margin andone pair closely spaced setae 3/11 from posteriormargin. Gap between mesopeltidial plates about 0.2anterior width of one plate. Metapeltidium dividedby narrow suture. Anterior sternum with nine setae,plus two sternapophysial setae. Posterior sternumwith three setae.Abdomen: Tergite I with two pairs anterior microsetae(in row) and one pair large posterior setae;tergite II with three pairs anterior microsetae (inrow) and one pair large posterior setae; tergitesIII-V with one pair dorsal setae each; tergite VI withone pair dorsal setae and one dorsolateral seta onright side; tergite VII with one pair dorsal and onepair dorsolateral setae; tergite VIII with one pairdorsal, one pair dorsolateral, and one pair lateral setae;tergite IX with one pair dorsal and one pair lateralsetae and one dorsolateral seta on left side.Segment X with one pair dorsal, one pair lateral,and four ventral setae; segment XI with one pairdorsolateral, one pair lateral, and six ventral setae;segment XII with one pair large dorsal, six smallerdorsal, and 10 ventral setae. Sternite VI about 2.4times as long as wide; width/length ratio versusbody length, 3.4. Sternites IV-IX with two distinctsubmarginal rows of setae. Flagellum (Figs. 32-33,82-84) 1.44 mrn long, 0.40 mrn wide, tubular withone pair short nonsclerotized retractable ventrolaterallobes; with complete set of setae.Pedipalps (Fig. 85): Trochanter not produced distally.Patella with three strong ventrolateral spinosesetae, increasing in length distally. Tibia with fivestrong ventrolateral spinose setae. Spur about 0.5,claw about 1.1 dorsal length of basitarsus-tarsus.Chelicerae: Serrula with eight teeth. Seta: 1=3;2=6; 3= 10; 4=2; 5=0; 6= 1.Legs: Leg I, including coxa, 14.82 mrn long;basitarsal-tarsal proportions: 31: 10:9:8: 11: 8:28.Femur IV about 5.3 times as long as deep.Female from Sotano de Yerbaniz: As in male exceptas follows. Length from distal edge of propeltidiumto base of flagellum, 12.40 mrn. Propeltidium3.02 mrn long, 1.46 mrn wide. Anteriorsternum with eight setae; posterior sternum withfour setae. Tergite VI with one pair dorsal setae andone dorsolateral seta on left side; tergite VIII withone extra dorsolateral seta on left side; tergite IXwith one pair dorsal, one pair lateral, and one pairsmall setae on lateral margins about midway alonglength of tergite. Sternite VI about 2.3 times as wideas long; width/length ratio versus body length, 5.4.Sternites V-IX with two distinct submarginal rowsof setae. Flagellum (Figs. 16-17, 86) 1.30 mrn long;dm2 seta missing; segments/articles I, III-V present.Sperrnathecae (Fig. 87) of one pair of gently curvedlobes with slightly enlarged apical bulb. Pedipalplength in proportion to body length about 0.9 malepedipalp length. Leg I, including coxa, 16.62 mrnlong; basitarsal-tarsal proportions: 33: 10: 10: 10:10: 10:30. Femur IV about 5.4 times as long asdeep.Variation: An examination of 12 adult or late instarspecimens reveals the following differencesfrom the holotype: Two specimens were found tohave only one pair of dorsal setae; a third specimenhad the anteriormost pair, but one seta was missingin the posteriormost pair. Three specimens had onlyone pair of setae on tergite VI. One specimen hadfour pairs of setae on tergite VIII, the extra pair beingvery small lateral setae distad to the posteriormargin; one specimen had one extra lateral seta distadto the posterior margin. One specimen had threepairs of setae on tergite IX, the extra pair being anteriorto the posterior margin and very small. Theanterior sternum included seven (one specimen),nine (one specimen), 10 (five specimens), and 11(five specimens) setae. One specimen had three setaeon the posterior sternum, seven specimens had foursetae, and four specimens had two setae. These variationscould not be correlated with sex, maturity/age,or locality.Measurements (mm).-Male holotype (female):Pedipalp: trochanter 0.34 (0.44); femur 1.42 (1.90);patella 1.30 (1.66); tibia 1.22 (1.78); basitarsustarsus0.72 (0.98); total 5.00 (6.76). Leg I:trochanter 0.70 (0.98); femur 3.64 (4.18); patella4.14 (4.54); tibia 3.38 (3.54); basitarsus 0.62(0.72); tarsus 1.50 (1.62); total 13.98 (15.58). LegII: trochanter 0.48 (0.68); femur 2.60 (3.08); patella1.40 (1.70); tibia 1.78 (2.00); basitarsus 1.30(1.56); tarsus 1.14 (1.38); total 8.70 (10.40). LegIII: trochanter 0.58 (0.82); femur 2.54 (2.86);patella 1.22 (1.44); tibia 1.70 (1.96); basitarsus1.48 (1.72); tarsus 1.22 (1.48); total 8.74 (10.28).Leg IV: trochanter 1.22 (1.64); femur 3.16 (3.78);patella 1.64 (1.82); tibia 2.68 (3.12); basitarsus1.82 (2.22); tarsus 1.34 (1.56); total 11.86 (14.14).Habitat.-The three caves from which thisspecies has been collected are located in the Sierrade EI Abra, a low mountain range extending fromsouth of Ciudad Valles north to Ciudad Mante. Allthree caves are extensive systems receiving massiveamounts of floodwater during the wet season(Mitchell et aI., 1977). Vegetation of the region ischaracterized as thorn forest. Specimens have beenfound actively roaming across silt banks in totaldarkness.65

Agastoschizomus huitzmolotitlensis RowlandFigs. la, 2-3, 34-35, 88-92Schizomus sp.: Reddell, 1967: 106; Reddell, 1971:28 (part-S6tano de Huitzmolotitla record only).Agastoschizomus huitzmolotitlensis Rowland,1975b:6, 8-10, fig. 3; Rowland, 1975a:28,44-50, 167-168, map 3, fig. 17; Rowland andReddell, 1977:80-82, fig. 1; Rowland and Reddell,1979a:162, 167, 169-170, 174, figs. 4, 7;Reddell, 1981:17,26, 35,65, 124-125, 320, fig.21.Material examined.-MEXICO: San Luis Potosi:S6tano de Huitzmolotitla, 1 km ESE of Tla-maya (=2 km NNW Xilitla), 600 m elev., 24 January1964 (T. Raines, T. Phillips), male holotype(AMNH).Diagnosis.-The more slender, apically pointedmale flagellum of this species serves to separate itfrom A. lucifer. The presence of a pair of setae onabdominal tergite I separates this species from A.stygius. The two species may also be separated bythe femur IV length/depth ratio (4.8 in A. huitzmolotitlensisversus 6.0 in A. stygius). The presenceof ventrolateral lobes on the male flagellum readilyseparates this species from A. patei.Description.-Male holotype (length from distaledge of propeltidium to base of flagellum, 7.0 mm);light orangish brown.888990Figs. 88-92.-AgaslOschizomus huilzmolotillensis: 88, male flagellum, dorsal aspect; 89, male flagellum, lateral aspect; 90, maleflagellum, ventral aspect; 91, male pedipalp, lateral aspect; 92, male femur IV, lateral aspect. Scale lines = 0.25 mm.66

Cephalothorax: Propeltidium 1.66 mm long, 0.90mm wide; with one seta on apical process and onepair of setae at base of process; with one pair ofdorsal setae located about 1/3 from anterior marginof propeltidium. Gap between mesopeltidial platesabout 0.2 anterior width of one plate. Metapeltidiumundivided. Anterior sternum with 11 setae; posteriorsternum with one seta, poorly sclerotized, triangular.Abdomen: Tergite I with two pairs anterior microsetae(in row) and one pair large posterior dorsalsetae; tergite II with three pairs anterior microsetae(in row) and one pair large posterior dorsal setae;tergites III-IV with one pair dorsal setae each; tergitesV-VI with one pair dorsal and one pair dorsolateralsetae each; tergite VII with one pair dorsaland one pair dorsolateral setae and one lateral setaon right side; tergite VIII with six setae near posterioredge of tergite and one pair dorsal and one pairdorsolateral setae near middle of tergite; tergite IXwith one pair long dorsal and one pair short lateralsetae near posterior margin, and one pair small dorsolateralsetae near middle of tergite. Sternites IVVIII with one row setae near middle of sternite andone row near posterior margin, IX with irregular anteriorrow. Sternite VI about 2.0 times as wide aslong; width/length ratio versus body length, 3.5.Segment X with one pair dorsolateral, one pair lateral,and row of six ventral setae. Segment XI withone pair lateral and row of seven ventral setae. SegmentXII with two very long posteriorly directeddorsal setae reaching about 2/3 length of flagellum,two shorter dorsolateral setae, and 11 ventral setae;longer than segments X-XI combined. Flagellum(Figs. 34-35, 88-90) 1.08 mm long, 0.26 mm wide;cylindrical, modified apically with pairs of retractableventrolateral lobes; vmS seta missing.Pedipalps (Fig. 91): Trochanter not produced distally.Patella with two strong ventrolateral spinosesetae, distalmost tapering to long pointed end. Tibiawith four strong ventrolateral spinose setae. Spurabout 0.5, claw about 1.2 times as long as dorsallength of basitarsus-tarsus.Chelicerae (Fig. la): Serrula with 7 teeth; setae:1=3; 2=1; 3=8; 4=2; 5=0; 6=1.Legs: Leg I, including coxa, about 10.94 mmlong; basitarsal-tarsal segment proportions:20:7:6:7:6:5:21. Femur IV (Fig. 92) about 4.8times as long as wide.Female unknown.Measurements (mm).-Male holotype: Pedipalp:trochanter 0.33; femur 1.02; patella 0.84; tibia0.96; basitarsus-tarsus 0.44; total 3.59. Leg I: trochanter0.56; femur 2.62; patella 3.06; tibia 2.48;basitarsus 0.46; tarsus 1.10; total 10.28. Leg II:trochanter 0.30; femur 2.22; patella 1.00; tibia1.28; basitarsus 0.90; tarsus 0.78; total 6.48. LegIII: trochanter 0.34; femur 2.20; patella 0.88; tibia1.26; basitarsus 1.08; tarsus 0.82; total 6.58. LegIV: trochanter 0.90; femur 2.32; patella 1.12; tibia1.90; basitarsus 1.40; tarsus 0.96; total 8.60.Habitat.-S6tano de Huitzmolotitla is an extensivecave which contains an active stream and receivesconsiderable floodwater during the wet season.The cave is 3,002 m long and 240 m deep(Russell and Raines, 1967). This species was collectedfrom the mud room near the back of the cave.Agastoschizomus stygius new speciesFigs. 2-3, 18-19,93-96Agastoschizomus sp.: Rowland and Reddell,1977:80, 82, fig. 1 (part-S6tano Hondo de Pinalitorecord only).Agastoschizomus n.sp.: Reddell, 1981:26, 124-125,fig. 21 (part).Material examined.-MEXICO: Hidalgo:S6tano Hondo de Pinalito, Pinalito (a village locatedat kilometer post 105 on highway 85 north ofJacala), 1,600 m elev., 1 Jan. 1976 (C. Soileau andP. Strickland), female holotype (AMNH), immatureparatype (TMM).Etymology.-The species name is from the Latinstygius, the lower world, referring to its cavernicolehabitat.Diagnosis.-The absence of the large posteriorsetae on abdominal tergite I separates this speciesfrom all other Agasroschizomus. The shape of thefemale spermathecae and femur IV length/width ratioalso distinguishes this species from other Agasroschizomus.This species also may be distinguishedfrom females of A. lucifer by possessing only segments/articlesIV and V in the female flagellum,rather than I and III-V. This species may be separatedfrom A. patei by the presence of segments/articlesin the female flagellum.Description.-Female holotype (length from distaledge of propeltidium to base of flagellum, 9.66mm). Abdomen and legs orangish brown; propeltidium,chelicerae, and pedipalps reddish brown.Cephalothorax: Propeltidium 2.50 mm long, 0.80mm wide; with one seta on anterior process and pairof setae at base of process and three dorsal setae(two close-set setae on right side, one seta on leftside). Gap between mesopeltidial plates about 0.3anterior width of one plate. Metapeltidium entire,with only a faint midline along the posterior margin67

of the plate; plate strongly depressed; greatest widthto length ratio about 1: 1. Anterior sternum withseven setae, plus two sternapophysial setae; posteriorsternum with three setae.Abdomen: Tergite I with two pairs anterior microsetae(in row), no setae on posterior margin;tergite II with two pairs anterior microsetae (in row)and one pair large posterior setae; tergites III-IVwith one pair dorsal setae each; tergites V-VII withtwo dorsal and two dorsolateral setae each; tergiteVIII with two dorsal, two dorsolateral, and two lateralsetae; tergite IX with two dorsal and two lateralsetae. Stemites V-IX with two distinct submarginalrows of setae. Sternite VI about 2.7 as wide as long;width/length ratio versus body length, 3.6. SegmentsX-XII telescoped; segment X with one pairlateral and five ventral setae; segment XI with onepair dorsolateral and six ventral setae; segment XIIwith one pair dorsal spinose setae, one pair smalland one pair large dorsolateral setae; one pair lateralsetae; and 10 ventral setae. Flagellum (Figs. 18-19,94) 1.44 mm long; with complete set of setae;~~---;d/ - ~ V~ \ -:::::±!fij, ~93articles IV and V present. Sperrnathecae (Figs.95-96) of one pair of straight lobes without apicalbulb and gently tapering from wider base towardsapex.Pedipalps (Fig. 93): Trochanter not produced distally.Patella with one strong spinose seta and onevery long whiplike seta on ventrolateral margin andwith six long mesal setae. Tibia with five strongspinose setae and one long whiplike seta on ventrolateralmargin and with five spinose setae tapering tolong pilose ends and irregular row of long whiplikepilose setae on ventromesal margin; one long pilosewhiplike seta ventrad to row of spinose setae. Spurabout 0.5, claw about 1.3 dorsal length ofbasitarsus-tarsus.Chelicerae: Serrula with 10 teeth; setae: 1=3,2=6,3=20,4=7,5=0,6=1; all setae except 6pilose.Legs: Leg I, including coxa, about 17.80 mmlong; basitarsal-tarsal segment proportions:26-7-7-7-7-25. Femur IV about 6.0 times as long asdeep...~" .., .. ,~...........•.......... ---,~.... ~~.:.~ .. - ~:. . .949695~-------Figs. 93-96.-Agastoschizomus srygius: 93, female pedipalp, lateral aspect; 94, female flagellum, lateral aspect; 95, femalespermathecae; 96, detail ofspermathecallobe. Scale lines = 0.25 mm for Figs. 93-94, 0.1 mm for Fig. 95, 0.01 mm for Fig. 96.68

Male adult unknown.Immature paratype male: As in holotype exceptthat there are only two dorsal setae on the propeltidium,the extra seta on the right side being absent;11 setae on anterior sternum; 4 setae onposterior sternum; chelicerae with 5 type 2 and 14type 3 setae; dm2 seta of flagellum missing.Measurements (mm}.-Female holotype: Pedipalp:trochanter 0.30; femur 1.98; patella 1.70; tibia1.90; basitarsus-tarsus 0.90; total 6.78. Leg I: trochanter0.98; femur 4.22; patella 5.36; tibia 3.92;basitarsus 0.60; tarsus 1.62; total 16.70. Leg II:trochanter 0.60; femur 3.18; patella 1.66; tibia2.28; basitarsus 1.62; tarsus 1.28; total 10.62. LegIII: trochanter 0.74; femur 3.00; patella 1.50; tibia2.34; basitarsus 1.84; tarsus 1.42; total 10.84. LegIV: trochanter 1.40; femur 3.82; patella 1.78; tibia3.28; basitarsus 2.42; tarsus 1.62; total 14.32.Habitat.-S6tano Hondo de Pinalito is a 91 mdeep, 198 m long cave (Bittinger, 1975; Sprouse,1982).Agastoschizomus patei new speciesFigs. 2-3, 20-21, 36-37,97-104New genus of schizomid: Pate et aI., 1987:91.Material examined.-MEXICO: Tamaulipas:Cueva de la Uorona, 3.5 km SSE Yerbabuena,1,860 m elev., 12-17 Oct. 1986 (P. Sprouse), maleholotype (AMNH), 2 female paratypes (TMM); 16Oct. 1985 (D. Pate), 1 female paratype (AMNH);13-17 Oct. 1985 (P. Sprouse), 1 immature (TMM);30 Dec. 1988 (A. Cobb), 1 female paratype(TMM); 15-17 March 1989 (P. Sprouse), 1 immature(TMM); 20 March 1989 (D. Pate), 2 immatures(TMM); March 1990 (P. Sprouse), 1 immature(TMM).Etymology.-This species is named in honor ofDale L. Pate in recognition of his outstanding collectionsof Mexican cave fauna.Diagnosis.-Agastoschizomus patei differs dramaticallyfrom other Agastoschizomus species: anteriorprocess not downturned apically; anterior sternumwith one sternapophysial seta. female and immatureflagella without segments or articles; maleflagellum without retractable ventrolateral lobes;and female pedipalp in proportion to body length1.2 times longer than male.Description.-Male holotype (length from distaledge of propeltidium to base of flagellum, 7.44mm). Body light orangish brown; abdomen tan.Cephalothorax: Propeltidium 1. 72 mm long; 1.00mm wide; with one seta on anterior process and onepair setae at base of anterior process; anterior processdrawn to point, not downturned. Propeltidiumwith one pair lateral setae about 1/3 from distalmargin, one seta on left in front of pair. Gap betweenmesopeltidial plates about 0.1 anterior widthof one plate. Metapeltidium undivided. Anteriorsternum with four setae plus one sternapophysialseta. Posterior sternum with four setae.Abdomen: Tergite I with row of two pairs (plusone extra seta on left side) anterior microsetae andone pair large dorsal setae on posterior margin; tergiteII with three pairs (plus one extra seta on leftside) anterior microsetae and one pair large dorsalposterior setae; tergites III-V with one pair dorsalsetae each; tergite VI with one pair large dorsal setaeand one small dorsolateral seta on left side; tergiteVII with one pair dorsal and one pair dorsolateralsetae, and one lateral seta on left side; tergiteVIII with one pair dorsal and one pair lateral setae;tergite IX with one pair lateral, and one dorsolateralseta on right side. Sternites III-IV with one rowsetae each; sternites V-IX with two distinctsubmarginal rows of setae. Sternite VI about twotimes as wide as long; width/length ratio versusbody length, 3.7. Segments X-XII not telescoped.Segment X with one pair lateral and six ventralsetae; segment XI with one pair lateral and sevenventral setae; segment XII with one pair dorsal, onepair dorsolateral, one pair lateral, and seven ventralsetae. Flagellum (Figs. 36-37, 97-99) 0.84 mmlong, 0.24 mm wide, tubular, without lateral lobes;vmS seta is either missing or it is one of theoff-centered pairs labeled d12, in which case one ofthe dl2 setae is missing (aberrant individual).Pedipalps (Fig. 100): Trochanter not produceddistally; with scattered setae on ventral and ventromesalsurfaces; femur, patella, tibia, andbasitarsus-tarsus with scattered long setae (somepilose) on mesal and ventral surfaces. Lateral anddorsal setation as in Fig. 100. Spur about 0.6, clawabout 1.33 times dorsal length of basitarsus-tarsus.Chelicerae: Serrula with eight rounded teeth; setae:1=3; 2=5; 3=16; 4=4; 5=0; 6=1. All withsmall barbs (6) or pilose (1-4).Legs: Leg I, including coxa, 16.98 mm long;basitarsal-tarsal proportions: 44:8: 11 :9: 10: 10:30.Femur IV about 8.25 times as long as deep.Female paratype (length from distal edge of propeltidiumto base of flagellum, 8.02 mm). Propeltidium2.20 mm long; 1.30 mm wide. Otherwiseas in male holotype except that the anterior rightseta is present. Tergal setation as in male except asfollows: tergite II with two triads (not in a row) ofmicrosetae; tergite V with one pair dorsal setae and69

one unmatched small dorsolateral seta on right side;tergite VI with one pair large dorsal and one pairsmall dorsolateral setae; tergite VII with one pairdorsal and one pair lateral setae, and one dorsolateralseta on right side; tergite VIII with one pairdorsal setae, one dorsolateral seta on right side, onepair lateral setae at posterior margin, and one smalllateral seta on left side about middle of tergite; tergiteIX with one pair dorsal and one pair lateral setae.Segment XI has only one lateral seta. SterniteVI 2.0 times as wide as long; Width/length ratio versusbody length, 4.0. Flagellum (Figs. 20-21, 102)0.86 rom long; without segments or articles; dm2seta missing. Sperrnathecae (Figs. 103-104) consistingof a single pair of slender straight lobes onlyslightly enlarging apically. Leg I, including coxa,16.12 rom long; basitarsal-tarsal proportions:40:9:10:9:11:10:29. Femur IV about 8.25 times aslong as deep. Pedipalp setation as in male (Fig.101). Pedipalp length in proportion to body length1.2 male pedipalp length.Variation: Some specimens occur with either twopairs dorsal setae or with one seta missing. Someindividuals have five setae on the anterior sternumand three on the posterior sternum. Tergal setationis highly variable, with individuals possessing orlacking either one or two small dorsolateral setae.Two individuals lack the large dorsal setae on tergiteIX.Measurements (mm).-Male holotype (femaleparatype): Pedipalp: trochanter 0.26 (0.36); femur1.16 (1.50); patella 1.02 (1.26); tibia 1.10 (1.38);basitarsus-tarsus 0.62 (0.78); total 4.16 (5.28). LegI: trochanter 0.72 (0.64); femur 3.86 (3.82); patella4.98 (5.02); tibia 4.16 (4.24); basitarsus 0.88(0.86); tarsus 1.58 (1.54); total 16.18 (16.12). LegII: trochanter 0.50 (0.36); femur 2.98 (3.08); patella1.46 (1.50); tibia 1.30 (2.36); basitarsus 1.38(1.42); tarsus 1.26 (1.34); total 7.88 (10.06). LegHI: trochanter 0.54 (0.48); femur 2.86 (3.06); patella1.40 (1.42); tibia 1.30 (2.54); basitarsus 1.68(1.68); tarsus 1.34 (1.38); total 9.12 (10.56). Legt \~._~ 99\Figs. 97-IOO.-AgaslOschizomus patei, male: 97, flagellum, dorsal aspect; 98, flagellum, lateral aspect; 99, flagellum, ventralaspect; 100, pedipalp, lateral aspect. Scale lines = 0.25 mrn.70

IV: trochanter 1.20 (1.10); femur 3.54 (3.96); patella1.80 (1.84); tibia 3.44 (3.62); basitarsus 2.24(2.34); tarsus 1.48 (1.56); total 13.70 (14.42).Habitat.-The specimens of A. pate; were collectedin the California Chamber, a large moistbreakdown-floored room about 300 m below the entrance.Cueva de la L1orona has a surveyed length of3,136 m and a depth of 432 m (Pate et aI., 1987).Agastoschizomus sp. cf. patei new speciesNew species of schizomid: T. Sprouse, 1985:38;Sprouse and Sprouse, 1985:79.Material examined.-MEXICO: Tamaulipas:S6tano de San Marcos, La Reforma, 800 m elev.,14-16 Oct. 1984 (P. Sprouse), 1 immature (TMM);Cueva del Tecolote, Los San Pedro, 1,450 m elev.,22-28 Nov. 1986 (T. Treacy Sprouse), 1 immature(TMM); 15-17 March 1989 (P. Sprouse), 1 immature(TMM); 20 March 1989 (D. Pate), 2 immatures(TMM).Comments.-The immature specimens fromthese two caves generally agree with A. patei, but inthe absence of adults we feel it premature to assignthem to that species. The specimen from S6tano deSan Marcos was collected at the back of thecave (Sprouse and Sprouse, 1985). This species issympatric with Protoschizomus sprousei in Cuevadel Tecolote. Two specimens from Cueva del Tecolotewere collected in the Russian Dancer Boreholenear the limit of exploration in the cave.Agastoschizomus sp.Agastoschizomus sp.: Rowland and Reddell,103Figs. 101-104.-AgaslOschizomus paId, female: 101, pedipalp, lateral aspect; 102, flagellum, lateral aspect; 103, spennathecae;104, detail ofspennathecallobe. Scale lines = 0.25 mm for Figs. 101-102,0.01 mm for Fig. 103,0.01 mm for Fig. 104.71

1977:80, 82, fig. 1 (part-Cueva Piedra Ancharecord only).Agastoschizomus n.sp.: Reddell, 1981:26, 124-125,fig. 21 (part).Material examined.-MEXICO: Hidalgo:Cueva Piedra Ancha, 19 Aug. 1965 (J. Reddell, J.Fish, W. Bell), 1 immature (TMM). San Luis PotOSI:Cueva de San Pedro, 4 Ian N of Tlamaya, 900m elev., 28 Dec. 1984 (P. Sprouse), 1 immature(TMM); S6tano de Tlamaya, Tlamaya, 650 m elev.,24 Dec. 1984 (P. Sprouse), 1 immature (TMM).Comments.-The specimen from Cueva PiedraAncha may belong to A. styg;us. Material fromCueva de San Pedro and S6tano de Tlamaya maybelong to A. huitzmolotitlensis. All of these records,however, are based on specimens too immature forpositive placement.Undetermined genus and speciesFig. 3Protoschizomidae: Elliott, 1984a: 15; Elliott,1984b:8.Schizomid: Elliott, 1984a: 15; Elliott, 1984b:8; Elliottand Reddell, 1985:212.Undetermined genus and species (troglobite): Elliottand Reddell, 1985:214.?Agastoschizomus n. sp.: Reddell, 1985:5.Material examined.-U.S.A.: Texas: Val VerdeCounty: Seminole Sink: [=Seminole Canyon Cave),Seminole Canyon State Historical Park, 27 May1984 (W. Elliott, L. Bement), entrance talus cone, 1immature (TMM).Comments.- This specimen measures 1.90 mmlong (including the flagellum). It is apparently anearly instar, but clearly a member of the Protoschizomidae.Generic placement is not possible withoutadult material. This is certainly a relict populationnow isolated in a cave in Texas. The only otherTexas schizomid is Stenochrus mulaiki (Gertsch)from the Rio Grande Valley. Seminole Sink: is animportant archeological site and the specimen wascollected from the underside of a small rock deeplyburied in the talus cone beneath the 10 m deep entrancesink. The cone was completely excavatedduring the archeological excavations in the cave buthas since been replaced. Attempts to rediscover thespecies shortly after completion of the archeologicalstudy were unsuccessful, probably because the rockshad not yet had time to settle. The cave is in a statepark and protected by a gate and by a policy of admittanceonly for scientific research. Hopefully itwill be possible to find additional specimens andproperly place this significant species. This cavealso harbors a tropical relict amblypygid, Phrynus n.sp.ACKNOWLEDGMENTSWe are especially grateful to Dale Pate, PeterSprouse, and Terri Treacy for making extra effortsto collect in the caves of the Purificaci6n region. Wealso thank the following for providing us with material:William R. Elliott, Patty Mothes Jameson,Roy Jameson, H. Lynn McCutchen, CarmenSoileau, Ned Strenth, and Peter Strickland. The loanof material by Norman 1. Platnick (American Museumof Natural History) and by Wojciech 1. Pulawski(California Academy of Sciences) is deeply appreciated.We are grateful to Mark S. Harvey(Western Australian Museum) for allowing us to examinehis unpublished manuscript on scbizomids ofAustralia. The first author was supported in part byTexas Tech University (Depts. of Biological Sciencesand Entomology), for which he is grateful.Jeffrey W. Shultz (University of Cincinnati) kindlyread an early draft of the section on phylogeny andoffered many useful comments. Norman 1. Platnickis thanked for his initial cladistic analysis usingHennig86 and for his encouragement that we learnto use Hennig86. Further reviews and useful commentscame from Charles D. Dondale(Biosystematics Research Center, Ottawa), Norman1. Platnick (American Museum of Natural History),W. David Sissom (Elon College), and Darrell Ubick(California Academy of Sciences). Their assistanceis greatly appreciated.LITERATURE CITEDAnonymous. 1982. Schizomid. The New Encyclopaedia Britannica.Chicago: The Encyclopaedia Britannica Co. 15th ed.,Micropaedia,8:961.Alberti, G., and J.G. Palacios-Vargas. 1987. Fine structure ofspermatozoa and spermatogenesis of Schizomus palaciosiReddell and Cokendolpher, 1986 (Arachnida: Uropygi, Schizomida).Protoplasma, 137:1-14.Bittinger, S. 1975. Sotano Hondo de Pinalito. Assoc. MexicanCave Stud. Activities Newsl., no. 3:13-14.Briggs, T.S., and K. Hom. 1972. A cavernicolous whip-scorpionfrom the northern Mojave Desert, California (Schizomida:Schizomidae). Occas. Papers California Acad. Sci., no.98, 7 pp.Brignoli, P.M. 1973. Note sulla morfologia dei genitali degliSchizomidi e diagnosi preliminari di due nuove specie delMessico (Arachnida, Schizomida). Frag. Entomol., 9:1-9.Brignoli, P.M. 1974. A contribution to the knowledge of theSchizomida of Mexico and Guatemala (Arachnida, Schizomida).Accad. Naz. Lincei, Probl. Att. Sci. Cult., Quad.,171(2):143-152.72

Cokendolpher, 1.C. 1981. The order Schizomida. AustralasianArachno\., no. 5, pp. 6-7.Cokendolpher, 1.C. 1988. Review of the Schizomidae(Arachnida, Schizomida) of lapan and Taiwan. Bull. Nat\.Sci. Mus., Tokyo, ser. A, 14(4):159-171.Dumitresco, M. 1973. Deux especes nouvelles du genre Schizomus(Schizomida), trouvees 8 Cuba. Resultats des expeditionsbiospeologiques cubano-roumaines 8 Cuba, 1:279-292.Elliott, W.R. 1984a. Grotto news: UTG news. Texas Caver,29:15-16.Ellliott), W.R. 1984b. A Texas cave schizomid? North AmericanBiospeleo\. News\., no. 30, p. 8.Elliott, W.R., and 1.R. Reddell. 1985. Appendix II. The biologyof Seminole Sink. pp. 211-216 in: S.A. Turpin, comp.,Seminole Sink: Excavation of a vertical shaft tomb, ValVerde County, Texas. Texas Archeo\. Surv. Res. Rept., no.93.Farris, 1.S. 1988. Hennig86 reference, version 1.5.Hansen, H. 1., and W. Sorensen. 1905. The Tartarides, a tribeof the order Pedipalpi. Ark. Zoo\., 2(8):1-78, pis. 1-7.Harvey, M.S. In press. The Schizomida (Chelicerata) of Australia.Invert. Taxon.Lawrence, R.F. 1958. Whipscorpions (Uropygi) from Angola,the Belgian Congo and Mossambique. Pub\. Cult. CompoDiamantes de Angola, no. 48, pp. 69-80.Lawrence, R.F. 1969. The trichoid structures on the cheliceraeof the short-tailed whip-scorpions (Schizomida; Arachnida).Trans. Roy. Soc. South Africa, 38:123-132.Levi, H.W. 1982. Schizomida. P. 76, pI. 94 in: McGraw HillSynopsis and Classification of Living Organisms. New York:McGraw-Hili Book Co.Maddison, W.P., and D.R. Maddison. 1987. MacClade. Version2.1, Cambridge, Massachusetts.Martin, P.S. 1958. A biogeography of reptiles and amphibians inthe Gomez Farias region, Tamaulipas, Mexico. Misc. Publ.,Mus. Zool., Univ. Michigan, no. 101. 102 pp., 7 pis.Millot, 1. 1942. L'Ordre des pedipalpes doit-il subsister? Bull.Soc. Zool. France, 67:141-145.Mitchell, R.W., W.H. Russell, and W.R. Elliott. 1977. Mexicaneyeless characin fishes, genus Asryanax: Environment, distribution,and evolution. Spec. Publ. Mus. Texas Tech Univ.,no. 12,89 pp.Mothes, P. 1982. Recon to Miquihuana. Assoc. Mexican CaveStud. Activities Newsl., no. 12:93-94.Pate, D., P.IS.] Sprouse, and T. Sprouse. 1987. Cueva de laLlorona. Assoc. Mexican Cave Stud. Activities News\., no.16:89-93.Reddell, 1.R. 1967. Cave biology of the Xilitla region. pp.106-107 in: W.H. Russell and T.W. Raines, eds., Caves ofthe InterAmerican Highway. Assoc. Mexican Cave Stud.Bull., I.Reddell, 1.R. 1971. A preliminary bibliography of Mexican cavebiology with a checklist of published records. Assoc. MexicanCave Stud. Bull., 3, 184 pp.Reddell, 1.R. 1973. Ten years of Mexican cave biology. Assoc.Mexican Cave Stud. News\., 4:31-43.Reddell, 1.R. 1981. A review of the cavernicole fauna of Mexico,Guatemala, and Belize. Texas Mem. Mus. Bull., 27, 327pp.Reddell, 1.[R.] 1985. Terrestrial tropical relicts in the Texas cavefauna. North American Biospeleo\. Newsl., no. 32, p. 5.Reddell, 1.R., and 1.C. Cokendolpher. 1984. Reports on ongoingresearch [The order Schizomida ...] American Arachnol.,no. 29, p. 5.Reddell, 1.R., and 1.C. Cokendolpher. 1985. Redescription ofTrithyreus grassii (Thorell) (Arachnida: Schizomida: Schizomidae).Oriental Insects, 18:43-52.Reddell, 1.R., and 1.C. Cokendolpher. 1991. Redescription ofSchizomus crassicaudatus (pickard-Cambridge) and diagnosesof Hubbardia Cook, Stenochms Chamberlin, and Sotanostenochmsnew genus, with description of a new speciesof Hubbardia from California (Arachnida: Schizomida: Hubbardiidae).Texas Mem. Mus., Pearce-Sellards Ser., no. 47,24 pp.Reddell, 1.R., and W.R. Elliott. 1973. A checklist of the cavefauna of Mexico. IV. Additional records from the Sierra deEI Abra, Taffi8ulipas and San Luis PotosI. Assoc. MexicanCave Stud. Bull., 5:171-180.Reddell, 1.R., and R.W. Mitchell. 1971a. A checklist of the cavefauna of Mexico. I. Sierra de EI Abra, Tamaulipas and SanLuis PotosI. Assoc. Mexican Cave Stud. Bull., 4:137-180.Reddell, 1.R., and R.W. Mitchell, eds. 1971b. Studies on thecavernicole fauna of Mexico. Errata. pp. 1-3 insert in:Assoc. Mexican Cave Stud. Bull., 4.Rowland, 1.M. 1971. Agastochizomus lucifer, a new genus andspecies of cavernicole schizomid (Arachnida, Schizomida)from Mexico. Assoc. Mexican Cave Stud. Bull., 4:13-17.Rowland, 1.M. 1973a. A new genus and several new species ofMexican schizomids (Schizomida: Arachnida). Occas. PapersMus. Texas Tech Univ., no. 11,23 pp.Rowland, 1.M. 1973b. Revision of the Schizomida (Arachnida).1. New York Entomo\. Soc., 80:195-204.Rowland, 1.M. 1973c. Three new Schizomida of the genusSchizomus from Mexican caves (Arachnida). Assoc. MexicanCave Stud. Bull., 5: 135-140.Rowland, 1.M. 1975a. Classification, phylogeny and zoogeographyof the American arachnids of the order Schizomida.Ph.D. Diss. Lubbock: Texas Tech Univ., ix + 415 pp.Rowland, 1.M. 1975b. A partial revision of Schizomida(Arachnida), with descriptions of new species, genus, andfamily. Occas. Papers Mus. Texas Tech Univ., 31, 21 pp.Rowland, 1.M., and 1.R. Reddell. 1977. A review of the cavernicoleSchizomida (Arachnida) of Mexico, Guatemala, andBelize. Assoc. Mexican Cave Stud. Bull., 6:79-102.Rowland, 1.M., and 1.R. Reddell. 1979a. The order Schizomida(Arachnida) in the New World. I. Protoschizomidae anddumitrescoae group (Schizomidae: Schizomus). 1. Arachnol.,6:161-196.Rowland, 1.M., and 1.R. Reddell. 1979b. The order Schizomida(Arachnida) in the New World. II. simonis and brasiliensisgroups (Schizomidae: Schizomus). 1. Arachnol., 7:89-119.Rowland, 1.M., and 1.R. Reddell. 1980. The order Schizomida(Arachnida) in the New World. m. mexicanus and pec/dgroups (Schizomidae: Schizomus). 1. Arachno\., 8:1-34.Rowland, 1.M., and 1.R. Reddell. 1981. The order Schizomida(Arachnida) in the New World. IV. goodnightorum andbriggsi groups and unplaced species (Schizomidae: Schizomus).1. Arachno\., 9: 19-46.Russell, W.H., and T.W. Raines, eds. 1967. Caves of the InterAmerican Highway, Nuevo Laredo, Tamaulipas, to Tamazunchale,San Luis PotosI. Assoc. Mexican Cave Stud. Bull.,1.126 pp.Shear, W.A., and 1. Gruber. 1983. The opilionid subfamilyOrtholasmatinae (Opiliones, Troguloidea, Nemastomatidae).American Mus. Novitates, no. 2757, 65 pp.Shear, W.A., P.A. Selden, W.D. Ian Rolfe, P.M. Bonamo, and1.D. Grierson. 1987. New terrestrial arachnids from theDevonian of Gilboa, New York (Arachnida, Trigonotarbida).American Mus. Novitates, no. 2901, 74 pp.Shultz, 1.W. 1989. Morphology of locomotor appendages inArachnida: evolutionary trends and phylogenetic implications.Zool. 1. Linnean Soc., 97:1-56.Shultz, 1.W. 1990. Evolutionary morphology and phylogeny ofArachnida. Cladistics, 6:1-38.73

Sprouse, P.S. 1982. Mexico news: Hidalgo: "The village ofPinalito, Hidalgo..." Assoc. Mexican Cave Stud. News!., no.12, pp. 3-4.Sprouse, P.[S.] 1985. Proyecto Espeleologico Purificacion 19841985. Assoc. Mexican Cave Stud. Activities News!., no.15:34-46.Sprouse, P.[S.], and T. Sprouse. 1985. Sotano de San Marcos.Assoc. Mexican Cave Stud. Activities News!., no. 15:76-81,2 maps.Sprouse, T. 1985. Sotano de San Marcos. Texas Caver,30:37-38.Sprouse, T., P.[S.] Sprouse, and C. Vesely. 1987. Cueva delTecolote. Assoc. Mexican Cave Stud. Activities News!., no.16:94-99.van der Hammen, L. 1986. Comparative studies in ChelicerataIV. Apatellata, Arachnida, Scorpionida, Xiphosura. Zoo!.Verhand!., no. 226, 52 pp.van der Hammen, L. 1989. An introduction to comparativearachnology. SPB Academic Pub!., Hague, x + 576 pp.Weygoldt, P. 1971. Notes on the life history and reproductivebiology of the giant whip scorpion, Mastigoproctus giganteus(Uropygi, Thelyphonidae) from Florida. J. Zoo!., London,164:137-147.Weygoldt, P., and H.F. Paulus. 1979. Untersuchungen zurMorphologie, Taxonomie und Phylogenie der Chelicerata II.Cladogramme und die Entfaltung der Chelicerata. Z. zoo!.Syst. Evolut.-forsch., 17: 177-200.This.is publication No. N.S.-56 of the Texas Memorial Museum.74

Gertsch, W.J. 1992. Distribution patterns and speciation in North American cave spiders with a list of the troglobites and revision ofthe cicurinas of the subgenus Cicurella. Texas Mem. Mus., Speleol. Monogr., 3:75-122.DISTRffiUTION PATTERNS AND SPECIATION IN NORTH AMERICANCAVE SPIDERS WITH A LIST OF THE TROGLOBITESAND REVISION OF THE CICURINAS OF THE SUBGENUS CICURELLAWillis J. Gertsch!Curator Emeritus, Department of EntomologyAmerican Museum of Natural HistoryNew York, New YorkABSTRACTThe cave spiders of North America are discussed, withspecial emphasis on the distribution of troglobitic species. Listsare provided for the troglobitic spiders of the three principalregions in North America. The Appalachian region contains 18species in four genera and four families. California contains aminimum of 20 species in seven genera and five families. TheTexas fauna includes 61 species in four genera and four families.The cicurinas are sedentary spiders that spin funnel webs andtangled lines of dry silk in and under surface detritus and inground openings and caves. Cicurina is mostly an Americangenus with numerous species in most parts of North America anda few representatives in Japan, Korea, and Europe; the Europeanspecies Cicurina cicur (Fabricius) is the type of the genus. Thepresent paper deals only with the taxa assigned to the subgenusCicurella by Chamberlin and Ivie in 1940, which comprises aseries of mostly small forms derived from standard eight-eyedtypes of which many are progressively losing or have lost theireyes. A few typical cicurellas with eight or six eyes occur inArkansas and adjacent states as well as in Texas. Eyeless speciesso far are known mainly from Texas and most of these fromcaves in the Edwards Plateau region. A single eyeless Cicurellais found in Alabama and a few have been found in Mexico. Thecicurellas can be characterized as a derivative American groupfeaturing small size and loss of eyes nurtured in a special epigeanand cavernicole habitat.ICurrent address: P. O. Box 673, Douglas, AZ 85608DISTRmUTION AND SPECIAnON INNORTH AMERICAN CAVE SPIDERSSpiders have exploited the monotonous securityof the cave habitat with some success and act thereas predators of crawling and flying invertebrates.They enter all parts of caves and use them as transitory,semipermanent or permanent habitations.During many years I have been bringing togetherdata on cave spiders from all parts of this countrywith the following results. More than 400 speciesbelonging to 28 or more families have penetratedcaves to variable distances and spend part or all oftheir lives there. This motley assemblage includedbig-eyed diurnal hunting spiders that depend onsight for their hunting and presumably could notexist long in caves. Many of the others are notbound by deep ties and move freely in caves andoutside habitats that offer suitable conditions. Thepresent discussion is concerned largely with spidersof the United States that seem to be obligative cavernicoles,that is those that live in caves andpresumably cannot survive outside for long.Information on such taxa is available only fromscattered publications so I have prepared a general75

list and offered an analysis of our obligative spidersto allow comparison with such animals of othergroups.Recognition of troglobites is not so precise ascould be wished and the certainty of this specialstatus is a matter requiring much more informationthan mere presence of a blind taxon in a cave. Thereare some spiders with attributes of troglobites thatdo not live in caves. Blind spiders are found in termitenests in Africa and others with greatly reducedor obsolete eyes live in ant nests or under moist humusof tropical and temperate forests on many continents.Obsolescence of the eyes is not necessarily ameasure of length of cave life or even indication ofit. A few cavernicoles with essentially normal eyeshave not been taken outside of caves; a prime exampleis Phanetta suhterranea (Emerton), a cavernicolewith small eyes studied by me on specimensfrom 141 caves ranging from Pennsylvania and Illinoissouth into Alabama, with so far no epigeanrecords, but which live deep within caves often underlitter and rocks. This taxon is listed arbitrarily asa troglophile. By contrast, in other instances I haveassumed a more generous attitude in identifyingtroglobites and list a number with degenerated eyesin trivial remnants that are seemingly limited tosingle caves or cave systems. Spider troglobitesshare most of the special features of cave animals ofother groups. The integument is pallid or whitishbecause of loss of pigment and often is less stronglysclerotized than in surface forms. The eyes showvarying degrees of regression from mere loss ofpigment to complete disappearance of all externalvestiges. This occurs in regular pattern in spiderswith loss first of the anterior median pair and thengeneral reduction and gradual disappearance of thethree remaining diads. The troglobite status is rarelyreached without strong modification of the eyes:most of our obligative cavernicoles are eyeless ornearly so. Finally, in spiders and many other groupsthe appendages tend to become long and much thinnerthan in surface forms.Spider troglobites have been derived from cryptozoictaxa of both primitive and specialized groups.Most are sedentary and spin webs in crevices oncave walls and ceilings or in or under detritus on thefloors. Within these webs spiders have minimizedneed for sight and substitute an increasingly keenchemotactic sense centered in the hairs and setae oftheir appendages. Such sedentary types move to newstations only to satisfy food and climate needs. Allof our American spider troglobites and most fromworld centers come from spiders of this type. Thefew hunting spiders that attain troglobite rank havelikewise de-emphasized sight as a life requirement.A prime example is an eyeless troglobitic wolf spider,Adelocosa anops Gertsch, which lives in KoloaCave, Kauai, Hawaii. Also in Hawaii occurs Lycosahowarthi Gertsch, a yellowish, essentially blindtroglobitic species with vestigial eyes, known fromseveral caves.In 1985 Barr and Holsinger published a basic paperentitled "Speciation in Cave Faunas" in which amap was included of the major limestone areas ofthe United States, in which most of our cavernicolespiders are to be found. Spider troglobites are nowknown from the three principle areas as follows: 1.The Appalachian Valley and Western Plateaus; 2.The Texas Region; and 3. The cave areas of CentralCalifornia. Absence of troglobites from variouslimestone nuclei outside of the cited districts reflectsmore than a failure of intensive collecting and indicatesunlikely presence of troglobites. From none ofthe caves in Utah, Idaho, Arizona, and New Mexicohave come spider troglobites in spite of intensivecollecting by arachnologists. Carlsbad Caverns, oneof our great caves, thus far lacks truly derivativetypes. It is likely that additional troglobites may befound in areas south of these three regions but fewwill be found much to the north of the indicatedlimits.Spider Troglobites of AppalachiaWithin this broad province are found manycaves, including such giants as Manunoth and manyothers famous for their faunas. The great wealth ofcave habitats of this region early became an explOrativearea for speleologists who found spidertroglobites relatively numerous: at least 18 are soclaimed and they belong to eight genera of fourfamilies. Especially numerous are taxa of the familiesLinyphiidae and Nesticidae and include 14 ofthe known troglobites. The first known Americanspider troglobite, Anthrobia mammouthia, a smallwhitish spider without trace of eyes, was describedby Tellkampf in 1844 from Manunoth Cave in Kentucky;it is still the only species of its genus. Severaladditional cavernicoles were added by various studentsso that in 1960 Nicholas was able to list ninepresumed obligative taxa from the United States:five of them do not now qualify for that status.American cave spiders are now known from quitelarge collections from many localities, so it is nowpossible to present a favorable picture of our faunaon the basis of new publications and much unpublisheddata in my possession. Our fauna showssimilarity to the European one, with many of the76

same families and genera, but also carries its distinctiveAmerican flavor.The list of names and authors that follows refrainsfrom duplicative listing available in our standardbibliographic journals. For the sake of completeness,undoubted new species taxa, some beingstudied elsewhere, are included for general interestand tentative completeness.List of the Spider Troglobites of AppalachiaFamily LinyphiidaeAnthrobia mammouthia Tellkampf, 1844. Kentucky,some bordering states.Porhomma cavernicolus (Keyserling), 1886.Widespread from Illinois, Arkansas, Virginia, toAlabama.lslandiana cavealis Ivie, 1965. Kentucky.lslandiana muma Ivie, 1965. Virginia.lslandiana speophila Ivie, 1965. West Virginia.Smilax, new species. Iowa, Wisconsin.Oreonetides, new species. Virginia.Family NesticidaeNesticus dilutus Gertsch, 1984. Tennessee.Nesticus furtivus Gertsch, 1984. Tennessee.Nesticus valentinei Gertsch, 1984. Tennessee.Nesticus barrowsi Gertsch, 1984. Tennessee.Nesticus jonesi Gertsch, 1984. Alabama.Nesticus stygius Gertsch, 1984. Tennessee.Nesticus georgia Gertsch, 1984. Georgia.Family LeptonetidaeAppaleptoneta georgia (Gertsch), 1974. Georgia.Family DictynidaeCicurina (Cicurina), new species. Georgia.Cicurina (Cicurina), new species. Alabama.Cicurina (Cicurella), new species. Alabama.Spider Troglobites of CaliforniaThe small cluster of spots in the middle of California(Barr and Holsinger, 1985, p. 313) identifiesrich cave localities in the foothills of the Sierran andCascadian Mountain Ranges. All the troglobitesbelong to different genera than those of Appalachiaand Texas. Most California caves are quite small butintensive collecting by Richard Graham, DarrellUbick, and a growing group of interested studentshave indicated presence of a rmmmum of 20troglobites belonging to five families. The familiesand seven genera are all strongly represented in thewhole region by numerous epigean species. TheCalifornia region has a very large fauna of crypticagelenines (these now assigned to two special families),so it is not surprising that at least nine speciesrepresenting three distinct genera have producedtroglobites. Missing from this list is any troglobiusspecies of Cicurina, which has produced a multiplicityof eyeless taxa in Texas and is also representedby three taxa in the Appalachia area.It need not be surprising that only one of thelisted troglobites has been formally named but therichness of the fauna in the cited area marks it as aspecial project. The presence of the genus Telema(its single eyeless species known only from a fewcaves in the French Pyrenees), earlier assigned tothe American genus Usofila, is now known fromCalifornia by numerous eyed species and severaleyeless ones.The listed undescribed taxa of the Californian regionare being studied by several students.List of the Spider Troglobites of CaliforniaFamily NesticidaeNesticus potterius (Chamberlin), 1933. ShastaCounty.Family LinyphiidaeLinyphiinae, new species. Amador County.Linyphiinae, new species. Tulare County.Family LeptonetidaeCalileptoneta, new species. Trinity County.Family CybaeidaeCybaeus, new species. Shasta County.Cybaeozyga, new species. Shasta County.Cybaeozyga, new species. Shasta County.Cybaeozyga, new species. Shasta County.Family DictynidaeBlabomma, new species. Shasta County.Blabomma, new species. Calaveras County.Blabomma, new species. Calaveras County.Blabomma, new species. Tulare County.Blabomma, new species. Calaveras County.Blabomma, new species. Amador County.77

Family TelemidaeTelema, new species. El Dorado County.Telema, new species. Calaveras County.Telema, new species. Shasta County.Telema, new species. Tuolumne County.Telema, new species. Tulare County.Telema, new species. Siskiyou County.Spider Troglobit~ of the Texas RegionA rich area of spider troglobite evolution is centeredin the Texas plateaus. For more than 20 yearsa wealth of material has been gleaned from scores ofTexas caves by James Reddell and numerous colleaguesand students of the Texas Speleological Surveyof Austin. The many troglobitic spiders so farfound (and they are continuing to find more) belongto most of the families of the Appalachian Provincebut the generic emphasis is quite distinct. No speciesof Nesticus has been so far recorded but the nesticidgenus Eidmannella (represented in Appalachia andTexas by pallidus which ranges deep into Mexico)has produced five troglobitic taxa of mostly eyelessstatus in Texas. The genus Neoleptoneta, comprisingtiny generalized, six-eyed spiders of the familyLeptonetidae, is relatively rich in species and themajority, perhaps 9 species, seem to be caveadapted. In that family we are now able to match theEuropean fauna with a number of well-markedtroglobites of this largely cavernicole family.The outstanding troglobites of the Texas regionare those of the genus Cicurina. This genus is richin species all over the United States and most arecryptic types that abound under ground cover inmesic situations. Some 46 completely eyeless speciesof the subgenus Cicurella have been found inTexas caves and an additional one occurs in Alabama.The total number of troglobites now claimedfor Texas is 61 and with the totals of 18 for Appalachiaand 20 for California now reaches 99.List of Spider Troglobites of the Texas RegionFamily LinyphiidaeIslandiana unicornis Ivie, 1965. Childress andWheeler Counties.Family NesticidaeEidmannella delicata Gertsch, 1984. Val VerdeCounty.Eidmannella bullata Gertsch, 1984. CulbersonCounty.Eidmannella nasuta Gertsch, 1984. Medina County.Eidmannella reclusa Gertsch, 1984. Travis County.Eidmannella rostrata Gertsch, 1984. EdwardsPlateau.Family LeptonetidaeNeoleptoneta anopica (Gertsch), 1974. WilliamsonCounty.Neoleptoneta coeca (Chamberlin and Ivie), 1942.Comal County.Neoleptoneta concinna (Gertsch), 1974. TravisCounty.Neoleptoneta devia (Gertsch), 1974. Travis County.Neoleptoneta microps (Gertsch), 1974. BexarCounty.Neoleptoneta myopica (Gertsch), 1974. TravisCounty.Neoleptoneta valverde (Gertsch), 1974. Val VerdeCounty.Neoleptoneta, new species. Hays County.Neoleptoneta, new species. Bexar County.Family DictynidaeCicurina buwata Chamberlin and Ivie, 1940. Travisand Williamson Counties.Cicurina (Cicurella) taxa listed in followingreVISIOn.CICURINAS OF THE SUBGENUSCICURELLA OF TEXAS AND NEARBYSTATES AND MEXICOINTRODUCTIONThe American Cicurina fauna was little exploitedby students until publication by Chamberlin and Iviein 1940 of ftAgelenid Spiders of the GenusCicurina." Wilton Ivie was assigned to preparationof a comprehensive review of the known species onthe basis of much new material. Most of the systematicadvances of that paper are to be credited tohim for clarifying the systematic status and faunalwealth of these ecribellates in North America. Hebuilt the genus Cicurina to outstanding size even atthis early date by doubling it to 47 taxa of which 23were described as new. His keen analyses of the taxaand meticulous drawings of the genitalia mark hisartistry. Wilton Ivie continued to collect and studyCicurina after joining me as a colleague at theAmerican Museum of Natural History in New York,where he continued to prepare a supplement to thebasic revisional paper. More than a dozen unknown78

species from the United States and many more fromMexican localities were recognized by him and afew of these tentatively described and illustrated.None of this unfinished material was published butnow forms part of the American Museum collectionhere with me in Portal, Arizona.The present paper continues study of thecicurinas on the basis of large new collections frommany epigean and cavemicole habitats of NorthAmerica and limits itself to the subgenus Cicurellaof the stem genus Cicurina. As now definedCicurella comprises 73 species of eyed and eyelesstaxa of which only 10 have so far been described.Within the United States most species occur in thestate of Texas, except for two from Arkansas andsingle ones from Oklahoma, Missouri and Alabama.A few additional species are known from Mexico.Very little is known about the biology of this groupof spiders. Typical species live in mesic conditionsin or under ground objects or debris or in groundopenings or caves. Often associated with thecicurellas as part of collections are numerous spidersin mature or developing stages of various familiesfavoring such cryptic habitats. Among suchneighbors but not necessarily living near them intheir microhabitats as neighbors are a few standardcicurinas. The most notable one is the largeeight-eyed Cicurina varians Gertsch and Mulaiknow known from a high percentage of the caves ofTexas.SYSTEMATIC SECTIONFAMILY DICTYNIDAEAgelenidae Simon, 1898: 193-308; Comstock, 1912:597; Chamberlin and Ivie, 1940:1-108; Petrunkevitch,1911:530.Cicurininae Lehtinen, 1967: 353.Discussion.-Cicurina are sedentary spiders withone pair of booklungs and a single tracheal spiracleimmediately in front of the spinnerets. Their eyestypically are eight in number and lie in two essentiallystraight rows close to the c1ypeal edge; theymay be reduced in size and number or even completelylost. Traces of an abortive cribellum (nowcalled coluius) are apparent in front of the two anteriorspinnerets. The cicurine genera are few in numberand form a series without obvious near relatives.The familial position of the genus Cicurina, held formany years to be in the family Agelenidae, waschanged by Lehtinen (1967, p. 353), withpersuasive arguments, to the family Dictynidae.This action was tentatively accepted by Forster(1970, pp. 12, 127-139) and recently by Brignoli(1983) and Platnick (1989) in their catalogues. Afteragain reviewing such dictynid genera as Blabomma,Lathys, and especially Brommella, I have tentativelyfollowed Lehtinen and now place Cicurina in thefamily Dictynidae within its own special subfamilyCicurininae. I doubt that even the present tendencyto break up unwieldy families into smaller units willsoon result in holding the group Cicurininae asworthy of its own special ecribellate family.Genus Cicurina MengeCicurina Menge, 1871:IV; Simon, 1898:265; Petrunkevitch,1911:530-531; Comstock, 1912:564,565, 596; Chamberlin and Ivie, 1940:1-108;Vogel, 1967:1-196; Bonnet, 1956:1086-1091.Diagnosis.-The cicurinas of the subgenusCicurella are pale sedentary spiders that offer fewcoloration or somatic features to allow easy identificationof the multiplicity of allied taxa. The cephalothoraxand appendages are pale yellowish toorange with head and distal parts of appendagessomewhat darker. The chelicerae and mouth partsare orange to pale brownish. In general males havethe carapace and legs somewhat darker, and eyelessspecies are paler than those of epigean habitats. Immaturespecies of most habitats are whitish or faintlyyellowish with some sheen on larger surfaces. Mostcicurellas are small spiders and range from 1.2 romto about 7 rom in total length with the average beingabout 3 rom and only a few range much beyond thisaverage. The males are larger than the females withheavier legs and stouter hairs and spines. Inconspicuouspale hairs lie in rows on the carapaceand are darker on the abdomen and leg segments.The true spines are little larger than some of thehairs in some females but they are darker and moredeveloped in males. Loss of individual spines isfrequent but here considered of minor systematicinterest.Description.-The carapace is longer than broad,moderately convex and well rounded on the sides,truncated across the c1ypeus, with the width of theeyes when present about half the greatest width ofthe carapace. Cephalic sutures are obsolete or faintlyindicated but the thoracic fovea is present as a darklinear, slightly indented groove behind the middle ofthe carapace. From the side the carapace is nearlyequal in height for five-sevenths of its length andfrom that point evenly sloping to the caudal margin.The c1ypeus in most eyed taxa is about equal in79

height to the full diameter of an anterior lateral eyebut becomes increasingly lower in those species thatare in the process of losing or have lost some or allof the eyes. The eyes lie in two close rows near thefront edge of the carapace and are of medium size.In eyed taxa the eyes are present as eight or six butall or some may be in various stages of reductionand disappearance, in extreme cases only evident astrivial black or whitish spots; all of the foregoingfeatures are listed among the eyed taxa and brieflynoted among the completely eyeless taxa. The eyesof the frent row when the number is eight are in amoderately procurved line typically narrower thanthe four of the posterior row which is mostlystraight or lightly curved. The triads of the six-eyedtaxa may be close together or widely separated.The sternum is longer than broad, suboval, withthe posterior coxae separated by their width in femalesbut somewhat less in males. The labium is aswide as long and about half as high as the endites.The chelicerae are stout, moderately geniculate, withsmall lateral condyle (boss) at base on the side. Eachchelicera has two rows of teeth margining the longfang furrow: on the promargin are usually threeteeth of which the middle one is often largest, andcontinued by a smooth or irregular carina nearly fulllength; on the retromargin is usually a variable seriesof four to ten teeth or denticles with those atbase longest.The legs are quite robust, especially in males,with the femora thickest and the fourth leg longestand more slender, in the leg formula 4123 of thesubgenus. The full leg charts of the measurements ofboth sexes of an eight-eyed species (see arkansa) -aregiven to illustrate the patterns of the group; withthem comes the conclusion that such data contributelittle to fuller understanding of the taxa. The femoraare subequal in length. The patella-tibia index offerssome indication of the comparative total leg lengths.The leg spines are well represented on all legs: theyare weaker in females and little larger than manyhairs; in males they are heavier and on some legsegments shorter and heavier. The typical spine patternsare presence of 1-1-1 median spines on dorsaof femora and tibiae and 2-2-2 or 2-2-0 ventralspines on tibiae and metatarsi of most legs. Specialspinal reference is given only when such data appearnecessary. The spine formulae of Chamberlin andIvie are useful and presumably valid. I have preferredto emphasize the characters of eyes whenpresent and genitalia of this distinctive group.The segments of the male palpus feature a slenderfemur, a very short patella and a similarly short tibiawhich bears a long flattened apophysis typical of theentire genus. A small angular spur or groove on theprolateral side of the tibia presents trivial differencesin size and position among the species. The cymbiumis a concave bowl, the alveolus, into which thepalpal bulb and elements are fixed. Most of thetegulum is flattened; the frontal smooth area is herecalled subtegulum and the large basal area oftenshowing internal tubular structures is the tegulum.The embolus originates at the base of the tegulum asa thin black tube and curves around the subtegulum,sometimes coming free of the restraining groove; itoffers little information for separation of the taxa.On the retrolateral side of the tegulum beyond theorigin of the embolus lies a conspicuous rounded ortwisted tegular process, the conductor, which continuesas a very long coil lying free on the side ofthe palpus. The conductor is a principle mark of thecicurellas but offers few features in form or lengthto identify the various taxa.The total number of mature males known fromCicurella collections, about a dozen, is only a fifthof the number of mature females, and this mightbring the implication that the ratio of the sexes is anunequal one. In general many spider studies assignnumber superiority to the males but in general it isbelieved in spiders that the ratio of the sexes is essentiallyequal, at least in the beginning. Only 15 ofthe about 70 known taxa have assignable males, 8among the eyed taxa, and 7 to the eyeless ones. Immaturespecimens of all grades of development cannotbe assigned to specific taxa, but the frequentpresence of obvious subadult males among immaturessuggests that the disparity is due to otherfactors.The abdomen is longer than wide, a suboval sacof medium height and is liberally clothed over mostof its surfaces by short soft gray or whitish hairs. Inmost members of the cicurellas it is whitish to grayin color without any distinctive darker pattern. Theepigynal area often shows some brownish color andhas on each side the pinkish book lung openings andthe central epigastric furrow through which the eggsare extruded. The internal tracheal system is presumedto be of the divided branch style described byForster (1970, p. 14).The six functional spinnerets are subapical in positionat the end of the abdomen. The two shortstout anterior spinnerets are well separated by nearlyhalf their length. In front of them lies the colulus(aborted cribellum) reduced to a narrow brown sclerotizedband and smooth area without setae about aswide as the space between the anterior lateral spinnerets.The two more slender posterior spinneretpairs lie close together in a slightly procurved row,80

the posterior median pair short and close together,and the posterior lateral pair with short apical segmentlonger and more widely separated.The epigynum is situated immediately in front ofthe genital orifice of the female, a narrow abdominalopening through which the eggs are extruded. Thestandard features of the Cicurina epigynum consistof a tubular element to receive the semen (the bursa)and conduct it to the storage unit (the spermathecum)to be held until the semen is delivered tothe eggs through a fertilization duct. This generalizedpattern fits the standard Cicurina and covers avariansvariansharrietaevariansrudimentopsarkansadelriosecretaarmadillof~00°0 \joyamicropso 0o 0 0 mirificaEyelessCicurellasintoniaChart I.-Eye and body patterns of Cicurina.81

connecting canalspermathecumsperm sac ~index coil -._----. bursa -~~~~~~-----atrium ----=-~~~~~delriomirificaminoratacuevasubtegulum ;:... emboluscoiltegulumsecretabaroniaarkansaChart 2.-Morphology of male and female genitalia of Cicurina (Cicurella) spp.82

great many species of North America as well as thetype of the genus, Cicurina cicur (Fabricius) ofEurope. The subgenera Cicurina and Cicurusta havefurther modified it by development of a bursal bulb,a secondary spermathecum, and further elongationand complication of the connecting coils.The present revision deals only with the taxa assignedto the subgenus Cicurella and comprehends agrowing number of closely allied forms so far equalin number to those of all the other subgenera. Noexotic taxa are known and it seems to be an exclusivelyAmerican group. In basic features Cicurelladiffers from Cicurina only in the subdivision of thespermathecum by development of a blind sac, insome cases of sizable dimension, at the base of theprinciple body. The fertilization duct runs from theprinciple body of the spermathecum into the atrium.The entire epigynum is covered by a thin sclerotizedsheet with only a single transverse opening, theatrium, in front of the genital orifice. In the varioustaxa the atria are typically narrow in width andshape but offer some detail for specific sorting ofthe taxa. The lightly sclerotized areas of the externalepigynum may show internal details rather clearly orthey may be largely undetailed or invisible, makingnecessary clearing with clove oil or other agents.The ventral (internal) view of the epigynum oftenbrings out details of size, proportion, separation ofparts, and good understanding of the whole structure.Each side of the epigynum is essentially a mirrorimage of the other but not rarely details of oneside may be modified by positional changes. Both•• ••• •• •• •••Map I.-Distribution of eyed Cillcurina (Cicllrella) spp.83

dorsal and ventral views are offered here for allepigyna to give exact information on the internalstructures. Care must be taken during study of thesesmall spiders: some epigyna measuring only 0.2 or0.3 nun may float away in the preservative or be lostby sticking to forceps or dropping on unfavorablesurfaces.The descriptions that follow list verbal and illustrativedata sufficient to make identification easy.The offered diagnosis aims to present a concisecompendium of features separating the taxon fromall known relatives: it is a succinct series ofmorphological, geographical and ecological charactersthat position the taxon among members of itsspecial group and may even identify it specifically.In many taxa its role is adequately served by mentionof a single genital feature or in cavernicoles of asingle cave. The aim of the formal description is tooffer a more complete picture of the taxon withsome mention of the basic characters of eye pattern,measurements, and locality information but withreluctance to duplicate the general data of the preliminarydiagnosis. In final analysis it has beenconcluded the specific characters are best shown inthe details of the epigyna of females where most ofthe pertinent information is found. The presence offew adult males in the collection illustrates their existencebut also shows that even among the few representativestheir palpal figures give little data onspecific identification. In that sex it is unfortunatethat outside of good characters in eyed taxa that thecharacters of their quite complicated palpi offer littleto separate the species.LIST OF THE EYED TAXADescriptions of the seven previously describedspecies and 16 additional new ones are listed withinformation on state and county localities as follows:ARKANSAS: Washington County: secreta, newspecies. Bradley County: arkansa, new species.MISSOURI: Oregon County: harrietae, new species.OKLAHOMA: Latimer County: oklahoma,new species. TEXAS: Bexar County: minorata(Gertsch and Davis); gatita, new species. BlancoCounty: blanco, new species. Burnet County: marmorea,new species. Comal County: joya, new species.Duval County: rudimentops Chamberlin andIvie. Hays County: aenigma, new species. KendallCounty: pampa Chamberlin and Ivie. Kerr County:microps Chamberlin and Ivie; dorothea, new species;modesta, new species. Kimble County: rosae,new species. Llano County: texana (Gertsch). PecosCounty: mirifica, new species. San Patricio County:sintonia, new species. Starr County: riograndeGertsch and Mulaik. Tom Green County: hexopsChamberlin and Ivie. Travis County: annadillo,new species. Val Verde County: delrio, new species.Cicurina arkansa, new speciesChart 1, Figs. 1-2,37-38Diagnosis.-Eight-eyed eplgean species ofBradley County, Arkansas; anterior median eyessubcontiguous; index coil of epigynum (Fig. 1) witha tight loop in subvertical position; conductor ofmale palpus (Fig. 37) with thin loop running to nearend of cymbium.Etymology.-Specific name for the State ofArkansas.Description.-Female holotype: Length 2.4 nun.Carapace 1.2 nun long, 0.7 nun wide. Abdomen 1.2nun long, 0.8 nun wide. Clypeal height full widthof anterior lateral eyes; both eye rows (Chart I) essentiallystraight with posterior medians separatedby full diameter and half as far from subequal lateraleyes. Chelicerae: retromargin of fang with 9 teeth inclose file.I II III IV PalpFemur 0.95 1.00 0.80 1.00 0.20Patella 0.20 0.20 0.15 0.20 0.12Tibia 0.80 0.55 0.50 0.80 0.15Metatarsus 0.55 0.58 0.60 1.00Tarsus 0.25 0.25 0.20 0.60 0.15Total 2.75 2.58 2.25 3.60 0.62Leg formula: 4123. First leg three times as longas carapace; fourth leg about four times as long ascarapace; patella-tibia of first and fourth legs shorterthan carapace; metatarsus-tarsus of fourth leg muchlonger than carapace; metatarsus of fourth legslightly longer than carapace.Epigynum (Figs. 1-2): spermathecum roundedlobe broadly jointed to smaller sac; thick connectingcanal closely encircling spermathecal parts.Male: Length 3.65 nun. Carapace 1.80 nun long,1 nun wide. Abdomen 1.85 nun long, 0.85 romwide. Eyes closer together than in female; both rowsgently procurved to essentially straight; posteriormedian eyes separated by less than full diameter,half as far from subequal lateral eyes. Spines of firstand fourth tibiae 2-2-2.I II III IV PalpFemur 1.13 0.95 1.00 1.15 0.50Patella 0.20 0.25 0.30 0.50 0.20Tibia 1.00 0.40 0.30 1.00 0.20Metatarsus 0.80 0.90 0.80 1.10Tarsus 0.60 0.55 0.50 0.63 0.60Total 3.73 3.05 2.90 4.38 1.5084

Key to the Eyed Females1. Eight eyes present. 2Six eyes present 62. Eight-eyed taxa with typical eyes and anterior median eyes of normal size 3Eight-eyed taxa with evanescent eyes; with anterior median eyesreduced to small dark or whitish spots 53. Index coil with wide loop across spermathecum (Figs. 9-10);epigean species of Oregon County, Missouriharrietae, new speciesIndex coil smaller loop 44. Eyes of front row subcontiguous; index coil with right loop(Figs. 1-2); epigean species of Bradley County, ArkansasEyes of front row moderately separated; index coil narrow procurvedloop (Figs. 3-4); epigean species of Starr County, Texas5. Anterior median eyes reduced to whitish spots; index coil ofepigynum procurved loop, connecting canals closely encirclingspermathecal parts (Figs. 19-20); epigean species of DuvalCounty, TexasAnterior median eyes small dark spots; index coil heavyprocurved loop (Figs. 21-22); cavernicole of caves ofVal Verde County, Texasarkansa, new speciesriogrande Gertsch & MulaikrudimeJlfops Chamberlin & Iviedelrio, new species6. Anterior lateral eyes separated by about diameter 7Anterior lateral eyes separated by two or more diameters 147. Index coil absent 8Index coil present 98. Large oblong spermathecum with small twisted sac (Figs. 7-8);epigean species of Blanco County, TexasOval spermathecum with large rounded sac (Fig. 15); epigeanspecies of Bexar County, Texasblanco, new speciesminorata, new species9. Index coil of typical transverse pattern 10Index coil of atypical pattern 1110. Index coil thin canal in vertical position (Figs. 27-28);cavernicole from Black Cat Cave, Bexar County, TexasIndex coil forming wide loop around spermathecal parts(Figs. 29-30); epigean species of Hays County, Texasgatita, new speciesaenigma, new species11. Index coil small oval canal (Figs. 25-26); epigean speciesof San Patricio County, Texassintonia, new speciesNot as above 1212. Index coil wide procurved canal (Figs. 35-36); Raven Ranch,Kerr County, Texasdorothea, new speciesNot as above 1313. Index coil short procurved canal (Figs. 11-12); epigeanspecies of Burnet County, TexasIndex coil thin transverse canal (Figs. 23-24); epigeanspecies of Kendall County, Texasmarmorea, new speciespampa Chamberlin & Ivie85

14. Index coil absent; round spermathecum with round sac(Figs. 33-34); epigean species of Travis County, Texasarmadillo, new speciesIndex coil present 1515. Index coil narrow canal in oblique position between spermathecal parts 16Index coil transverse canal 1716. Round spermathecum with small round sac (Figs. 13-14);cavernicole of Heidrich's Cave, Comal County, TexasOblong spermathecum with small round sac (Fig. 16); epigeanspecies of Kerr County, Texasjoya, new speciesmicrops Chamberlin & Ivie17. Connecting canal closely encircling spermathecal parts 18Connecting canal widely encircling spermathecal parts 1918. Index coil deeply procurved (Figs. 19-20); epigean speciesof Duval County, Texasrudimentops Chamberlin & IvieIndex coil short transverse canal 2019. Suboval spermathecum with round sac (Figs. 17-18); epigeanspecies of Washington County, ArkansasOval spermathecum with twisted sac (Figs. 31-32); epigeanspecies of Kimble County, Texas20. Index coil deeply procurved (Figs. 5-6); eyes evanescent,six pale eye spots; cavernicole of Amazing Maze Cave,Pecos County, TexasIndex coil heavy procurved loop joint S-shaped figure(Figs. 21-22); cavernicole of caves of Val Verde County, Texassecreta, new speciesrosae, new speciesmirifica, new speciesdelrio, new speciesKey to the Eyed Males1. Eight-eyed species with eyes of anterior row subcontiguous;Bradley County, Arkansasarkansa, new speciesSix-eyed species " 22. Anterior lateral eyes separated by about full diameter. 3Anterior lateral eyes separated by two or more diameters 63. Tarsal fold about half width of tarsal process 4Tarsal fold less than half width of process 64. Tarsal process evenly curved at apex (Figs. 41-42) oklahoma, new speciesTarsal process widened at apex 55. Tarsal process (Figs. 43-44) texana (Gertsch)Tarsal process (Figs. 39-40)secreta, new species6. Coil of conductor short, not much longer than base; TomGreen County, Texashexops Chamberlin & IvieCoil of conductor long fine spines continuing to near endof tegulum 77. Tarsal process with small notch at apex modesta, new speciesTarsal process without apical notch 88. See details of Figs. 51 and 52; from Kerr County, Texas microps Chamberlin & IvieSee details of Figs. 47 and 48; San Patricio County, Texassintonia, new species86

39111012Figs. 1-12.-Ventral and dorsal views of epigyna of eyed Cicurina (Cicurella): 1-2, arkansa, new species; 3-4, riogrande Gertschand Mulaik; 5-6, mirifica, new species; 7-8, blanco, new species; 9-10, harrietae, new species; 11-12, marmorea, new species.87

Leg formula: 4123. First leg twice as long ascarapace; fourth leg two and half times as long ascarapace; patella-tibia of first and fourth legs shorterthan carapace; metatarsus-tarsus of fourth leg aboutas long as carapace.Male palpus (Figs. 37-38): tegulum with threeincomplete tubules; tarsal process of tibia with longfold less than half its width; adjacent transverse spurof typical size.Type-data.-Female holotype from Sumpter,Bradley County, Arkansas, 17 November 1963(Leslie, collected from pine-oak woods) (AMNH).Distribution.-Known only from Sumpter. Otherrecords: September and November 1963 and 1964(Leslie) (AMNH); September 1963, probably fromcan traps, in Exline-Peck Collection (CAS).Cicurina harrietae, new speciesChart 1, Figs. 9-10Diagnosis.-Eight-eyed epigean species of Alton,Oregon County, Missouri; eyes of front row contiguous(Chart 1); index coil of epigynum (Figs.9-10) wide canal lying across spermathecum. Maleunknown.Etymology.-Specific name for the late HarrietE. Frizzell, friend and student of the genusCicurina.Description.-Female holotype: Length 4 mm.Carapace 1.7 mm long, 1 mm wide. Abdomen 2.4mm long, 1.5 mID wide. Eye group half as wide ascarapace and clypeus as high as full diameter of anteriorlateral eye; eyes large (Chart 1) with tuberclesblack; anterior eye row moderately recurved withlarger eyes contiguous; posterior eye row essentiallystraight with median eyes separated by less than fulldiameter and half as far from lateral eyes.Chelicerae: retromargin with 8 teeth, three of themlarger. Leg lengths: first femur 1.3 mm, fourth femur1.4 mm; first patella-tibia 1.7 mm; fourthpatella-tibia 1.8 mm.Epigynum (Figs. 9-10): spermathecum roundwith broad juncture to smaller round sac; connectingcanals thick tubes rather closely encircling the spermathecum.Type-data.-Female holotype from Alton, OregonCounty, Missouri, May 1956

1419 23Figs. 13-24.-Ventral and dorsal views of epigyna of eyed Cicurina (Cicurella): 13-14, joya, new species; 15, minorataChamberlin and lvie; 16, microps Chamberlin and lvie; 17-18, secreta, new species; 19-20, rndimentops Chamberlin and lvie; 21-22,delrio, new species; 23-24, pampa Chamberlin and lvie. [Figs. 16, 19,23 from Chamberlin and lvie, 1940)89

Type-data.-Female holotype from Cove CreekValley, Washington County, Arkansas, 15 milessouth of Prairie Grove, December 1955 (F. Hite)(AMNH).Distribution.-Known only from Cove CreekValley. Other records: males, August 1952, 1961,1962 (F. Hite) (AMNH); 6 October 1962, 2 males,1 female (Exline Peck) (CAS).Cicurina mannorea, new speciesFigs. 11-12Diagnosis.-Six-eyed epigean species fromMarble Falls, Burnet County, Texas; anterior medianeyes full diameter apart; index coil small procurvedcanal lying across spermathecal sac. Maleunknown.Etymology.-Specific name from Latin marmorea,marble, for Marble Falls.Description.-Female holotype: Length 3.9 mm.Carapace 1.7 mm long, 1.3 mm wide. Abdomen 2.2mm long, 1.35 mm wide. Clypeus narrow, abouthalf diameter of anterior lateral eye. Posterior eyerow essentially straight; median eyes slightly morethan full diameter apart, their radius from posteriorlaterals. Chelicerae: retromargin of fang with 5teeth. Leg segments: first femur 0.7 mm, fourth femur0.75 mm; parts of legs missing.Epigynum (Figs. 11-12): subround spermathecum,broadly joined to small round sac; broadconnecting canals tightly encircling spermathecalparts.Type-data.-Female holotype from 8 miles northof Marble Falls, Burnet County, Texas, 8 November1964 (J. Reddell) (AMNH).Cicurina pampa Chamberlin and IvieFigs. 23-24Cicurina pampa Chamberlin and Ivie, 1940:79, PI.VIII, fig. 60.Diagnosis.-Six-eyed epigean species from KendallCounty, Texas; anterior lateral eyes separatedby full diameter; index coil of epigynum thin tubecrossing between sperrnathecal parts. Male unknown.Etymology.-Specific name from Spanishpampa, a grassy plain.Description.-Female holotype: Length 2.1 mm.Carapace 0.95 mm long, 0.68 mm wide. Abdomen1.15 mm long, 0.70 mm wide. Clypeal height abouthalf diameter of anterior lateral eye; posterior eyerow straight with median eyes 1.5 diameters apart,scarcely radius from lateral eyes. Chelicerae: retromarginof fang with 5 teeth. Leg lengths: firstpatella-tibia 0.77 mm, fourth patella-tibia 0.8 mm.Epigynum (Figs. 23-24): oval spermathecumbroadly joined to rounded sac; index coil and connectingcanal closely encircling spermathecal parts.Type-data.-Female holotype from KendallCounty, Texas, December 1939 (D. and S. Mulaik)(AMNH).Cicurina modesta, new speciesFigs. 45-46Diagnosis.-Six-eyed epigean species from CampVerde, Kerr County, Texas; anterior lateral eyesseparated by full diameter; conductor of male palpuswith long coil reaching end of subtegulum. Femaleunknown.Etymology.-Specific name from Latin modestus,modest.Description.-Male holotype: Length 2.7 mm.Carapace 1.5 mm long, 1 mm wide. Abdomen 1.2mm long and wide. Clypeus as wide as anterior lateraleye. Posterior eye row straight; median eyesseparated by full diameter. Chelicerae: retromarginwith 5 teeth. Leg lengths: first femur 0.9 mm,fourth femur 0.9 mm; first patella-tibia 0.95 mm,fourth patella-tibia 1 mm. Leg spines: first tibia2-2-0, fourth tibia 2-2-2.Male palpus (Figs. 45-46): tegulum with threetubules; tarsal process with trivial SUbapical spurand long thin fold less than third its width and longtransverse spur.Type-data.-Male holotype and male from CampVerde, Kerr County, Texas (W. Rogers), frompitfall traps (AMNH).Cicurina hexops Chamberlin and IvieFigs. 49-50Cicurina hexops Chamberlin and Ivie, 1940:79, PI.XII, fig. 92.Diagnosis.-Six-eyed epigean species fromWater Valley, Tom Green County, Texas; anteriorlateral eyes separated by two diameters; conductorof male palpus with short coil. Female unknown.Etymology.-Specific name from Latin hexops,six eyes.Description.-Male holotype: Length 2.1 mm.Carapace 0.9 mm long, 0.8 mm wide. Abdomen 1.2mm long, 0.7 mm wide. Clypeus less than full diameterof anterior lateral eye. Posterior eye rowstraight; median eyes separated by 1.5 diameters,90

27 293334Figs. 25-36.-Ventral and dorsal views of epigyna of eyed Cicurina (Cicurella): 25-26, simonia, new species; 27-28, garita, newspecies; 29-30, aenigma, new species; 31-32, rosae, new species; 33-34, armadillo, new species; 35-36, dorothea, new species.91

about half as far from side eyes. Chelicerae: retromarginof fang with 5 teeth. Leg lengths: first femur0.58 mm, fourth femur 0.6 mm; first patella-tibia0.65 mm, fourth patella-tibia 0.67 mm. Leg spines:first tibia 2-2-0, fourth tibia 2-2-1.Male palpus (Figs. 49-50): tegulum with parts ofthree tubules; conductor with rather short coil, itslength about that of tegulum; tarsal process of tibiawith thin fold about fourth its width and adjacentthin transverse spur.Type-data.-Male holotype and additional malefrom Water Valley, Tom Green County, Texas,December 1939 (S. and D. Mulaik) (AMNH).Cicurina oklahoma, new speciesFigs. 41-42Diagnosis.-Six-eyed epigean species from RedOak, Latimer County, Oklahoma; clypeus as high asanterior lateral eye; conductor of male palpus withlong coil with tip near end of subtegulum. Femaleunknown.Etymology.-Specific name for the state ofOklahoma.Description.-Male holotype: Length 2.02 mm.Carapace 1.2 mm long, 0.82 mm wide. Abdomen0.82 mm long, 0.50 mm wide. Posterior eye rowslightly procurved; median eyes separated by fulldiameter, nearly touching lateral eyes. Chelicerae:retromargin of fang with 6 teeth. Leg lengths: firstfemur 0.95 mm, fourth femur 1.05 mm; firstpatella-tibia 1.10 mm, fourth patella-tibia 1.20 mm.Leg spines: first tibia 2-2-0, fourth tibia 2-2-1.Male palpus (Figs. 41-42): tegulum with two·incompletetubules; tarsal process of tibia with longfold about half its width and narrow transverse spur.Type-data.-Male holotype from Red Oak,Latimer County, Oklahoma, 8-9 October 1976 (V.Roth) (AMNH).Cicurina minorata (Gertsch and Davis)Fig. 15Chorizomma minorata Gertsch and Davis, 1936:6,fig. 8.Cicurina minorata: Chamberlin and Ivie, 1940:80,pI. VIII, fig. 63, pI. XII, fig. 96.Diagnosis.-Six-eyed epigean species from SanAntonio, Bexar County, Texas; anterior lateral eyesseparated by diameter; index coil completely lackingin usual position. Male unknown.Etymology.-Specific name from Latin, minimus,very small.Description.-Female holotype: Length 1.6 mm.Carapace 0.7 mm long, 0.47 mm wide. Posterioreye row straight with median eyes separated by longdiameter, half as far from lateral eyes. Chelicerae:retromargin of fang with 5 small teeth. Leg lengths:first femur 0.46 mm; fourth leg missing.Epigynum (Fig. 15): spermathecum oval withsmall laterally directed sac; connecting canalsclosely appressed to sperrnathecum.Type-data.-Female holotype from San Antonio,Bexar County, Texas, December 1934 (L. I. Davis)(AMNH).Cicurina texana (Gertsch)Figs. 43-44Chorizomma texana Gertsch, 1935:15, figs. 36-37.Cicurina texana: Chamberlin and Ivie, 1940:78, PI.XII, fig. 90, male only.Diagnosis.-Six-eyed epigean species of Llano,Texas; anterior lateral eyes separated by long diameter;conductor of male palpus with long coil. Femaleunknown.Etymology.-Specific name for the State ofTexas.Description.-Male holotype: Length 2.2 mm.Carapace 1 mm long, 0.75 mm wide. Abdomen1.15 mm long, 0.78 mm wide. Narrow clypeus halfdiameter of anterior lateral eye. Posterior eye rowessentially straight; slightly smaller median eyesseparated by little more than diameter, half as farfrom lateral eyes, which are equal in size to anteriorlateral eyes. Chelicerae: retromargin of fang with 4small teeth. Leg lengths: first femur 0.73 mm,fourth femur 0.75 mm; first patella-tibia 0.9 mm,fourth patella-tibia 0.97 mm. Leg spines: first andfourth tibiae 2-2-2.Male palpus (Figs. 43-44): tegulum with two incompletetubules; heavy conductor with long coilreaching end of subtegulum; tarsal process with foldnearly half its width and thin adjacent transversespur.Type-data.-Male holotype from Llano, LlanoCounty, Texas, December 1934 (L. Irvy Davis)(AMNH).Cicurina gatita, new speciesFigs. 27-28Diagnosis.-Six-eyed cavemicole from Black CatCave, Bexar County, Texas; anterior lateral eyesseparated by full diameter; index coil of epigynum92

Figs. 37-48.-Ventral and retrolateral views of male palpi of Cicurina (Cicurella): 37-38, arkansa, new species; 39-40, secreta,new species; 41-42, oklahoma, new species; 43-44, texana Gertsch; 45- 46, modesta, new species; 47-48, sintonia, new species.93

thin canal directed vertically forward for full lengthof spermathecum.Etymology.-Specific name from Spanish gatita,little cat.Description.-Female holotype: Length 1. 86mm. Carapace 0.66 mm long, 0.50 mm wide. Abdomen1.2 mm long, 0.8 mm wide. Clypeus narrow,about half diameter of anterior lateral eye. Eyegroup nearly two-thirds width of carapace; posterioreye row essentially straight with median eyes separatedby 1.5 diameters, half as far from lateral eyes.Chelicerae: promargin of fang with 6 teeth. Leglengths: first femur 0.5 mm, fourth femur 0.5 mm;first patella-tibia 0.6 mm, fourth patella-tibia 0.7mm. Leg spines: first tibia 2-2-0, fourth tibia 2-2-2.Epigynum (Figs. 27-28): spermathecum ovalwith broadly attached oval sac; atrium slender transversegroove.Type-data.-Female holotype from Black CatCave, Bexar County, Texas, 27 January 1987 (J.Reddell, M. Reyes) (AMNH).CiCUriTUl aenigma, new speciesFigs. 29-30Diagnosis.-Six-eyed epigean species from HaysCounty, Texas; spermathecum elongated andbroadly joined to long sac; index coil wide ringaround spermathecal parts. Male unknown.Etymology.-Specific name from Latin aenigma,enigma, secret.Description.-Female holotype: Length of damagedspecimen 2 mm. Carapace 1 mm long, 0.66mm wide. Clypeus as high as one diameter of anteriorlateral eye. Anterior lateral eyes separated byfull diameter. Posterior eye row essentially straightwith median eyes separated by diameter. Chelicerae:retromargin of fang with 5 teeth. Leg lengths: firstfemur 0.45 mm, fourth femur 0.5 mm; firstpatella-tibia 0.54 mm, fourth patella-tibia 0.66 mm.Epigynum (Figs. 29-30): spermathecum elongatedin mixture of canals and spermathecal elements,with much uncertainty of details.Type-data.-Female holotype from HaysCounty, Texas, 15 April 1939 (D. and S. Mulaik)(AMNH).Cicurina dorothea, new speciesFigs. 35-36Cicurina texana: Chamberlin and Ivie, 1940:78, pI.VIII, fig. 64 (female only).Diagnosis.-Six-eyed eplgean species fromRaven Ranch, Kerr County, Texas; anterior lateraleyes separated by diameter; spermathecum longoval, broadly joined to smaller rounded sac. Maleunknown.Etymology.-Specific name for Dorothea Mulaik,collector of many Texas spiders.Description.-Female holotype: Length 2.5 mm.Carapace 1 mm long, 0.65 mm wide. Abdomen 1.5mm long, 0.65 mm wide. Clypeus half diameter ofanterior lateral eye. Posterior eye row straight, withmedian eyes less than diameter apart, a radius fromside eyes. Chelicerae: retromargin with 5 teeth. Leglengths: first femur 0.7 mm, fourth femur 0.7 mm;first patella-tibia 0.8 mm, fourth patella-tibia 0.85mm. Leg spines: first tibia 2-2-0, fourth tibia 2-2-2.Epigynum (Figs. 35-36): index coil broad moderatelyprocurved canal across spermathecum;connecting canals thick, closely margining spermathecum.Type-data.-Female holotype from RavenRanch, Kerr County, Texas, August 1939 (D. andS. Mulaik) (AMNH).Distribution.-Known only from Raven Ranch.Other record: December 1939 (D. and S. Mulaik),female.Cicurina rosae, new speciesFigs. 31-32Diagnosis.-Six-eyed epigean species from 7miles east of Junction, Kimble County, Texas, withanterior lateral eyes separated by 1.5 diameters; indexcoil of epigynum small, slightly procurved canalacross spermathecum. Male unknown.Diagnosis.-Specific name for Rose Carpenter,friend and collector of many Texas spiders.Description.-Female holotype: Length 2.9 mm.Carapace 1.3 mm long, 0.9 mm wide. Abdomen 1.6mm long, 1 mm wide. Clypeal margin less than diameterof anterior lateral eye, which are separatedby 1.5 diameters; posterior eye row straight, withsmall middle eyes separated by 1.5 diameters, lessthan diameter from lateral eyes. Chelicerae: retromarginwith 5 teeth. Leg lengths: first femur 0.8mm, fourth femur 0.8 mm; first patella-tibia 1 mm,fourth patella-tibia 1. I mm. Leg spines: first tibia 22-0, fourth tibia 2-2-2.Epigynum (Figs. 31-32): Spermathecum ovalwith short curved sac, appearing in ventral view assmall round lobe; index coil and connecting canalthick and tightly margining the spermathecum.Type-data.-Female holotype from 7 miles eastof Junction, Kimble County, Texas, 19 November1967 (Rose Carpenter) (AMNH).94

Cicurina blanco, new speciesFigs. 7-8Diagnosis.-Six-eyed epigean species from JohnsonCity, Blanco County, Texas; anterior eyes separatedby full diameter; index coil of epigynum notdetectable in ventral view. Male unknown.Etymology.-Specific name from Spanishblanco, white, in reference to Blanco County, usedin apposition.Description.-Female holotype: Length 2.8 mm.Carapace 1.1 mm long, 0.9 mm wide. Abdomen 1.7mm long, 1.15 mm wide. Clypeal margin equal tohalf diameter of anterior lateral eyes, with medianeyes separated by full diameter, about radius fromlateral eye. Chelicerae: retromargin with 5 teeth.Leg lengths: first femur 0.7 mm, fourth femur 0.75mm; first patella-tibia 1 mm, fourth patella-tibia 1mm.Epigynum (Figs. 7-8): Spermathecum elongate,suboval, with small twisted sac; connecting canalthick, closely appressed to spermathecal parts.Type-data.-Female holotype from 10 miles eastof Johnson City, near Pedemales River, BlancoCounty, Texas, 23 February 1986 (S. J. Harden)(AMNH).Cicurina annadillo, new speciesChart I, Figs. 33-34Diagnosis.-Six-eyed epigean species from nearAustin, Travis County, Texas; eyes of both rowsseparated by two diameters (Chart 1); index coil ofepigynum not present between spermathecal lobes.Male unknown.Etymology.-Specific name from Spanish annadillo,used in apposition.Description.-Female holotype: Length 2.6 mm.Carapace 1 mm long, 0.7 mm wide. Abdomen 1.6mm long, 0.85 mm wide. Clypeus narrow, abouthalf diameter of anterior lateral eye. Anterior lateraleyes separated by two diameters. Posterior eye rowmoderately procurved with median eyes separated byabout two diameters, a radius from posterior laterals.Chelicerae: retromargin of fang with 5 teeth.Leg lengths: first femur 0.55 mm, fourth femur0.58 mm; first patella-tibia 0.6 mm, fourthpatella-tibia 0.65 mm. Leg spines: first and fourthtibiae 2-2-0.Epigynum (Figs. 33-34): large round spermathecumwith small rounded sac; connecting canalclosely appressed to spermathecal parts.Type-data.-Female holotype from near Austin,Travis County, Texas, 8 January 1948 (Cheldon),from armadillo nest (AMNH).Cicurina rudimentops Chamberlin and IvieChart 1, Figs. 19-20Cicurina rudimentops Chamberlin and Ivie,1940:76, pI. VIII, fig. 59.Diagnosis.-Eight-eyed epigean species of Alice,Duval County, Texas; anterior median eyes reducedto whitish spots; dark anterior lateral eyes nearlytwo diameters apart; index coil of epigynum slenderwidely procurved canal across sperrnathecal sac.Male unknown.Etymology.-Specific name from Latin rudimentumand ops, rudimentary eyes.Description.-Female holotype: Length 2.1 mm.Carapace 1.05 mm long, 0.73 mm wide. Abdomen1 mm long, 0.7 mm wide. Clypeal height less thanfull diameter of anterior lateral eye. Anterior eyerow essentially straight. Posterior eye row straight;posterior median eyes separated by nearly two diameters.Chelicerae: retromargin with 5 teeth. Leglengths: first femur 0.6 mm, fourth femur 0.6 mm;first patella-tibia 0.8 mm, fourth patella-tibia 0.9mm. Leg spines: first tibia 2-2-0, fourth tibia 2-2-2.Epigynum (Figs. 19-20): round sperrnathecumnarrowly joined to small oval sac; thick connectingcanal narrowly ringing spermathecal parts.Type-data.-Female holotype from 17 milesnorth of Alice, Duval County, Texas, December1939 (D. and S. Mulaik) (AMNH).Cicurina sintonia, new speciesChart 1, Figs. 25-26, 47-48Diagnosis.-Six-eyed epigean species of Sinton,San Patricio County, Texas; anterior lateral eyesseparated by full diameter; index coil short oval canallying across sperrnathecum. Male palpus (Figs.47-48).Etymology.-Specific name for Sinton, Texas.Description.-Female holotype: Length 2.7 mm.Carapace 1.35 mm long, 1 mm wide. Abdomen1.35 mm long, 1 mm wide. Eyes small, evanescent;posterior row straight with smaller median eyesseparated by about diameter. Patella-tibia of firstand fourth legs 1 mm long.Epigynum (Figs. 25-26): spermathecum oval,broadly joined to small rounded sac; connecting canalclosely encircling spermathecum.Male: Length 2.75 mm. Carapace 1 mm long andwide. Abdomen 0.82 mm long, 0.52 mm wide.95

Eyes closer together than those of female but patternessentially the same. Color mostly whitish but onemale with carapace pale yellow. Chelicerae: retromarginof fang with 5 teeth. Leg lengths: first femur0.85 rom, fourth femur 0.95 rom; first patella-tibia1.2 rom, fourth patella-tibia 1.4 mm.Male palpus (Figs. 47-48): tegulum with onevisible tubule and adjacent one on subtegulum; conductorwith twisted base and long coil running nearend of tegulum; tarsal process with thin fold third itswidth.Type-data.-Female holotype and 4 males fromSinton, San Patricio County, Texas, 20 November1959 (H. E. Laughlin) (AMNH).CicurilUl delrio, new speciesCharts 1,2, Figs. 21-22Diagnosis.-Eight or six-eyed cavernicole of ValVerde County, Texas; eyes evanescent with widelyseparated anterior median eye spots present or absent(Chart 1); index coil of epigynum tight procurvedhook across small spermathecal sac. Maleunknown.Etymology.-Specific name for Del Rio, Texas,used in apposition.Description.-Female holotype: Length 4 rom.Carapace 1.9 mm long, 1.25 mm wide. Abdomen2.1 mm long, 1.5 mm wide. Clypeal margin wide,about two diameters of anterior lateral eye. Anterioreye row straight with tiny medians 1.5 rom apart,about same distance from anterior lateral eyes. Posterioreye row slightly recurved; small median eyesabout three diameters apart, two diameters from anteriorlateral eyes. Chelicerae: retromargin of fangwith 6 teeth. Leg lengths: first and fourth femora1.6 mm. Leg spines: first tibia 2-2-0, fourth tibia2-2-2.Epigynum (Figs. 21-22): round spermathecumwidely joined to smaller rounded sac; thick connectingcanal forming widely separated loop aroundspermathecal parts.Type-data.-Female holotype from Sunset Cave,12 miles NW Del Rio, Val Verde County, Texas,on Ellison Brite Ranch, 14 December 1962 (J. Reddell,W. Russell), on wall 50 feet from cave entrance(AMNH).Distribution.-TEXAS: Val Verde County:Diablo Cave, Calyx Hole entrance, 12 August 1963(J. Reddell, D. McKenzie), 1 female, 1 penultimatemale from under rotting shirt. Unnamed cave (No.8), half mile from Ladder Cave, 12 August 1963 (J.Reddell, D. McKenzie), 6 immature and 3 penultimatemales, mostly from under rocks.CicurilUljoya, new speciesChart 1, Figs. 13-14Diagnosis.-Six-eyed cavernicole of Heidrich'sCave, Comal County, Texas; anterior lateral eyesseparated by nearly two diameters; index coil ofepigynum thin tube crossing between spermathecalelements. Male unknown.Etymology.-Specific name from Spanish joya,jewel, used in apposition.Description.-Female holotype: Length 1.7 mm.Carapace 0.7 mm long, 0.53 mm wide. Clypealmargin equal in height to diameter of anterior lateraleye; posterior eye row slightly procurved; posteriormedian eyes separated by diameter, about two-thirdsdiameter from posterior lateral eyes. Chelicerae:retromargin of fang with 5 teeth. Leg lengths: firstfemur 0.5 rom, fourth femur 0.52 mm; firstpatella-tibia 0.62 mm, fourth patella-tibia 0.7 mm.Leg spines: first tibia 2-2-0; fourth tibia 1-1-1.Epigynum (Figs. 13-14): spermathecum round,broadly joined to small round sac; connecting canalsclosely encircling spermathecal parts.49 50Figs. 49-52.-Ventral and retrolateral views of male palpi of Cicurina (Cicurella): 49-50, hexops Chamberlin and Ivie; 51-52,microps, new species.96

Type-data.-Female holotype from inside entranceof Heidrich's Cave, northwest of New Braunfels,Comal County, Texas, 19 March 1960 (yV. J.Gertsch, Wilton Ivie) (AMNH).Cicurina riogrande Gertsch and MulaikFigs. 3-4Cicurina riogrande Gertsch and Mulaik, in Chamberlinand Ivie, 1940, p. 76, pI. VIII, figs.57-58.Diagnosis.-Eight-eyed epigean species from RioGrande City, Starr County, Texas; small anteriormedian eyes half as large as anterior laterals; indexcoil of epigynum thin procurved loop. Male unknown.Etymology.-Specific name for Rio GrandeCity, Texas.Description.-Female holotype: Length 2.8 mIDlong. Carapace 1.55 mID long, 0.9 mID wide. Abdomen1.25 mID long, 1 mID wide. Clypeus narrow,about half diameter of anterior lateral eye. Eye rowsslightly procurved; posterior median eyes about twodiameters apart, about diameter from side eyes.Chelicerae: retromargin of fang with 5 teeth. Leglengths: first patella-tibia 1.1 mID, fourthpatella-tibia 1.15 mID.Epigynum (Figs. 3-4): round spermathecum narrowlyjoined to smaller suboval sac; index coil thinprocurved canal around spermathecal sac; connectingcanals rather closely circling spermathecum.Type-data.-Female holotype from 5 miles eastof Rio Grande City, Starr County, Texas, 12 January1939 (S. Mulaik) (AMNH).Cicurina microps Chamberlin and IvieChart 1, Fig. 16,51-52Cicurina microps Chamberlin and Ivie, 1940:77, pI.VIII, figs. 61-62, pI. XII, fig. 91.Diagnosis.-Six-eyed epigean species fromRaven Ranch, Kerr County, Texas; eyes (Chart 1)small, evanescent, pearly white spots in two separatedtriads; slender conductor of male palpus withvery long coil reaching end of cymbium (Figs.51-52). Assigned female from Brady, Texas, unavailable,known from Ivie illustrations (see Fig.16).Etymology.-Specific name from Greek microps,small eyes.Description.-Male holotype: Length 3.2 mID.Carapace 1.5 mID long, I mID wide. Abdomen 1.7mID long, 1 mID wide. Clypeus about diameter ofanterior lateral eye. Anterior lateral eyes 3 diametersapart, about radius from other eyes. Posterior eyerow moderately procurved; median eyes about twodiameters apart, about diameter from side eyes. Leglengths: first patella- tibia 1.1 mID, fourthpatella-tibia 1.4 mID.Male palpus: tegulum with two incomplete tubulesand one visible one on subtegulum; tarsalprocess with fold about third its width and adjacentspur of median size (Figs. 51-52).Female allotype: Length 2.8 mm. Carapace 1. 19mID long, 0.7 mID wide. Eyes and other characterspresumed to be similar to those of male.Type-data.-Male holotype from Raven Ranch,south of Kerrville, Kerr County, Texas, 16 December1939 (D. and S. Mulaik) (AMNH).Distribution.-As above for male only, and presumedfemale allotype: South of Brady, McCullochCounty, Texas, 12 December 1939 (D. and S. Mulaik)(AMNH).LIST AND DISCUSSION OF THEEYELESS TAXAThe subgenera Cicurata and Cicurella of Chamberlinand Ivie featured spiders of relatively smallsize with progressive loss of eyes nurtured in specialepigean and cavernicole habitats. The taxa studiedmainly by Ivie included his first completely eyelessspecies, which he named buwata and assigned to thestem genus Cicurina with special knowledge of themeaning of that genus. Inasmuch as buwata is anunknown taxon with unknown type locality comingfrom an area where numerous species are known tooccur, and not one can be singled out even as aplausible type species, the name buwata must bedropped as a nomina inquirienda of unknown status,and with it goes the unusable subgeneric name Cicurata.The subgenus Cicurella, with its well knowntype species, Cicurina microps Chamberlin and Ivie,has largely been used in a general but never in a genericsense by Ivie or following students. For thesake of editorial completeness the name buwata islisted at the end of the valid taxa of the genus.A list of the eyeless taxa follows: descriptions ofthe 50 valid species are given with information onstate and county localities of the United States andknown data on the Mexican records. In this paperthe generic name Cicurina covers all the taxa.Descriptions of the 50 eyeless taxa follows:TEXAS: Williamson County: browni, vibora,elliotti. new species, and buwata Chamberlin andIvie. Coryell County: coryelli, new species. Hays97

County: ezelli, russelli, and ubicki, new species.Travis County: travisae, reddelli, bandida, reyesi,wartoni, and cueva, new species. Bexar County:baronia, madia, vespera, and venii, new species.Comal County: reclusa and puentecilla, newspecIes. Medina County: medina, new species.Menard County: menardia, new species. RealCounty: sheari and orellia, new species. BanderaCounty: bandera, obscura, sprousei, and mckenzie;,new species. Kerr County: pastura and stowersi,new species. Val Verde County: patei, holsingeri,and porteri, new species. Sutton County: suttoni andbarri, new species. San Saba County: sansaba andmachete, new species. Terrell County: venefica,new species. Kimble County: caverna, new species.Edwards County: raines; and gruta, new species.Uvalde County: uvalde, watersi, pablo, serena, andselecta, new species. ALABAMA: JeffersonCounty: wiltoni, new species. YUCATAN: mayaGertsch. COAHUILA: coahuila Gertsch. NUEVOLEON: leona, new species.Cicurina browni, new speciesFigs. 53-54Diagnosis.-Eyeless troglobite of Brown's Cave,Williamson County, Texas; epigynum like that ofvibora with elongate spermathecum and large sac insuboblique position; two index coils crossing middleof sac; heavy connecting canals running beyond sacarea. Male unknown.Etymology.-Named for Brown's Cave.Description.-Female holotype: Length 5.6 mm.Carapace 2.25 mm long, 1.6 mm wide. Abdomen3.2 mm long, 2.4 mm wide. Chelicerae: retromarginof left side with 6 teeth, of right side 5. Leg lengths:first femur 2 mm, fourth femur 2.2 mm; firstpatella-tibia 2.4 mm, fourth patella-tibia 2.5 mm.Leg spines of first and fourth tibiae 2-2-2.Type-data.-Female holotype, subadult femaleand 3 immature from Brown's Cave, WilliamsonCounty, Texas, 23 April 1989 (W. Elliott, J. Reddell,M. Reyes) (AMNH).Cicurina vibora, new speciesFigs. 55-56Diagnosis.-Eyeless troglobite of RattlesnakeFilled Cave, Williamson County, Texas; ovatesperrnathecum with rounded sac close together atmidline, turned in oblique position; two index coilsdraped together around sac; connecting canals limitedto sac area. Male unknown.Etymology.-Specific name for Mexican vibora,viper.Description.-Female holotype: Length 5 mm.Carapace 2.5 mm long, 1.2 mm wide. Abdomen 2.5mm long, 1.6 mm wide. Chelicerae: retromargin offang with 5 teeth. Leg lengths: first femur 2 mm,fourth femur 2.1 mm; first patella-tibia 2.5 mm,fourth patella-tibia 2.8 mm. Leg spines: first andfourth tibiae 2-2-2.Type-data.-Female holotype and 2 immaturefrom Rattlesnake Filled Cave (southwest of CricketCave), 24 August 1963 (J. Reddell, W. Russell)(AMNH).Distribution.-TEXAS: Williamson County:Temples of Thor Cave, 13 May 1991 (1. Reddell,M. Reyes), 2 females, 5 immature.Cicurina holsingeri, new speciesFigs. 57-58Diagnosis.-Eyeless troglobite of Seminole CanyonCave, Val Verde County, Texas; round spermathecumwith short rounded sac; index coil a thinloop the length of the spermathecum; connectingcanal continuous with the index coil and formingloop around spermathecal parts. Male unknown.Etymology.-Named for Dr. John R. Holsingerof Old Dominion University, specialist on manycave animals.Description.-Female holotype: Length 2.7 mm.Carapace 1.4 mm long, 0.85 mm wide. Abdomen1.3 mm long, 0.8 mm wide. Chelicerae: retromarginof fang with 5 teeth. Leg lengths: first femur 1.1mm, fourth femur 1.3 mm; first patella-tibia 1.4mm, fourth patella-tibia 1.5 mm. Leg spines: firsttibia 2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from SeminoleCanyon Cave, Seminole State Park, Val VerdeCounty, Texas, 4 March 1983 (W.R. Elliott)(AMNH).Distribution.-Known only from Seminole CanyonCave. Other record: 21 May 1984 (Ralph L.Bement), 2 immature.Cicurina menardia, new speciesFigs. 59-60, 157-158Diagnosis.-Eyeless troglobite of Powell's Cave,Menard County, Texas; oval spermathecum withbroadly rounded sac; index coil heavy procurved canalacross sac; connecting canal fonning close looparound spermathecal parts. Male palpus (Figs.157-158).Etymology.-Specific name for Menard County.98

Description.-Female holotype: Length 3.9 mm.Carapace 1.9 mm long, 1 nun wide. Abdomen 2 mmlong, 1.2 rom wide. Chelicerae: retromargin of fangwith 5 teeth. Leg lengths: first femur 1.4 mm,fourth femur 1.6 mm; first patella-tibia 1.8 mm,fourth patella-tibia 2 mm. Leg spines: first andfourth tibiae 2-2-0.Male: Length 2.8 mm. Carapace 1.4 mm long,1.1 mm wide. Abdomen 1.4 rom long, 1.1 mmwide. Chelicerae: retromargin of fang with 5 teeth.Leg lengths: first femur 1.2 mm, fourth femur 1.4mm; first patella-tibia 1.5 mm, fourth patella-tibia1.5 mm. Leg spines: first and fourth tibiae 2-2-2.Male palpus (Figs. 157-158).Type-data.-Female holotype from Powell'sCave, 8 miles west of Menard, Menard County,Texas, 16 September 1978 (1. Reddell), from Msection maze (AMNH).Distribution.-Known only from Powell's Cave.Other records, 30 September 1967 (1. Reddell, D.Meredith), 2 females; 16 September 1964 (1. Reddell,D. McKenzie), 1 male, 3 females, 2 immature;26 January 1989 (W.R. Elliott), I penultimate male;28 January 1989 (W.R. Elliott), I immature; 28 October1989 (George Veni), 1 immature.Cicurina ezelli, new speciesFigs. 61-62Diagnosis.-Eyeless troglobite of Ezell's Cave,Hays County, Texas; ovate spermathecum with largerounded sac; index coil of epigynum with largegently procurved canal across sac in oblique position;connecting canal forming quite close looparound spermathecal parts. Male unknown.Etymology.-Specific name for Ezell's Cave.•• •• •••••• •••• • ••_-I~""'-"'-•••• ••••••••Map 2.-Distribution of eyeless Cicurina (Cicurella) spp.in Texas.99

555754565861 6364Figs. 53-64.-Ventral and dorsal views ofepigyna of eyeless Cicurina (Cicurella): 53-54, browni, new species; 55-56, vibora, newspecies; 57-58, holsingeri, new species; 59-60; menardia, new species; 61-62, ezelli, new species; 63-64, travisae, new species.100

Description.-Female holotype: Length 2.6 mm.Carapace 1.5 mm long, 0.8 mm wide. Abdomen 1.1mm long, 0.8 mm wide. Chelicerae: retromarginwith three teeth. Leg lengths: first femur 1.2 mm,fourth femur 1.4 mm; first patella-tibia 2.3 mm,fourth patella-tibia 1.5 mm. Leg spines: first tibia2-2-0, fourth tibia 2-2-1.Type-data.-Female holotype from Ezell's Cave,San Marcos, Hays County, Texas, 7 September1963 (J. Reddell, D. McKenzie, R. Ballinger)(AMNH).Distribution.-Known only from Ezell's Cave.Other records: 3 July 1978 (James C. Davis), 1 female,18 July 1978 (James C. Davis), 1 female, 16January 1978 (James C. Davis), 4 immature; 30January 1965 (J. Reddell), 1 immature; 26 October1967 (J. Reddell), 1 immature.Cicurina travisae, new speciesFigs. 63-70Diagnosis.-Eyeless troglobite of Tooth Caveand adjacent caves of Travis County, Texas; oblongsperrnathecum broadly joined to rounded or laterallyproduced sac; index coil rounded to angled loop inprocurved position; connecting canal of medium sizeforming loop well outside of sperrnathecal parts.Male unknown.Etymology.-Specific name for Mrs. NevennaTsanoff Travis, tireless sponsor for preservation ofcaves.Description.-Female holotype: Length 5 mm.Carapace 2.2 mm long, 1.42 mm wide. Abdomen2.9 mm long, 2.2 mm wide. Chelicerae: retromarginof fang with 5 teeth. Leg lengths: first femur 2.5mm, fourth femur 2.7 mm.Type-data.-Female holotype, female, and 10immature from Tooth Cave, 15 miles NW ofAustin, Travis County, Texas, 5 August 1963 (J.Reddell) (AMNH).Distribution.-TEXAS: Travis County: ToothCave, 5 March 1964 (1. Reddell, D. McKenzie, T.Phillips), 2 females, 8 immature; 9 June 1964 (R.Mitchell), 6 immature; 14 May 1966 (1. Reddell), Ifemale, 6 immature; 9 June 1967 (R.W. Mitchell), 6immature; 7 April 1988 (D. Pate, W. Elliott), 4immature; 19 August 1970 (J. Reddell), 1 female.Kretschmarr Cave, 2 March 1963 (1. Reddell, D.McKenzie), 5 immature; 2 June 1963 (J. Reddell,W.H. Russell), I female, 2 immature; 13 September1963 (J. Reddell, B. Russell), 1 immature; 23 June1968 (J. Reddell, R. W. Mitchell), 1 female, 2 immature;21 February 1988 (M. Reyes), 1 female, 1immature; 20 January 1963 (D. McKenzie), 1female. Pisarowicz Cave, 21 April 1984 (J. Reddell,M. Reyes), 1 female, 1 immature; SalamanderCave, 26 April 1963 (J. Reddell, B. Russell), 11immature; 6 April 1966 (J. Reddell), 1 female; RootCave, 12 July 1984 (J. Reddell, M. Reyes), 4immature; 1 April 1989 (J. Reddell, M. Reyes), Ifemale, 2 immature; McDonald Cave (=SchulzeCave), 21 August 1963 (B. Russell), 12 immature; 4October 1964 (J. Reddell), 1 immature; 12 May1984 (J. Reddell), 1 immature; 18 May 1984 (D.Pate, J. Reddell, M. Reyes), 1 female, 2 immature;15 December 1988 (J. Reddell, M. Reyes), 2 females,numerous immatures; 29 May 1989 (W.Elliott, J. Reddell, M. Reyes), 5 females, 14immature. Amber Cave, 8 April 1984 (J. Reddell,M. Reyes), 1 female.Cicurina wartoni, new speciesFigs. 75-76Diagnosis.-Eyeless troglobite of Pickle PitCave, Travis County, Texas; elongate spermathecumwith large rounded sac in inclined positionwith sacs widely separated; index coil shortprocurved canal lying across sac; connecting canalforming loose loop around sperrnathecal parts. Maleunknown.Etymology.-Named for Mike Warton, diligentstudent of caves.Description.-Female holotype: Length 6 mm.Carapace 3 mm long, 2.5 mm wide. Abdomen 3 mmlong, 2 mm wide. Chelicerae: retromargin of fangwith 5 teeth. Leg lengths: first femur 2 mm, fourthfemur 2.3 mm; first patella-tibia 3 mm, fourthpatella-tibia 3.5 mm. Leg spines: ventral spines offirst and fourth tibiae 2-2-2.Type-data.-Female holotype from Pickle PitCave, 21 May 1990 (J. Reddell, M. Reyes, L. Sherrod)(AMNH).Cicurina elliotti, new speciesFigs. 73-74Diagnosis.-Eyeless troglobite of Travis andWilliamson Counties; elongate spermathecum andlarge rounded sac in mostly erect position; indexcoil heavy canal forming rounded or angled loop;connecting canal forming quite wide loop aroundsperrnathecal parts. Male unknown.Etymology.-Named for Dr. William R. Elliott.Description.-Female (Beck's Sewer Cave):Length 5.2 mm. Carapace 2.6 mm long, 1.6 mmwide. Abdomen 2.6 mm long, 2 mm wide.Chelicerae: retromargin of fang with 6 teeth. Leg101

65 6768 707172Figs. 65-76.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 65-66, travisae, new species; 67-68, travisae,new species; 69-70, travisae, new species, aberrant; 71-72, coryelli, new species; 73-74, elliotti, new species; 75-76, wartoni, newspecies.102

lengths: first femur 2 rom, fourth femur 2.3 mm;first and fourth tibiae 2-2-2.Type-data.-Female holotype from Beck'sSewer Cave, 27 January 1965 (1. Calvert, J. Reddell)(AMNH).Distribution.-TEXAS: Travis County: CotterellCave, 18 May 1988 (W. Elliott, J. Reddell, M.Reyes), female and 4 immature. Fossil GardenCave, 6 June 1990 (1. Reddell, M. Reyes), 1 female,1 penultimate male, 1 immature. GalliferCave, (1. Reddell, M. Reyes), 20 April 1991, 2females. Williamson County: Bev's Grotto, 16 April1989 (W. Elliott, J. Reddell, M. Reyes), 1 female.Beck's Sewer Cave, 23 January 1965 (J. Reddell, R.Mitchell), 1 female; 27 January 1965 (J. Calvert, J.Reddell), 1 female; 28 February 1987 (J. Reddell,M. Reyes), 1 female,S immature. Buttercup RiverCave (B. Larsen, W. Russell), 1 female, 1 immature.Good Friday Cave, 7 March 1989 (W. Elliott,M. Reyes), 2 females. McNeil Quarry Cave, 22 July1963 (1. Reddell, B. Russell), I immature. MarigoldCave, 6 August 1988 (1. Reddell, B. Larsen), 1 female,1 immature; 18 September 1988 (PeterSprouse), 1 female, 2 immature. Testudo Tube, 29May 1991 (J. Reddell, M. Reyes), 1 female, 1 immature;11 May 1991 (J. Reddell, M. Reyes), 12 ormore immature. T.W.A.S. A Cave (W. Elliott, J.Reddell, M. Reyes), 15 April 1989, I female, 1immature.CiCUri.1Ul coryelli, new speciesFigs. 71-72Diagnosis.-Eyeless troglobite of Tippit Cave,Coryell County, Texas; elongated sperrnathecumwith round sac; index coil narrow lightly procurvedband across sac; connecting canal forming closeloop partially covering base of sperrnathecal parts.Male unknown.Etymology.-Specific name for Coryell County,Texas.Description.-Female holotype: Length 4.3 rom.Carapace 1.3 rom long, 1 rom wide. Abdomen 3 mmlong, 1.6 mm wide. Chelicera: retromargin with 6teeth. Leg lengths: first femur 1.65 mm long, fourthfemur 2.5 rom long; first patella-tibia 1.6 mm long,fourth patella-tibia 2 rom long. Leg spines: ventralspines of first and fourth tibiae 2-2-2.Type-data.-Female holotype, 2 subadultfemales from Tippit Cave, Coryell County, Texas,31 January 1992 (1. Reddell, M. Reyes) (AMNH).Distribution.-Known only from Tippit Cave.Other records: 9 March 1963 (1. Reddell, D.McKenzie), 1 immature; 24 January 1992 (D.McKenzie, J. Reddell, M. Reyes), 2 females.,immature.Cicuri.1Ul uvalde, new speciesFigs. 101-102Diagnosis.-Eyeless troglobite of Rambie'sCave, Uvalde County, Texas; oval sperrnathecumset in oblique position, with large oval sac; indexcoil heavy straight canal in oblique position; connectingcanal forming close loop around spermathecalparts. Male unknown.Etymology.-Specific name for Uvalde County.Description.-Female holotype: Length 5.5 mm.Carapace 2.5 mm long, 1.16 mm wide. Abdomen 3mm long, 1.9 mm wide. Chelicerae: retromargin offang with 6 or 7 teeth. Leg lengths: first femur 2.2rom, fourth femur 2.4 rom; first patella-tibia 3 mm,fourth patella-tibia 2.4 mm. Leg spines: first andfourth tibiae with 2-2-2 ventral spines.Type-data.-Female holotype and 18 immaturefrom Rambie's Cave, 8 miles N of Uvalde, UvaldeCounty, Texas, 6 April 1963 (J. Reddell, D.McKenzie) (AMNH).Distribution.-Known only from Rambie'sCave. Other records: 14 August 1976 (J. Reddell,C. Yates, R. Fieseler, D. Pate), 1 female, 4 immature;4-5 September 1976 (D. Pate, C. Yates), 1female, 1 immature.Cicuri.1Ul watersi, new speciesFigs. 103-104Diagnosis.-Eyeless troglobite of Frio QueenCave, Uvalde County, Texas; oval spermathecumwith rounded sac; index coil transverse, moderatelyprocurved canal across sac; thin connecting canalclosely encircling sperrnathecal parts. Male unknown.Etymology.-Specific name for the collector,Randy M. Waters.Description.-Female holotype: Length 4.6 mm.Carapace 2.4 mm long, 1.5 rom wide. Abdomen 2.2mm long, 1.6 mm wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 1.6 mm,fourth femur 1.9 rom; first patella-tibia 2.8 mm,fourth patella-tibia 3 rom. Ventral spines of first andfourth tibiae 2-2-2.Type-data.-Female holotype and 1 immaturefrom Frio Queen Cave, Uvalde County, Texas,Summer 1983 (Randy M. Waters) (AMNH).103

828587848688Figs. 77-88.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 77-78, reddelli, new species; 79-80, bandida,new species; 81-82, cueva, new species; 83-84, russelli, new species; 85-86, reyesi, new species; 87-88, ubicki, new species.104

Cicurina pablo, new speciesFigs. 105-106Diagnosis.-Eyeless troglobite of Pablo's Cave,Uvalde County, Texas; elongate spermathecum withlarge rounded sac; index coil heavy transverse canalin oblique position; heavy connecting canal enclosingspermathecal parts. Male unknown.Etymology.-Specific name for Pablo's Cave,used in apposition.Description.-Female holotype: Length 5 mrn.Carapace 2.5 mrn long, 1.8 mrn wide. Abdomen 2.5mrn long, 1.6 mrn wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 2.3 mrn,fourth femur 2.6 mrn; first patella-tibia 2.6 mrn,fourth patella-tibia 3.2 mrn. Ventral spines of firsttibia 2-2-1, offourth tibia 2-2-2.Type-data.-Female holotype and 4 immaturefrom Pablo's Cave (2 mi. N Burial Cave), 5 April1963 (J. Reddell, D. McKenzie), in total darkness50 feet from entrance (AMNH).Cicurina orellia,new speciesFigs. 107-108Diagnosis.-Eyeless troglobite of Orell CreviceCave, Real County, Texas; epigynum broader thanlong; spermathecum elongated with apical endsnearly touching in oblique position; index coil smallcurved canal across sac; connecting canal formingwide loop around spermathecal parts. Male unknown.Etymology.-Named for Orell Crevice Cave.Description.-Female holotype: Length 5.2 mrn.Carapace 2.6 mrn long, 1.5 mrn wide. Abdomen 2.6mrn long, 0.13 mrn wide. Chelicerae: retromargin offang with 5 teeth, three of them enlarged. Leglengths: first femur 2.4 mm, fourth femur 2.6 mrn;first patella-tibia 3 mm. Leg spines: first tibia 2-21, fourth tibia 2-2-2.Type-data.-Female holotype and 2 immaturefrom Orell Crevice Cave, 100 yards west of OrellBat Cave, 18 August 1963 (J. Reddell, D. McKenzie)(AMNH).Cicurina serena, new speciesFigs. 109-110Diagnosis.-Eyeless troglobite of Picture CaveNo.1, Uvalde County, Texas; oblong spermathecumwith rounded sac; index coil thickstraight bar; connecting canal closely enclosingspermathecal parts. Male unknown.Etymology.-Species name for Latin serenus,serene.Description.-Female holotype: Length 3.7 mm.Carapace 1.7 mm long, 1.2 mm wide. Abdomen 2mm long, 1.4 mm wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 1.6 mm,fourth femur 1.8 mm; first patella-tibia 1.6 mm,fourth patella-tibia 12.8 mrn. Leg spines: first tibia2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from Picture CaveNo.1, 23 miles NW Uvalde, 3 November 1962 (J.Reddell) (AMNH).Distribution.-Uvalde County: Picture Cave No.1, 7 March 1965 (J. Reddell), 2 immature. NorthWell Cave, 2 mi. N of Pablo's Cave, 3 April 1963(1. Reddell, D. McKenzie), 1 female.Cicurina selecta, new speciesFigs. 111-112Diagnosis.-Eyeless troglobite of SandtlebenCave (=Davy Crockett Cave), Uvalde County,Texas; short oval spermathecum with wide sac inoblique position; index coil narrow transverse canal;connecting canal forming loose loop around spermathecalparts. Male unknown.Etymology.-Specific name from Latin selectus,to choose.Description.-Female holotype: Length 4.4 mm.Carapace 2.2 mrn long, 1.6 mm wide. Abdomen 2.2mrn long, 1.7 mm wide. Chelicerae: retromargin offang with 5 teeth. Leg lengths: first femur 2 mm,fourth femur 2.4 mm; first patella-tibia 2.8 mm,fourth patella-tibia 3.2 mm. Leg spines: first andfourth tibiae 2-2-2.Type-data.-Female holotype from SandtlebenCave (=Davy Crockett Cave), Uvalde County,Texas, 18 October 1964 (J. Reddell) (AMNH).Distribution.-Known only from SandtlebenCave. Other record: 13 August 1965 (1. Fish, J.Reddell), 1 immature.Cicurina reddelli, new speciesFigs. 77-78Diagnosis.-Eyeless troglobite of Cotterell Cave,Travis County, Texas; long ovate spermathecumwith rounded, laterally directed sac; index coil ofepigynum an angled procurved loop, and connectingcanal of medium size lying outside of spermathecalparts. Male unknown.Etymology.-Named for James R. Reddell, specialiston Texas caves.105

89909998100Figs. 89-100.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 89-90, baronia, new species; 91-92, madia,new species; 93-94, vespera, new species; 95-96, venii, new species; 97-98, reclusa, new species; 99-100, puentecilla, new species.106

Description.-Female holotype: Length 6.9 mm.Carapace 3.6 mm long, 2.4 mm wide. Abdomen 3.3mm long, 2 mm wide. Chelicerae: retromargin with7 teeth, basal 3 large and with 4 denticles. Leglengths: first femur 3.2 mm, fourth femur 3.6 mm.Leg spines: first and fourth tibiae with 2-2-2 ventralspmes.Type-data.-Female holotype from CotterellCave, near Spicewood Springs Road and MesaDrive, Austin, Travis County, Texas, 11 March1964 (W. Russell) (AMNH).Cicurina banditkl, new speciesFigs. 79-80Diagnosis.-Eyeless troglobite of Bandit Cave,Travis County, Texas; large round spermathecumwith broad oval sac in oblique position; index coilof epigynum short canal at oblique angle across sac;connecting canal heavy tube closely marginingsperrnathecal parts. Male unknown.Etymology.-Specific name from Spanish bandido,bandit, named for Bandit Cave.Description.-Female holotype: Length 5.6 mm.Carapace 2.6 mm long, 2 mm wide. Abdomen 3 mmlong, 2.4 mm wide. Chelicerae: retromargin with 6teeth. Leg lengths: first femur 2.8 mm, fourth femur3 mm; first patella-tibia 3.6 mm; fourth patella-tibia3.9 mm; fourth metatarsus and tarsus 4.6 mm. Legspines: first and fourth tibiae 2-2-2.Type-data.-Female holotype and 2 immaturefrom Bandit Cave, Rollingwood, Travis County,Texas, 26 May 1966 (J. Reddell, J. Fish) (AMNH).Distribution.-Travis County: Bandit Cave, 27May 1963 (J. Reddell, B. Frank), 1 female, 7immature; 20 May 1965 (J. Reddell), 2 immature;13 September 1988 (W. Elliott), 1 immature. Ireland'sCave, 14 March 1964 (W. Russell), 1 female,8 immature; 10 April 1964 (W. Russell), 1 immature;1 March 1986 (Dale L. Pate), 2 immature; 23January 1989 (1. Reddell, M. Reyes, E. Grimm, M.Grimm), 1 aberrant female, 1 immature.Cicurina cueva, new speciesFigs. 81-82Diagnosis.-Eyeless troglobite of Cave X, TravisCounty, Texas; rounded spermathecum with longround sac in oblique position; index coil ofepigynum curved canal as long as sperrnathecum, setin subvertical position; connecting canal lying closeto sperrnathecal parts. Male unknown.Etymology.-Specific name from Spanish cueva,cave.Description.-Female holotype: Length 5.4 mm.Carapace 2.3 mm long, 1.3 mm wide. Abdomen 2.4mm long, 1.3 mm wide. Chelicerae: retromargin offang with 6 teeth, basal three enlarged and wellspaced. Leg lengths: first femur 2.1 mm, fourth femur2.3 mm; first patella-tibia 2.8 mm, fourthpatella-tibia 3 mm. Leg spines: first and fourthtibiae 2-2-2.Type-data.-Female holotype from Cave X,Travis County, Texas, September 1962 (B. Bell, S.Woolsey) (AMNH).Distribution.-Travis County: Cave X, 8 March1963 (D. McKenzie, J. Reddell), 1 female, 2immature; 14 March 1964 (J. Reddell), 1 immature;October 1970 (J. Reddell), 2 immature. Flint RidgeCave, August 1989 (M. Grimm), 1 aberrant female,1 immature; April 1989 (J. Reddell, M. Reyes), 3immature; 19 June 1989 (M. Grimm, J. Reddell, M.Reyes), 1 female, 3 immature.Cicurina russelli, new speciesFigs. 83-84Diagnosis.-Eyeless troglobite of Boyett's Cave,Hays County, Texas; ovoid spermathecum withlarge round twisted sac in oblique position; indexcoil heavy canal moderately curved to top ofsperrnathecum; connecting canal forming ratherclose loop around spermathecal parts. Maleunknown.Etymology.-Named for William Russell.Description.-Female holotype: Length 5.8 mm.Carapace 2.8 mm long, 2.3 mm wide. Abdomen 3mm long, 2.3 mm wide. Chelicerae: retromarginwith 5 stout teeth. Leg lengths: first femur 2.8 mm,fourth femur 3 mm; first patella-tibia 3 mm, fourthpatella-tibia 3.6 mm. Leg spines: first tibia 2- 2-2,fourth tibia 2-2-1.Type-data.-Female holotype and 3 immaturefrom Boyett's Cave, south of Wimberley onHays-Comal County line, 30 March 1963 (J.Reddell, W. Russell) (AMNH).Distribution.-Known only from Boyett's Cave.Other records: 19 October 1963 (1. Reddell, J.Porter), 1 immature; 23 June 1978 (W. Elliott, J.Holsinger, T. Poulson, F. Howarth), 2 immature.Cicurina reyesi, new speciesFigs. 85-86Diagnosis.-Eyeless troglobite of Airman'sCave, Travis County, Texas; ovate sperrnathecumwith rounded sac; index coil thin slightly curved107

103104109Figs. 101-112.-Venlral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 101-102, uvalde, new species; 103-104,watersi, new species; 105-106, pablo, new species; 107-108, orellia, new species; 109-110, serena, new species; 111-112, selecta,new species.108

canal in nearly vertical posItion; thin connectingcanal lying just outside of spermathecum.Etymology.-Named for Marcelino Reyes.Description.-Female holotype: Length 5 mm.Carapace 2.5 mm long, 1.5 mm wide. Abdomen 2.4mm long, 1.5 mm wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 2 mm,fourth femur 2.1 mm. Leg spines: first and fourthtibiae with 1-2-2.Type-data.-Female holotype, damaged maleand 5 immature from Airman's Cave, Austin, TravisCounty, Texas, 3 September 1989 (J. Reddell, M.Reyes) (AMNH).Distribution.-Known only from Airman'sCave. Other records: 27 September 1975 (A.Grubbs, L. Wilk), 3 immature; 13 March 1982 (S.Robertson, M. Williams), 1 immature; 14 May1984 (J. Reddell, M. Reyes), 2 immature; 1 June1984 (J. Reddell, M. Reyes), 5 immature.Cicurina ubicki, new speciesFigs. 87-88Diagnosis.-Eyeless troglobite of Fern andMcGlothlin Caves, Hays County, Texas; ovatespermathecum with heavy oval sac in suhlateralposition; index coil heavy canal in nearly verticalposition; heavy connecting canal closely marginingspermathecal parts. Male unknown.Etymology.-Specific name for Darrell Ubick,collector of many cave spiders.Description.-Female ho1otype: Carapace 2.6mm long, 1.7 mm wide. Abdomen 2.6 mm long,1.7 mm wide. Chelicera: promargin with 5 teeth.Leg lengths: first femur 2.5 mm, fourth femur 2.7mm; first patella-tibia 2.6 mm, fourth patella-tibia3.0 mm. Leg spines: first and fourth tibiae 2-2-2.Type-data.-Female holotype and 2 immaturefrom Fern Cave, Hays County, Texas, 2 September1989 (D. Ubick, S. Fend, S. Renkes), deposited inAMNH, courtesy of Darrell Ubick.Distribution.-Hays County: Fern Cave, 15 July1988 (A. Grubbs, J. Evans, L. Schneider), 2immature. McGlothlin Cave, 26 May 1989 (A.Grubbs, J. Reddell, M. Reyes), 1 female, 2immature; 3 September 1989 (D. Ubick, S. F('Jld, S.Renkes), 1 female, 7 immature.Cicurina baronia, new speciesFigs. 89-90, 155-156Diagnosis.-Eyeless troglobite of Robber BaronCave, Bexar County, Texas; epigynum wider thanlong; broader than long spermathecum with roundedsac; index coil heavy procurved bar across sac;heavy connecting canal forming loose loop aroundspermathecal parts. Male palpus (Figs. 155-156).Etymology.-Specific name for Robber BaronCave.Description.-Female holotype: Length 6 mm.Carapace 2.6 mm long, 1.7 mm wide. Abdomen 3.4mm long, 2.6 mm wide. Chelicerae: retromarginwith 4 teeth. Leg lengths: first femur 2.2 mm,fourth femur 2.25 mm; first patella-tibia 2.4 mm,fourth patella-tibia 2.6 mm. Leg spines: first tibia2-2-1, fourth tibia 2-2-2 ventral spines.Male: Length 3 mm. Carapace 1.4 mm long, 1.1mm wide. Abdomen 1.6 mm long, 1.1 mm wide.Chelicerae: retromargin with 6 teeth. Leg lengths:first femur 1.5 mm, fourth femur 1.6 mm; firstpatella-tibia 1.4 mm, fourth patella-tibia 1.7 mm.Leg spines: first femur 2-2-2, fourth femur 2-2-2.Male palpus (Figs. 155-156).Type-data.-Female holotype from RobberBaron Cave, San Antonio, Bexar County, Texas,April 1969 (Roger Bartholomew) (AMNH).Distribution.-Known only from Robber BaronCave. Other records: 9, 11 December 1983 (ScottHarden, Randy Waters), 2 males, 2 immature; 10March 1982 (A.G. Grubbs, B. Steele, R. Waters), 1immature; 3 April 1982 (A. Grubbs), 2 females, 8immature; 6 April 1983 (Randy Waters), 1immature; 11 December 1982 (Randy Waters), 1immature; 3 September 1987 (Allan Cobb, GeorgeVeni), 3 immature.Cicurina madia, new speciesFigs. 91-92Diagnosis.-Eyeless troglobite from Madia'sCave, Bexar County, Texas; elongate spermathecumwith rounded sac; index coil straight horizontal barwith downward turn; connecting canal loosely coiledaround spermathecal parts. Male unknown.Etymology.-Specific name for Madia's Cave.Description.-Female holotype: Length 5.8 mm.Carapace 2.4 mm long, 1.7 mm wide. Abdomen 3.4mm long, 2 mm wide. Chelicerae: retromargin offang with 4 teeth. Leg lengths: first femur 2.6 mm,fourth femur 2.8 nun; first patella-tibia 3 rom,fourth patella-tibia 3.3 rom. Leg spines: first tibia2-2--, fourth tibia 2-2-2.Type-data.-Female holotype and 5 immaturefrom Madia's Cave, 4 miles NW Helotes, BexarCounty, Texas, 4 October 1963 (J. Reddell, D.McKenzie) (AMNH).109

113114123120Figs. I13-124.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 113-114, bandera, new species; 115-116,obscura, new species; 117-118, palei, new species; 119-120, sprousei, new species; 121-122, slowersi, new species; 123-124, paslura,new species.110

Cicurina vespera, new speciesFigs. 93-94Diagnosis.-Eyeless troglobite from GovernmentCanyon Bat Cave, Bexar County, Texas; roundspermathecum with small rounded sac; index coilrecurved bar across sac; connecting canal formingquite close loop around spermathecal parts.Epigynum broader than long. Male unknown.Etymology.-Specific name from Latin vespera,in the evening.Description.-Female holotype: Length 2.7 mm.Carapace 1 mm long, 0.45 mm wide. Abdomen 1.7mm long, 1.2 mm wide. Chelicerae: retromarginwith 4 teeth. Leg lengths: first femur 0.8 mm,fourth femur 0.9 mm; first tibia-patella 0.9 mm,fourth patella-tibia 1 mm. Leg spines: first andfourth tibiae 2-2-1.Type-data.-Female holotype from GovernmentCanyon Bat Cave, Bexar County, Texas, 11 August1965 (J. Reddell, J. Fish) (AMNH).Cicurina venii, new speciesFigs. 95-96Diagnosis.-Eyeless troglobite from Braken BatCave, Bexar County, Texas; elongate spermathecummoderately curved inward, with rounded sac; indexcoil an essentially straight bar across sac; connectingcanal forming loose loop around spermathecal parts.Male unknown.Etymology.-Named for George Veni, student ofTexas caves.Description.-Female holotype: Length 3.4 mm.Carapace 1.7 mm long, 1.3 mm wide. Abdomen 1.7mm long, 1 mm wide. Chelicerae with 4 or 5 teeth.Leg lengths: first femur 1.8 mm; fourth femur 2mm; first patella-tibia 1.8 mm, fourth patella-tibia 2mm. Leg spines: first tibia 2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from Braken BatCave, Bexar County, Texas, 22 November 1980(George Veni) (AMNH).Cicurina reclusa, new speciesFigs. 97-98Diagnosis.-Eyeless troglobite of KappelmanSalamander Cave, Comal County, Texas; epigynumbroader than long; oval spermathecum with smalltwisted sac; index coil heavy moderately curvedcanal across sac; connecting canal forming quitetight loop around spermathecal parts. Maleunknown.Etymology.-Specific name from Latin recludere,a recluse.Description.-Female holotype: Length 3.9 mm.Carapace 1.8 mm long, 1.2 mm wide. Abdomen 2.1mm long, 1.4 mm wide. Chelicerae: retromargin offang with 3 teeth on left, and nodule on right. Leglengths: First femur 1.5 mm, fourth femur 1.8 mm;first patella-tibia 2 mm, fourth patella-tibia 2.2 mm.Leg spines: first tibia 2-2-1, fourth tibia 2-2-2.Type-data.-Female holotype from KappelmanSalamander Cave, 100 yards NW of KappelmanCave, Comal County, Texas, 15 March 1964 (W.Russell, J. Reddell) (AMNH).Distribution.-TEXAS: Comal County: KappelmanCave, 16 mi. NW of New Braunfels, 9March 1968 (1. Reddell), 1 female. KappelmanSalamander Cave, 14 January 1965 (J. Reddell, T.Raines), penultimate male.Cicurina puentecilla, new speciesFigs. 99-100Diagnosis.-Eyeless troglobite from NaturalBridge Caverns, Comal County, Texas; epigynumbroader than long; round spermathecum with sac inlateral position; index coil heavy slightly curvedcanal in vertical position; heavy connecting canalrunning full width of spermathecum. Male unknown.Etymology.-Specific name from Spanish puenrecilia,little bridge.Description.-Female holotype: Length 5.8 mm.Carapace 2.8 mm long, 2 mm wide. Abdomen 3 mmlong, 1.8 mm wide. Chelicerae: promargin with 4widely spaced teeth. Leg lengths: first femur 2.6mm, fourth femur 3 mm; first patella-tibia 3.8 mm,fourth patella-tibia 4 mm. Leg spines: first tibia2-2-1, fourth tibia 2-2-2.Type-data.-Female holotype and two immaturefrom Natural Bridge Caverns, Comal County,Texas, 2 September 1978 (A.G. Grubbs) (AMNH).Distribution.-Known only from Natural BridgeCaverns. Other records: 1 April 1965 (J. Reddell,T. Raines), 2 immature; 23 February 1963 (J. Reddell,D. McKenzie), 2 immature; 23 September1989 (0. Knox), 1 immature; 1 March 1990 (0.Knox, J. Reddell, M. Reyes), 4 immature.Cicurina bandera, new speciesFigs. 113-114Diagnosis.-Eyeless troglobite from Fossil Cave,Bandera County, Texas; elongate spermathecumwith rounded sac; index coil thick moderatelyIII

125126131132 136Figs. 125-136.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 125-126, machete, new species; 127-128,sansaba, new species; 129-130, venefica, new species; 131-132, cavema, new species; 133-134, porteri, new species; 135-136,sheari, new species.112

procurved canal across sac; connecting canal widelyencircling spermathecal parts. Male unknown.Etymology.-Named for Bandera County.Description.-Female holotype: Length 4.7 mm.Carapace 2.3 mm long, 1.6 mm wide. Abdomen 2.4mm long, 1.6 mm wide. Chelicerae: retromargin offang with 4 or 5 teeth. Leg lengths: first femur 2.2mm, fourth femur 2.4 mm; first patella-tibia 2.4mm, fourth patella-tibia 2.8 rom. Leg spines: firsttibia 2-2-2, fourth tibia 2-2-1.Type-data.-Female holotype and 6 immaturefrom Fossil Cave, Bandera County, Texas, 23 July1966 (J. Reddell, D. McKenzie) (AMNH).Distribution.-Known only from Fossil Cave.Other record: 21 March 1971 (1. Reddell, T. Mollhagen,S. Wiley), 1 female, 2 immature.Cicurina obscura, new speciesFigs. 115-116Diagnosis.-Eyeless troglobite from SutherlandHollow Cave, Bandera County, Texas; ovate spermathecumwith larger rounded sac; index canalheavy procurved canal forming deep inverted loopthick on inner side; connecting canal fonning wideloop around spermathecal parts. Male unknown.Etymology.-Specific name from Latin obscurus,obscure.Description.-Female holotype: Length about 3mm. Carapace 1.8 mm long, 1.2 mm wide. Abdomenmissing. Chelicerae: retromargin of fang with 5teeth. Leg lengths: first femur 2 mm, fourth femur2.2 mm; first patella-tibia 2.5 mm, fourthpatella-tibia 2.6 rom. Leg spines: first tibia 2-2-1,fourth tibia 2-2-2.Type-data.-Female holotype from SutherlandHollow Cave, Bandera County, Texas, 4 August1974 (S. Sweet) (AMNH).Cicurina patei, new speciesFigs. 117-118Diagnosis.-Eyeless troglobite of Fawcett'sCave, Val Verde County, Texas; elongated spermathecumwith laterally curved sac; index coilbroad essentially straight canal crossing sac; connectingcanal widely encircling spermathecal parts.Male unknown.Etymology.-Named for Dale Pate.Description.-Female holotype: Length 4.4 mm.Carapace 2 mm long, 1.5 mm wide. Abdomen 2.4mm long, 1.6 mm wide. Chelicerae: retromargin offang with 5 teeth. Leg lengths: first femur 2.1 mm,fourth femur 2.1 rom; first patella-tibia 2.3 mm,fourth patella-tibia 2.5 mm. Leg spines: first tibia2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from Fawcett'sCave, 36 miles N of Del Rio, Val Verde County,Texas, 8 August 1987 (Dale Pate) (AMNH).Distribution.-Known only from Fawcett'sCave. Other record: 10 April 1968 (J. Reddell), 1female, 2 immature in AMNH.Cicurina sprousei, new speciesFigs. 119-120Diagnosis.-Eyeless troglobite of Station "C"Cave #1, Bandera County, Texas; oval spermathecumwith small rounded sac; index coil quitethick procurved canal; connecting canal closelyringing spermathecal parts. Male unknown.Etymology.-Specific name for Peter Sprouse,student of caves.Description.-Female holotype: Length 3.85rom. Carapace 2.6 mm long, 1.6 mm wide. Abdomen1.25 mm long, 1.6 mm wide. Chelicerae:retromargin of fang with 6 teeth. Leg lengths: firstfemur 1 mm, fourth femur 1.2 mm; firstpatella-tibia 2.8 mm, fourth patella-tibia 3 mm. Legspines: first tibia 2-2-1, fourth tibia 2-2-2.Type-data.-Female holotype, 7 immature fromStation "CO Cave #1, 15 miles N of Vanderpool,Bandera County, Texas, 4 September 1988 (P.Sprouse) (AMNH).Distribution.-Known only from Station "C"Cave. Other record: August 1962 (J. Reddell), 2immature; 30 October 1963 (D. McKenzie), 1 female,I immature.Cicurina stowersi, new speciesFigs. 121-122Diagnosis.-Eyeless troglobite of Stowers Cave,Kerr County, Texas; oval spermathecum with broadrounded sac; index coil thick procurved canal; slenderconnecting canal forming wide loop aroundspermathecal parts. Male unknown.Etymology.-Specific name for Stowers Cave.Description.-Female holotype: Length 3.3 mm.Carapace 1.6 mm long, 1.1 nun wide. Abdomen 1.7mm long, 0.8 nun wide. Chelicerae: retromargin offang with 6 small teeth. Leg lengths: first femur 1.4mm, fourth femur 1.8 mm; first patella-tibia 2 mm,fourth patella-tibia 2.1 mm. Leg spines: first andfourth tibiae 2-2-0.113

Type-data.-Female holotype from StowersCave, 24 miles W Kerrville, Kerr County, Texas, 3May 1969 (Roger Bartholomew) (AMNH).Distribution.-Known only from Stowers Cave.Other records: 20 March 1965 (J. Reddell), 5 immature;25 March 1972 (S. Wiley, T. Mollhagen), 1immatureCicurina pastura, new speciesFigs. 123-124Diagnosis.-Eyeless troglobite from Water PondPasture Cave, Kerr County, Texas; oval spermathecumwith rounded sac set in oblique position,with tips nearly touching; index coil thick canal inclinedoutward; connecting canal loosely enclosingspermathecal parts. Male unknown.Etymology.-Specific name from Latin pasture,pasture.Description.-Female holotype: Length 4 mm.Carapace 2 mm long, 1.4 mm wide. Abdomen 2 mmlong, 1.4 mm wide. Chelicerae: retromargin with 5teeth. Leg lengths: first femur 1.8 mm, fourth femur2 mm; first patella-tibia 2.2 mm, fourth patella-tibia2.5 mm. Leg spines: first tibia 2-2-0, fourth tibia2-2-2.Type-data.-Female bolotype from Water PondPasture Cave, Kerr County, Texas, 16 October 1976(D. Pate, R. Fieseler, C. Yates) (AMNH).Cicurina machete, new speciesFigs. 125-126Diagnosis.-Eyeless troglobite of WhitefaceCave, San Saba County, Texas; oval sperrnathecumwith twisted sac; index coil heavy procurvedblade-like canal crossing sac; connecting canalheavy loop around sperrnathecal parts. Male unknown.Etymology.-Specific name for Spanish machete,cutlass, used in apposition.Description.-Female holotype: Length 4.6 mm.Carapace 2.4 mm long, 1.6 mm wide. Abdomen 2.2mm long, 1.4 mm wide. Chelicerae: retromargin offang with 8 teeth. Leg lengths: first femur 2.4 mm,fourth femur 2.8 mm; first patella-tibia 3.2 mm,fourth patella-tibia 3.4 mm. Leg spines: first tibia2-2-2, fourth tibia 2-2-0.Type-data.-Female holotype and 4 immaturefrom Whiteface Cave, 30 mi. S Richland Springs, 9February 1964 (J. Reddell, D. McKenzie, K. Garrett)(AMNH).Cicurina sansaba, new speciesFigs. 127-128, 153-154Diagnosis.-Eyeless troglobite of Gorman Cave,San Saba County, Texas; elongate sperrnathecumwith broad rounded sac; index coil with transversecanal across sac; heavy connecting canal formingloose loop around sperrnathecal parts. Male palpus:Figs. 153-154.Etymology.-Specific name for San SabaCounty.Description.-Female holotype: Length 3.65mm. Carapace 1.65 mm long, 1.15 mm wide. Abdomen3.5 mm long, 2.2 mm wide. Chelicerae:retromargin of fang with 6 teeth. Leg lengths: firstfemur 1.85 mm, fourth femur 1.95 mm; firstpatella-tibia 1.3 mm, fourth patella-tibia 3.25 mm.Ventral leg spines: first tibia 2-2-1, fourth tibia2-2-2.Male: Length 2.7 mm, specimen dried andshriveled. Male palpus (Figs. 153-154): tegulumwith three tubules; tarsal process about half width oftarsus; coil of conductor about length of basal process.Type-data.-Female holotype and 3 immaturefrom Gorman Cave, 6 mi. SW of Bend, San SabaCounty, Texas, 15 March 1963 (J. Reddell, D.McKenzie) (AMNH).Distribution.-Known only from Gorman Cave.Other records: 12 June 1978 (J. Reddell), 1 male, 2females, 2 immature, from Beyond Breakdown; 20October 1962 (J. Reddell), 1 immature.Cicurina venefica, new speciesFigs. 129-130Diagnosis.-Eyeless troglobite of Wizard's Well,Terrell County, Texas; elongate spermathecum withrounded sac, set in oblique position with tips nearlytouching; index coil large canal in inclined position;connecting canal forming wide coil around spermathecalparts. Male unknown.Etymology.-Specific name from Latin venefica,a witch.Description.-Female holotype: Length 4.6 mm.Carapace 2.3 mm long, 1.6 mm wide. Chelicerae:retromargin of fang with 6 teeth. Leg lengths: firstfemur 2.2 mm, fourth femur 2.6 mm; firstpatella-tibia 3 mm, fourth patella-tibia 3.3 mm. Legspines: first and fourth tibiae with 2-2-0.Type-data.-Female holotype from Wizard'sWell, Terrell County, Texas, 12-13 February 1983(Eric Short, Randy Waters) (AMNH).114

Cicurina caverna, new speciesFigs. 131-132Diagnosis.-Eyeless troglobite of Flemming'sBat Cave, Kimble County, Texas; elongate ovalspermathecum with rounded sac; index coil procurvedcanal crowning sac; slender connecting canalfonning loose loop around sperrnathecal parts. Maleunknown.Etymology.-Specific name from Latin caverna,a cavern.Description.-Female holotype: Length 3 rnm.Carapace 2 mm long, 1.4 mm wide. Abdomen 2 rnmlong, 1.2 mm wide. Chelicerae: retromargin of fangwith 5 teeth. Leg lengths: first femur 1.6 mm,fourth femur 1.8 mm; first patella-tibia 1.2 rnm,fourth patella-tibia 1.4 mm. Leg spines: first andfourth tibiae 2-2-2.Type-data.-Female holotype and 5 immaturefrom Flemming's Bat Cave, 4 mi. N Telegraph,Kimble County, Texas, 21 February 1964 (W.H.Russell) (AMNH).Cicurina porteri, new speciesFigs. 133-134Diagnosis.-Eyeless troglobite of Oriente MilestoneMolasses Bat Cave, Val Verde County, Texas;elongate sperrnathecum with rounded sac; index coilheavy procurved canal across sac; connecting canalloosely enclosing spermathecal parts. Male unknown.Etymology.-Named for John Porter, student ofcaves.Description.-Female holotype: Length 4.4 rnm.Carapace 2.4 mm long, 1.8 mm wide. Abdomen 2mm long, 2.4 mm wide. Chelicerae: retromargin offang with 5 or 6 teeth. Leg lengths: first femur 2.2mm, fourth femur 2.4 mm; first patella-tibia 2.6mm, fourth patella-tibia 2.7 mm. Leg spines: firsttibia 2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype and 1 immaturefrom Oriente Milestone Molasses Bat Cave, about20 mi. NE Del Rio, Val Verde County, Texas, 25January 1964 (1. Reddell, D. McKenzie, J. Porter),under rock beyond bat room (AMNH).Cicurina sheari, new speciesFigs. 135-136Diagnosis.-Eyeless troglobite of Ramsey BatCave, Real County, Texas; small coiled spermathecumand rounded sac in oblique position; indexcoil heavy recurved canal forming invertedV-shaped figure; connecting canal fonning wideloop around sperrnathecal parts. Male unknown.Etymology.-Named for William A. Shear, studentof spider behavior and evolution.Description.-Female holotype: Length 4.7 mm.Carapace 2.2 mm long, 1.8 mm wide. Abdomen 2.5mm long, 1.8 mm wide. Chelicerae: retromarginwith 4 large teeth on left chelicera and 6 in scatteredrow on right chelicera. Leg lengths: first femur 2.2mm, fourth femur 2.5 mm; first patella-tibia 2.6mm, fourth patella-tibia 2.8 mm. Leg spines: firsttibia 2-2-1, fourth tibia 2-2-2.Type-data.-Female holotype from Ramsey BatCave, Real County, Texas, 2 October 1976 (D.Pate, R. Hemperly, K. Heuss) (AMNH).Cicurina suttoni, new speciesFigs. 137-138, 151-152Diagnosis.-Eyeless troglobite of Felton Cave,Sutton County, Texas; elongate sperrnathecum withbroadly rounded sac; index coil transverse inwardlyturned canal around sac; connecting canal fonning aloose loop around spermathecal parts. Male (Figs.151-152).Etymology.-Specific name for Sutton County.Description.-Female holotype: Length 5.8 mm.Carapace 2.6 mm long, 2 mm wide. Abdomen 3.2mm long, 2 mm wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 2.4 mm,fourth femur 2.5 mm; first patella-tibia 2.4 mm,fourth patella-tibia 2.5 mm. Leg spines: first tibia2-2-0, fourth tibia 2-2-2.Male: Length 5.4 mm. Carapace 2.6 mm long, 2mm wide. Abdomen 2.8 mm long, 2 mm wide.Chelicerae: retromargin of fang with 6 teeth. Leglengths: first femur 2 mm, fourth femur 2.4 mm;first patella-tibia 2.6 mm, fourth patella-tibia 2.8mm. Male palpus (Figs. 151-152).Type-data.-Female holotype from Felton Cave,15 mi. SW Sonora, Sutton County, Texas, 4 July1964 (J. Reddell), in rotting root 2000 feet from entrance(AMNH).Distribution.-Known only from Felton Cave.Other records: 4 JulY 1964 (J. Reddell), 6 immaturefrom near Bat Room, 600 ft. from entrance; 14 October1928 (O.G. Babcock), 940 feet from entrance,1 male.Cicurina mckenziei, new speciesFigs. 139-140Diagnosis.-Eyeless troglobite of Fog Fissure,Bandera County, Texas; rounded spermathecum115

140138142145147144Figs. 137-148.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 137-138, sultoni, new species; 139-140,mckenziei, new species; 141-142, barri, new species; 143-144, rainesi, new species; 145-146, wi/loni, new species; 147-148, grnla,new species.116

with sac of similar size set in oblique position; indexcoil heavy canal inclined outward and joined to connectingcanal to moderately enclose spermathecalparts. Male unknown.Etymology.-Named for David McKenzie, studentof caves.Description.-Female holotype: Length 5 nun.Carapace 2.5 nun long, 1.4 nun wide. Abdomen 2.5nun long, 1.7 nun wide. Chelicerae: retromargin offang with 5 widely spaced teeth. Leg lengths: firstfemur 2 nun, fourth femur 2.2 nun; firstpatella-tibia 2.5 nun, fourth patella-tibia 2.8 nun.Leg spines: first tibia 2-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from Fog Fissure,5 mi. N of Vanderpool, Bandera County, Texas, 30October 1963 (D. McKenzie) (AMNH).Cicurina barri, new speciesFigs. 141-142Diagnosis.-Eyeless troglobite of Caverns ofSonora (=Mayfield Cave), Sutton County, Texas;oval spermathecum with broadly rounded sac; indexcoil short recurved canal; heavy connecting canalrather closely looped around spermathecal parts.Male unknown.Etymology.-Specific name for Thomas Barr,dean of American speleologists.Description.-Female holotype: Length 5 nun.Carapace 2.4 nun long, 1.4 nun wide. Abdomen 2.6nun long, 1.4 nun wide. Chelicerae: retromarginwith 5 separated teeth. Leg lengths: first femur 2nun, fourth femur 2.4 mm; first patella-tibia 2.8nun, fourth patella-tibia 3 nun. Leg spines: first andfourth tibiae 2-2-0.Type-data.-Female holotype and 1 immaturefrom Caverns of Sonora (=Mayfield Cave), 29 August1959 (T. Barr) (AMNH).Cicurina rainesi, new speciesFigs. 143-144Diagnosis.-Eyeless troglobite of 3-Bounce Pit,Edwards County, Texas; oval spermathecum withrounded sac; index coil heavy canal crossing sac;connecting canal forming open loop around spermathecalparts. Male unknown.Etymology.-Named for Terry Raines, studentof caves.Description.-Female holotype: Length 4.6 nun.Carapace 2.3 rom long, 1.3 mm wide. Abdomen 2.3nun long, 1.6 rom wide. Chelicerae: retromarginwith 7 small teeth. Leg lengths: first femur 1.6 nun,fourth femur 1.8 nun; first patella-tibia 2 nun,fourth patella-tibia 2.2 nun. Leg spines: first tibia 22-0, fourth tibia 2-2-1.Type-data.-Female holotype from 3-BouncePit, Carta Valley, Edwards County, Texas, February1974 (T. Raines, J. Lewis, R. Fieseler)(AMNH).Distribution.-Known only from 3-Bounce Pit.Other record: 14 July 1976 (yV. Elliott, D. McKenzie),3 females, 9 immature.Cicurina gruta, new speciesFigs. 147-148Diagnosis.-Eyeless troglobite of Dunbar Cave,Edwards County, Texas; elongate spermathecumwith widely rounded sac; index coil slightly procurvedcanal across sac; connecting canal formingloose loop around spermathecal parts. Male unknown.Etymology.-Specific name for Spanish gruta,cave.Description.-Female holotype: Length 4.75nun. Carapace 2.25 nun long, 1.55 nun wide. Abdomen2.5 nun long, 1 rom wide. Chelicerae:retromargin of fang with 5 teeth. Leg lengths: firstfemur 2.2 nun long, 1 nun wide; first patella-tibia2.8 nun, fourth patella-tibia 2.95 rom. Leg spines:first tibia 1-2-0, fourth tibia 2-2-2.Type-data.-Female holotype from DunbarCave, 20 mi. SW of Rocksprings, Edwards County,Texas, 29 September 1956 (yV. McAlister)(AMNH).Distribution.-Known only from Dunbar Cave.Other record: 30 August 1964 (D. McKenzie, T.Raines), 1 immature.Cicurina medina, new speciesFigs. 149-150Diagnosis.-Eyeless troglobite of Boehme'sCave, Medina County, Texas; conductor of malepalpus forming short curved loop as shown in figures.Female unknown.Etymology.-Named for Medina County.Description.-Male holotype: Length 3.6 rom.Carapace 1.8 rom long, 1.4 rom wide. Abdomen 1.8nun long, 1.4 nun wide. Chelicerae: retromargin offang with 6 teeth. Leg lengths: first femur 1 nun,fourth femur I. 1 nun; first patella-tibia 2 nun,fourth patella-tibia 2.4 rom. Leg spines: first andfourth tibiae 2-2-2.Type-data.-Male holotype from Boehme'sCave, 3 mi. S main dam on Lake Medina, 16117

150154158Figs. 149-160.-Ventral and retrolateral views of male palpi of eyeless Cicurina (Cicurella): 149-150, medina, new species;151-152, sUlloni; 153-154, sansaba, new species; 155-156, baronia, new species; 157-158, menardia, new species; 159-160, wiltoni,new species.118

February 1964 (J. Reddell, D. McKenzie, J. Porter)(AMNH).Cicurina wi/toni, new speciesFigs. 145-146, 159-160Diagnosis.-Eyeless troglobite of Crystal Cavernsand nearby cave of Jefferson County, Alabama;large round spermathecum with small oval sac; indexcoil thin canal crossing small sac; connectingcanal closely encircling spermathecal parts. Malepalpus (Figs. 159-160).Etymology.-Specific name for the late WiltonIvie.Description.-Female holotype: Length 5 mm.Carapace 2.5 mm long, 1.75 mm wide. Abdomen2.5 mm long, 1.7 mm wide. Chelicerae: retromarginof fang with 9 teeth. Leg lengths: first femur 2 mm,fourth femur 2.5 mm; first patella-tibia 2.6 mm,fourth patella-tibia 3 mm. Leg spines: first tibia 22-0, fourth tibia 2-2-2.Male: Length 4 mm. Carapace 2 mm long, 1.5mm wide. Abdomen 2 nun long, 1.5 mm wide.Chelicerae: retromargin with 8 teeth. Leg lengths:first femur 1.8 mm, fourth femur 2 mm; firstpatella-tibia 2.1 mm, fourth patella-tibia 2.4 mm.Male palpus (Figs. 159-160).Type-data.-Female holotype from Crystal Caverns,1 mi. N of Clay, Jefferson County, Alabama,12 July 1951 (W. B. Jones, 1. M. Valentine)(AMNH).Distribution.-ALABAMA: Jefferson County:Crystal Caverns, 29 June 1958 (W.B. Jones, TomSernnes, T.W. Daniel, Jr.), 1 female, 4 immature;18 March 1966 (S. Peck), 2 females, 4 immature;23 July 1965 (S. Peck), 1 male, 2 immature.McClunney-Alabama Caverns, near Clay (H.E.Steeves), 2 females.Cicurina maya GertschFigs. 161-162Cicurina maya Gertsch, 1977:127, Fig. 86.Diagnosis.-Eyeless troglobite from Cueva(Actlin) Tucil, Yucatan, Mexico; oval spermathecumin oblique position with small roundedsac, index coil hidden behind sac; connecting canal161163165162 166Figs. 161-166.-Ventral and dorsal views of epigyna of eyeless Cicurina (Cicurella): 161-162, maya Gertsch; 163-164, coahuilaGertsch; 165-166, leona, new species.119

forming close ring around spermathecal parts. Maleunknown.Etymology.-Named for the Maya people of Yucatan.Description.-Female holotype: Length 1.6 mm.Carapace 1.6 mm long, 1.08 mm wide. Abdomen1.9 mm long, 1.1 mm wide. Chelicerae: retromarginwith 8 teeth.Type-data.-Female holotype and subadult malefrom Cueva (Actlin) Tucil, 2 km S Muna, Yucatan,Mexico, 27 March 1973 (J. Reddell) (AMNH).Cicurina coahuila GertschFigs. 163-164Cicurina coahuila Gertsch, 1971: 110, fig. 108.Diagnosis.-Eyeless troglobite of Cueva de losLagos, Coahuila, Mexico; oval spermathecum withsmall oval sac; index coil thin procurved canal runninglength of spermathecum; connecting canalloosely encircling large spermathecum. Male unknown.Etymology.-Named for the State of Coahuila,Mexico.Description.-Female holotype: Length 3.8 mm.Carapace 1.6 mm long, 1.5 mm wide. Abdomen 2.2mm long, 1.5 mm wide. Chelicerae: retromarginwith 5 teeth.Type-data.-Female holotype from Cueva de losLagos, 24 km west of Ciudad Acuna, Coahuila,Mexico, 24 January 1969 (1. Reddell, D. McKenzie,J. Porter) (AMNH).Cicurina leona, new speciesFigs. 165-166Diagnosis.-Eyeless troglobite of Cueva deCuchillo, Nuevo Le6n, Mexico; ovate spermathecumwith rounded sac; index coil large canalforming prominent loop; thin connecting canalclosely draped around spermathecal parts. Male unknown.Etymology.-Specific name for State of NuevoLe6n, Mexico.Description.-Female holotype: Length 3.3 mm.Carapace 1.44 mm long, 1.2 mm wide. Abdomen1.9 mm long, 1.4 mm wide. Chelicerae: retromarginof fang with 6 teeth. Leg lengths: first femur 1.3mm, fourth femur 1.4 mm; first patella-tibia 1.5mm, fourth patella-tibia 1.8 mm. Leg spines: firsttibia 2-2-1, fourth tibia 1-2-1.Type-data.-Female holotype from Cueva deCuchillo, 2.5 km S Minas Viejas, Nuevo Le6n,Mexico, 22 April 1988 (Peter Sprouse) (AMNH).Cicurina buwata Chamberlin and IvieCicurina buwata Chamberlin and Ivie, 1940: 15, 75,pI. XIII, fig. 94.Diagnosis.-Eyeless troglobite from cave nearAustin, Texas; immature specimen of uncertain sexfrom uncertain cave habitat, here regarded as a nominainquirienda without specific status.Description.-Coloration and structural detailsare offered with drawing (Chamberlin and Ivie,1940:fig. 94) of the immature specimen in dorsalview. Eyes are entirely absent; measurements andratios of leg segments are included.Type-data.-Immature type from cave nearAustin, Texas, March 12-18, collector J. H. Comstock(Cornell University).Cicurina sp.The following records of immature eyelessCicurina are included to better indicate the range ofthe group in Texas and to encourage attempts to obtainadult specimens from these caves.Records.-TEXAS: Bandera County: HabySwallow Cave, 21 March 1971 (J. Reddell), 2 immature;Keese Cave, 11 July 1974 (D. McKenzie,W. Elliott), 3 immature. Bell County: Adam's GoldMine, 27 July 1963 (J. Reddell, D. McKenzie), 1immature. Bexar County: Black Cat Cave, 27 January1987 (J. Reddell, M. Reyes), 1 immature; 8March 1987 (J. Reddell, M. Reyes), 2 immature;Genesis Cave, June 1985 (R.M. Waters, A. Cobb),1 immature; Headquarters Cave, 24 April 1966 (B.Russell, D. McKenzie), 2 immature; Helotes Blowhole,25 December 1982 (R.M. Waters), 4 immature;John Wagner Ranch Cave No.3 (=Adam Wilson'sCave, Jr.); 23 December 1963 (C. Huebner,O. Knox), 1 immature; 4 October 1963 (1. Reddell,D. McKenzie), 2 immature; Kamikazi Cricket Cave,19 January 1986 (S.J. Harden), 1 immature; Robber'sCave, 3 September 1987 (A. Cobb, G. Veni),3 immature; Young Cave No.1, 6 August 1983 (G.Veni, J. Ivy), 1 immature. Blanco County: ForestView Cave, 20 February 1983 (W.R. Elliott), 1immature. Comal County: Double Decker Cave, 5February 1978 (G. Darilek, D. Montgomery, T.Mills), 1 immature; Lewis Cave, 8 March 1968 (1.Reddell, J. Fish, S. Fowler), 1 immature; StartzvilleBat Cave, 1982 (A.G. Grubbs), 1 immature. Coryell120

County: Diamond Cave, 16 August 1964 (J. Reddell,D. McKenzie), 3 immature. Edwards County:Deep Cave, 4 September 1965 (J. Reddell, D.McKenzie), 2 immature; Devil's Sinkhole, 26 October1963 (J. Reddell, J. Porter), 1 immature; 14February 1965 (J. Reddell), 2 immature; DunbarCave, 30 August 1964 (D. McKenzie, T. Raines), 2immature; Jacoby Cave, 22 September 1963 (J.Reddell, D. McKenzie), 2 immature; Wyatt Cave,21 September 1963 (J. Reddell, D. McKenzie), 4immature. Hays County: Boggus Cave, 21 June1985 (A. Cobb), 3 immature; 16 January 1986 (S.J.Harden), 2 immature; Donaldson Cave, 15 August1965 (B. Frank, B. Benfer), 1 immature; GrapevineCave, 6 February 1988 (A.G. Grubbs, L.J. Graves,C. Clayton, M. Bearll), 6 immature; Ladder Cave, 2September 1989 (J. Reddell, M. Reyes), 2 immature;26 May 1989 (A. Grubbs, J. Reddell, M.Reyes), 9 immature; McCarty Cave, 1988 (A.Grubbs), 2 immature; Michaelis Cave, January 1990(A.G. Grubbs, L.J. Graves), 2 immature; Morton'sCave, 9 September 1963 (D. McKenzie, B.Russell), 2 immature. Kendall County:Cave-Without-A-Name (=Century Caverns), 30July 1969 (J. Reddell, D. McKenzie), 1 immature.Kerr County: Mingus Swallow Cave, 28 April 1968(J. Reddell, S. Fowler), 1 immature. MedinaCounty: Davenport Cave, 10 July 1966 (1. and J.Reddell), 6 immature; Marguerite Cave, 28 April1984 (S. Harden), 1 immature; 5 May 1984 (R.M.Waters), 2 immature; Valdina Farms Sinkhole, 12January 1964 (J. Reddell, D. McKenzie, J. Porter),2 immature; Weynand Cave, 12 August 1965 (1.Reddell, J. Fish), 6 immature. Real County: Cave ofthe Lakes (=Haby Cave), 30 October 1963 (D.McKenzie), 2 immature; Skeleton Cave, 18 August1963 (J. Reddell, D. McKenzie), 12 immature. SanSaba County: Cicurina Cave, 17 February 1963 (1.Reddell, D. McKenzie), 1 immature; Clark's BranchWell Cave, 13 October 1962 (1. Reddell), 1 immature;Harrell's Cave, 9 February 1964 (J. Reddell,D. McKenzie, K. Garrett), 1 immature; Lemon'sCave, 1 September 1963 (J. Reddell, D. McKenzie),5 immature; Suprise Cave, 15 February 1964 (J.Reddell, D. McKenzie, J. Porter), 2 immature;Unknown Cave, 8 April 1989 (K. Markette), 1 immature;Wedge Cave, 11 June 1978 (J. Reddell, R.Fieseler, E. Kastning), 1 immature. Travis County:Backyard Cave, 18 January 1963 (W. Russell), 1immature; Beckett's Cave, 8 March 1963 (B. Russell),3 immature; 5 December 1965 (B. Russell), 1immature; Bee Creek Cave, 2 October 1963 (J.Reddell, D. McKenzie), 9 immature; 7 June 1965(1. Fish, J. Reddell), 2 immature; Broken StrawCave, 27 September 1965 (J. Reddell), 3 immature;Dead Dog Cave No.2, October 1963 (B. Russell), 1immature; Gallifer Cave, 28 August 1988 (J. Reddell,M. Reyes), 1 penultimate male, 1 immature;Goat Cave, 4 December 1983 (D.L. Pate), 4 immature;Lost Gold Cave, 27 May 1963 (J. Reddell, B.Frank), 3 immature; October 1963 (D. McKenzie),4 immature; Lunsford Cave, April 1963 (B. Russell),1 immature; McNeil Bat Cave, 18 January1963 (W. Russell), 1 immature; Midnight Cave, 6March 1966 (B. Russell), 1 penultimate male; NewComanche Trail Cave, 11 January 1989 (J. Reddell,M. Reyes), 1 immature; Weldon Cave, 7 January1965 (B. Russell), 1 immature. Uvalde County:Cement Tank Cave, 12 July 1974 (W. Elliott, D.McKenzie, P. Lynn), 4 immature; Dripstone Cave,13 August 1965 (J. Fish, J. Reddell), 5 immature;22 March 1983 (R.M. Waters, G.A. Poole), 1 immature;July 1983 (R. Waters, E. Short), 1 immature;Frio Bat Cave, January 1984 (R.M. Waters), 1immature; Frio King Cave, 8 June 1985 (A.Grubbs, R. Waters, A. Cobb), 1 immature; GrapeHollow Cave, 26 October 1965 (1. Reddell, J. Calvert),1 immature; Indian Creek Cave, 3 December1963 (J. Reddell), 1 immature; Moss Pit Cave, 13May 1989 (M. Warton), 5 immature; Story Cave,18 October 1964 (J. Reddell, D. McKenzie), 1 immature;Tampke Ranch Cave, 11 February 1966 (J.Reddell, D. McKenzie), 7 immature; 25 July 1974(W. Elliott, S. Sweet), 2 immature; Whitecotton BatCave, 24 April 1966 (J. Reddell, E. Alexander), 1immature. Val Verde County: Emerald Sink, 24January 1964 (J. Reddell, D. McKenzie, J. Porter),3 immature; Ladder Cave, 11 August 1963 (J. Reddell,D. McKenzie), 5 immature. WilliamsonCounty: Bat Well, 4 March 1988 (J. Reddell, M.Reyes), 2 immature; Beck Bat Cave, 24 August1963 (J. Reddell, B. Russell), 1 immature; BeckRanch Cave (=Beck's Tin Can Cave), November1962 (1. Reddell), 2 immature; 12 March 1963 (B.Russell), 1 immature; 23 June 1968 (1. Reddell), 4immature; 27 February 1972 (J. Reddell), 1 immature;9 March 1988 (J. Reddell, M. Reyes), 2 immature;Bone Cave, 4 August 1963 (J. Reddell), 1immature; Cobb Cavern, 31 March 1963 (1. Reddell),5 immature; Cricket Cave, 30 March 1965 (J.Reddell), 2 immature; Grimace Cave, 16 April 1989(W. Elliott), 1 immature; Hex Cave, 1 June 1989 (1.Reddell, M. Reyes, M. Warton), 3 immature; InnerSpace Caverns (=Laubach Cave), 9 July 1965 (1.Reddell), 3 immature; 2 November 1968 (W. Elliott),1 immature; Kamikazi Crack Cave, 16 April1989 (1. Reddell, M. Reyes), 1 immature; LakeLineCave, 21 January 1990 (1. Reddell), 1 immature; 7121

February 1990 (J. Reddell, M. Reyes), 4 immature;16 February 1990 (J. Reddell, M. Reyes), 2 immature;Off Campus Cave, 8 April 1989 (W. Elliott, J.Reddell, M. Reyes), 2 immature; Steam Cave, 7July 1963 (J. Reddell, B. Russell), 2 immature;Three-Mile Cave, 30 March 1965 (J. Reddell), Iimmature; Walsh Ranch Cave, 24 August 1963 (J.Reddell, B. Russell), 1 immature; Williams Cave,24 August 1963 (J. Reddell, B. Russell), I immature;Wolf's Cave, 7 August 1983 (W. Elliott, B.Vinson, D. Pate), penultimate male.ACKNOWLEDGMENTSThe research material on which this paper isbased has been brought together by friends and colleaguesover a period of many years. One of the firststudents was Wilton Ivie who was responsible forbasic work on the cicurine spiders and began studyof the Cicurella complex until his untimely death.The active field contributors have been speleologistslong devoted to systematic study of caves and cavefaunas. In this case no better reason for intensivestudy of cave cicurellas are the spiders themselveswhich have responded to cave stimuli to produce anunparalleled total of more than fifty completelyeyeless taxa within a quite limited geographical area.Caving demands devotion to arduous and often dangerousexcursions into many kinds of caves. Amongthe first Texas cave students were James R. Reddelland Robert W. Mitchell, who have devoted much oftheir field life to cave research in the United Statesand Mexico. We salute these and the many otherstudents as assistants during years of caving; manyof their names will be found among the describedtaxa. I am grateful to the following Museum caretakersfor loans of types and other valuable specimens,and for the many special favors during thisstudy: Dr. W. G. Reeder, Texas Memorial Museum(TMM), Austin, Texas; Dr. Norman Platnick,American Museum of Natural History (AMNH),New York; Mr. Darrell Ubick, California Academyof Sciences (CAS), San Francisco.I am especially grateful for the cooperation andkindness of James R. Reddell during the process oforganizing this study and bringing it to completion.LITERATURE CITEDBarr, T.C., Jr.• and J.R. Holsinger. 1985. Speciation in cavefaunas. Ann. Rev. Ecol. Syst., 16:313-337.Bonnet, P. 1956. Bibliographia Araneorum, 2(G-M), pp.1927-3026. Toulouse.Brignoli, P.M. 1983. A catalogue of the Araneae describedbetween 1940 and 1981. London: Manchester Univ. Press.518 pp.Chamberlin, R.V. 1933. On a new eyeless spider of the familyLinyphiidae from Potter Creek Cave, California. Pan-PacificEntomol.,9:122-124.Chamberlin, R.V., and W. Ivie. 1940. Agelenid spiders of thegenus Cicurina. Bull. Univ. Utah, BioI. Ser., 30:1-107.Comstock, J.H. 1912. The spider book. Garden City, NewYork: Doubleday, Page & Co. 721 pp., 771 figs.Exline, H. 1936. Nearctic spiders of the genus Cicurina Menge.American Mus. Novitates, no. 850. 25 pp.Forster, R.R. 1970. The spiders of New Zealand, Part III. OtagoMus. Bull., 3:1-184.Gertsch, W.J. 1935. Spiders from the southwestern United Stateswith descriptions of new species. American Mus. Novitates,no. 792. 31 pp.Gertsch, W.J. 1971. A report on some Mexican cave spiders.Assoc. Mexican Cave Stud. Bull., 4:47-111.Gertsch, W.J. 1974. The spider family Leptonetidae in NorthAmerica. J. Arachnol., 1: 145-203.Gertsch, W.J. 1977. Report on cavernicole and epigean spidersfrom the Yucatan Peninsula. Assoc. Mexican Cave Stud.Bull., 6:103-131.Gertsch, W.J. 1984. The spider family Nesticidae (Araneae) inNorth America, Central America, and the West Indies. TexasMem. Mus. Bull., 31. 91 pp.Gertsch, W.J., and L.I. Davis. 1936. Spiders of the southwesternUnited States with descriptions of new spiders. AmericanMus. Novitates, no. 1936, pp. 1-31,39 figs.Ivie, W. 1965. The spiders of the genus lslandiana (Linyphiidae,Erigoninae). American Mus. Novitates, no. 2221. 25 pp.Keyserling, E. 1886. Die Spinnen Amerikas. Theridiidae, partU. Nurnberg, 1886,2(2):1-295, pis. I-XI.Lehtinen, P.T. 1967. Classification of the cribellate spiders andsome allied families, with notes on the evolution of the suborderAraneomorphae. Annal. ZooI. Fennici, 4:199-468,figs. 1-524.Menge, A. 1871. Preussische Spinnen, IV. Schr. naturf. Ges.Danzig, N.F., 2:265-296.Nicholas, Bro. G. 1960. Checklist of macroscopic troglobiticorganisms of the United States. American MidI. Nat., 64:123-160.Petrunkevitch, A. 1911. A synonymic index-catalogue of spidersof North, Central and South America, etc .... Bull. AmericanMus. Nat. Hist., 19.791 pp.Platnick, N.I. 1989. Advances in spider taxonomy. Manchester:Manchester Univ. Press. 720 pp.Simon, E. 1898. Histoire naturelle des Araignees, 2(2): 193-380.Paris.Tellkampf, T. 1844. Beschreibung einiger neuer in der MammuthH6hlein Kentucky aufgefundener Gattungen von Gliedertieren.Arch. Naturg., 10(1):318-322, pI. VIII.Vogel, B.R. 1967. A list of new North American spiders.American Entomol. Soc., Acad. Nat. Sci., Philadelphia. 196pp.This is publication No. N.S.-57 of the Texas Memorial Museum.122

Roth, V.D. 1992. A new and first troglobitic spider from Arizona (Thymoires, Theridiidae). Texas Mem. Mus., Speleol. Monogr.,3:123-126A NEW AND FIRST TROGLOBITIC SPIDER FROM ARIZONA(THYMOITES, THERIDIIDAE)Vincent D. RothSpider Lane # IBox 136Portal, AZ 85632ABSTRACTThe first troglobitic spider from Arizona, Thymoires minero,new species, is described from Southwestern Cave, CochiseCounty. This is the first troglobitic Thymoires and the firsttroglobitic species of the Theridiidae in North America.INTRODUCTIONArizona has a depauperate cave fauna mainlyconsisting of troxloxenes and troglophiles (Roth,unpublished; Peck 1980) and a few troglobites.Barr's (1963) definition of troglobites would includeonly the following blind forms: the isopod Brackenridgiasphinxensis Schultz (1984), an amphipod,Stygobromus arizonensis Holsinger (1974), a phreatobite,and possibly the collembolan, Tomocerussp., and campodeid, Haplocampa sp., the latter twolisted by Peck.Peck presented a list of species taken from cavesin the Grand Canyon, used a broadened definition ofthe term "troglobite," and included a "cave adapted... low level or relatively unspecialized" spider(Telema sp., Telernidae) as a troglobite. Barr wouldidentify this as a troglophile. He describes troglobitesas "those obligative cavernicoles -usuallydistinguished morphologically by regression of pigmentand photoreceptors, and frequently by longer,more slender appendages than their epigean congeners.In the absence of regressive andlor adaptivemodifications frequently associated with troglobites,an animal, even though known only from caves, isusually (and probably should be) considered atroglophile. " This is the criteria followed in this paperand accordingly some of Peck's "troglobites ordisjunct troglobites" must be considered troglophilesor disjunct troglophiles, or, at the most, incipienttroglobites.Undescribed female telernids have a FIC index of260, (femur I length/carapace length X 100), for aneyeless troglobite, 218 carapace length for an eyedtroglophile, and 170 for an epigean species. Peck'sTelema sp. showed the presence of pigmented eyes,not reduced in size, integument slightly pigmented,and not much longer legs than normal with a F/Cindex of 195, obviously a troglophile.This suggests that the new species of Thymoites(Theridiidae) is the first spider troglobite knownfrom Arizona. It is unpigmented, has only pale eyespots except for the AME which do not appearfunctional and are irregularly pigmented and moreslender legs with a F/C index of 169 - 197 (males)and 194 - 219 (females) compared with eight epigeanspecies with an index of 94 - 161 (males) and96-136 (females).Few Theridiidae are troglobites. Styposis havethe AME reduced or lost (Levi 1959a, 1964a) butare not cave species.The 6-eyed Comaroma (= Archerius) is not atheridiid and its transfer to the Anapidae by Wunderlich(1986) was overlooked by both Peck andShear (1987) and Merrett and Ashmole (1989). Theydescribed the first theridiid troglobites, the blindTheridion streptipes (Peck and Shear) from the123

Galapagos Islands and one with eye spots from theAzores, T. pico (Merrett and Ashmole).This third species was found in a moist limestonecave, the Southwestern Cave at Bisbee, Arizona.Troglophiles in the habitat included a pale psocid,collembolans, and mycetophilid flies. The spiderswere brought to my attention by Ralph Luetckewhose keen interest in the natural history of the areamade the find possible. His assistance in guidingour group and helping with collections is herebygratefully acknowledged. Special thanks go to Dr.Herbert Levi who confirmed the identity of thisspecies and proofed the manuscript and to Dr. ToddRiddell of the U.S. Army for the original drawings.The females key out to couplet 6 of Levi's(1964b) key but lack the lateral ducts; the maleskey to couplet 33, however their carapace setae arenot transverse but form three longitudinal rows onthe pars cephalica.DESCRIPTIONThe following description follow those of Levi(1957) to provide a comparison of species.Thymoites minero n. sp.Figs. 1-5Type-data.-Male holotype, female allotype and3 male and 7 female paratypes from Southwest Cavein Bisbee, Cochise County, Arizona, Dec. 15, 1983,Vincent and Barbara Roth, Ralph and SuzanneLuetcke, collectors.The holotype and allotype will be deposited inthe Museum of Comparative Zoology, paratypes inthe California Academy of Sciences, Texas MemorialMuseum, Museum of Natural History,Washington D. C., The American Museum ofNatural History, and the author's collection.Etymology.-The specific name "minero" isSpanish for "one who digs for metals" and is used tohonor the miners of Bisbee, Arizona who opened theunderground cavities where these spiders occur.Diagnosis.-This is the only troglobitic Thymoites,differing from the epigean species by longerlegs, femora I, 1 1/2 to 2 X as long as the carapace,and the tiny colorless (except AME) eyes. Themales have three rows of long setae on the parscephalica (Fig. 2) and the females can be recognizedby the knob-like epigynum (Fig. 3) with very shortducts on the spermatheca (Fig. 5).The epigynum is similar to that of T. missionen-sis (Levi, 1959b) but the copulatory opening issmaller, round and internally the spermatheca bearsa short blind (?) duct off the ventral wall. Themales do not seem similar to any other Thymoites.The females have relatively longer legs than themales, an unusual feature among spiders but anothercase was noted in the same genus. One pair of T.madera Gertsch and Archer (Levi, 1957) had a FICindex of 95 (male) and 115 (female).Description.-Pale yellow spiders lacking marksexcept for irregular dark pigment in the AME.Carapace high (Fig. 2), equal to length ofchelicera exposed below lower edge of c1ypeus. Thec1ypeus about 2/3 as high as width of AER. Eyesreduced in size, about 0.03 mm diameter, pale, exceptAME with scattered pigment. Viewed fromfront, AER slightly procurved to straight; viewedfrom above PER strongly procurved, PLE advancedone diameter anterior to PME; latter separatedslightly more than AME and about half that distancefrom them. Pars cephalica (Fig. 2) encircled by 15long (0.25-0.3 mm) curved setae, 4 on each side, 3on the center line, plus four on the anterior part ofthe pars thoracica. Chelicerae with large promarginaltooth, none on retromargin. Legs with fewlarge dorsal spines, two on patellae and one nearbase of tibiae.Palpus as illustrated (Fig. 1) with median apophysisextending out from edge of bulb. Palpal femurcurved, patella continuing the curve almost180 0 to parallel the femur. Tibia constricted nearbase.Total length of males 1.7-2.0 mm. Measurementsof male holotype: total length 1.8 mm; carapacelength 0.82 mm; width 0.69 mm. Leg I totallength 5.11 mm; first femur 1.64 mrn; patella-tibia1.57 mm; metatarsus 1.31 mm; tarsus 0.59 mm;second patella-tibia 1.29 mrn; third 0.90 mm; fourth1.29 mrn.Female: As in male except setae on carapacemuch shorter (0.18 -0.26 nun) and few on a medianline. Carapace not as high in front, 2/3 the lengthof the exposed part of the chelicerae, c1ypeus heightabout 315 the width of the AER.Epigynum raised posteriorly, knob-like (Figs. 3,4) with an opening at the tip. Vulva simple (Fig.5), two large bulbous spermathecae with short ducts.Total lengths of females, 2.5 mrn. Measurementsof female allotype: total length 2.5 mm; carapacelength 0.85 mm, width 0.72 mm; leg I length 5.47mm; first femur 1.62 mm; patella-tibia 1.80 mm;metatarsus 1.41 mm; tarsus 0.64 mm; secondpatella-tibia 1.41 mrn; third 1.08 mrn; fourth 1.57mrn.124

LITERATURE CITEDBarr, T.C. 1963. Ecological classification of cavernicoles.Cave Notes, 5(2):9-12.Holsinger, J.R. 1974. Systematics of the subterranean amphipodgenus Stygobromus (Gammaridae), Part I: Species ofthe Western United States. Smithsonian Contributions toZoo!., no. 160,63 pp.Levi, H.W. 1957. The spider genera Enoplognatha, Theridion,and Paidisca in America north of Mexico (Araneae,Theridiidae). Bull. American Mus. Nat. Hist., 112(1):1-124,figs. 1-421, maps 1-41.Levi, H.W. 1959a. The spider genus Styposis (Araneae,Theridiidae). Psyche, 66(1-2):13-19.Levi, H.W. 1959b. The spider genera Achaearanea, Theridionand Sphyrotinus from Mexico, Central America and the WestIndies. (Araneae, Theridiidae). Bull. Mus. Compo Zoo!.,Harvard Univ., 121(3):1-163, figs. 1-430.Levi, H.W. 1964a. The American spiders of the genera Styposisand Pholcomma (Araneae, Theridiidae). Psyche,71(1):32-39.Levi, H.W. 1964b. The spider genus Thymoites in America(Araneae: Theridiidae). Bull. Mus. Compo Zoo!., HarvardUniv., 130(7):445-471.Merrett, P., and N. P. Ashmole. 1989. A new trog10bitic---(//IfJ\\\\\\\\\I"\1IJ/I___---'--I 4III5Figs. 1-5.-Thymoites minero, new species: I, left male palpus, ventral view; 2, Male carapace; 3, lateral view ofepigynum; 4,epigynum, ventral view; 5, epigynum, dorsal view.125

1heridion (Araneae: Theridiidae) from the Azores. Bull.British Arachnol. Soc., 8(2):51-54.Peck, S.B. 1980. Climatic Change and the Evolution of CaveInvertebrates in the Grand Canyon, Arizona. NSS Bulletin,42(3):53-60.Peck, S.B., and W.A. Shear. 1987. A new eyeless, stridulating1heridion spider from caves in the Galapagos Islands(Araneae, Theridiidae). Canadian Entomol., 119:881-885.Schultz, G.A. 1984. Brackenridgia sphinxensis n. sp. from acave with notes on other species from Arizona and California(Isopoda, Oniscoidea). The Southwestern Naturalist,29(3) :309-319.Wunderlich, J. 1986. Spinnenfauna gestern and heute: FossileSpinnen in Bernstein und ihre heute lebenden Verwandten.Wiesbaden: Quelle & Meyer. 293 pp.This is publication No. N.S.-58 of the Texas Memorial Museum.126

Muchmore, W.B. 1992. Cavernicolous pseudoscorpions from Texas and New Mexico (Arachnida: Pseudoscorpionida). Texas Mem.Mus., Speleol. Monogr., 3:127-153CAVERNICOLOUS PSEUDOSCORPIONS FROM TEXAS AND NEW MEXICO(ARACHNIDA: PSEUDOSCORPIONIDA)William B. MuchmoreDepartment of BiologyUniversity of RochesterRochester, New York 14627ABSTRACTAll known cavernicolous pseudoscorpions from Texas andNew Mexico have been reviewed and studied. A key to representedgenera is provided. In addition to the 9 identified speciespreviously reported, 17 others are here recorded, including 9newly described. The new species are: Tyrannochthonius texanus,Tartarocreagris comanche, Chitrella we/boumi, C. major,C. el/iotti, Cheiridium reyesi, Apocheiridium reddel/i, Neoallochemes(?)incertus, and Dinocheirus cavicolus. The genusTartarocreagris Curcic is discussed in regard to newly discoveredmales, and 2 species are transferred therein. Also, the genusNeoallochemes Hoff is redefined; a new species, N. cubanus, isdescribed from Cuba and Dinocheirus stercoreus Turk, fromTexas bat caves is reassigned to that genus. In an Addendum,another new species from a cave in Texas is described: Tartarocreagrisintennedia.INTRODUCTIONUntil now, reports of cavernicolous pseudoscorpionsin Texas and New Mexico have been widelyscattered in the literature (Turk, 1949; Hoff, 1957;Reddell, 1965, 1970; Barr and Reddell, 1967;Muchmore, 1969, 1976, 1981, 1986). Seven identifiedspecies have been reported from 8 caves inTexas and 2 identified species from 2 caves in NewMexico; additional tentatively identified or unidentifiedpseudoscorpions were known from about 15other caves in Texas and one in New Mexico. Recentcollections, especially by James Reddell andcolleagues in Texas and W.C. Weibourn in NewMexico, now allow the addition of 17 species(including 9 newly described below) to the list ofthose found in caves in the two states.Unless otherwise noted, the materials upon whichthis report is based are deposited in the Florida StateCollection of Arthropods (FSCA), Gainesville,Florida. Some specimens are housed in the AmericanMuseum of Natural History (AMNH), NewYork, New York; the Canadian National Collectionof Insects and Arachnids (CNC), Ottawa, Canada;the J.C. Chamberlin Collection (JCC), ForestGrove, Oregon; the Museum of Comparative Zoology(MCZ), Cambridge, Massachusetts; and theUnited States National Museum of Natural History(USNM), Washington, D.C.FAMILY CHTHONIIDAE HANSENGenus Aphrastochthonius ChamberlinAphrastochthonius Chamberlin, 1962:307; Muchmore,1986:17.127

Key to Genera of Cavernicolous Pseudoscorpions in Texas and New Mexico1. Legs III and IV with more segments (6 beyond coxa) than legs Iand II (5 beyond coxa) (family Chthoniidae) 2All legs with same number of segments (either 5 or 6 beyond coxa) 42. Coxal spines only on coxa of leg I ApochthoniusCoxal spines not confined to leg I 33. Coxal spines only on coxa of leg II TyrannochthoniusCoxal spines on coxae of legs I and IIAphrastochthonius4. Tarsus of each leg divided into 2 parts, each leg with 6segments beyond coxa 5Tarsus of each leg undivided, each leg with 5 segments beyond coxa 95. Chelicera with inner margin of movable finger distinctly dentate 6Chelicera with inner margin of movable finger essentially smooth(family Garypidae)Archeolarca6. Abdomen with pleural membranes striate (family Syarinidae) ChitrellaAbdomen with pleural membranes granulate 77. Trichobothrium ib in middle of dorsum of chelal hand, isb onbase of fixed finger (family Bochicidae)LeucohyaTrichobothrium ib on base of fixed chelal finger along with isb(family Neobisiidae) 88. Movable finger of chelicera with spinneret in form of lowelevationMovable finger of chelicera with spinneret in form of long,usually branched galeaMicrobisiumTartarocreagris9. Femora of all legs similar, articulations between basifemora andtelofemora obsolete (family Cheiridiidae) 10Femora of legs I and II quite different from those of legs III andIV, articulations well developed (family Chemetidae) 1110. Movable finger of palpal chela with 2 trichobothria CheiridiumMovable finger of palpal chela with only 1 trichobothriumApocheiridium11. Chelicera with flagellum composed of 3 setae; hand of chelicerawith 4 setaeNeoallochernesChelicera with flagellum composed of 4 setae; hand of chelicerawith 5 setae 1212. Tarsus of leg IV with an acuminate tactile seta DinocheirusTarsus of leg IV without an acuminate tactile setaHesperochernes128

Species of Aphrastochthonius are known fromAlabama, Cuba, Central America, New Mexico, andCalifornia. Except for 2 species from Cuba andChiapas, Mexico, all are cavernicoles.Aphrastochthonius pachysetus MuchmoreAphrastochthonius pachysetus Muchmore, 1976:361, Figs. 1-3.The holotype female of this species was found inDoc Brito Cave, near White's City, Eddy County,New Mexico. No other representative is known.Genus Apochthonius ChamberlinApochthonius Chamberlin, 1929a:66; Muchmoreand Benedict, 1976:68.Apochthonius is distributed over most of theUnited States and Canada. Representatives are to befound in moist litter, both epigean and hypogean.Apochthonius magnanimus HoffApochthonius magnanimus Hoff, 1956a:4, Figs.2-4; Hoff, 1958:6.A male belonging to Apochthonius was collectedin Ogle Cave, Carlsbad Caverns National Park,Eddy County, New Mexico, 25 May 1974, by W.C.Welbourn. Though it is a little smaller than thetypes described by Hoff, it is, no doubt, arepresentative of A. magnanimus (see Welbourn,1978:31), accidental in the cave.Genus Tyrannochthonius ChamberlinTyrannochthonius Chamberlin, 1929a:74; Muchmore,1986:20.This genus has a wide distribution in the wannerparts of the world; in North America it is foundthrough the southern United States and into the Caribbeanarea and Central America (mostly unpublishedrecords). While Chamberlin and Malcolm(1960) made mention of troglobitic forms ofTyrannochthonius from Alabama, only 2 namedspecies have been reported from the United States,an epigean fonn from Florida and a troglobite fromTexas. Epigean representatives are found In moistlitter.Tyrannochthonius troglodytes MuchmoreTyrannochthonius troglodytes Muchmore, 1986:23,Fig. 11.The holotype and allotype of this species werecollected in Rock Slab Cave, on Enchanted Rock,Llano County, Texas. No other representative isknown.Tyrannochthonius texanus, new speciesFigs. 1, 2Tyrannochthonius spp.: Reddell, 1970:403.Type-data.-Holotype male (WMI078.01001)and allotype female (WM1078.01002) from ToothCave, Travis County, Texas, 14 May 1966 (J. Reddell).Paratype female from Ezell's Cave, San Marcos,Hays County, Texas, 7 September 1963 (J.Reddell); paratype male from Deep Cave, EdwardsCounty, Texas, 4 September 1965 (J. Reddell); 2paratypes (1 male, 1 female) from Wren Cave, ValVerde County, Texas, 9 April 1968 (J. Reddell andT. Mollhagen). All specimens mounted on slides(FSCA).Diagnosis.-Smaller than T. troglodytes (chelalength

typical; one spinelike seta on hand. Fixed chelalfinger with 18-23 spaced macrodenticles and 9-12interspersed microdenticles; movable finger with 7-9spaced macrodenticles, 5-8 interspersed microdenticles,and 15-18 low, rounded teeth proximally.Sensillum on movable finger proximal to lastmacrodenticle.Legs relatively robust; leg IV with entire femur2.3-2.5 and tibia 4.0-4.3 times as long as deep.Long tactile setae on tibia and both tarsi of leg IV.Measurements (mm).-Figures given first forholotype followed in parentheses by ranges for theallotype and paratypes. Body length 1.08(1.08-1.41). Carapace length 0.385 (0.355-0.45).Chelicera 0.285 (0.27-0.34). Palpal femur 0.415(0.38-0.47) by 0.095 (0.085-0.11); tibia 0.165(0.165-0.21) by 0.105 (0.095-0.12); chela 0.625(0.585-0.725) by 0.13 (0.12-0.16); hand 0.255(0.23-0.26) by 0.135 (0.1260.16); movable finger0.415 (0.37-0.46) long. Leg IV: entire femur 0.37(0.34-0.43) by 0.16 (0.15-0.17); tibia 0.265(0.235-0.31) by 0.06 (0.55-0.75).Etymology.- The new species is named texanusfor the state in which it has been found.Remarks.-This species is very similar to T.floridensis Malcolm and Muchmore (1985), but canbe differentiated easily from that species by the occurrenceof 6 setae on tergite 4 and the lack of a setaon the apical projection of coxa I. Although it hasbeen found only in caves it is not much adapted forcavemicolous existence; the only modification appearsto be the slight reduction in development ofthe posterior eyes. It is at best a troglophile andperhaps only accidental in caves. No epigeanrepresentatives of thi~ genus have been reportedfrom Texas, but this fact may be due simply to thepaucity of collections of the litter dwelling fauna ofthe state.Also at hand are 4 representatives of Tyrannochthoniusfrom Arrowhead Cave, Hays County,Texas, 1983 (A.G. Grubbs), which probably belongto this species. Unfortunately, these specimens areso broken and dismembered that they cannot beidentified with certainty./134Figs. 1-4.-1-2, Tyrannochlhonius lexanus, new species, holotype male: I, epistome and flanking setae on carapace; 2, left palp,dorsal view; 3-4, Tanarocreagris lexana (Muchmore), male from Amber Cave: 3, chaetotaxy of genital opercula; 4, internal genitalia,ventral view.b130

FAMILY NEOBISIIDAE CHAMBERLINGenus Microbisium ChamberlinMicrobisium Chamberlin, 1930:20; Nelson, 1984:341.This genus is represented in North America by 2litter-dwelling species, Microbisium brunneium(Hagen) and M. parvulum (Banks). The latter iswidespread through the United States and is occasionallyfound in caves.Microbisium parvulum (Banks)Obisium parvulum Banks, 1895: 12.Microbisium parvulum: Chamberlin, 1930:21; Hoff,1946:495; Hoff, 1956b:3; Hoff, 1958:9; Nelson,1984:341.Microbisium confusum Hoff, 1946:496; Hoff,1958:9; Nelson, 1975:280.Microbisium sp.: Reddell, 1970:403.This species is apparently distributed widely overNorth and Central America (Nelson, 1984), usuallybeing found in moist ground litter. Occasionalspecimens have been found in Texas and 4specimens are here reported from caves: 2 femalesfrom Mold Hole, Travis County, 8 June 1966 (J.Reddell); 1 female from Copperhead Cave, San SabaCounty, 12 June 1978 (J. Reddell and R. Fieseler);and 1 female from the entrance sink of RobberBaron Cave, San Antonio, Bexar County, 21February 1986 (S.J. Harden).Genus Tartarocreagris CurcicTartarocreagris Curcic, 1984: 163.The genus is based on the species Microcreagrisillfernalis Muchmore, 1969, which is known onlyfrom a single female specimen. With the addition of3 more species, our understanding of the genus isincreased.Diagnosis (revised).-A genus of the familyNeobisiidae Chamberlin (1930:9). Carapace longerthan broad; with epistome small to absent; eyes 4 orabsent; chaetotaxy 20-30, with 4 at anterior and 4-6at posterior margin. Palpal coxa with 3-4(occasionally 5) apical setae. Middle tergites with12-15 setae; middle sternites with 15-17 setae(including 2 discal setae on sternites 6-8). Femaleanterior operculum with 3-4 small setae on each sideof midline; posterior operculum with marginal rowof 12-15, few sometimes displaced anteriorly nearmiddle. Male anterior operculum with about 24 scatteredsetae; posterior operculum with 6-12 on faceand 12-15 in marginal row. Internal genitalia ofmale with conspicuous round ventral sac, 2 smallerdorsal sacs, and 2 long, narrow, wrinkled lateralsacs. Cheliceral hand with 6-7 setae; flagellum of 8serrate setae; galea bifurcate distad of middle, eachramus with 0-2 spinules, smaller in male than infemale. Palp with small granulations on medial sidesof trochanter, femur, and chelal hand, andoccasionally on tibia; chelal fingers with numerouscontiguous marginal teeth, the distalmost 10-20 withcusps, the others rounded. Fixed finger of chelawith trichobothria et, it and est in distal half and ist,isb, and ib in proximal half; esb and eb placedlaterally on distal part of hand. Subterminal tarsalsetae of legs with spine near middle and fewspinules near tip (not "furcate," as stated by Curcic,1984: 164).Distribution.-As presently understood, the genusTartarocreagris is represented only by troglobiticforms in Williamson and Travis Counties, Texas.Tartarocreagris illfernalis (Muchmore)Microcreagris illfernalis Muchmore, 1969:15, fig.12.Microcreagris. spp. (in part): Reddell, 1970:403.Tartarocreagris infernalis (Muchmore): Curcic,1984:163, figs. 21, 40.The holotype female of this species was collectedfrom under a rock, several thousand feet from entrance,in Core Hole Cave (Inner Space Cavern)near Georgetown, Williamson County, Texas. Noother representative is known.Microcreagris infernalis was designated the typespecies of the new genus Tartarocreagris by Curcic(1984).Tartarocreagris texalla (Muchmore),NEW COMBINATIONFigs. 3,4Microcreagris texana Muchmore, 1969: 18, figs. 13,14; Chambers and Jahrsdoerfer, 1988:36029.Microcreagris. spp. (in part): Reddell, 1970:403.Australillocreagris texalla (Muchmore): Curcic,1989:360, figs. 8, 15.This species was described from the female holotypecollected in Tooth Cave, Travis County, Texas.131

Recently, in nearby Amber Cave, a male was foundwhich appears conspecific with the holotype.New record.-One male from Amber Cave,Travis County, Texas, 8 April 1984 (1. Reddell andM. Reyes); mounted on slide (FSCA). As this is thefirst male known for the genus TartarocreagrisCurcic, it is appropriate to provide a shortdescription. Very similar to female in most respects.Carapace with epistome much reduced; no eyes;23 setae, 4 at anterior and 6 at posteriormargin. Palpal coxa with 4 apical setae. Tergalchaetotaxy 8: 10: 12: 11: 12: 12: 13: 13: 12: 10:3:2. Sternalchaetotaxy 24:[2-5]:(5)6/12(5):(5)1/12(5): 15:2/15:2/14:2/13: 1/12: 12:9:2 (Fig. 3). Internal genitalia(Fig. 4) much like those of Microcreagrisphyllisae Chamberlin (1962:fig. 12), with 2conspicuous, rounded sacs lying dorsal to a largeround ventral sac, and the 2 lateral sacs long, rathernarrow, and wrinkled. Cheliceral hand with 7 setaeon right and 6 on left; flagellum of 9 serrate setae;galea bifurcated just distad of middle, each ramus intum bifurcated. Palp with femur 5.75, tibia 4.55,and chela (with pedicel) 6.1 times as long as broad;hand (with pedicel) 2.75 times as long as deep;movable fmger 1.34 times as long as hand withpedicel. Fixed finger with 97 and movable fingerwith 108 marginal teeth of which only the distal10-15 are cusped. Leg IV with entire femur 5.15and tibia 8.25 times as long as deep; subterminaltarsal setae with a small median spine and 1-2spinules toward tip.Measurements (mm).-Body length 3.96. Carapacelength 1.08. Chelicera length 0.68. Palpal trochanter0.71 by 0.28; femur 1.52 by 0.265; tibia1.36 by 0.30; chela (without pedicel) 2.47 by0.435; hand (without pedicel) 0.975 by 0.42;pedicel 0.185 long; movable finger 1.55 long. LegIV: entire femur 1.18 by 0.23; tibia 1.07 by 0.13.Remarks.-In his revision of some North Americanspecies of Microcreagris, Curcic (1984) did notdeal with M. texana. However, his new genus Tartarocreagrisis based on M. infernalis from InnerSpace Cavern, about 25 miles north of Tooth Cave,the type locality of M. texana. The holotypes ofboth M. infernalis and M. texana are females, whichare quite similar to one another in basicmorphology, differing only in the degree ofattenuation of their appendages and in small detailsof chaetotaxy. The male from Amber Cave is verylike the females except in sex-related characters. It isclear that these specimens are congeneric and texanais assigned to Tartarocreagris.The male described above is the first male knownin the genus. It is interesting and intriguing that itsgenitalia are quite similar to those of Microcreagrisphyllisae, which Curcic has placed in the genusSaetigerocreagris (1984: 158). Further study will berequired to resolve the questions posed by this similarity,as very little has yet been recorded about thegenitalia of neobisioid pseudoscorpions.Curcic (1989:360) has assigned M. texana to thegenus Australinocreagris Curcic (1984), which isbased on M. grahami Muchmore from CalaverasCounty, California. I fmd that assignment untenablebecause the internal genitalia of the male texana arequite different from those of male grahami, whichlack the conspicuous rounded dorsal sacs (personalobservation, to be published elsewhere).It should be noted that T. texana has been declaredan endangered species (see Chambers andJahrsdoerfer, 1988) along with some other invertebratesin Travis and Williamson Counties.Tartarocreagris reddelli (Muchmore),NEW COMBINATIONMicrocreagris reddelli Muchmore, 1969:17, fig. 11.Microcreagris spp. (in part): Reddell, 1970:403.Australinocreagris reddelli (Muchmore): Curcic,1989:360, figs. 7, 16.This species was described from the holotype femalecollected in McDonald (Schulze) Cave, TravisCounty, Texas. More recently, two other specimensreferable to this species have been found.New records.-A topotype male from McDonaldCave, Travis County, Texas, 15 December 1988 (J.Reddell and M. Reyes) and a female from Beck'sRanch Cave, Williamson County, Texas, 23 June1968 (J. Reddell and R.W. Mitchell); mounted onslides (FSCA). Both the female and the male arevery similar to the holotype female in most respects,with the following noteworthy features. While theholotype has 21 setae on the carapace, with 4 at bothanterior and posterior margins, the new specimenshave 24 (male) and 25 (female) setae, with 4 at anteriorand 5 at posterior margins. The holotype has 3long setae on the apex of each palpal coxa, but eachof the new specimens has 4 setae in that position.Anterior genital operculum of female with 8 setae,posterior operculum with 12. Male genitalia muchlike those of T. texana, with 2 rounded sacs lyingdorsal to a single ventral sac; chaetotaxy23:[3-3]:(6)12/14(7): (4)11(4):-. Both of the newspecimens have 6 (rather than 7) setae on the handof the chelicera; galea smaller in male than female,bifurcate in distal half, each ramus with 0-2spinules. Ratios of palpal segments, male (female):132

trochanter 2.5 (2.6), femur 5.55 (5.55), tibia 4.05(4.0), chela (with pedicel) 5.05 (5.1), hand (withpedicel) 2.2 (2.25), movable finger/hand withpedicel 1.53 (1.43).Measurements (mm), male (female).-Bodylength 3.42 (3.08). Carapace length 1.04 (1.08).Chelicera 0.605 (0.67) long. Palpal trochanter 0.63by 0.25 (0.70 by 0.27); femur 1.39 by 0.25 (1.47by 0.265); tibia 1.26 by 0.31 (1.325 by 0.33); chela(including pedicel) 2.43 by 0.48 (2.65 by 0.52);hand (including pedicel) 1.00 by 0.45 (1.13 by0.50); movable finger 1.53 (1.62) long. Leg IV:entire femur 1.10 by 0.225 (1.16 by 0.25); tibia1.07 by 0.125 (1.10 by 0.13).Remarks.-As with M. texana Curcic (1989:360) has assigned this species to the genus Australi-nocreagris. However, the internal genitalia of themale are very similar to those of texana, and showthat reddelli is congeneric with texana and not withgrahami.It should be noted that Schulze Cave and McDonald Cave are the same; the preferred name isMcDonald Cave.Tartarocreagris comanche, new speciesFigs. 5-7Type-data.-Holotype female (WM7367.01001)from New Comanche Trail Cave, Travis County,Texas, 26 January 1989 (J. Reddell and M. Reyes);mounted on slide (FSCA).Diagnosis.-A moderate sized species (palpal5Figs. 5-7.-Tanarocreagris comanche, new species, holotype female: 5, chaetotaxy of genital opercula; 6, right palp, dorsal view;7, left chela, lateral view (darkened areoles are underneath).133

chela 1.75 mm long), with relatively robust appendages(chela LIB = 3.25), and with poorly developedeyes.Description of female (male unknown).-Bodytan, palps light brown. Carapace longer than broad;with a low, triangular epistome; 4 small, scarcelycomeate eyes; about 30 setae, 4 at anterior and 7 atposterior margin. Coxal area typical; palpal coxaewith 4 apical setae on right and 5 on left.Abdomen long ovate. Tergal chaetotaxy 8: 13: 13:13:13:16:12:16:12:11:3:2. Anterior genital operculumwith 3 or 4 setae on either side of midline, posterioroperculum with 13 marginal setae (Fig. 5);each spiracle with 6 or 7 guard setae; stemites with14-15 marginal setae, and, in addition, stemites 6-8with 2 setae on face near middle.Chelicera 0.65 as long as carapace; hand with 7setae; galea bifurcated near end, each ramus with2-3 terminal branches; flagellum of 8 pinnate setae.Palp rather short and heavy (Fig. 6); femur 1.15and chela 1.67 times as long as carapace. Trochanter2.0, femur 3.8, tibia 2.7, and chela (includingpedicel) 3.25 times as long as broad; hand(including pedicel) 1.55 times as long as deep; depthof hand greater (1.15 times) than breadth; movablefinger 0.93 as long as hand with pedicel. Femur andchelal hand at base of fingers moderately granulate,other parts smooth. Trichobothria as shown in Fig.7. Fixed fmger with 44 and movable finger with 55marginal teeth, only 8-10 at distal end of each fingerwith cusps.Legs normal. Leg IV with entire femur 3.95 andtibia 5.75 times as long as deep. Tibia and each tarsuswith a tactile seta. Subterminal tarsal setaespinous in distal third.Measurements (mm).-Body length 3.51. Carapacelength 0.96. Chelicera length 0.62. Palpal trochanter0.56 by 0.28; femur 1.10 by 0.29; tibia0.97 by 0.36; chela (without pedicel) 1.61 by 0.54;hand (without pedicel) 0.805 by 0.62; pedicel 0.155long; movable finger 0.895 long. Leg IV: entirefemur 0.91 by 0.23; tibia 0.805 by 0.14.Etymology.-The species is named comanche forthe cave in which it is found, New Comanche TrailCave.Remarks.-Tartarocreagris comanche is distinguishedfrom the other known species of the genusby its lesser modification for cave existence; it issmaller and more robust, is darker in color, and has4 (albeit small) eyes. It is almost like an epigeanform, but (in the absence of a non-cave representativeof Tartarocreagris) compared to Cryptocreagrislaudabilis (Hoff) it is much lighter and has poorlydeveloped eyes.In most neobisiid pseudoscorpions the hand ofthe palpal chela is almost round in cross section or alittle flattened in a dorsal-ventral direction, that is,the depth is equal to or less than the breadth. In T.comanche, however, the depth of the hand is greaterthan the breadth, by a significant factor of 1.15. Theimportance of this unusual construction is unknown,but it may be analogous to the situation in some 01piids (e.g. AldabriJlus aldabrinus Chamberlin) andsome chemetids (e.g. Dinocheirus tenoch Chamberlin)where similar deep chelal hands are developed.FAMILY BOCHICIDAE CHAMBERLINGenus Leucohya ChamberlinLeucohya Chamberlin, 1946:7; Muchmore, 1986:26.This genus is represented by only 3 species, fromNuevo LeOn, Mexico, and Texas. All are troglobites.Leucohya texana MuchmoreLeucohya texana Muchmore, 1986:26, figs. 16-18.The holotype female of this species was found inFrio Queen Cave, Uvalde County, Texas. No otherrepresentative is known.FAMILY SYARINIDAE CHAMBERLINGenus Chitrella BeierChitrella Beier, 1932a: 165; Malcolm and Chamberlin,1960:2; Muchmore, 1973:183; 1982:218,219.Representatives of this genus have been recordedfrom the states of Califomia, Utah, New Mexico,Colorado, Tennessee, Virginia, and West Virginia.Three of the 7 known species are troglobitic. Epigeanforms are found in moist litter.Chitrella welbourni, new speciesFigs. 8-10Chitrella sp.: Welboum, 1978:31.Type-data.-Holotype male (WM3757.01001)and paratype male in Berlese of material from134

ottom of entrance pit, Ogle Cave (Ogle Section),Carlsbad Caverns National Park, Eddy County,New Mexico, 2 September 1974 (yV.C. Welbourn);allotype female (WM3340.01001) found on dampflowstone in twilight, Ogle Cave (Rainbow Section),CCNP, Eddy County, New Mexico, 27 May 1973(yV.R. Elliott); all mounted on slides (FSCA).Diagnosis.-Much like Chitrella transversa(Banks) but larger (chela length 1.13-1.41 mmrather than 0.78-1.00 mm) and with more slenderappendages (chela lengthlbreadth 3.55-3.9 ratherthan 2.8-3.35).Description.-Male and female similar, but femaledistinctly larger. Carapace and palps lightbrown, other parts lighter. Carapace longer thanbroad; surface smooth; with 2 indistinct transversefurrows; 4 corneate eyes; about 30 setae, 6 at anteriorand 7-8 at posterior margin.Abdomen long, narrow. Tergal chaetotaxy ofholotype 7:10:11:14:13:14:15:15:15:11:7:2; otherssimilar. Sternal chaetotaxy of male like that ofC. transversa (see Muchmore, 1973:190); that ofholotype 14:[1-1 ]:(3)9114(4):(3)24(3):0/16: 17: 18:17:16:14:4:2; paratype similar but with 4 setae onall stigmatic plates; there is no sensory area onsternite 6; sternites 6-10 with 2 larger setae nearmiddle of row. Anterior sternal chaetotaxy of female7:(4)16(4):(4)12(4):-; sternites 6-10 with 2enlarged setae as in male.Chelicera 0.55-0.6 as long as carapace; hand with5 setae; flagellum apparently of 5 serrate setae; nospinneret visible.Palp rather slender (Fig. 8); femur 4.2 (male),5.05 (female), tibia 2.55-2.7 (male), 3.2 (female),and chela (without pedicel) 3.55-3.75 (male), 3.9(female) times as long as broad; hand (withoutpedicel) 1.55-1.6 (male), 1. 85 (female) times aslong as deep; movable finger 1.35 (male), 1.3(female) times as long as hand. Surfaces smoothexcept few small granules on medial sides of femur,tibia, and chelal hand. Trichobothria typical (Fig.9); t, on movable fmger, is flattened or lanceolatetoward distal end. Fixed fmger with 52-65 cuspedmarginal teeth; movable finger with 52-63 teeth, ofwhich only the distal 10-12 are cusped, the otherslower and rounded. Movable finger with a sensillumbetween trichobothria st and sb. In addition,movable finger with a low thickening of the dentalmargin (denticle?) lateral to proximal teeth (4thfrom proximal end of row in holotype, 6th inallotype) (Fig. 10); this structure not evident inparatype male.Legs rather slender; leg IV with entire femur3.25-3.4 and tibia 5.1- 6.0 times as long as deep. Along tactile seta on tibia and basitarsus of leg IV.Subterminal tarsal setae dentate.Measurements (mm).-Figures given first forholotype male, followed in parentheses by those forparatype male and allotype female. Body length 3.0(3.0-3.05). Carapace length 0.76 (0.79-0.82).Chelicera 0.41 (0.42-0.48) long. Palpal trochanter0.36 (0.35-0.49) by 0.19 (0.21-0.21); femur 0.75(0.76-0.96) by 0.18 (0.18-0.19); tibia 0.63(0.64-0.80) by 0.235 (0.25-0.25); chela (withoutpedicel) 1.13 (1.20-1.41) by 0.30 (0.34-0.36); hand(without pedicel) 0.50 (0.52-0.65) by 0.31(0.33-0.35); pedicel 0.07 (0.075-0.09) long; movablefinger 0.68 (0.69-0.83) long. Leg IV: entirefemur 0.69 (0.695-0.79) by 0.20 (0.215-0.23); tibia0.56 (0.585-0.72) by 0.11 (0.115-0.12).Etymology.-The species is named for W.C.Welboum, who has collected most of the pseudoscorpionmaterial known from caves in NewMexico.Remarks.-It is interesting to note that C.welbourni is similar to C. transversa in not having aspecialized sensory area on the 6th sternite of males.On the other hand, 9 specimens of C. transversaexamined fail to show the lateral thickening(denticle?) of the proximal dental margin as seen inC. welbourni.In C. welbourni (and the other species of Chitrellatreated below) there are 2 transverse furrowson the carapace, rather than one as seen in C. transversa;one (equivalent to that in C. transversa) is locatedjust posterior to the middle of the carapace,while the second is closer to the posterior margin.As Hoff pointed out (1956b:22), these "furrows"are noticeable more because of underlyingthickening of the cuticle than depression of thesurface.Chitrella major, new speciesFig. 11Type-data.- Holotype female (WM4571.01002)and 2 paratypes (1 female, 1 tritonymph) from FernCave, Val Verde County, Texas, 14 April 1973(T.R. Mollhagen); mounted on slides (FSCA).Diagnosis.-This is the largest known species ofChitrella in the United States (palpal chela > 1.4mm in length).Description of female (male unknown).-Carapaceand palps reddish brown, other parts tan. Carapacelonger than broad; surface smooth; with 2 distincttransverse furrows; 4 corneate eyes, posteriorpair smaller; 38-40 setae, with 7-9 at anterior and 8at posterior margin135

Abdomen long, narrow. Tergal chaetotaxy ofholotype 7: 11: 13: 12: 14: 17: 15: 15: 15: 13: 11 :2. Sternalchaetotaxy of holotype 9:(4)19(4):4)15(4):18:18:17:18:17:14:4:2; stemites 7-10 with 2 largersetae near middle of row.Chelicera about 0.55 as long as carapace; handwith 5 acuminate setae; flagellum of 5 denticulatesetae; no spinneret visible.Palp rather slender (Fig. 11); femur 4.25, tibia2.75-2.8, and chela (without pedicel) 3.1-3.2 timesas long as broad; hand (without pedicel 1.6 times aslong as deep; movable finger about 1.15 times aslong as hand. Surfaces smooth except few smallgranules on medial sides of femur, tibia and chelalhand. Trichobothria typical; t flattened or lanceolatetoward distal end. Fixed chelal fmger with 56-61and movable finger with 55-62 marginal teeth.Movable finger with sensillum distad oftrichobothrium sb; also with a low thickening of thedental margin as seen in C. welbourni, nearpenultimate tooth in holotype and 7th tooth inparatype female (this structure not apparent intritonymph paratype, but present in tritonymph fromH.T. Miers Cave).Legs rather slender; leg IV with entire femur 3.3and tibia 5.7-6.0 times as long as deep. A long~10Figs. 8-13.-8-10, Chitre/la we/bourni, new species: 8, right palp of holotype male, dorsal view; 9, left chela of allotype female,lateral view (darkened areoles are underneath); 10, most proximal teeth on movable finger of holotype male, showing lateral thickening(denticle?) of finger margin; II, Chilre/la major, new species, female: right palp, dorsal view; 12-13, Chilre/la e/liolli, new species,holotype male: 12, sternites 4-6; 13, right palp, dorsal view.136

tactile seta on tibia and basitarsus. Subterminaltarsal setae denticulate.Tritonymph.-Similar to adults but less heavilysclerotized, smaller, and with less slender appendages.Carapace without the transverse furrows soevident in adults; 4 corneate eyes, posterior pairsmaller; 6 setae at anterior and 7 at posteriormargin. Chelicera with 5 setae on hand and 5 setaein flagellum. Palp with femur 3.65, tibia 2.45, andchela (without pedicel) 3.0 times as long as broad.Fixed chelal finger with 46 and movable finger with44 marginal teeth; movable finger with sensillum onlateral surface between trichobothria st and b, closerto st; the paratype without a proximal "denticle" onthe dental margin. Leg IV with entire femur 3.0times as long as deep.In addition to the paratype tritonymph, there is athand a very similar tritonymph from H.T. MiersCave, Del Rio, Val Verde County, Texas, 29 January1984 (P. Sprouse); mounted on slide (FSCA).Measurements (mm).-Adults: Figures givenfirst for holotype, followed in parentheses by thosefor paratype. Body length 3.84 (3.83). Carapacelength 0.975 (1.00). Chelicera 0.52 (0.54) long.Palpal trochanter 0.56 (0.56) by 0.26 (0.26); femur1.00 (1.00) by 0.235 (0.235); tibia 0.87 (0.89) by0.31 (0.325); chela (without pedicel) 1.44 (1.50) by0.465 (0.47); hand (without pedicel) 0.69 (0.73) by0.43 (0.45); pedicel 0.11 (0.12) long; movable finger0.805 (0.815) long. Leg IV: entire femur 0.83(0.83) by 0.25 (0.25); tibia 0.75 (0.74) by 0.125(0.13).Tritonymph: Body length 2.57. Carapace length0.68. Chelicera 0.37. Palpal trochanter 0.36 by0.17; femur 0.62 by 0.17; tibia 0.54 by 0.22; chela(without pedicel) 0.955 by 0.32; hand (withoutpedicel) 0.465 by 0.29; pedicel 0.08 long; movablefinger 0.555 long. Leg IV: entire femur 0.55 by0.185; tibia 0.46 by 0.095.Etymology.-The species is named major for itslarge size among western U.S. species of Chitrella.Chitre/La elliotti, new speciesFigs. 12, 13Type-data.-Holotype male (WM4568.01OO1)and allotype female (WM4568.01002) from FeltonCave, Sutton County, Texas, 26 October 1974(W.R. Elliott); mounted on slides (FSCA).Diagnosis.-Similar to C. frallSversa (Banks) inmany respects; intermediate in size between C.traIlSversa and C. major, with palpal chela about 1.2mm long.Description.-Male and female essentially alike.Carapace and palps light brown, other parts lighter.Carapace longer than broad; surface smooth; with 2distinct transverse "furrows"; 4 corneate eyes; about36 setae, with 6 at anterior and 8 at posterior marglO.Abdomen long, narrow. Tergal chaetotaxy of holotype8: 12: 12: 14: 14: 15: 15: 14: 15: 11 :7:2; allotypesimilar. Sternal chaetotaxy of male 7:[1-1]:(3)8/14(3):(4)23(3): 13/16: 16: 18: 15: 16: 14:5:2; nosensory area on sternite 6 (Fig. 12); sternites 7-10with 2 enlarged setae near middle of row. Anteriorsternal chaetotaxy of female 8:(4)19(4):(3)14(4):-;sternites 7-10 with 2 larger setae as in male.Chelicera about Ih as long as carapace; hand with5 acuminate setae; flagellum of 5 serrate setae; nospinneret visible.Palp moderately slender (Fig. 13); femur3.8-3.9; tibia 2.35-2.5, and chela (without pedicel)2.85-3.15 times as long as broad; hand (withoutpedicel) 1.4-1.55 times as long as deep; movablefinger 1.17-1.22 times as long as hand. Surfacessmooth, except few small granules on medial sidesof femur, tibia and chelal hand. Trichobothriatypical; t, on movable finger, flattened or lanceolatetoward distal end. Fixed finger with 53-56 andmovable finger with 54-61 marginal teeth. Sensillumon movable finger between trichobothria st and sb;no thickening of dental margin (denticle?) such asthat seen in C. welbourni and C. major.Legs moderately slender; leg IV with entirefemur 3.05-3.15 and tibia 4.6-4.85 times as long asdeep. A long tactile seta on tibia and basitarsus ofleg IV. Subterminal tarsal setae dentate.Measurements (mm).-Figures given first forholotype male, followed in parentheses by those forallotype female. Body length 2.78 (3.42). Carapacelength 0.82 (0.83). Chelicera 0.42 (0.46) long. Palpaltrochanter 0.445 (0.48) by 0.21 (0.22); femur0.82 (0.83) by 0.21 (0.22); tibia 0.70 (0.70) by0.28 (0.295); chela (without pedicel) 1.20 (1.22) by0.38 (0.43); hand (without pedicel) 0.555 (0.59) by0.36 (0.42); pedicel 0.11 long; movable finger 0.68(0.69) long. Leg IV: entire femur 0.695 (0.70) by0.22 (0.23); tibia 0.57 (0.58) by 0.12 (0.125).Etymology.-Tbe new species is named for WilliamR. Elliott, who collected the type specimensand many other interesting pseudoscorpions in thesouthwestern states.Chitrella, sp. indet.A single male was found in Lower Sloth Cave,Guadalupe Mountains National Park, CulbersonCounty, Texas (W.C. Weibourn), which IS137

mutilated and has important parts missing or broken.From the data available it seems close to C.welboumi, with which it may be conspecific.However, a fmal determination will have to awaitcollection and study of additional material.Lower Sloth Cave also supports a population ofArcheolarca guadalupensis (see below).FAMILY GARYPIDAE HANSENGenus Archeolarca Hoff and ClawsonArcheolarca Hoff and Clawson, 1952:2.Five species of Archeolarca are known, all frompack rat nests and/or caves in Utah, Oregon, California,Arizona, New Mexico and Texas.Archeolarca guadalupensis MuchmoreArcheolarca guadalupensis Muchmore, 1981:54,figs. 9, 10.The 7 types (2 males, 2 females, 2 tritonymphs,1 deutonymph) were collected in Lower Sloth Cave,Guadalupe Mountains National Park, CulbersonCounty, Texas. No other representative is known.Archeolarca, sp. indet.Three nymphs belonging to this genus were collectedby Berlese separation of bat guano in TruckettGuano Cave, Valencia County, New Mexico, 11September 1976, by W.C. Welbourn. It is impossibleto determine the specific identification of theseimmature specimens.FAMILY CHEIRlDIIDAE CHAMBERLINGenus Cheiridium MengeCheiridium Menge, 1855:36; Hoff, 1952: 188.Four species of Cheiridium are presently knownfrom the United States, as discussed below. All aretiny creatures, usually found in rather dry litter.Cheiridium reyesi, new speciesFigs. 14, 15Type-data.-Holotype female (WM7363.01001)from Cot Cave, Kinney County, Texas, 7 October1987 (M. Reyes); mounted on slide (FSCA).Diagnosis.-Similar to C. finnum Hoff (1952)from which it can be distinguished by its larger size(palpal chela 0.40 mm long vs. 0.33-0.36 mm) andmore slender appendages (palpal femur 4.25 timesas long as broad vs. 3.25-3.5 times).Description of female (male unknown).-Withthe general characters of the genus. All parts lightbrown. Carapace subtriangular, with a median transversefurrow and a shallow depression near posteriormargin; surface granulate anterior to transverse furrow,reticulate behind; 2 small, corneate eyes; setaeslender, curved, 4 at anterior and 8-10 at posteriormargin. Abdomen ovate; tergites divided. About 20slender curved setae on middle tergites. Chaetotaxyof sternites not observable.Chelicera about JA as long as carapace; setae onhand not discernible; flagellum of 4 setae; galealong, slender, with 2 or 3 small terminal rami.Palp rather slender (Fig. 14); femur 4.25, tibia2.65, and chela (without pedicel) 2.85 times as longas broad; hand (without pedicel) 1.55 times as longas deep; movable finger 1.05 times as long as hand.Surfaces heavily granulate, with many arcuate setaewith midlateral spinules. Trichobothria as usual inthe genus, fixed fmger with 7 and movable fingerwith 2; et just proximad of middle of fixed fmger,much closer to est than to finger tip (Fig. 15). Eachfinger with 8-10 small marginal teeth, conical orrounded distally, becoming flattened proximally.Legs rather slender; leg IV with entire femur 5.3and tibia 4.65 times as long as deep. Articulationbetween basifemur and telofemur of all legs faintlyvisible (as in Figs. 17 and 18).Measurements (mm).-Body length 1.04. Carapacelength 0.34, posterior breadth 0.41. Chelicera0.09 long. Palpal trochanter 0.16 by 0.085; femur0.32 by 0.075; tibia 0.25 by 0.095; chela (withoutpedicel) 0.40 by 0.14; hand (without pedicel) 0.20by 0.13; pedicel 0.03 long; movable finger 0.21long. Leg IV: entire femur 0.265 by 0.05; tibia 0.21by 0.045; tarsus 0.185 by 0.03.Etymology.- The new species is named for MarcelinoReyes who collected the type specimen.Remarks.-Until now, only 3 species ofCheiridium have been reported from the UnitedStates: C. finnum Hoff (1952), C. insperatum Hoffand Clawson (1952), and C. museorum (Leach) (seeMuchmore, 1972). Of these, C. museorum is animmigrant from Europe found in Massachusetts andis larger and has more slender palps than the newspecies. C. insperatum was described from Utah andhas subsequently been identified in California(unpublished); it is even larger and has even slendererpalps. Cheiridium finnum is recorded onlyfrom Illinois, but a specimen from a cave in138

Missouri may be conspecific (unpublished); it issmaller and more robust than the new species.Cheiridium reyesi is apparently more closely relatedto the midwestern forms and may be somewhatmodified for cavernicolous existence.It is interesting to note that the articulations betweenfemoral segments are distinctly visible in alllegs. This is at variance with the condition in manyspecies of Cheiridium and related genera, where thefemora appear completely fused, with no visible suturebetween basifemur and telofemur.Genus Apocheiridium ChamberlinApocheiridium Chamberlin, 1924:34; Benedict,1978:231.Seven species of Apocheiridium have been knownfrom across the United States. As representatives areusually found under the bark of living trees, it issurprising that a new form from Texas IScavernicolous.Apocheiridium reddelli, new speciesFigs. 16-18Apocheiridium sp.: Reddell, 1970:402.Type-data.-Holotype male (WM900.01001)from Devil's Sinkhole, near Rocksprings, EdwardsCounty, Texas, 14 March 1966 (J. Reddell);mounted on slide (FSCA).Diagnosis.-Similar to A. stanndardi Hoff(1952) in general shape, but a little smaller (palpalchela 0.33 rom long), with surfaces much lessgranulate, and with chelal fingers shorter than hand.Description of male (female unknown).-Withthe general characters of the genus (see Benedict,1978). Carapace and tergites not granulate, but distinctlyreticulate; setae small, acuminate. Carapacewith shallow, median transverse furrow; 2prominent corneate eyes; 4 setae at anterior and 12at posterior margin. Abdomen ovate, 11 segmentsvisible from above; tergites 1-9 divided; tergalchaetotaxy 15:16: 15:16:21:21:19:19:-; sternal1518Figs. 14-18.-14-15, Cheiridium reyesi, new species, ho!otype female: 14, left palp, dorsal view; IS, right chela, dorsolateral view(darkened areoles are underneath); 16-18, Apocheiridium reddelli, new species, ho(otype male: 16, right palp, dorsal view; 17, leg I,139

chaetotaxy not discernible. Internal genitalia likethat shown by Chamberlin (1931:fig. SID).Chelicera \4 as long as carapace; 4 setae on hand;flagellum of 3 unequal setae; galea short, simple.Palp relatively slender (Fig. 16): femur 5.2, tibia3.25, and chela (without pedicel) 3.45 times as longas broad; hand (without pedicel) 2.4 times as long asdeep; movable finger 0.89 as long as hand. Femurwith a small, proximal bulge. Surfaces with sparse,small granules, no large granules. Some setaearcuate and with midlateral spinules. Trichobothriaas usual in the genus, 7 on fixed finger and one onmovable finger; et well distad of middle of fixedfinger, much closer to finger tip than to est.Marginal teeth on fingers not discernible.Legs rather slender: leg IV with entire femur 4.5and tibia 4.5 times as long as deep. Femora of alllegs show the articulation between basifemur andtelofemur (Figs. 17-18).Measurements (mm).-Body length LIS. Carapacelength 0.325, posterior breadth 0.35. Chelicera0.08 long. Palpal trochanter 0.12 by 0.075; femur0.265 by 0.05; tibia 0.21 by 0.065; chela (withoutpedicel) 0.33 by 0.095; hand 0.18 by 0.075; pedicel0.03 long; movable finger 0.16 long. Leg IV: entirefemur 0.22 by 0.05; tibia 0.155 by 0.035; tarsus0.13 by 0.025.Etymology.-The species is named for James R.Reddell in honor of his great contributions to thestudy of the cave fauna ofTexas.Remarks.-Apocheiridium reddelli appears mostclosely related to A. stannardi Hoff (1952, 1961),which is known from Illinois, Michigan and Colorado.Its lighter color, reduction of granulation, andmore slender palps are probably adaptations to thecave habitat. In the reduction of granulation A. reddelliresembles A. inexpectum Chamberlin fromsouthern California and Baja California, Mexico(see Benedict, 1978); from that tree-dwelling speciesit is distinguished by its smaller size and the chelalfingers shorter than the hand. Other unusual, but unexplained,characters of A. reddelli are the relativelyfar distal position of trichobothrium et and theobvious articulations between the femoral segmentsof all legs (which can also be seen in Cheiridiumreyesi, above).FAMILY CHERNETIDAE CHAMBERLINGenus Neoallochernes HoffNeoallochernes Hoff, 1947:499.Tejachernes Hoff, 1957:83 (NEW SYNONYMY).The genus Neoallochernes is based on Chelanopsgarcianus Banks, 1909, from Havana, Cuba. Hoff(1947) studied the type collection (in MCZ) consisting,according to him, of 3 females and 1 tritonymph(?);one of the adults was mounted on aslide and designated the lectotype. Redescription ofthe type species and diagnosis of the genus wasbased on the mounted lectotype, supplemented bycertain data from the 2 other adults. I have had theopportunity to restudy the lectotype and to mountand study the 2 other adult specimens. In most respects,the descriptions by Hoff are found to be accurate;however, a few corrections and additions areprovided below.In addition, there is available from Las Villas,Cuba, a collection of pseudoscorpions which arecertainly congeneric with Chelanops garcianus.They are described below as N. cubanus.On the basis of new information gained recently,it is now possible to diagnose the genus Neoallochernesin up-to-date terms.Diagnosis (revised).-A genus of the familyChernetidae. Carapace longer than broad; surfacestrongly granulate; 2 conspicuous transverse furrows;2 distinct eyespots; setae narrow clavodentate.Abdominal tergites 1-10 and stemites 5-10 divided;tergal surfaces granulate, setae narrow clavodentateto terminally denticulate, no acuminate tactile setaeon tergite II; sternal surfaces smooth, setae mostlyacuminate, lIth sternite with 2 (lateral) long, acuminatetactile setae. Anterior operculum of male withgroup of 40-50 setae, including 4 elongated ones;posterior operculum with 4-8 small, internal guardsetae on each side and a marginal row of 10-15larger setae. Anterior operculum of female with ahorseshoe-shaped group of 30-40 setae; posterioroperculum with row of 10-15. Internal genitalia ofmale large and heavily sclerotized (Fig. 19), withlarge wrinkled lateral sacs. Each spermatheca of femalein form of an elongated sac with a small terminalcribriform plate and a short slender tubule leadinginto a single median chamber with a large ovoidcribriform plate (Fig. 20). Chelicera with 4 setae onhand (sbs absent), bs moderately long, denticulate,and the others long, acuminate; flagellum of 3 setae,at least the distal 2 denticulate; galea fairly robustbut with only a few small rami, smaller in male thanfemale. Palp fairly robust, less so in female thanmale; surfaces partly granulate; setae strong, narrowclavodentate to denticulate; both chelal fingers withseveral accessory teeth; venom apparatus well developedin movable fmger, with nodus ramosus at orproximad of level of trichobothrium st; usually avestigial venom duct apparent in fixed finger.140

Trichobothriotaxy of chela generally typical forchernetids (Fig. 22); movable finger with t and Sfclose together in distal half and b and sb close togetherin proximal half; fixed finger with ist a littledistad of est, which is proximad of middle, and iband isb near level of esb at base of finger. Legs typical;surfaces of femora granulate to scaly. Tarsus ofleg IV with a long, terminally denticulate seta about3/4 length of segment from proximal end.Remarks.-Dinocheirus stercoreus Turk, thetype species of the genus Tejachernes Hoff, is,without doubt, a representative of Neoallochernes(see below).Its peculiar combination of characters, especiallythe unique spermathecae of the female, placesNeoallochernes well apart from other genera in theChernetidae, according to our present knowledge.However, further study of the genitalia of the manyother relatively unknown American forms can beexpected to shed more light on its affinities.Geographic distribution.-In addition to therepresentatives from Cuba and Texas treated below,many other specimens belonging to Neoallochernesare presently under study. These are from Mexicoand several Central American countries.Neoallochernes garcianus (Banks)Chelanops garcianus Banks, 1909:147.Dinocheirus garcianus (Banks): Beier, 1932b:139.Neoallochernes garcianus (Banks): Hoff, 1947:500,figs. 13, 14.Types examined.-Lectotype male and 2paratype males from Havana, Cuba (Baker); lectotypemounted on slide by Hoff and paratypesmounted by me (MCZ).As mentioned above, Hoff has given a full andmostly accurate account of this species. However, acouple of corrections and additions should be mentioned,based upon my restudy of the type material.Firstly, the adult specimens, including the lectotype,are all males, rather than females as Hoffstated (1947:500); the genital features are typicallymale. Thus it is the female which is absent from thetype collection, making a characterization of thefemale genitalia impossible.Hoff was correct in stating "Tarsus of fourth legwithout tactile seta" (1947:499), if "tactile seta" isunderstood to be an acuminate seta larger than theusual investing setae. However, careful examinationshows that the lectotype does possess, on the distalend of the fourth tarsus, a conspicuously larger setawhich bears a few spinulations near the tip (a"pseudotactile seta").Hoff reported (1947:500) that the cheliceral flagellumis composed of 3 setae, but he failed to mentionthe number and nature of the setae on the handof the chelicera. Reexamination confirms that thereare 3 setae in the flagellum in all of the types andshows that the cheliceral hand bears only 4 setae,with Is, is and es long, acuminate, and bs shorterand denticulate.Hoff mentioned (1947:499, 501) that tactile seta(trichobothrium) ist is near level of est. This is correct,but it should be noted that ist is always a littledistal to level of est, which is proximad of middle offinger.Neoallochernes cubanus, new speciesFigs. 19-22Type-data.-Holotype male (WM4169.01oo2),allotype female (WM4169.01008) and about 90paratypes "en nido de murci6lagos Tadarida minutay T. laticauda en palma Copernicia vespenilioflum"at Estero Real, Mayajigua, Las Villas, Cuba, 8March 1973 (J. de la Cruz). The holotype, allotypeand 7 paratypes (4 males, 3 females) mounted onslides; all these and most paratypes in alcohol depositedin the Academia de Ciencias de Cuba, 15paratypes in alcohol in FSCA.Diagnosis.-Generally like N. garcianus butlarger (chela of male 1.04-1.27 mm long, ratherthan 0.87-0.95) and with more slender appendages(chela of male 2.95-3.25 times as long as broad,rather than 2.65-2.85).Description.-Male and female generally similar,but female a little smaller and with more slenderappendages. Carapace, tergites and palps wellsclerotized and light brown in color, other partsthinner, tan. Carapace a little longer than broad;surface strongly granulate; 2 deep transversefurrows; 2 distinct eyespots; setae denticulate tonarrow clavodentate. Tergites 1-10 divided; surfacesheavily granulate; setae long clavodentate;chaetotaxy of holotype 18:21: 18:20:24:23:20:23:20: 16: 15:2; no tactile setae on tergite 11. Sternites5-10 divided; surfaces smooth; setae mostlyacuminate; chaetotaxy of holotype (male) 55:[6-8]:(3)1O(4):(3)11(1):22:18:18:17:17:12:T3T:2; with 4long setae among the 55 on the anterior operculumand 2 long, acuminate tactile setae on sternite 11;anterior sternal chaetotaxy of allotype (female)36:(4)8(2):(3)13(2):20:20:-. Internal genitalia ofmale (Fig. 19) heavily sclerotized and large,0.55-0.62 as long as carapace. Each spermatheca of141

female in form of an elongated sac with a small terminalcribriform plate and a short slender tubuleleading into a single median chamber with a largeovoid cribriform plate (Fig. 20).Chelicera with 4 setae on hand, bs denticulate,others long, acuminate; flagellum of 3 setae, at leastthe distal 2 denticulate; galea fairly robust, but withonly a few small rami, smaller in male than in female.Palp rather robust (Fig. 21), more so in malethan in female; trochanter 1.45-1.8, femur 2.55-3.2,tibia 2.1-2.7, and chela (without pedicel) 2.95-3.3times as long as broad; hand (without pedicel)1.4-1.7 times as long as deep; movable finger0.94-1.04 times as long as hand. Surfaces partlygranulate; setae strong, narrow clavodentate todenticulate. Trichobothria as shown in Fig. 22; onmovable finger st close to t and sb close to b, thedistances between each pair about equal; on fixedfinger ist distal to level of est near middle of fingerand eb, esb, ib and isb close together on base offmger, at or proximal to level of last marginal tooth.Each fmger with 35-45 cusped marginal teeth andseveral accessory teeth, internal and external.Movable finger with well developed venedens andvenom duct, nodus ramosus at or proximal to levelof trichobothrium st; fixed finger with shortvenedens and vestigial venom duct.Legs rather slender; leg IV with entire femur3.6-4.1 and tibia 5.3-6.2 times as long as deep. Tarsusof leg IV with a long, denticulate"pseudotactile" seta about 3/4 length of segmentfrom proximal end.Measurements (mm).-Male (figures given firstfor holotype, followed in parentheses by ranges forthe 4 mounted paratypes). Body length 3.02(2.38-2.97). Carapace length 0.95 (0.79-0.95).Chelicera length 0.265 (0.245-0.27). Palpal trochanter0.48 (0.40-0.49) by 0.33 (0.24-0.30); femur0.93 (0.75-0.90) by 0.37 (0.27-0.35); tibia 0.82(0.67-0.81) by 0.39 (0.28-0.36); chela (withoutpedicel) 1.27 (1.04-1.25) by 0.44 (0.33-0.41); hand(without pedicel) 0.66 (0.53-0.67) by 0.48(0.34-0.44); pedicel 0.08-0.11 long; movable fingerlength 0.63 (0.54-0.65). Leg IV: entire femur 0.87(0.71-0.87) by 0.24 (0.19-0.24); tibia 0.71(0.58-0.71) by 0.13 (0.10-0.13); tarsus 0.52(0.45-0.525) by 0.085 (0.08-0.085).I192120Figs. 19-22.-Neoallochemes cubanus, new species: 19, internal genitalia ofholotype male, ventral view; 20, spermathecae of al1otypefemale; 21, right palp of holotype male,. dorsal view; 22, left chela of holotype, lateral view (darkened areoles are underneath).Scale lines = 0.1 mm for Figs. 19 and 20, 0.5 mm for Figs. 21 and 22.142

Female (figures given first for allotype, followedin parentheses by ranges for the 3 mountedparatypes). Body length 2.90 (2.88-2.95). Carapacelength 0.83 (0.81-0.86). Chelicera length 0.26(0.245-0.25). Palpal trochanter 0.40 (0.39-0.41) by0.23 (0.22-0.24); femur 0.73 (0.73-0.74) by 0.24(0.23-0.24); tibia 0.61 (0.64) by 0.24 (0.24-0.26);chela (without pedicel) 0.98 (1.01) by 0.32(0.30-0.33); hand (without pedicel) 0.53(0.53-0.54) by 0.33 (0.32-0.33); pedicel 0.07-0.08long; movable finger length 0.50 (0.50-0.53). LegIV: entire femur 0.73 (0.70-0.75) by 0.19(0.17-0.20); tibia 0.57 (0.60-0.62) by 0.10 (0.10);tarsus 0.42 (0.42-0.45) by 0.075 (0.075-0.08).Etymology.-The species is named for Cuba,where it has been found.Remarks.-Though the males of N. cubanus arelarger and have more slender appendages than themales of N. garcianus, they are quite similar to thelatter in all qualitative characters. Therefore, it iscertain that they are congeneric; and the accompanyingfemales can then illustrate the female characteristicsof the genus, particularly the unique spermathecae.It is interesting to note that the specimens of N.cubanus were found in or at (en) a roost (nido) ofthe bats Tadarida minuta (Miller) and T. latieaudata(E. Geoffroy) in a palm, Coperniea vespertilionumLe6n (see Silva-Taboada and Koopman, 1964). Noecological data are available for the types of N.garcianus. All other known specimens of Neoalloehernes,from southwestern United States andCentral America, have been collected in caves. Mostof these collections are not accompanied byecological data, but of those that are all are from batguano, usually of Tadarida brasiliensis (I. Geof.St.-Hilaire).Neoalloehernes stereoreus (Turk),NEW COMBINATIONDinoeheirus stereoreus Turk, 1949:121; Hoff,1958: 28.Tejaehernes stereoreus (Turk): Hoff, 1957:84, figs.1-5; Reddell, 1965:167.Tejachernes sp.: Reddell, 1970:403.This species was described by Turk (1949) on thebasis of 2 poorly prepared specimens from BrackenCave, Comal County, Texas, and was placed by himin the genus Dinocheirus. Hoff (1957), after studyof 11 topotypes (3 males, 8 females) and severalspecimens from Frio Cave, Uvalde County, Texas,recognized that stereoreus does not belong inDinoeheirus and he erected a new genus,Tejaehernes, to accommodate it.The descriptions by Turk and Hoff have characterizedthe species fairly well, but a few additionaldata are warranted here, based upon study of themany specimens listed below.There is much more variation in measurementsthan Hoff observed (1957:87). On average, malesare larger and more robust than females, but there isconsiderable overlap between the 2 sexes (Table 1).Though the species is quite variable in size andproportions, even for a chernetid, it is remarkablyconsistent in qualitative features. The distinctive genericcharacters are always evident: 3 setae in thecheliceral flagellum; 4 setae on the cheliceral hand,with bs denticulate and es longer, acuminate; arrangementof trichobothria on palpal chela, movablefmger with t and st close together in distal half andb and sb close together in proximal half, and fixedfinger with est proximad of middle (see Hoff,1957:figs. 2, 5); movable chelal finger with well developedvenom apparatus, nodus ramosus usuallyproximad of trichobothrium st, and small, vestigialvenom duct in fixed finger; tarsus of leg IV withoutan acuminate tactile seta but with a prominent denticulateseta about 3/4 length of segment fromproximal end; spermathecae of female "relativelyshort, a little less than the distal one-half of each inform of a weakly swollen, pyriform or cylindricalsac capped by a small cribriform plate," the 2 tubesconnected to a central sac with a conspicuous, ovoidcribriform plate (Hoff, 1957:84, figs. 3a-c). In addition,carapace and palps strongly granulate, the carapacewith 2 distinct transverse furrows and 2Table I.-Ranges of measurements (mm) and proportions of43 mounted males, 19 mounted females and 5mounted.tritonymphs of Neoallochemes stercoreus from 10caves in Texas. Abbreviations: L=Length, B=Breadth,D = Depth; • indicates length exclusive of pedicel.Males Females TritonymphsBody L 2.05-3.10 2.15-3.05 1.60-2.40Carapace L 0.73-0.94 0.71-0.90 0.64-0.68Chelicera L 0.20-0.28 0.20-0.26 0.19-0.21Palpal femur L 0.56-0.87 0.51-0.69 0.45-0.52LIB 2.15-2.70 2.60-2.95 2.10-2.20Palpal tibia L 0.52-0.79 0.46-0.62 0.41-0.47LIB 1.90-2.25 2.15-2.45 1.95-2.10Palpal chela L· 0.86-1.23 0.77-0.97 0.69-0.80L·/B 2.40-3.05 2.75-3.15 2.50-2.90Chelal hand L· 0.40-0.59 0.37-0.48 0.36-0.41L·/D 1.09-1.40 1.33-1.50 1.25-1.45Movable finger L 0.47-0.68 0.43-0.53 0.40-0.44finger Llhand L· 0.98-1.20 1.03-1.23 1.06-1.14Leg IV femur L 0.56-0.78 0.50-0.69 0.45-0.49LID 3.15-3.60 2.90-3.60 2.80-3.10143

eyespots; internal genitalia of male large and heavilysclerotized, with large, wrinkled lateral sacs; lithsternite with 2 lateral, acuminate tactile setae, lithtergite lacking such setae.Tritonymph (based on 5 mounted specimens fromComal, Kinney and Val Verde Counties).-Muchlike adults but smaller and with more robustappendages (Table 1). Carapace with heavilygranulate surface; 2 distinct transverse furrows; 2small eyespots. Chelicera with flagellum of 3 setae;4 setae on hand, bs denticulate, es long acuminate;galea like that of female. Palpal surfaces mostlygranulate; chela lacking trichobothria sb and ist, tand st close together; est proximad of middle offinger; vestigial venom duct in fixed finger. Tarsusof leg IV with rather long, denticulate"pseudotactile" seta about 2/3 length of segmentfrom proximal end.New records.-TEXAS: Blanco County: DavisBlowout Cave, 12 April 1970 (Becker and Howden),15 males, 2 females, 1 tritonymph (CNC); 3March 1984 (W.R. Elliott and D. Pate), 5 males, 2females, 1 nymph; 24 July 1984 (W.R. Elliott), 3males. Burnet County: Beaver Creek Bat Cave, Fall1977 (W.R. Elliott), 5 males, 1 female; 28 March1987 (W.R. Elliott), 1 male. Comal County:Bracken Bat Cave, 6 mi. N of Bracken, 19 January1963 (J. Reddell and D. McKenzie), 2 males, 1 tritonymph(1CC); January 1963 (M. Tandy), onguano, 2 males, 2 females, 1 tritonymph (JCC); 5April 1983 (R.M. Waters), 1 male, 2 tritonymphs;10 October 1987 (W.R. Elliott), 1 male, 3 females.Edwards County: Punkin Cave, 9 April 1965 (D.Dickey and J. Reddell), 6 males, 3 females. KendallCounty: Two Step Cave, 20 June 1987 (A. Cobb), 1male, 2 nymphs. Kerr County: Stowers Cave, 25March 1971 (S. Wiley and T. Mollhagen), 1 male, 1female; date? (R. Bartholomew), 7 males, 2 females,2 nymphs. Kinney County: Cricket SiphonCave, 27 February 1988 (J. Ivy, M. Ulmer, D.Pearson), 2 males, 2 females; Porcupine Cave, 17October 1987 (G. Veni and J. Ivy), 1 male; WebbCave, 10 mi. N of Brackettville (W. Russell), 2males, 3 females, 1 tritonymph (JCC). MasonCounty: James River Bat Cave, 29 May 1988 (W.R.Elliott), 3 males, 2 females. Medina County:Valdina Farms Sinkhole, 12 January 1963, 1 female(1CC); Ney Cave, 21 June 1968 (J. Reddell), inguano of Tadarida brasiliensis mexicana (Saussure),17 males, 5 females, 1 tritonymph; 14 April 1972(S. Wiley, T. Mollhagen, B. Davis), from batguano, 9 males, 5 females. Terrell County: AdamsCave (=Sorcerer's Cave), 16 September 1978 (G.Veni) , 1 female. Uvalde County: Frio Bat Cave, 24January 1970 (B. David and R. McDaniel), 3 males,10 nymphs; 24 March 1971 (S. Wiley and T. Mollhagen),from bat guano, 3 males, 1 female, 1 tritonymph;January 1984 (R.M. Waters), 15 males, 2females; 10 March 1984 (S. Harden), 10 males, 1female. Val Verde County: Cave Hollow Cave,12-13 July 1974 (W.R. Elliott), 3 males, 1 female;Fern Cave, 30 September 1962 (1. Reddell), onguano, 1 tritonymph (ICC); 12 June 1966 (R.W.Mitchell), from guano, 11 males, 30 females, 40nymphs; 19 July 1968 (R.W. Mitchell), "a small[sic!] sample from guano," largely of Tadaridabrasiliensis, 200-300 specimens, all stages; 14 April1973 (T.R. Mollhagen), 1 female; Twin Tree Cave,12 July 1969 (W. Russell and C. Kunath), 4 males;abandoned railroad tunnel 11 miles W Comstock,largely inhabited by Tadarida brasiliensis, 13 April1968 (J. Reddell and T. Mollhagen), berlese sampleof guano, 1 tritonymph.Remarks.-Neoallochernes stercoreus is, withoutany doubt, congeneric with N. garcianus and N.cubanus; it shares with those species all of the importantgeneric characters of Neoallochernes.Neoallochernes stercoreus is common in manybat caves in Texas as far west as Terrell County.However, in spite of considerable collecting, it hasnot been found in the extensive bat caves in EddyCounty, New Mexico, where one might expect conditionssimilar to those in Texas. Perhaps the Carlsbadcaves are actually different in some way(temperature, humidity, kinds of bats, kinds of foodorganisms, etc.), or perhaps N. stercoreus has beenexcluded from these caves by competition with Dinocheirusastutus, which is commonly found there.Neoallochernes(?) incertus, new speciesFigs. 23, 24Type-data.-Holotype male (DMI81.01001)from "bottom of 100 foot entrance drop" in LonesomeRidge Deep Pit, Eddy County, New Mexico,14 April 1963 (B. Bell); mounted on slide(AMNH).Diagnosis.-Generally similar to males of N.stercoreus but, most obviously, with more slenderappendages (palpal femur LIB = 3.05, rather than2.5 or less). More subtle differences include the natureof setae bs and es on the cheliceral hand and the"tactile seta" on the tarsus of leg IV, and the placementof trichobothria on the chelal fingers, etc., asdiscussed below.Description.-Rather lightly sclerotized, all partstan in color. Carapace longer than broad; surfacefinely granulate, especially laterally; 2 distinct144

transverse furrows; eyespots not evident; setae clavodentate.Tergites 1-10 and sternites 4-10 divided;surfaces of tergites finely granulate to scaly, sternitessmooth; dorsal setae clavodentate, ventral setaedenticulate. Tergal chaetotaxy 15:16:15:18:18:20:18:17:16:14:12:2; sternal chaetotaxy ?:(2)18(2):(2)14(2):22:21 :22:21 :20: 16: 12:2; presence of tactilesetae on tergite and sternite 11 uncertain, as lateralsetae are missing. Genitalia large and heavy but otherwisenot describable, as the abdomen has beenseparated from the cephalothorax in this specimenand the genital region is damaged.Chelicera with 4 setae on hand, all acuminate, esshort, less than Ih as long as bs; flagellum of 3 setae;both galeae are broken from the movable fmgersof this specimen.Palp (Fig. 23) more slender than that of N. stercoreus:trochanter 1.95, femur 3.05, tibia 2.4, andchela (without pedicel) 3.15 times as long as broad;hand (without pedicel) 1.65 times as long as deep;movable finger 1.09 times as long as hand. Surfacesfinely granulate, except chelal fingers smooth; mostsetae short, denticulate. Trichobothria as shown inFig. 24; on movable finger the distance between tand st is about 2 times the distance between b andsb; on fixed fmger est lies distad of middle, nearlyat same level as ist; eb, esb, ib and isb close togetheron base of finger, at or distal to level of lastmarginal tooth. Fixed fmger with 45 and movablefinger with 47 cusped marginal teeth; each fingerwith 5-7 internal and external accessory teeth.Venom apparatus well developed in movable finger,with nodus ramosus midway between trichobothria tand st; only a short venedens and vestigial venomduct in fixed finger.Legs rather slender; leg IV with entire femur 4.22324Figs. 23-24.-Neoallochemes(?) incertus, new species, holotype male: 23, left palp, dorsal view; 24, right chela, lateral view(darkened areoles are underneath).145

and tibia 5.4 times as long as deep. Tarsus of leg IVwith a short, acuminate tactile seta 3/4 length ofsegment from proximal end.Measurements (mm).-Body length 2.75. Carapacelength 0.805. Chelicera length 0.275. Palpaltrochanter 0.435 by 0.225; femur 0.715 by 0.235;tibia 0.62 by 0.26; chela (without pedicel) 1.18 by0.375; hand (without pedicel) 0.58 by 0.35; pedicel0.13 long; movable finger length 0.62. Leg IV: entirefemur 0.65 by 0.155; tibia 0.54 by 0.10; tarsus0.43 by 0.08.Etymology.-The species is named incertus becauseof its uncertain taxonomic position.Remarks.-This species is tentatively placed inNeoallochernes on the basis of these characters: 3setae in the cheliceral flagellum; only 4 setae oncheliceral hand, sbs lacking; location of "tactileseta" on tarsus of leg IV at distal 3/4; and possessionof a vestigial venom duct in the fixed chelalfinger. However, it differs from others in that genusin several important characters and may well belongto an hitherto undescribed genus. Some differencesfrom other known members of Neoallochernes arethe following: on the cheliceral hand, seta bs isacuminate rather than denticulate and es is shorterthan bs rather than longer; on the tarsus of leg IVthe "tactile seta" is acuminate rather than denticulate;on the movable chelal finger, the distance betweentrichobothria t and st is twice as great as thedistance between band sb rather than the 2 distancesessentially equal, and the nodus ramosus of thevenom dust is midway between t and st rather thanproximad of st; on the fixed chelal finger, est is distadof the middle rather than proximad, and thegroup eb, esb, ib and isb lies mostly distal to thelevel of the last marginal tooth rather than proximalto the last tooth; and eyespots are apparently absentrather than present. These differences probably indicatethat the new species represents a new genus, butI prefer not to name it now on the basis of a single,damaged male specimen. The differences should beconfirmed on additional material and the nature ofthe genitalia, especially of the female, should beelucidated. At hand are a few specimens from CentralAmerica which appear to be congeneric and mayallow a proper characterization of the taxon.Genus Hesperochernes ChamberlinHesperochernes Chamberlin, 1924b:89; Muchmore,1974:27.Representatives of the genus Hesperochernes arewidely distributed through the temperate andtropical parts of North and Central America. Threespecies have been reported previously from NewMexico, but only one from Texas (Hoff, 1958).Many of the forms are found in close associationwith small mammals, especially packrats, andseveral are recorded from caves, usually on batguano.Hesperochernes occidentalis (Hoff and Bolsterli)Pseudozaona occidentalis Hoff and Bolsterli,1956: 170, figs. 1-3; Hoff, 1958:24.Hesperochernes occidentalis (Hoff and Bolsterli):Muchmore, 1974:30.This species was first described from caves inWashington County, Arkansas. It has since beenfound in many other caves in the Ozark Region ofArkansas, Missouri and Oklahoma (unpublished).There are 2 collections of H. occidentalis fromTexas, both from Edwards County: 4 specimens (1male, 2 females, 1 tritonymph) from Wyatt Cave, 2miles north of Wheat Cave, 21 September 1963 (J.Reddell and D. McKenzie); 3 specimens (1 male, 1female, 1 tritonymph from nest of Petrochelidon fulva(Vieillot) in Dunbar Cave, 22 July 1976 (R.Martin). The species is highly variable in size andproportions and these individuals fit easily into theknown ranges of measurements.Hesperochernes riograndensis Hoff and ClawsonHesperochernes riograndensis Hoff and Clawson,1952:19, figs. 11-12; Hoff, 1958:23.A number of specimens collected from caves inwestern Texas and eastern New Mexico appear tobelong to this species, which has been known previouslyonly from "food storage of a kangaroo rat(Dipodomys) " in Socorro County, New Mexico(Hoff and Clawson, 1952:23). Most of them arelarger than the types described by Hoff and Clawson,but they have the same general shape and proportions.On the other hand, some of them havesome characters similar to those of H. canadensisHoff and H. utahensis Hoff and Clawson, both ofwhich have been reported in the southern RockyMountain region. Until a comprehensive revision ofthe western species of Hesperochernes is completed,it is impossible to be certain of the identities of mostindividuals.Present collections include: 1 female from"below dome 250 feet from entrance" in BlackstoneCave, Terrell County, Texas, 4 June 1963 (J.146

Reddell and W. Russell); I female from MurphyWells Cave, Irion County, Texas, 24 February 1974(R. Ballinger); 3 males from dark zone of WindCave (upper level), Eddy County, New Mexico, 31December 1973 (W.C. Welbourn); 1 female fromHelen's Cave, Carlsbad Caverns National Park,Eddy County, New Mexico, 31 August 1974 (W.e.Welbourn); 1 female near bat guano in EndlessCave, Eddy County, New Mexico, 17 February1975 (W.C. Welbourn); 3 males, 1 female fromThree Fingers Cave, Lincoln National Forest, EddyCounty, New Mexico, 28 May 1978 (W.C. Welbourn);1 female in "twilight zone near porcupinelair," Serpentine Root Cave, NW Tinnie, LincolnCounty, New Mexico, 27 July 1973 (W.C. Welbourn).All are mounted on slides (FSCA).Hesperochernes molestus HoffHesperochernes molestus Hoff, 1956c:33, figs.12-15; Hoff, 1958:24; Hoff, 1959:32.?Hesperochernes sp.: Reddell, 1970:403.This species has been known previously onlyfrom central New Mexico; found in close associationwith rodents, in nests of Neotoma sp., Dipodomyssp., and Perognathus flavus Baird, and onthe hair of Onychomys leucogaster (Wied-Neuwied)(Hoff, 1956c:38).At hand is one collection of 26 specimens (21males, 4 females, 1 deutonymph) which appear tobelong to this species. They were taken from "batguano in the Bat Room," Four Mile Cave, 4 milesN Del Rio, Val Verde County, Texas, 24 April1966 (1. Reddell and E. Alexander); 10 adults (6males, 4 females) mounted on slides (FSCA). Theytend to be a little larger on average than the typesreported by Hoff from New Mexico, but in generalshape and proportions they are much the same.A single conspecific female was found "attachedto the right rear leg of a cranefly" at College Station,Brazos County, Texas, II March 1985 (S.W.Taber). This sort of phoretic association may assist,at least in part, in the spread of the species fromplace to place.Hesperochernes unicolor (Banks)Chelanops unicolor Banks, 1908:39.Hesperochernes unicolor (Banks): Hoff, 1947:5II;Hoff and Clawson, 1952:23.A single female referable to this species wasfound in Ezell's Cave, Hays County, Texas, 15August 1978 (J.e. Davis); mounted on slide(FSCA). It is a little larger than 3 femalespreviously reported by Hoff (1947) and Hoff andClawson (1952), but otherwise fits the description ofthe species very well. It was obviously accidental inthe cave, not at all modified for cave life.Genus Dinocheirus ChamberlinDinocheirus Chamberlin, 1929b: 171; Muchmore,1974:31.Dinocheirus, as presently (imperfectly) understood,is widely distributed in Europe and North andCentral America. Five species have been reportedpreviously from New Mexico and 2 from Texas(Hoff, 1958). Representatives are usually found inrich forest litter or in the nests of small mammals;one species has been reported from bat guano inCarlsbad Caverns.Dinocheirus astutus HoffDinocheirus astutus Hoff, 1956c:44, figs. 18-20;Hoff, 1958:28; Hoff, 1959:27, 33; Barr andReddell, 1967:259.Dinocheirus astutus was described by Hoff(1956c) from specimens taken mainly from nests ofNeotoma sp. in west-central New Mexico. It wassubsequently reported (del. Hoff) from the BatCave, Carlsbad Caverns, in southeastern New Mexico(Barr and Reddell, 1967). More recent collectionshave provided many more specimens fromCarlsbad Caverns and a single female from EllisCave in west-central New Mexico.Nine males and 10 females from the recent collectionshave been mounted and studied. In mostparticulars they agree closely with Hoff's descriptions.However, both males and females tend to be alittle larger and have slightly more slender appendagesthan Hoff's. These differences may be an artifactof different techniques of measurement or mayindicate some small adaptations of the cavernicolousforms.New records.-NEW MEXICO: Eddy County:Carlsbad Caverns National Park, Bat Cave, berlesesample of guano of Tadarida brasiliensis, 25-27November 1967 (T. Rossen), about 150 specimens,all stages; berlese samples, 14 February 1974, 27June 1974, 28 November 1974, 26-27 April 1975(W.C. WeIbourn), about 150 specimens, all stages.147

Smuioval County: Ellis Cave, under rock nearrodent feces, 3 August 1976 (W.C. Welbourn), 1female.Dinocheirus te.xanus Hoff and ClawsonDinocheirus te.xanus Hoff and Clawson, 1952:27,figs. 16-19; Hoff, 1958:28.This species was described from 2 small collectionsfrom nests of Neotoma micropus Baird atLaguna Madre, Cameron County, Texas.Present material consists of a single male fromBFS Cave, Uvalde County, Texas, 8 June 1985 (A.Grubbs). This specimen is a little larger than thosedescribed by Hoff and Clawson but otherwise isvery similar.Dinocheirus venustus Hoff and ClawsonDinocheirus venustus Hoff and Clawson, 1952:31,figs. 20-23; Hoff and Bolsterli, 1956: 174; Hoff,1958:28. .?Acuminochernes sp.: Reddell, 1970:403.?Dinocheirus sp.: Reddell, 1970:403.Dinocheirus venustus was originally found in apack-rat (Neotoma) nest at Lawrence, Kansas (Hoffand Clawson, 1952). It was subsequently reportedfrom Iron County, Missouri, with no ecological data(Hoff and Bolsterli, 1956). The present record extendits range far to the west.The original description by Hoff and Clawson,supplemented by data in Hoff and Bolsterli, characterizethe species very well. However, a few pertinentadditions and comments may be made here,based on 23 mounted specimens (11 males, 12 females).The new specimens, all from caves inTexas, are on average a bit larger and have somewhatmore slender appendages than those reportedfrom Kansas and Missouri. There are 2 distincttransverse furrows on the carapace; no eyes evident.Male anterior genital operculum with 20-32 setae,including 4 long ones; posterior operculum with19-27 setae on face and 2-3 very small setae on eachside of middle of anterior margin; internal genitaliafairly large and heavily sclerotized. Female anteriorgenital operculum with 15-24 setae, posterior operculumwith 11-15; spermathecae are long, slendertubes without any terminal enlargement. Cheliceralhand with 5 setae, bs always acuminate, sbs usuallyterminally and subterminally denticulate but sometimesacuminate; flagellum of 4 setae. Palp of malegenerally heavier than that of female: male femur2.3-2.75 (female 2.35-3.0), male tibia 2.05-2.35(female 2.1-2.4), and male chela (without pedicel)2.05-2.5 (female 2.45-2.75) times as long as broad;male hand (without pedicel) 1.2-1.35 (female1.4-1.6) times as long as deep; male movable finger0.95-1.01 (female 0.93-1.00) as long as hand. Maleleg IV with entire femur 3.5-3.9 (female 3.65-4.0)and male tibia 4.8-5.5 (female 4.8-5.5) times aslong as deep. Tarsus of leg IV with short, acuminate,tactile seta 0.72-0.79 length of segment fromproximal end.Measurements (mm).-Ranges given first formales, followed in parentheses by those for females.Body length 2.42-3.32 (2.77-3.55). Carapace length0.795-1.00 (0.83-0.96). Palpal femur 0.70-0.89(0.63-0.86) by 0.275-0.35 (0.265-0.32); tibia0.57-0.79 (0.59-0.73) by 0.275-0.36 (0.28-0.32);chela (without pedicel) 1.12-1.36 (1.08-1.30) by0.49-0.615 (0.415-0.495); hand (without pedicel)0.585-0.76 (0.61-0.70) by 0.485-0.62 (0.415-0.49);pedicel length about O. 10; movable finger length0.59-0.72 (0.51-0.695). Leg IV: entire femur0.66-0.80 (0.67-0.79) by 0.175-0.23 (0.175-0.215);tibia 0.55-0.695 (0.565-0.65) by 0.11-0.13(0.105-0.13).New records.-TEXAS: Comal County: BrackenBat Cave, 19 January 1963 (J. Reddell and D.McKenzie), 1 female (JCC). Edwards County: DeepCave, 1,4 mile E Punkin Cave, 4 September 1965 (1.Reddell and D. McKenzie), on guano in upper levels,1 female; Devil's Sinkhole, 8 mi. NE Rocksprings,14 February 1965 (J. Reddell), 1 male, 1female; 14 March 1966 (1. Reddell), "under rockson guano and silt below the cave entrance," 1 male,4 females; 8 March 1968 (J. Reddell), 1 male.Medina County: Weynand Cave, 2 August 1965 (1.Reddell and J. Fish), 1 male. Val Verde County:Fern Cave, 14 April 1973 (T.R. Mollhagen), 16males, 12 females, 3 tritonymphs.Remarks.-Because of the acuminate form ofseta sbs on the cheliceral hand of some, these specimenswere at first thought to belong to the genusAcuminochemes (hence the tentative identification ofthe Deep Cave specimen in Reddell, 1970). Furtherstudy, however, has shown their similarity to Dinocheirusvenustus. The variability in the denticulationof sbs here raises a question about the distinctnessof Acuminochemes and Dinocheirus. Additionally,though, the absence of terminal enlargementsof the spermathecae sets this taxon apart fromboth Dinocheirus and Acuminochemes, where terminalenlargements of spermathecae were consideredthe rule. These problems further point up the needfor a thorough revision of the genera Dinocheirus148

and Acuminochernes, along with the closely relatedHesperochernes and Chernes.Dinocheirus cavicolus, new speciesFigs. 25-27Type-data.-Holotype male (WMI744.01002),allotype female (WM1744.01005) and 11 paratypes(2 males, 8 females, 1 nymph) in guano of Myotisvelifer (J.A. Allen), Fawcett's Cave, 36 miles N DelRio, Val Verde County, Texas, 10 April 1968 (J.Reddell) (FSCA); 2 males, 2 females, same locality,29 March 1975 (R.W. Mitchell) (FSCA); 1 female,Sumac Pit, Kinney County, Texas, October 1987(A.G. Grubbs) (FSCA); 2 males, 3 females, 2nymphs, Rattlesnake Cave, Kinney County, Texas,3 May 1964 (J. Reddell and D. McKenzie) (JCC).Most adults mounted on slides.Diagnosis.-Similar to D. athleticus Hoff inbody size but with more slender appendages, especiallythe legs (femur of leg IV LIB = 4.5-5.4,rather than 3.3-3.5).Description.-Male and female generally similar,both sexes quite varied. Carapace light brown, palpsbrown, other parts much lighter. Carapace longerthan broad; surface with low granules, 2 distincttransverse furrows; no eyes; about 75 narrow c1avodentatesetae, 4 at anterior and 7-9 at posteriormargin. Tergites 1-10 and sternites 4-10 divided;cuticle thin and surfaces slightly granulate tosmooth; dorsal setae narrow c1avodentate, ventralsetae acuminate to multidenticulate. Tergal chaetotaxyof holotype male 10: 10:9: 10: 10: 12: 11: 12: 12:12: 10:2, others generally similar; tergite 11 withoutacuminate tactile setae. Sternal chaetotaxy of holotypemale 20:[2-2]:(3)15(3):(1)8(1): 13: 12: 13:11: 12:9:2T2T2:2, other males similar; the 2 acuminatetactile setae of sternite 11 are near the middle.Anterior sternal chaetotaxy of allotype female17:(3)11(3):(1)6(1):11:12:-, others similar. Internalgenitalia of male typical, that of holotype 0.45as long as carapace. Spermathecae of female arelong, thin tubules with expanded, saclike ends.Chelicera with 5 setae on hand, sbs denticulate,others acuminate; flagellum of 4 setae; galea ratherslender, with 4-6 small rami.26Figs. 25-27.-Dinocheirus cavicolus, new species, holotype male: 25, right palp, dorsal view; 26, left chela, lateral view (darkenedareoles are underneath); 27, leg IV, lateral view.149

Palp (Fig. 25) rather slender, a little more so infemales than in males; trochanter 1.75-2.1, femur2.6-3.25, tibia 2.3-2.7, and chela (without pedicel)2.8-3.4 times as long as broad; hand (withoutpedicel) 1.45-1.75 times as long as deep; movablefinger 0.92-1.05 as long as hand. Surfaces finelygranulate except chelal fingers smooth; setae mostlyshort, denticulate. Trichobothria as shown in Fig.26. Fixed chelal finger with 38-46 cusped marginalteeth; movable fmger with 41-50 teeth, similar exceptproximal 8-10 rounded; each finger with 3-6accessory teeth, both internal and external. Venomapparatus well developed in movable finger; nodusramosus usually just distad of trichobothrium sl, butjust proximad of sl in holotype.Legs quite slender; leg IV (Fig. 27) with entirefemur 4.5-5.4 and tibia 5.7-7.0 times as long asdeep. Tarsus of leg IV with a short, acuminate tactileseta about 3/4 length of segment from proximalend.Measurements (mm).-Figures given first forholotype followed in parentheses by ranges of allotypeand mounted paratypes. Body length 2.25(2.04-2.81). Carapace length 0.805 (0.74-0.88).Chelicera length 0.265 (0.23-0.29). Palpal trochanter0.39 (0.34-0.445) by 0.22 (0.185-0.245);femur 0.74 (0.67-0.82) by 0.27 (0.27-0.29); tibia0.665 (0.60-0.725) by 0.285 (0.24-0.29); chela(without pedicel) 1.07 (0.98-1.30) by 0.38 (0.330.41); hand (without pedicel) 0.57 (0.50-0.64) by0.385 (0.33-0.41); pedicel about 0.10 long; movablefinger length 0.56 (0.50-0.62). Leg IV: entirefemur 0.65 (0.60-0.75) by 0.14 (0.12-0.15); tibia0.57 (0.51-0.64) by 0.09 (0.08-0.095); tarsus 0.465(0.40-0.48) by 0.065 (0.06-0.075).Etymology.-Tbe new species is named cavicolusfor its habitat in a cave.Remarks.-Dinocheirus cavicolus appears to bemodified for cave existence in the same way thatHesperochernes occidenralis and H. mirabilis are,i.e. by having a thinner cuticle than epigean fonus,by having more attenuated appendages, and by reductionin the number of setae on the body.Dinocheirus, sp. indet.In the collections from Texas caves are twospecimens which cannot be identified with certaintyat this time: one tritonymph from Wheat Cave, NWcomer of Edwards County, Texas, 21 September1963 (J. Reddell and D. McKenzie) (lCC); one femalefrom Arrowhead Cave, 3 mi. N San Marcos,Hays County, Texas, 1983 (A.G. Grubbs) (FSCA).Chernetidae, gen. et sp. indet.Several unidentifiable nymphs are at hand: onefrom 400 Foot Cave, Brewster County, Texas, 30June 1985 (A.G. Grubbs); one from Harrell's Cave,on Pete Sloan Ranch about 2 mi. W Chappell, SanSaba County, Texas; 2 from Springdale RanchCave, San Saba County, Texas, 22 October 1989(R.C. Matthews, Jr.); 3 from Secret Valley Cave,Uvalde County, Texas, February 1984 (R.M.Waters).ACKNOWLEDGMENTSMany sincere thanks are due to the collectors ofthe specimens reported herein, and especially toJames Reddell, who not only collected most of thematerial but also strongly encouraged and supportedthis publication. I am also indebted to C.R. Hignuttfor drawing some of the figures. An anonymous reviewermade many comments which have helped toimprove the manuscript.LITERATURE CITEDBanks, N. 1895. Notes on the Pseudoscorpionida. J. New YorkEntomo!' Soc., 3:1-13.Banks, N. 1909. New tropical pseudoscorpions. J. New YorkEntomol. Soc., 17:145-148.Banks, N. 1908. The pseudoscorpions of Texas. Bull. WisconsinNat. His!. Soc., 6:39-42.Barr, T.C., Jr., and J.R. Reddell. 1967. The arthropod cavefauna of the Carlsbad Caverns region, New Mexico. SouthwesternNat., 12:253-274.Beier, M. 1932a. Pseudoscorpionidea I. Subord. Chthoniinea etNeobisiinea. Tierreich, 57:1-258.Beier, M. 1932b. Pseudoscorpionidea ll. Subord. C. Cheliferinea.Tierreich, 58:1-294.Benedict, E.M. 1978. False scorpions of the genusApocheiridium Chamberlin from western North America(pseudoscorpionida, Cheiridiidae). J. Arachno!., 5 :231-241.Chamberlin, J.C. 1924a. The Cheiridiinae of North America.Pan-Pacific Entomo!., 1:32-40.Chamberlin, LC. 1924b. Hesperochemes [aurae, a new speciesof false scorpion from California inhabiting the nest ofVespa. Pan-Pacific Entomol., 1:89-92.Chamberlin, LC. 1929a. A synoptic classification of the falsescorpions or chela-spinners, with a report on a cosmopolitancollection of the same.- Part I. The Heterosphyronida(Chthoniidae) (Arachnida-Chelonethida). Ann. Mag. Nat.Hist., (\0),4:50-80.Chamberlin, LC. 1929b. Dinocheirns lenoch, an hitherto undescribedgenus and species of false scorpion from Mexico(Arachnida-Chelonethida). Pan-Pacific Entomol., 5: 171-173.Chamberlin, J.C. 1930. A synoptic classification of the falsescorpions or chela-spinners, with a report on a cosmopolitancollection of the same.- Part ll. The Diplosphyronida(Arachnida-Chelonethida). Ann. Mag. Nat. His!., (10),5:1-48,585-620.Chamberlin, J.C. 1931. The arachnid order Chelonethida.Stanford Univ. Pub!. Bio!' Sci., 7:1-284.150

Chamberlin, J.C. 1946. The genera and species of the Hyidae, afamily of the arachnid order Chelonethida. Bull. Univ. Utah,37(6):1-16.Chamberlin, J.C. 1962. New and little-known false scorpions,principally from caves, belonging to the families Chthoniidaeand Neobisiidae (Arachnida, Chelonethida). Bull. AmericanMus. Nat. Hist., 123:303-352.Chamberlin, 1.C., and D.R. Malcolm. 1960. The occurrence offalse scorpions in caves with special reference to cavernicolousadaptation and to cave species in the North Americanfauna. American Mid\. Nat., 64:105-115.Chambers, S.M., and S. Jahrsdoerfer. 1988. Endangered andthreatened wildlife and plants; final rule to determine fiveTexas cave invertebrates to be endangered species. FederalRegister, 53(180):36029-36033.Curcic, B.P.M. 1984. A revision of some North Americanspecies of Microcreagris Balzan, 1892 (Arachnida: Pseudoscorpiones:Neobisiidae). Bul!. British Arachno!. Soc.,6:149-166.Curcic, B.P.M. 1989. Further revision of some North Americanfalse scorpions originally assigned to Microereagris Balzan(pseudoscorpiones, Neobisiidae). 1. Arachno\., 17:351-362.Hoff, C.C. 1946. American species of the pseudoscorpion genusMicrobisium Chamberlin, 1930. Bull. Chicago Acad. Sci.,7:493-497.Hoff, C.C. 1947. The species of the pseudoscorpion genus Che/anops described by Banks. Bull. Mus. Compo Zoo!., 98:473-550.Hoff, C.C. 1952. Two new species of pseudoscorpions fromIllinois. Trans. minois Acad. Sci., 45: 188-195.Hoff, C.C. 1956a. The heterosphyronid pseudoscorpions of NewMexico. American Mus. Novitates, 1772:1-13.Hoff, C.C. 1956b. Diplosphyronid pseudoscorpions from NewMexico. American Mus. Novitates, 1780:1-49.Hoff, C.C. 1956c. Pseudoscorpions of the family Chernetidaefrom New Mexico. American Mus. Novitates, 18001-66.Hoff, C.C. 1957. Tejachemes (Arachnida-Chelonethida,Chernetidae-Chernetinae), a new genus of pseudoscorpionbased on Dinoeheirus stereoreus. Southwestern Nat.,2:83-88.Hoff, C.C. 1958. List of the pseudoscorpions of North Americanorth of Mexico. American Mus. Novitates, 1875:1-50.Hoff, C.C. 1959. The ecology and distribution of the pseudoscorpionsof north-central New Mexico. Univ. New MexicoPub\. Bio\., 8:1-68.Hoff, C.C. 1961. Pseudoscorpions from Colorado. Bull. AmericanMus. Nat. Hist., 122:409-464.Hoff, C.C., and J.E. Bolsterli. 1956. Pseudoscorpions of theMississippi River drainage basin area. Trans. American Microse.Soc., 75:155-179.Hoff, C.C., and D.L. Clawson. 1952. Pseudoscorpions fromrodent nests. American Mus. Novitates, 1585:1-38.Malcolm, D.R., and J.C. Chamberlin. 1960. The pseudoscorpiongenus Chitrella (Chelonethida, Syarinidae). AmericanMus. Novitates, 1989: 1-19Malcolm, D.R., and W.B. Muchmore. 1985. An unusual speciesof Tyrannochthonius from Florida (pseudoscorpionida,Chthoniidae).1. Arachno!., 13:403-405.Menge, A. 1855. Ueber die Scheerenspinnen. N. Schr. Naturforsch.Ges., Danzig, 5(2):1-43.Muchmore, W.B. 1969. New species and records of cavernicolouspseudoscorpions of the genus Microcreagris(Arachnida, Chelonethida, Neobisiidae, Ideobisiinae). AmericanMus. Novitates, 2392:1-21.Muchmore, W.B. 1972. European pseudoscorpions from NewEngland. 1. New York Entomo\. Soc., 80:109-110.Muchmore, W.B. 1973. The genus Chitrella in America(pseudoscorpionida, Syarinidae). J. New York Entomo\.Soc., 81:183-192.Muchmore, W.B. 1974. Clarification of the genera Hesperoehernesand Dinoeheirus (Pseudoscorpionida, Chernetidae). 1.Arachno\., 2:25-36.Muchmore, W.B. 1976. Aphrastochthonius paehysetus, a newcavernicolous species from New Mexico (pseudoscorpionida,Chthoniidae). Proc. Bio\. Soc. Washington, 89:361-364.Muchmore, W.B. 1981. Cavernicolous species of Larea,Archeo/arca, and Pseudogarypus with notes on the genera(pseudoscorpionida, Garypidae and Pseudogarypidae). J.Arachno!.,9:47-60.Muchmore, W.B. 1982. The genera Ideobisium and Ideob/othrus,with remarks on the family Syarinidae(pseudoscorpionida). 1. Arachno\., 10: 193-221.Muchmore, W.B. 1986. Additional pseudoscorpions, mostlyfrom caves, in Mexico and Texas (Arachnida: Pseudoscorpionida).Texas Mem. Mus., Speleo!. Monogr., 1:17-30.Muchmore, W.B., and E.M. Benedict. 1976. Redescription ofApochthonius moestus (Banks), type of the genus ApoehthoniusChamberlin (pseudoscorpionida, Chthoniidae). J.New York Entomo\. Soc., 84:67-74.Nelson, S., Jr. 1975. A systematic study of Michigan Pseudoscorpionida(Arachnida). American MidI. Nat., 93:257-301.Nelson, S., Jr. 1984. The genus Microbisium in North andCentral America (pseudoscorpionida, Neobisiidae). J. Arachnol.,12:341-350.Reddell, 1.R. 1965. A checklist of the cave fauna of Texas. I.The Invertebrata (exclusive of Insecta). Texas 1. Sci.,17:143-187.Reddell, 1.R. 1970. A checklist of the cave fauna of Texas. IV.Additional records of Invertebrata (exclusive of Insecta).Texas 1. Sci., 21:389-415.Silva-Taboada, G., and K.F. Koopman. 1964. Notes on theoccurrence and ecology of Tadarida laticaudata yucatanicain eastern Cuba. American Mus. Novitates, 2174:1-6.Turk, F.A. 1949. Dinocheirus stercoreus, a new pseudoscorpionfrom the Bracken Cave, Texas, U.S.A. Ann. Mag. Nat.Hist., (12), 2:120-126.Welbourn, W.C. 1978. Biology of Ogle Cave, with a list of thecave fauna of Slaughter Canyon. NSS Bull., 40:27-34.ADDENDUMAfter the manuscript for this paper was completed,some additional cavernicolous pseudoscorpionsbecame available. In order to bring the accountup-to-date, this material is treated below.Tanarocreagris infernalis (Muchmore)One female, probably referable to this species,was collected in Off Campus Cave, WilliamsonCounty, Texas, 8 April 1989, by W. Elliott, J. Reddelland M. Reyes (mounted on slide-FSCA). Thisspecimen differs in several minor respects from, butis generally similar to, the holotype female of thespecies, which was found in Inner Space Cavern,only about a mile south of Off Campus Cave. In151

spite of the differences, it seems best to consider it arepresentative of T. infernalis.Tartarocreagris intennedia, new speciesFigs. 28-29Type-data.-Holotype male (WM7453.01002)and paratype male from Airman's Cave, TravisCounty, Texas, 3 September 1989 (1. Reddell andM. Reyes); mounted on slides (FSCA).Diagnosis.-A moderately large species of thegenus (palpal chela 2.1- 2.2 mm long), with ratherslender appendages (chela LIB = 3.7-3.8), and withvery poorly developed eyes.Description of male (female unknown).-Palpslight brown, carapace tan, other parts lighter. Carapacelonger than broad; with a small epistome, triangularor rounded; 4 barely discernible eye spots;24-26 setae, 4 at anterior and 6-7 at posterior margin.Coxal area typical; palpal coxa with 3 or 4 apicalsetae.Abdomen long ovate. Tergal chaetotaxy of holotype8: 12: 12: 13: 13: 13: 14: 12: 12: 11 :T2T:3; paratypesimilar but with 10 setae on 1st tergite and 2 onanal operculum. Sternal chaetotaxy of holotype25: [4-5]:(6)8/12(6):(6)11(5): 14:2/14:2/15:2/15: 14:14:TlT2TIT:2; paratype much the same. Internalgenitalia much like those of T. texana (cf. Fig. 4).Chelicera 0.6 as long as carapace; hand with 7setae; galea bifurcated near tip; flagellum of 8 pinnatesetae.Palp rather slender (Fig. 28); femur about 1.3and chela (including pedicel) 2.15-2.2 times as longas carapace. Trochanter 2.15-2.3, femur 4.25-4.35,tibia 3.15-3.2, and chela (including pedicel) 3.7-3.8times as long as broad; hand (including pedicel)1.6-1.65 times as long as deep; movable finger1.08-1.13 times as long as hand with pedicel. Trochanter,femur and chelal hand lightly granulate.Trichobothria as shown in Fig. 29. Fixed fingerwith 59-60 and movable finger with 68-70 marginalteeth, only the distal 12-14 with cusps.Legs rather slender; leg IV with entire femur4.25-4.35 and tibia 6.8- 7.2 times as long as deep.Tibia and each tarsus with a tactile seta. Subterminaltarsal setae spinous in distal third.Measurements (mm).-Figures given first forholotype, followed in parentheses by those forparatype. Body length 3.23 (3.51). Carapace length1.01 (0.96). Chelicera length 0.615 (0.615). Palpaltrochanter 0.64 (0.61) by 0.295 (0.265); femur 1.29(1.26) by 0.295 (0.295); tibia 1.18 (1.13) by 0.37oo29oFigs. 28-29.-Tarrarocreagris intennedia, new species, holotype male: 28, right palp, dorsal view; 29, left chela, lateral view(darkened areoles are underneath).152

(0.36); chela (without pedicel) 2.00 (1.91) by 0.59(0.555); hand (without pedicel) 0.895 (0.84) by0.665 (0.635); pedicel 0.19 (0.19) long; movablefinger length 1.17 (1.16). Leg IV: entire femur 1.04(1.02) by 0.245 (0.235); tibia 0.955 (0.96) by 0.14(0.13).Etymology.-The new species is named intermediafor its morphological characters intermediate betweenthose of T. comanche and the other morehigWy cave-adapted forms.Remarks.-It is interesting to note that, as in T.comanche (above), the depth of the chelal hand ofthis species is greater than the breadth. As no noncavespecies of the genus is yet known, it is impossibleto judge whether or not this is a primitivecharacter.The discovery of yet another species of Tartarocreagrisin the area of Travis and Williamson Countiesemphasizes the fact that much speciation hasgone on in the caves here. It is to be expected thatfurther collecting will yield more new species and itis to be hoped that collections in epigean habitatswill turn up some "ancestral" surface forms.Tartarocreagris, sp. indet.A lone tritonymph was collected in LakeLineCave, Williamson County, Texas, 7 February 1990,by M. Reyes and J. Reddell mounted on slide(FSCA). As LakeLine Cave lies about midway betweenMcDonald Cave and Beck Ranch Cave whereadults of Tartarocreagris reddelli have been found,it is reasonable to believe that this nymph belongs tothat species; however, a definite identification is impossibleat this time.Hesperochernes unicolor (Banks)A male of tills species was taken by Berlese separationof leaf litter from the entrance area of LakeLine Cave, Williamson County, Texas, 13 February1990, by J. Reddell and M. Reyes; mounted onslide (FSCA). This specimen is smaller than, butgenerally similar to, the female from Ezell's cavementioned above. Also, it is quite similar to themale from a pack rat nest in Cameron County,Texas, which was tentatively assigned to this speciesby Hoff and Clawson 1952).This is Publication No. N.S.-59 of the Texas Memorial Museum.153

cavemicolous, it lacked the troglomorphic modificationsof the latter, being distinctly pigmented andhaving well developed eyes, a robust eye mound,and relatively short appendages. Of particular interestwere the sexually dimorphic structures; maleshaving both a trochanteral spur, an elongate processon the ventral surface of the fourth trochanter, and apostopercular process, a conical outgrowth just posteriorof the genital operculum (Fig. 2). Althoughsimilar, but smaller, spurs are known from both aNearctic (Sitalcina Banks) and Palearctic genus(Scotolemon Lucas), the postopercular process isunique, as is the combined presence of two dimorphicstructures, and was the first clear indication ofa derived character defining the genus.In addition to the topotypes of T. reddelli, thesecollections also included specimens from severalother caves along the Balcones Escarpment. All ofthese specimens were troglomorphically intermediatebetween the two species. For example, they lackedthe retina (unlike T. reddelli) but retained at least avestigial cornea (unlike T. mulaiki). On the basis ofsomatic characters these additional specimens wereassigned to the two species: the most troglomorphicto T. mulaiki, the others to T. reddelli.Continued sampling of the Texas cave fauna producedadditional specimens of Texella, spreading theknown distribution of the genus over much of theBalcones Escarpment (Mitchell and Reddell, 1971),but turned up no further species. The presence ofonly two species of Texella now seems incongruous.On the one hand, this region is extremely rich inother cavernicolous arachnids; for example, themany species of the spider genera NeoleptonetaBrignoli (Gertsch, 1974), Eidmanella Roewer(Gertsch, 1984), and Cicurina (Cicurella) Menge(Gertsch, 1992). Furthermore, the other knownphalangodid faunas of the Nearctic region are quitespeciose. An unusually rich fauna is known fromthe Californian region (Briggs, 1968; Briggs andHom, 1966, 1967; Briggs and Ubick, 1981, 1989;Ubick and Briggs, 1989) and the Appalachian fauna,Figs. 1-4.-Texel/a spp., lateral views: I, T. bifurcata (Briggs), male topotype; 2, T. spinoperca, new species, maleparatopotype; 3, T. mulaild Goodnight and Goodnight, male, Fern Cave; 4, T. mulaild Goodnight and Goodnight, female, FernCave.156

although not recently studied is nonetheless representedby a dozen nominal species (Goodnight andGoodnight, 1942).The present study, based on the examination ofan extensive collection of these relatively rare harvestmen,made available largely through the effortsof James Reddell and field associates, shows thatTexella is far richer and more complex than previouslyimagined. In terms of species richness, the 21species now recognized make Texella one of thelargest Nearctic phalangodid genera (second in sizeonly to Calicina Ubick and Briggs (1989) with 25species). The high degree of complexity is evidentin the pronounced interspecific variation of bothsomatic and genitalic characters as well as in the patternsof relationship which emerge upon theiranalysis.Although this study greatly increases our knowledgeof Texella, there are probably few to manyadditional species to be discovered. This is suggestedby the high isolation of the species (abouttwo thirds are known from single localities) and thehighly disjunct distribution of the species groups.Even along the Balcones Escarpment, by far themost heavily sampled region, many caves andpatches of epigean habitat have not been adequatelysearched. Also, several of the species descriptionsare not adequate, being based on only a few specimens(three species are represented by single specimens),and give little or no indication of intraspecificvariation. Additionally, several populations ofTexella are currently known from only female andjuvenile specimens and, thus, can not be identifiedwith certainty. Clearly, much basic work remains tobe done on this genus and it is hoped that this studyencourages biologists, both cavernicole and epigean,to gather the additional material necessary to fillexisting gaps and test the various hypotheses presentedhere.Another aim of this study is to urge the conservationof Texella. There seems little doubt that thebiological complexity of Texella is a consequence ofa high degree of species isolation which is now evidentin the widespread pattern of localized endemism.Unfortunately, localized endemicity alsomay have negative ramifications, namely the increasedpotential for extinction. This is especiallyof concern in regions of rapid urbanization, as alongthe Balcones Escarpment, where development is apotential threat to the many biologically rich caves.At present, only two species of Texella (reddelli andreyesi) are afforded federal protection. However, ifrestricted distribution is an important prerequisitefor listing species, then comparable protectionshould be afforded all of the remalmng specieswhich, with the exception of T. bijurcata, have evenmore restricted ranges.MATERIALS AND METHODSThis study is based on the examination of about300 specimens of Texella, which (along with relatedPhalangodidae) were borrowed from the followingpersons and institutions:AMNH: American Museum of Natural History,Dr. Norman I. PlatnickCAS (including TSB collection): CaliforniaAcademy of Sciences, Dr. Wojciech J. PulawskiCDU: Mr. Darrell Ubick collectionMLG: Dr. Marie L. Goodnight collectionNMNH: National Museum of Natural History,Dr. Jonathan CoddingtonTMM: Texas Memorial Museum, Mr. James R.Reddell (including the collection formerly at TexasTech University and made available by Mr. JamesC. Cokendolpher)WAS: Dr. William A. Shear collectionSpecimens were prepared for study as were thoseof Calicina (Ubick and Briggs, 1989). Measurementswere made with ocular reticles in Olympusand Leitz dissecting microscopes, at 80 X, and arebelieved accurate to about ±0.03 (and from ±0.08to ±0.12 for leg II, depending on length). Detailedobservations and drawings were made with the helpof a compound, phase-contrast (Zeiss) microscopeequipped with a grid.Scanning electron micrographs were taken with aHitachi S-520 Scanning Electron Microscope. Specimenpreparation consisted of dehydration (soakingspecimens in serial baths from 75% to 100% ethanol),critical point drying (using liquid carbon dioxide),and mounting onto a stub. Specimens werecleaned manually, using eye lash brushes, both beforedehydration and upon being mounted. It waseventually discovered that the most effective methodof mounting specimens was to glue the fourth legs toa fine wire attached to the stub. This leaves thebody unspoiled by the adhesive and permitsmovement of the specimen into virtually anyposition. It should be noted that due to photoreductionof the micrographs, the original magnificationsgiven are useful only for facilitating comparison ofthe images; the actual dimensions can be determinedusing the accompanying scale bars. All measurementsare in millimeters, unless indicated otherwise.Tenninology.-Tubercles. This commonly usedterm refers to a wide variety of cuticular outgrowthsin harvestmen. To avoid unnecessary redundancy, it157

is here used in a more restricted sense to includeonly medium-sized structures (of about 20 /-Lm inbasal diameter) having rounded crowns. Included inthis definition are the paraocular anterior tubercles(AT), tubercles located on the posterior tergitemargins, ventral trochanteral tubercles, and bodytubercles scattered over dorsal and ventral surfaces.The smaller tuberculate structures, which constitutethe background texture of the phalangodid bodycuticle Cmicrotubercles' of Murphree, 1988), arereferred to as 'fine rugosity' (for microtubercles ofabout 5 /-Lm in diameter) or 'coarse rugosity' (formicrotubercles of about 10 /-Lm in diameter). Forlarger structures, the terms 'eye mound' and'megaspine' (here used somewhat loosely to refer toboth the macroseta and the basal, cuticular process)are used to substitute 'eye tubercle' and 'palpalspine-bearing tubercle', respectively.Glans Nomenclature. The terminology used forthe male genitalia is largely descriptive and novel.Most of the terms used previously for Calicina(Ubick and Briggs, 1989) are not repeated here inorder to avoid making premature assumptions abouthomology. Positional adjectives of glans structures,such as "the ventral surface of the stylus," refer to afully expanded glans where the stylus is apically directed.Tarsal Count. The tarsal count notation is abbreviated,as in our previous studies, with the numberof tarsomeres of legs I through IV listed seriallyand separated by a "-" (e.g., 3-5-5-5). However, inmany individuals of Texella the tarsal count is notbilaterally symmetrical and this variation is shownby a "I", with the value for the left leg given first.For example, "3-5-615-5" indicates asymmetry inleg III, the left leg having six tarsomeres and theright having five.Abbreviations:AS = apical spineAT = anterior tuberclesBF = basal fold of stylusBK = basal knob of glansBS = basal segment of glansDS = dorsal setae of VPPGO = genital operculumLIIISL = leg III scute lengthLS = lateral setae of VPPML = middle lobe of glansPOP = postopercular processPSL = parastylar lobe(s)PSL2 = secondary parastylar lobe(s)PTb = palpal tibiaS = stylusSA = stylar apophysisTC = tarsal countTrIV = trochanter of leg IVVC = ventral carina of stylusVP = ventral plateVPP = ventral plate prongVS = ventral setae of VPPMORPHOLOGYOf the 29 characters used in this analysis the vastmajority (20) are from the male genitalia. Themorphologically simpler female genitalia werecomparably poorer in apparent characters (2). Inaddition, secondary sexual characters (3) and somesomatic characters (4) were found useful.Male genitalia.-(Figs. 6-7). The penis ofTexella consists of an inflatable basal sac, a sclerotizedtruncus, and a distal glans. The ventroapicalextension of the truncus, the ventral plate (VP), isdeeply bifurcate. The resulting prongs (VPP) arelateral in position and widely separated, permittingthe ventral protrusion of the glans. The prongs bearsetae arranged on the dorsal, lateral, and ventralsurfaces (OS, LS, and VS, respectively). Also onthe lateral surface, and somewhat distad of thelateral setae, is a stout cuticular outgrowth, theapical spine (AS).The glans of Texella is robust and appearssigmoid in lateral view. The basal segment (BS) islarge, often approaching the apex of the VPP, and,in roughly half of the species, bears on itsventroapical margin a pointed extension, the basalknob (BK). The middle segment is split apically toform two lobes, the parastylar lobes (PSL), and, insome species, an additional pair of lobes, thesecondary parastylar lobes (PSL2). The middle lobe(ML) is a thin flap which separates the middlesegment of the glans from the apical segment, thestylus (S). The stylus is either simple or has a basalenlargement (the basal fold, BF), a lateral process(the stylar apophysis, SA), andlor a membranousventral carina (VC).Of all the Nearctic phalangodid genera, the penisin Banksula (Fig. 5) is most similar to that of TexellaoOne obvious difference is the placement of theVPP, which are relatively contiguous and positionedon the ventral margin of the truncus. Also, theglans is considerably simpler than in Texella: theglans has only one axis of rotation, the basalsegment (BS) is reduced, and the apical elementslack accessory structures (except for a pair of PSL158

and stylar outgrowths, present in some species,which mayor may not be homologs to the SA).Function: The glans of Texella is of the foldedtype. In its unexpanded form, the stylus and the(deflated) middle segment are pressed against the(deflated) basal segment. During expansion, assimulated by treatment in KOH, the membranousportions of the glans inflate and the distal elementsrotate along a dorsoapical arc. There are two axesof rotation: the primary, at the junction of the basaland middle segments, and the secondary at the baseof the stylus. With complete expansion the styluspoints apically (roughly 180 0 from the unfoldedcondition) and the entire glans protrudes ventrallybeyond the VPP. At this stage, the stylus is typicallyseparated from the parastylar lobes, which abductand (in some species) inflate and contort topermit passage of the stylus.It is worth noting that, in virtually all instancesof unexpanded and partially expanded glandes observed,the parastylar lobes are in intimate contactwith the stylus. It appears that the PSL lock the S inposition. Three types of these apparent lock mechanismswere found: 1) PSL-S. In this, the mostcommon type of lock, the PSL adduct, pinching theS (Figs. 68, 87, 96). 2) PSL-SA. In this type,found only in species of the spinoperca infragroup,the PSL grip the attenuated prongs of the SA (Fig.180). 3) SA-PSL. In this type, found only in thereddelli infragroup, the clip-like SA grip the basalportions of the PSL (Figs. 120, 145).Female genitalia.-The ovipositor in Texella, asin other Nearctic Phalangodidae, is structurallyrelatively simple (Fig. 18). The cuticle is conspicuouslywrinkled (with longitudinal folds on the ventraland dorsal surfaces and transverse folds alongthe lateral surfaces) and sculptured (with minutetubercles and spines, often arranged in linear serrations).The apical surface bears a pair of cuticularprojections (= apical teeth) and 7 (rarely 6) pairs ofsetae, arranged as in Fig. 20.The ovipositor of BanksuLa resembles that ofTexella but lacks apical teeth (at least in the threespecies closely examined: B. grahami Briggs,meLones Briggs, and rudoLphi Briggs and Ubick).One noteworthy observation is the presence of ahyperexpanded ovipositor (Figs. 152-155) whichdiffers from the typical form in having an additionalapical segment terminating in four lobes. Thishyperexpanded state, most likely being the conditionassumed during egg laying, was also observed by usin a few specimens of Bishopella sp. and"SitaLcina" cockerelli Goodnight and Goodnight butwas not recorded by Martens, Hoheisel, and Gatze(1981) in their study of opilionid ovipositor morphology.However, thanks to the detailed drawingsVSVPP,..ASDSVS, /LS , ,LS, , ,,, .5 6 7Figs. 5-7.-Phalangodid penes, dorsolateral views: 5, Bank.mla galilei Briggs; 6, Texella bifurcara (Briggs); 7. Texella reyesi,new species. Styli shaded. Abbreviations: AS = apical spine, BF = basal fold of stylus, BK = basal knob of glans, BS = basalsegment of glans, DS = dorsal setae, LS = lateral setae, ML = middle lobe of glans, PSL = parastylar lobes, PSL2 = secondaryparastylar lobes, SA = stylar apophysis, VC = ventral carina of stylus, VPP = ventral plate prong, VS = ventral setae.159

of ovipositor sections presented in that study, it ispossible to conclude that the hyperexpanded segmentis actually the everted vagina (ibid, figs. 30-33) andthat the openings to the seminal receptacles are to befound at the bases of the apical lobes.Sexual dimorphisms.-The most conspicuoussexual dimorphism is the male trochanteral spur, alarge, robust process located ventroapically on thefourth trochanter (Figs. 24, 25). In several speciesthis structure is reduced in size (Figs. 91, 192), inonly one (T. mulaiki) it is completely absent (Fig.104), and in another (T. bifurcata) it is representedby one to three small, rounded tubercles (Fig. 11).(The female homolog of the spur is represented byone to two small tubercles, for example in Fig. 17.)Similar, although smaller, trochanteral spurs arepresent in some species of Sitalcina and the EuropeanScotolemon Lucas.The second conspicuous dimorphic structure, thepostopercular process (POP), appears to be uniqueto Texella. The POP is a conical cuticularoutgrowth located immediately behind the genitaloperculum and is found in most members of themulaiki species group (Figs. 132, 133). Thereappears to be no female counterpart to this unusualstructure [although one aberrant female of T. reyesidoes possess both a vestigial POP and spur (Figs.130, 134, 135)).Another interesting dimorphism is found in thefemales of the spinoperca infragroup. In thesespecies the genital operculum is modified apically;bearing 1-2 pairs of short spines (Figs. 171, 173,182, 183, 194). Males lack these spines, except forT. spinoperca where some males have homologsrepresented by short, blunt tubercles (Fig. 181).Similar apical tubercles are found in both sexes of T.jungi and shoshone (Figs. 64, 71).Finally, in one species, T. jungi, the number ofparaocular AT appears to be sexually dimorphic:males have 6-7 pairs, females 2-3 pairs (Figs.60-62).Somatic characters.-The somatic charactersused in the analysis were: the number and arrangementof anterior tubercles, the tarsal count, and thenumber of palpal megaspines. Most of these andother somatic characters are discussed in the sectionon Troglomorphy.PHYLOGENYThe character numbers in the following discussioncorrespond to those listed in Table 1.SISTER GROUPI) Ventral plate (VP) bifurcation. Three cladesof Nearctic harvestmen have deeply bifurcate ventralplates: Phalangodes and the nominal genera of theAppalachian region [henceforth referred to as Phalangodes(et al.); Briggs, 1974: fig. 7], Banksula(Fig. 5), and Texella (Figs. 6, 7). In the remaininggenera, which includes Calicina, the probable plesiomorphicsister of all Nearctic Phalangodidae, theVP is entire, suggesting that a bifurcate VP isderived.We are not sufficiently familiar with the Europeanphalangodids, the probable closest relatives of theNearctic fauna, to fully include them in the discussion.However, we have examined specimensand/or published illustrations of representatives ofall Mediterranean genera, with the exception of LolaKratochvil and Paralola Kratochvil, and none has abifurcate VP. However, an apically notched VPoccurs in some (but not all) species of Scotolemon(doriae, Martens, 1978, and terricola, Brignoli,1968) and Ausobskya (athos, Martens 1972). Thenotch is only about 1,4 the length of the VP and is,therefore, quite distinct from the bifurcate conditionin the Nearctic genera. Even assuming homology,the notch appears to be plesiomorphic relative to thebifurcation and so would not alter the monophyly ofthe Nearctic clade.2) Apical spine (AS). The AS appears to bepresent only in, and is considered a synapomorphyfor, Banksula and Texella (Figs. 5-7).3) Ventral plate prong (VPP) position. InBanksula the VPP are located ventrally, as is the VPof other Californian genera, suggesting that thelateral position of the VPP in Texella and Phalangodes(et al.) is derived.The sister group of Texella cannot be establishedwith much certainty at this time. However, takinginto consideration the relative merit of characters 2and 3, the shared presence of a unique structure(AS) appears to offer the stronger argument in supportof a sister relationship with Banksula. Additionalcharacters will obviously be necessary to resolvethis issue. On the other hand, there seems littledoubt that, of the three clades, Banksula is themost plesiomorphic. In addition to the placement ofthe VPP, several other character states of Banksulaare best interpreted as plesiomorphies: VP bifurcationnarrow; AS short, apically unmodified;VP setae short and slender; glans folding simple;glans with few accessory structures; and glans basalsegment small. By contrast, Phalangodes (et al.) isclearly and strongly derived. For example, the VPP160

in Phalangodes (et al.) is thick and swollen,whereas in Texella, Banksula, and the ventral plateof all remaining genera is thin and platelike. Furthermore,Phalangodes (et al.) seems to be autapomorphicin several additional characters of genericsignificance, such as: width of VPP (wide, virtuallysurrounding glans); size of glans (strongly reduced);folding mechanism of glans (basal segment expandsaccordion-like); form of glans lobes (thin,scale-like, reduced); movement of glans lobes(presumably stationary); placement of ovipositor setae(on protruding lobes); position of eye mound(posteriorly removed from anterior margin).MONOPHYLY4) VP bifurcation width. In Banksula, the VP bifurcation(i.e., distance between VP prongs) is narrow,less than the width of a single prong, and doesnot permit the ventral protrusion of the glans. InTexella, the VP bifurcation is wide, greater (usuallymuch greater) than the width of a single prong, andpermits the protrusion of the glans ventrally beyondthe VP. Given the probably plesiomorphic state ofthe Banksula VPP position, it seems likely that thetransformation series for the ventral plate is: VPentire-VP narrowly bifurcate-VP widely bifurcate.In this case a wide bifurcation would be autapomorphicfor Texella.5) Glans shape. With the exception of Calicina,all Nearctic phalangodids have folded glandes. Inmost genera the glans unfolds along a single axis(appearing V-shaped in lateral view; see Fig. 5),whereas in Texella there are two axes (the glans appearingW-shaped, or sigmoid, in lateral view; as inFigs. 6, 7). On the basis of uniqueness and greatercomplexity, the sigmoid glans is considered derived.6) Ovipositor teeth. The ovipositor in virtuallyall species of Texella has a pair of apical teeth,which appear to be derived. Apical teeth are lackingin Banksula, Sitalcina, Microcina, and the severalspecies of Phalangodes (et al.) examined. The presenceof apical teeth in a few species (5) of Calicinais most parsimoniously interpreted as a parallelism.Apical teeth are absent in T. kokoweef, diplospina,and in all members of the mulaiki infragroup; this isinterpreted as a reversal and an additional synapomorphyfor the mulaiki infragroup.7) Apical spine (AS) length. In Banksula the ASis short, being less than two thirds the width of theVPP; in Texella the AS length/VPP width ratioranges from 1.1 to 2.9. The longer AS length isconsidered a synapomorphy for. Texella. Here, theAS is longest in T. bifurcata (AS/vPP = 2.5-2.9),of moderate length in T. kokoweef, longistyla,reyesi, and the spinoperca infragroup (1.5-2.5), andshortest in the remaining species (1.1-1.5).SUBGROUPSThe bifurcata group8) VPP apex. In T. bifurcata the VPP apex isstrongly bent so that the ventral margin appearsnotched (Fig. 15); in other Texella and all Banksulathe VPP apex is straight or, at best, only moderatelybent. The widespread occurrence of the latter statesuggests plesiomorphy.9) Ovipositor shape. In T. bifurcata the ovipositoris bent basally (Fig. 18); it is straight for theother Texella and Banksula examined (i.e., thosewith fully extruded ovipositors). A more completesurvey will be necessary to resolve this character,although the data so far suggest that a bent ovipositoris another potential autapomorphy for T. bifurcata.10) Male trochanteral spur. Another possiblesynapomorphy for Texella is the male trochanteralspur (provisionally assuming a parallelism for thespur in some species of Sitalcina and Scotolemon,which will be explored in a later study). In mostspecies the spur is moderate to quite large. Spur reductionis most evident in the mulaiki infragroup,being completely absent in T. mulaiki (Fig. 104) andreduced to vestiges in the other species (Fig. 91).Strongly reduced spurs are also found in T. bilobata,fendi, and hom;; all species with relativelyderived genitalia, suggesting that spur reduction issecondary. This leaves a single species, T. bifurcata,where the spur is represented by 1-3 tubercles.These tubercles are smooth and rounded (Fig. 11)and appear quite different from small spurs, whichhave a textured cuticle (as do typical spurs) and aregenerally more elongate (Fig. 192).The phylogenetic implications of this characterdistribution are somewhat ambiguous. One possibilityis that the trochanteral spur is a synapomorphyfor the genus, with the tubercles of T. bifurcata representingextremely reduced spurs. However, giventheir morphological simplicity, it appears morelikely that the tubercles in T. bifurcata areprimitively simple. In this case, a well developedspur would be a synapomorphy for all of the remainingspecies of Texella (i.e., the kokoweef plus mulaikispecies groups).11) Anterior tubercle (AT) number. The greatestnumber of AT, ~ 9 pairs, is found in T.bifurcata (Fig. 9). Moderately high numbers (5-8161

....~...,"'" c .::: s: .::: ~'" '"~ " c .::: .::: 'E ..9- 0().S " l::§ '-' Q Q c.~,!;lt:~ ]~] '" ~ ~ ;g.:::.~"'";;. '" "l:: i;.~ .::; t: .;;;c '5(, '":s! ] ~0()oJ::>~ 1:: ~ .::: oJ::> ;> ...,'-' '- oJ::>...,"'""'"1•2 --CJ • 3 ---.04 -0---0 5 ---0 • • • • • • • • • • 6 --0---0-.--0• • •D-D--D • • • •0• • • 7•8910 -----D-D---O IEi1 I!!I IIilI I!il m •m 0• • • •ii!I I!!I ~ Il1iI !liI 11• • • • • • • D-04iI ~ JlII D IiIIlI IIIilI CJ---IlI--tB m liI12• • D--D D--O 0 0 13~14 0• • • • • • • • • • • • • • • • • • • •1516 --D--D----D-D-- • • • • • • • • • • • • • • • • •17181920 0• • D-D • • • • • • • • • • • • •21 D--D ~ I11III• •2223242526272829m rm l!!!! I11III mJI I!m D-ITnelenath: '36 IC.I.: 0.5023 T;jx;j29 ChdrsTable I.-Character table with cladogram of possible relationships in Texella.Characters and polarities discussed in text.162

pairs) are found in six species: T. kokoweef (Fig.22), brevistyla (Fig. 48), T. jungi (males only, Fig.62), welbourni, reddelli (Figs. 118, 119; althoughsome individuals have only 4 pairs), and diplospina(Fig. 166). The wide distribution, across the entirecladogram, of the multituberculate state suggestsplesiomorphy. This in tum suggests that the highestnumber of AT represents the most plesiomorphiccondition.12) AT rows. A biserial arrangement of AT isfound in species with high numbers of AT (~ 5pairs). Outgroup comparison cannot be used withBanksula or Phalangodes (et al.) where the AT areabsent (or represented by 1-2 tiny tubercles). However,the 6-8 pairs of AT present in Calicina mariposaare arranged in a single row. This suggeststhat the biserial arrangement is a synapomorphy forTexella. Within Texella, the loss of the biserialcondition would be derived.13) Secondary parastylar lobes (PSL2). ThePSL2 are found only in the bifurcata and kokoweefgroups and because of their uniqueness may be consideredto be synapomorpic for them. However,there appears to be some evidence which suggeststhat the PSL2 are plesiomorphic within Texella.First, there is a trend of reduction of the PSL2 in thekokoweef group, the largest PSL2 being found in T.kokoweef (Fig. 15) and the smallest in deserticola,where they appear to be reduced to mere vestiges(Fig. 45). Also, in virtually all species of Texellathe lateral faces of the middle segment bear ratherprominent folds (for example Figs. 53, 96, 106,136, 163, 177, 181, 187). It seems at least possiblethat these folds represent vestigial PSL2. If this isthe case, the PSL2 would most parsimoniously beinterpreted as a synapomorphy for the entire genus(and the loss of prominent PSL2 as another synapomorphyfor the mulaiki group).The several characters discussed above suggest aunique position of T. bifurcata and the likelihoodthat this species represents the plesiomorphic sistergroup of the remaining Texella.The kokoweefgroup14) PSL2 direction. In T. bifurcata, the PSL2point along the mesal axis of the glans (towards theventral surface of the stylus; Fig. 15); in the kokoweefgroup the PSL2 point along the longitudinalaxis (towards the tip of the stylus; Fig. 27, 41, 45).Using T. bifurcata as the outgroup, the longitudinalalignment is taken to be a synapomorphy for thekokoweef group.The mulaiki group15) Postopercular process (POP). The POP isabsent in the bifurcata and kokoweef species groupsbut is found in virtually all members of the T. mulaikigroup, for which it is considered a synapomorphy.In some species (of the mulaiki and spinopercainfragroups) the POP is absent (Fig. 104) and inothers (of the brevistyla subgroup and of the reddelliand mulaiki infragroups) it may be reduced in size(Figs. 50, 92). This structure also exhibits variabledevelopment intraspecifically. In the species withthe largest available series, T. reyesi, the POP mayvary from being extremely large to small (Fig. 132,133), even within a single cave population, and(along with the trochanteral spur) appears to varyallometrically.16) Basal fold of stylus (BF). The BF is foundonly in members of the T. mulaiki species group(Figs. 53, 68, 77, 83, 87, 96, 107, 116, 120, 136,161, 164, 169, 177, 181, 187, 201), where it isconsidered to be derived.17) Basal knob (BK). The BK is absent inBanksula and the T. bifurcata and kokoweef speciesgroups and is, therefore, a synapomorphy for the T.mulaiki species group (Figs. 66, 77, 83, 86, 95,107, 114, 121, 136, 161, 163). Its absence from T.brevistyla (Fig. 76) and the T. spinoperca infragroup(Figs. 174, 180, 188, 198) is regarded as asecondary loss.The brevistyla and mulaiki subgroups18) Stylus length. In the T. bifurcata andkokoweef species groups (except in T. deserticola )and most species of Banksula the stylus is subequalin length to the PSL. Thus, both a longer stylus (asin T. deserticola (Fig. 45) and the mulaiki subgroup(Figs. 77, 83, 96, 107, 138, 161, 163, 168, 175,180) and an extremely short one [in the brevistylasubgroup (Figs. 53, 75) and in T. homi (Fig. 201)]can be considered derived.19) Tarsal count (TC). The lowest TC, 3-5-4-5,is found in all members of the bifurcata and kokoweefgroups and the brevistyla subgroup. A TC of3-5-5-5 or more is found in all remaining species(except T. bi/obata and homi). Given the presenceof additional plesiomorphies in the former group,the higher tarsal count is presumed to be derived.However, an increased TC is clearly a troglomorphiccharacter, best demonstrated by the clinal variationin the reddelli infragroup which is discussed163

later. Evidence that an increased TC is not strictlytroglomorphic comes from: 1) cave species with TC= 3-5-4-5 (T. kokoweef, shoshone, brevistyla, andthe cavernicole specimen of T. bifureata); and 2)epigean species with TC = 3-5-5-5 (T. fendi and theepigean population of T. longistyla). A TC of3-5-5-5 is therefore regarded as a synapomorphy forthe mulaiki subgroup (with a reversal in T. bi/obataand T. homi), with further increases in TC beingrelated to cave adaptation.20) Stylar apophysis (SA). The SA is absent inbifureata (Fig. 14) but appears to be represented inthe other Californian species (except possibly deserticola)as a weakly developed laterobasal carina(Figs. 28, 41). The SA is absent in the two speciesof the brevistyla subgroup (Figs. 54, 75), but is welldeveloped in the remaining species of the mulaikigroup, where it is typically produced into atooth-like process (Figs. 78, 84, 87, 96, 107, 117,121, 136, 161, 164, 168, 176, 180, 187,201). It isnot clear whether the SA is a synapomorphy for thekokoweef group plus the mulaiki group (with a reversalin the brevistyla subgroup) or whether theSA is independently derived (autapomorphic) in thetwo clades. Given the structural differences betweenthe two types of SA, the latter possibility seemsmore plausible.21) Stylar apophysis shape. The simplest formof the SA is found in T. kokoweef and T. shoshone(Figs. 28, 41). A more prominent SA, which bearsa tooth (reduced to a tubercle in T. hardeni, Fig.88), seems to be synapomorphic for the mulaikisubgroup. In the mulaiki infragroup (Figs. 96, 107)and T. longistyla (Fig. 78) the SA consists of a basaltooth and a lateral carina. The absence of a carinain the remaining species may be derived. Long,attenuated SA teeth are a synapomorphy for thespinoperea infragroup (Figs. 168, 176, 180, 187);here the unique SA in homi, consisting of arectangular lateral carina, appears to beautapomorphic (Fig. 201). A triangular, apicallyscrolled SA (which clips onto the PSL) is asynapomorphy for the reddelli infragroup (Figs.121, 136-147).22) Stylus shape. The stylus is compressed inthe bifureata group (Fig. 14), the kokoweef group(Figs. 26, 42, 45), and the longistyla infragroup(Figs. 78, 84). If this compressed state is plesiomorphic,then the tubular stylus in the brevistylasubgroup (Figs. 53, 54, 74, 75), the mulaikiinfragroup (Figs. 86, 87, 95, 96, 107, 108), thebi/obata infragroup (Figs. 116, 117), T. fendi (Fig.187) and T. homi (Fig. 201) could be consideredderived, as could the apically spatulate stylus in thereddelli subgroup (Figs. 121, 141, 147, 161, 164,168, 176, 180).23) Parastylar lobe (PSL) shape. The PSL ofBanksula are claw-like: curved and tapering to apoint (Fig. 5). Similar PSL are found in many speciesof Texel/a, including T. desertieola (Fig. 45),jungi (Fig. 67) and most species of the reddelli subgroup(Figs. 120, 136, 161, 163, 169, 177, 180)and suggest plesiomorphy. However, in T. bifureata(Fig. 15), the remaining two species of thekokoweef group (Figs. 27, 41), and T. brevistyla(Fig. 53) the PSL are attenuated and ribbon-like,suggesting that this is the plesiomorphic condition.The PSL in the longistyla and mulaiki infragroupsare lobe-like (a robust, rectangular lobe with folds;Figs. 77, 83, 86, 94, 108) and appear derived. Thethree additional forms of PSL in Texella [spiral inbrevistyla (Fig. 53), hooked claw in bi/obata (Fig.116), and vermiform infendi and homi (Figs. 187,198)] are regarded as autapomorphies.24) Mesal megaspines of palpal tibia. In Banksulaand most species of Texella the palpal tibia hasthree mesal megaspines (Fig. 10). The presence ofonly two megaspines in the longistyla and mulaikiinfragroups (Fig. 97) seems to be synapomorphic.The presence of two megaspines in desertieola andshoshone is regarded as a parallelism (autapomorphicfor the two species).25) Ventral plate (VP) setal length. The lateralsetae of the VP are short in Banksula and of moderatelength in all species of Texella (Figs. 15, 42-47,74-79, 81, 86, 94, 106, 120, 136) except the brevidentaand spinoperea infragroups, where they aredistinctly longer (Figs. 160-165, 169, 177, 181,186, 200).26) Apical spine (AS) tip. The apical spine issimple or bifurcate (Figs. 15, 29, 47, 63, 76, 79,82,87, 123, 148, 149, 151) [or, rarely, trifurcate inT. mulaiki (Fig. 102) and some specimens of T.reyesi (Fig. 150)] in all species except the brevidentaand spinoperea infragroups, where it is polyfurcateto plumose and considered a synapomorphy(Figs. 160-165, 170, 172, 179, 191, 198-200).27) Genital opercular (GO) spines. The apicalmargin of the GO is unmodified in most species. Inthe spinoperea infragroup the apical margin of thefemale GO bears a pair of spines (Figs. 173, 183) [2pairs in T. diplospina (Fig. 171) and a pair of blunttubercles in T. fendi (Fig. 194)] which are consideredsynapomorphic. A few additional species havea pair of apical tubercles: males and females of T.jungi (Figs. 64, 71) and T. shoshone (Fig. 38) andsome males of T. spinoperea (Fig. 181). The presenceof these modifications in the more distantly164

elated species IS probably best interpreted as aparallelism.28) Stylar apophysis (SA) ongm. The SA ofmost species originates laterally on the stylus (Figs.77, 84, 88, 96, 107, 116, 120, 136, 164). In thespinoperea infragroup (except T. homi, Fig. 201)the SA is clearly in a more ventral position (Figs.169, 177, 180, 187), which is considered derived.29) Ventral plate setae number. The lowestnumber of < 15 setae per prong is found in Banksulaand the species of the bifureata and kokoweefgroups (Figs. 5, 6, 15, 27, 41, 42-47); the highest(> 25), apparently derived, in the brevidenta andspinoperea infragroups (Figs. 160-164, 169, 177,181, 188,200).The character transformations were studied withthe help of MacClade (Maddison and Maddison,1987). The preferred tree (Table 1) has a length of96 steps and a character index of 0.50.Some ambiguities exist within the mulaiki speciesgroup. The relationships within the mulaiki subgroupare uncertain as two characters, 23 (PSLshape) and 24 (palpal megaspines), cluster the mulaikiand longistyla infragroups whereas a third (22,S shape) clusters the mulaiki and bilobata infragroups.This problem may be resolved as additionaland fresh material of T. bilobata becomes availableto permit more detailed examination of that unusualspecies. The relationships of the three species in themulaiki infragroup are likewise unresolved. However,the form of the SA varies among the threespecies (see discussion under the infragroup) and themost strongly developed SA tooth in T. mulaikimay be plesiomorphic. Finally, the spinop'erea infragroup,with the exception of the synapomorphiesfor T. fendi and homi (10, reduced TrIV spur; 15,absence of POP; 22, tubular S; 23, attenuated PSL)is unresolved. However, the high number (11) andbiserial arrangement (12) of the AT in T. diplospinasuggests plesiomorphy.BIOGEOGRAPHYThe species of Texella are strongly allopatric intheir distribution and closely resemble the westernNearctic phalangodid genera recently studied. Thebifureata and kokoweef groups are Californian andoccur at the extreme parts of the state, both spatiallyand ecologically (Map 1). Markedly disjunct is themulaiki group (Maps 2, 5) which occurs in SE NewMexico and adjacent Texas (longistyla infragroup)and central Texas (all remaining species). With oneexception, all species are geographically isolated, asare all higher level clusters, from infragroup tospecies group. This high degree of disjunction immediatelysuggests a vicariant model of speciation.The presumed barriers, and their relative appearance,are immediately evident (Map 3). Given ourphylogenetic interpretation, the initial barrierswould have been in what is now California: the firstbarrier separating Texella from Banksula (currentlyrestricted to caves of the central Sierran foothills;see map in Briggs and Ubick, 1981); the secondseparating T. bifureata from the kokoweef group; thethird separating the Californian species from the mulaikigroup. All subsequent barriers would havebeen in Texas, the fourth isolating the brevistylasubgroup and the fifth separating the mulaiki andreddelli subgroups.Of some interest, but of uncertain significance, isthe fact that the relatively plesiomorphic elements(Banksula and T. bifureata) occur on presumed exoticterranes, whereas the relatively apomorphicelements (all remaining Texella) occur on the NorthAmerican Plate (Silberling et aI., 1984).However, some dispersal is nonetheless requiredto arrive at the present distribution, as indicated bythe single instance of sympatry: that of T. mulaikiwith diplospina and renkesae. Interestingly, theformer species is the most troglomorphic, whereasthe others (especially T. diplospina) are the leasttroglomorphic of all cavernicolous Texella. Giventhe presumably lower dispersal potential of troglobites,the simplest hypothesis requires the dispersalof the two species of the spinoperea infragroup intothe range of T. mulaiki.TROGLOMORPHYOver two thirds of the species of Texella occur incaves and, as expected, show at least some morphologicalmodification to that environment. The severaltroglomorphic characters identified in Texella(most of which are given in Table 2) appear torepresent four basic types:1) Appendage elongation. The most apparenttroglomorphy is leg elongation. To remove the effectsof body size on leg length the ratio of leg IIlength (the longest leg) to the scute length (whichexhibits less intraspecific variation than the totalbody length) was used. The lowest LII/SL valuesare found in epigean species (2.3-3.1, except for 3.4in T. desertieola); the longest in the mulaiki infragroup(LII/SL = 9.9-15.3), T. reyesi (4.3-8.7), andT. welbourni (4.6), the species here consideredtroglobites. Intermediate leg lengths occur in allremaining (cavernicolous) species (3.2-4.3) exceptfor T. brevistyla and diplospina which have shorter165

•• .c•..oMap I.-Map of western North America showing distribution of Texella bifurcata (dots), T. deserlicola (circle), T. shoshone (solidtriangle), and T. kokoweef(open triangle).166

legs than expected (2.6-3.2). This character is themost sensitive indicator of troglomorphy, apparentlydiscriminating between epigean and cavemicolouspopulations of the same species. In T. bifurcata, thecavemicolous specimen has a higher LIIISL value(3.3) than the epigean specimens (2.6-3.1) and in T.longistyla, the epigean specimen has a lower LIIISLvalue (2.8) than the cavemicole specimens(3.5-3.6).Another character, related to leg length, is thetarsal count (TC). As discussed earlier, a low TC(3-5-4-5) appears to be plesiomorphic in Texella andan increased TC (3-5-5-5) is most parsimoniouslyinterpreted as a synapomorphy for the mulaiki subgroup.However, further increases in the number oftarsomeres appear to be troglomorphic. In fourspecies of troglobites, the mulaiki infragroup and T.reyesi, the TC is distinctly higher (as well as beingpolymorphic); the remaining troglobite, T. welboumi,shows no increase in TC.Also related to appendage elongation is the attenuationof palpal elements, including palpalmegaspines, present only in the most troglomorphicspecies, the mulaiki infragroup.2) Eye reduction. All epigean species and mostcavernicoles have well developed eyes. The fivetroglobitic species lack a retina; four of these alsolack all traces of a cornea. In T. reyesi the conditionof the cornea varies from being well developed tocompletely absent. Retinal loss thus precedes cornealloss as appears to be the case in all other Phalangodidae.The reduction of the eye mound seems to be thefinal step in this transformation series (at least in13'~5 11 : 146\ \ ••1-153~•••r·"" 9 16• • •4/ \ I7 8/10 \ -- ./18/•17Map 2.-Map of western Texas and southeastern New Mexico showing distribution of the species of the Texella mu/aiki group.Numbers: I, T. /ongisry/a; 2, T. we/bourni; 3, T. jungi; 4, T. brevisry/a; 5, T. bi/obara; 6, T. hardeni (I-laby Salamander Cave, only);7, Marguerite Cave; 8, T. cokendo/pheri JRobber Baron Cave, only); 9, T. mu/aiki; 10, T. brevidenta; 11, T. gmbbsi; 12, T. redde/li;13, T. reyesi; 14, T.. spinoperca; 15, T. dip/ospina; 16, T. renkesi; 17, T.fendi; 18, T. homi.167

Texella). This is found in the three most troglomorphicspecies, the mulaiki infragroup, and is most extremein T. mulaiki.3) Reduction of protuberances. Troglobiticspecies are noticeably smoother in appearance thanother Texella as a result of an overall reduction ofbody sculpturing and other protuberances. Thebody surfaces are almost completely devoid of tubercles(including anterior tubercles). The cuticularmicrotubercles are small, giving a fine rugosity.Sexually dimorphic structures, female TrIV tubercle,male TrIV spur, and POP, are reduced in sizeor completely lost.4) Depigmentation. A loss of pigmentation ismost apparent in the troglobitic species.It is possible to arrange the species of Texella onthe basis of increasing troglomorphy. The leasttroglomorphic are the epigean species. Of these T.deserticola, with the highest LII/SL ratio and somereduction of anterior tubercles and depigmentation,would be most troglomorphic. Of intermediatetroglomorphy are the troglophiles, cavernicolousspecies having well developed eyes. In this groupT. brevistyla and diplospina have the lowest LII/SLratios and a relatively high number of anterior tubercles.The troglobitic species range from the moderatelymodified, T. welbourni and reyesi, to the extremelymodified, the mulaiki infragroup.The fact that troglomorphy appears to be derived,using T. bifurcata as outgroup, suggests the use ofthese characters in phylogenetic analysis. Using themost obvious gaps in the troglomorphic characterspectrum, three potential clades can be identified:all cavemicolous species (cave habitation and LII/SL2I@_..........I3oMap 3.-Distribution and area c1adogram of the subgroups and infragroups of the Texella mulaiki species group. Numbers: I,brevistyla subgroup; 2, longistyla infragroup; 3, mulaiki infragroup; 4, bi/obara infragroup; 5, brevidenta infragroup; 6, spinopercainfragroup; and 7, reddelli infragroup.168

~ 3.3 are synapomorphic); all troglobitic species(eye loss and LII/SL ~ 4.3 are synapomorphic); andthe mulaiki infragroup (reduction of eye mound,LII/SL ~ 9.9, etc. are synapomorphic). The factthat only the last group appears to be monophyleticsuggests that troglomorphic characters are primarilyadaptive.Indeed, it appears most probable that troglomorphyevolved several times in Texella. For example,the kokoweef group includes two troglophilic(T. kokoweef and shoshone) and one epigean species(T. desenicola). Since T. shoshone appears to bemore closely related to desenicola, it would seemmost parsimonious that troglomorphy evolved oncein the species group and was subsequently lost indesenicola. This may be further supported by thefact that desenicola, as noted above, is the mosttroglomorphic of the epigean species. However,one weakness in this argument is that it fails to takeinto account the relative time frame of phylogeneticand troglomorphic characters. In all probability thespeciation events in this group, judging by themorphological distinctness of the species and thebasal position of the group, must be old. If thedegree of troglomorphy is a function of time, thenthe presumed antiquity of these species should bereflected in their troglomorphic characters, which isnot the case (the species, especially kokoweef, arenot strongly troglomorphic). Thus, an equally, ifnot more, parsimonious argument would be that thecavemicolous habit and subsequent troglomorphicmodification is a more recent event, and one whichoccurred independently in the two species. Usingsimilar reasoning, it could be argued thattroglomorphy evolved independently at least seventimes in the mulaiki species group.Furthermore, and of possible use in understandingthe evolutionary patterns in Texella, is the clinalvariation in troglomorphic characters. This isspecies LnlSL(N) tarsal count ret/cor ant tubs TrlV f TrlV m popbifurcata (E) 2.6-3.1(22) 3-5-4-5 +/+ 9-12prs 0-2 tubs 1-3 tubs -bifurcata 3.3(1 ) 3-5-4-5 +/+lwkoweef 3.3-3.8(6) 3-5-4-5 +/+ 4-8prs 2 tubs long -shoshone 3.9-4.2(3) 3-5-4-5 +/+ 2prs 1 tub lon2 -desenicola (E) 3.4(\) 3-5-4-5 +\+ 2prs ? lon2 -junRi (E) 2.3-2.7(6) 3-5-4-5 +/+ 2-6prs I tub lon2 lar2ebrevistyla 2.6-3.2(1 \) 3-5-4-5 +/+ 8prs 2 tubs lon2 md-12longisrvla (E) 2.8(\) 3-5-5-5 +1+ 2prs ? lon2 medlonRisrvla 3.5-3.6(4) 3-5-5-5 +/+ 2-3prs I tub long medwelbourni 4.6(\) 3-5-5-5 -/- 5prs ? short smallhardeni 9.9-11.6(3) 3-6/8-4-5 -/- 1pr none tiny noneMarguerite Cave 9.4-11.2(2) 3-617-4-5 -/- ODr 1 tub ? ?watersi 11.1-12.9(5) 4-618-5-5 -/- 1pr none tiny smallmulaild 12.5-15.3(8) 4-5/8-5/6-5/6 -/- 0-lpr none none nonebi/obata (E) 2.6-2.8(3) 3-5-4-5 +/+ 2prs 1 tub short nonereddelli 3.8-4.2(7) 3-5-5-6 +/+ 4-5prs 2 tubs lon2 lar2ereyesi 4.3-8.7(71) 3/4-4/8-5/6-5/8 -/+ 0-2prs 0-1 tub md-lg sm-lgbrevidenta 3.3(1) 3-5-5-6 +/+ 4prs ? lon2 medgrubbsi 3.6-3.7(4) 3-5-5-5 +/+ 2prs 2 tubs lon2 md-12diDlosDina 2.6-3. \(9) 3-5-5-5/6 +/+ 4-6prs I tub lon2 md-12renkesae 3.2-3.4(5) 3-5-5-5 +/+ 4prs I tub lon2 md-12sDinoperca 3.4-4.3(13) 3-5-5-5 +/+ 3-4prs I tub 10nl( largefendi (E) 2.5-2.7(8) 3-5-5-5 +/+ 2-4prs 1 tub short nonehomi (E) 2.5-2.9(4) 3-5-4-5 +1+ 3prs I tub short noneTable 2.-Table of troglomorphic characters. Abbreviations: E = epigean species or populations; LlVSL = leg ill scute length; N= number of specimens measured; ret = retina; cor = cornea; ant tubs = number of paraocular anterior tubercles; pres) = pair(s);TrlV = trochanter rv; f = female; m = male; tubes) = tubercle(s); Ig = long, large; md, med = medium-sized; sm = small.169

evident in virtually all (nonmonotypic) clades. Inthe kokoweef group, the direction of least to mosttroglomorphic goes from T. deserticola to shoshone.A clearly defined cline is not evident in thebrevistyla subgroup. Although the cavernicolous T.brevistyla has a slightly higher LII/SL ratio, T.jungi has a more reduced AT and female TrIVtubercle count. In the longistyla infragroup,troglomorphy proceeds from the epigean populationof T. longistyla to the troglobitic T. welbourni. Inthe mulaiki infragroup, the direction is from T.hardeni and the species from Marguerite Cave tomulaiki. In the reddelli infragroup, the highesttroglomorphy is in the northernmost populations ofT. reyesi. In the brevidenta infragroup, T. grubbsi isclearly the more troglomorphic of the two species.Finally, the situation in the spinoperca infragroup issomewhat more complex. Although thecavemicolous species clearly form a cline (fromdiplospina to spinoperca, the most troglomorphic),the position of the epigean species is not clear.Contrary to expectation, these species appear moretroglomorphic than the cavernicolous species insome characters (Table 2).Finally, of biogeographic interest is the spatialdistribution of troglomorphy. The resulting pattern(Map 4) is unusually orderly. First, there appears tobe a directionality in troglomorphism. In virtuallyall clades, the least and the most troglomorphicharvestmen are at opposite ends of the clade'sdistribution. Second, and what seems quiteremarkable is that, in virtually all clades, the mosttroglomorphic element is at the N to NE part of thedistribution. The apparent nonrandornness of thesepatterns begs for an explanation. Hopefully one willcome as the distribution patterns ofthe rich Nearcticcavernicole fauna, especially that of central Texas,become better known.~-----'Map 4.-Distribution of troglomorphy in the Texella mulaiki species group. Symbols: open circle = epigean species (and epigeanpopulation of T. longisryla ); black dot = troglobitic species; mixed symbol = troglophilic species (and troglophilic populations of T.longisryla). Arrows give the direction towards increasing troglomorphy; double arrows indicate ambiguity in polarity.170

TAXONOMYTEXELLA Goodnight and GoodnightTexella Goodnight and Goodnight, 1942: 10 (typespecies by monotypy and original designationTexella mulaiki Goodnight and Goodnight 1942).Diagnosis.-Species of Texella may be distinguishedfrom those of Banksula in lacking spiniferoustubercles dorsally on the palpal femora, fromCalicina and Microcina in having tubercles or spursventrally on the fourth trochanters, and from Phalangodesand other nominate genera from the easternNearctic in having the eye mound situated on theanterior margin of the scute. The presence of threemesal megaspines on the palpal tibia and/or the absenceof a retina will distinguish most species ofTexella from those of Sitalcina.Males of Texella may be distinguished from allother Nearctic phalangodid genera by the followingcombination of genitalic characters: ventral platewith bifurcation wider than glans, ventral plate withwell developed apical spine, and glans doublefolded, appearing sigmoid in lateral view. Mostfemales may be distinguished from those of othergenera in having ovipositors with transverse wrinklesand a pair of apical teeth.Description.-Total body length 1.2-2.7; scutelength 0.8-1.7. Body color from yellowish white tobrownish orange, appendages relatively paler. Bodycuticle with uniform layer of hemispherical to ovoidmicrotubercles, appearing smooth (microtuberclesabout 5 p.m in diameter) to rugose (about 10 p.m) atnormal magnifications, and scattered patches oflarger tubercles (from about 20 p.m), which may beelongate, scale-like, and setiferous; setation light tomoderate, most pronounced on the abdomen. Scutewith eye mound at anterior margin; thoracic regionset off dorsally by transverse grooves and laterally(at insertion of leg III) by pronounced constriction;abdominal region typically with five transverse rowsof coarse tubercles representing fused tergites I-V.Eye mound roughly conical, reduced in troglomorphicspecies, rugosity smooth to coarse; eyes welldeveloped, reduced, or absent. Anterior tuberclespresent or absent. Ozopores well developed, on anteriolateralmargin, with distinct posterior channel.Tergites VI-VIII free, typically with median, transverserows of coarse tubercles interspersed withshort setae. Venter textured similarly as scute butmore densely setose. Coxae with coarse tuberculationtypically on ventral surfaces, retrolaterally oncoxae II, and prolaterally on coxae IV; enditesmesoapically situated on palpal coxae, coxae I, andcoxae II; anteriorly directed spur on ventral surfaceof palpal coxae. Genital operculum between coxaeIV, roughly semicircular, may be armed apicallywith 1-2 pairs of spines or tubercles. Large to smallconical, cuticular outgrowth (= postopercular process)immediately behind genital operculum presentin males of most species. Sternites typically withtransverse rows of tubercles; sternite I with spiraclenear distal part of coxa IV; sternites II-V free; sternitesVI-VII fused. Anal operculum coarsely tubercuiateand setose.Chelicerae with smooth cuticle, basal segmentwith dorsoapical swelling, apical segments dorsallysetose. Palpi with smooth cuticle of appressedscales, apically setose, typically robust but conspicuouslyattenuated in highly troglomorphicspecies; megaspine number and distribution: trochanter:1-2 ventral (small); femur: 3 ecto(ventro)basal, 1-2 mesoapical; patella: 1 ectal, 1-2 mesal;tibia: 2 ectal, 2-3 mesal; tarsus: 2 ectal, 2 mesal.Legs with cuticle of apically rebordered scalesgiving honeycomb appearance; calcanei and tarsismooth; lightly to moderately setose; leg II longest,leg formula (longest to shortest) II-IV-I-III; leg IIlength / scute length = 2-15; tarsal counts typically3-5-4-5, 3-5-5-5, and 3-5-5-6, but up to 4-8-6-8 inhighly troglomorphic species; tarsal claws single onlegs I and II, paired on legs III and IV; legs ofjuveniles with posterior claws on onychium andenclosing arolium. Trochanter IV with ventraltubercles (females, some males) or elongate process(=spur) (most males); spur generally slightlycurved, ventrally with smooth knobs, apically bentor enlarged.Penis without muscles, composed of basal sac,truncus, and apical glans. Ventral plate bifurcate;incision at least as wide as glans width; each prongwith a rigid, cuticular, apical spine and movable setae;setae arranged in dorsal, lateral, and ventral series.Glans double folded, appearing sigmoid inlateral view, expansion along two axes of rotation:1) junction of basal and middle segment and 2)junction of apical segment and stylus; basal segmenttypically with ventroapical extension (= basalknob); middle segment with a pair of distal lobes(parastylar lobes), occasionally with a pair of secondaryparastylar lobes; stylus apical, may bear a pairof basolateral processes (stylar apophyses) and ventralcarina. Ovipositor bent or straight; cuticlewrinkled, with longitudinal and transverse folds;microsculpturing of small, thin spines, serrations, orrounded tubercles, rarely smooth; apex with dorsoventralslit, typically with a pair of teeth; setal171

Key to the species of TexellaMales1. POP absent (Figs. 1, 3, 104) 2POP present (Figs. 2, 50, 64, 92, 132, 133) 102. Trochanter IV with ventral spur longer than trochanter (Figs. 24, 37). Distribution California 3Trochanter IV with spur (or tubercles) shorter than trochanter (Figs. 11, 192)or absent (Fig. 104). Distribution California and Texas 53. Glans with well developed PSL2; S subequal to PSL (Figs. 27, 39). Cavernicole 4Glans with tiny PSL2. S distinctly longer than PSL (Fig. 45). Epigeandeserticola n. sp.4. PSL2 robust; VPP wide (Fig. 27). Palpal tibia with three mesal megaspines.Scute with 4-8 pairs of AT (Fig. 22)PSL2 slender; VPP narrow (Fig. 39). Palpal tibia with two mesal megaspines.Scute with 2-3 pairs of AT (Fig. 34)kokoweefn. sp.shoshone n. sp.5. Eyes well developed. TC = 3-5-4-5 or 3-5-5-5. LII/SL = 2.5-2.9 (rarely to3.3). Epigean (rarely troglophilic) 6Eyes absent, lacking retinae and corneae. TC higher. LII/SL = 9.9-15.3.Cavernicole (troglobitic) 96. Trochanter IV with 1-3 ventral tubercles (Fig. 11). Glans with 2 pairs ofPSL (Fig. 15).SA absent (Fig. 14). Distribution NW California and SW Oregonbifurcata (Briggs)Trochanter IV with short, ventral spur (Fig. 192). Glans with 1 pair PSL. SA present.Distribution central Texas 77. TC = 3-5-5-5. Both PSL and SA attenuated (Fig. 187) .fendi n. sp.TC = 3-5-4-5. SA and sometimes PSL not attenuated 88. BK present, bilobed; PSL hooked (Figs. 113-117) bi/obata n. sp.BK absent; PSL attenuated (Figs. 198-201)homi n. sp.9. Trochanter IV lacking spur or tubercle (Fig. 104). Tarsus I with fourtarsomeresTrochanter IV with tiny ventral spur. Tarsus I with three tarsomeresmulaiki Goodnight and Goodnighthardeni n. sp.10. Eyes well developed 11Eyes reduced, lacking at least retinae 1.911. TC = 3-5-4-5. S shorter than PSL (Figs. 74-76) 12TC = 3-5-5-5 or 3-5-5-6. S longer than PSL 1312. PSL short, claw-shaped (Figs. 67, 74). Scute with 3-6 pairs of AT (Figs. 60-62).EpigeanPSL long, spiraled (Figs. 52-55, 76). Scute with 7-8 pairs of AT (Fig. 48).Cavernicolejungi n. sp.brevistyla n. sp.13. S compressed (Figs. 77, 78). Distribution Eddy County, New Mexico, andCulberson County, Texaslongistyla n. sp.S spatulate apically (Figs. 121, 164, 168, 176, 180). Distribution central Texas 14172

14. BK absent (Figs. 177, 180). SA with moderately to extremely long prongs(Figs. 168, 176, 180) 15BK present (Figs. 121, 160-163). SA with short prongs (Fig. 164) orclip-like (Fig. 120) 1715. Scute with 3-4 pairs of AT (Fig. 178). LII/SL = 3.4-4.3. DistributionTravis County, Texasspinoperca n. sp.Scute with 4-6 pairs of AT (Fig. 166). LII/SL = 2.6-3.4. DistributionHays County, Texas 1616. SA originating medially on S (Fig. 168) diplospina n. sp.SA originating basally on S (Fig. 176)renkesae n. sp.17. SA clip-like (Fig. 120). UIISL = 3.8-5.2 reddelli Goodnight and GoodnightSA with a pair of short, blunt prongs (Fig. 164). UIISL = 3.3-3.7 1818. TC = 3-5-5-6. 4 pairs of AT. LII/SL = 3.3 brevidenta n. sp.TC = 3-5-5-5. 2 pairs of AT. LII/SL = 3.6-3.7 grubbsi n. sp.19. SA clip-like (Fig. 136). POP medium to long (Figs. 132, 133). LII/SL = 4.3-8.7 reyesi n.sp.SA with prongs (Figs. 84, 96). POP small to medium sized 2020. POP small (Fig. 92). LII/SL = 9.9-11.6 cokendolpheri n. sp.POP medium sized. LII/SL = 4.6 welboumi n. sp.Females(not known for T. deserticola, welboumi, and brevidenta)1. TC = 3-5-4-5 2TC = 3-5-5-5, or higher. 82. Scute with at least 4 pairs of AT (Fig. 16) 3Scute with at most 3 pairs of AT (Fig. 61) 53. Ovipositor bent, with a pair of large apical teeth (Figs. 18, 19). DistributionNW California to SW Oregonbifurcata (Briggs)Ovipositor straight; apical teeth small or absent (Figs. 30, 56) 44. LII/SL = 3.3-3.8. Ovipositor lacking apical teeth. Distribution S California kokoweefn. sp.LII/SL = 2.6-3.2. Ovipositor with small apical teeth. Distribution Texas brevistyla n. sp.5. Genital operculum apically unmodified bilobata n. sp.Genital operculum with a pair of apical tubercles or spines 66. Palpal tibia with 2 mesal megaspines. Distribution California shoshone n.sp.Palpal tibia with 3 mesal megaspines. Distribution Texas 77. Genital operculum with a pair of apical tubercles (Fig. 71) jungi n. sp.Genital operculum with a pair of apical spineshomi n. sp.8. Palpal tibia with two mesal megaspines 9Palpal tibia with three mesal megaspines 12173

9. Retina present. LII/SL = 2.8-3.6 longistyla n. sp.Retina absent. LII/SL = 10-15 1010. Tarsus I with three tarsomeres. LII/SL = 10-12 hardeni n. sp.Tarsus I with four tarsomeres 1111. Palpal patel1a with one mesal megaspine. Eye mound with several prominentconical tubercles (Fig. 90). LII/SL = 11-13 cokendolpheri n. sp.Palpal patel1a with two mesal megaspines. Eye mound lackingprominent tubercles (Fig. 101). LII/SL = 12-15 mulaiki Goodnight and Goodnight12. Genital operculum with spines or tubercles on anterior margin 13Genital operculum with anterior margin entire 1613. Genital operculum with 1 pair of apical tubercles (Fig. 194). Epigean fendi n. sp.Genital operculum with 1-2 pairs of apical spines. Cavernicolous 1414. Genital operculum with 2 pairs of apical spines (Fig. 171) diplospina n. sp.Genital operculum with 1 pair of apical spines (Fig. 183) 1515. LII/SL = 3.2-3.4. Distribution Hays County, Texas renkesae n. sp.LII/SL = 3.4-4.3. Distribution Travis County, Texas spinoperca n. sp.16. Retina absent reyesin. sp.Retina present 1717. TC = 3-5-5-5. LII/SL = 3.6-3.7 grubbsi n. sp.TC = 3-5-5-6. LII/SL = 3.8-4.9 reddelli Goodnight and Goodnightpattern: 4 pairs lateral, 1 pair ventral, subapical,and 2 pairs dorsal, subapical; subapical setae locatedin cuticular folds. Hyperexpanded ovipositor withfour apical valves and a patch of smal1 basolateralspines.Range.- Known from southwestern Oregon tosouthern California and southern New Mexico tocentral Texas.Texella Goodnight and GoodnightThe bifurcata GroupDiagnosis.-The presence of at least 9 pairs ofAT (Fig. 9) separates this group from the others.The males may be further distinguished by thefol1owing combination of characters: ventral surfaceof TrIV with tubercles, lacking spur (Fig. 11); POPabsent (Fig. 1); PSL2 present, mesal1y directed; Scompressed, basal1y unmodified, lacking SA; PSLlong, ribbon-like, apical1y bifurcate; apex of VPPwith ventral notch; AS long (2.5 times VP width)(Figs. 6, 12-15). The females appear to be unique inhaving a bent ovipositor (Fig. 18).Range.-Known from southwestern Oregon andnorthwestern California.Texella bifurcata (Briggs),New CombinationFigs. 1,6, 8-21Sitalcina bifurcata Briggs, 1968:29, figs. 31, 61,92, and 105 (male holotype from near westentrance of Castle Crags State Park, ShastaCounty, California, in CAS, examined).Diagnosis.-Males of this species are most easilydistinguished from other Texella by the presence oftransversely (rather than 10ngitudinal1y) directedPSL2 (Figs. 6, 15). Females have an ovipositor,possibly unique in the genus, which is bent (ratherthan straight) when ful1y extended (Fig. 18). Also,T. bifurcata differs from other species in the highernumber of AT (9-12 pairs, compared to 8-0 pairs inother Texella) which are arranged in two rows (abiserial arrangement is also found in T. kokoweef,jungi, brevistyla, reddelli, and diplospina) (Fig. 9).174

Figs. 8-11.-Texella bijUrcata (Briggs), male topotype: 8, anterior half of body, lateral view; 9, enlargement of previous imageshowing cornea and anterior tubercles; 10, palpi and chelicerae, ventrolateral view; II, trochanter IV, lateral view.175

Figs. 12-15.-Texella bifurcata (Briggs), male topotype: 12, penis, ventrolateral view; 13, 15, penis, dorsolateral view; 14,penis, apical view.176

Description.-Total body length, 1.51-2.10.Scute length, 1.00-1.31. Leg II length, 2.92-4.03.Leg II/Scute length, 2.60-3.27. (N=23).Color orange to brownish orange. Body coarselyrugose; tubercles present on eye mound, dorsal andlateral surfaces of pars thoracica, tergite margins,and ventral and lateral faces of coxae. Carapacewith 9-12 pairs of anterior tubercles arranged indouble rows. Eye mound broadly conical, eyeswell developed. Palpal megaspines: trochanter, 1ventral; femur, 1 mesoapical; patella, 2 mesal; tibia,3 mesal. Tarsal count: 3-5-4-5.Male (holotype): Total body length, 1.72. Scutelength, 1.18; width, 1.26. Eye mound length, 0.31;width,0.26. Leg II length, 3.31. TrIV with threetubercles. POP absent.Penis (Figs. 12-15): VPP with apex stronglybent; 2 dorsal, 8 lateral, and 4 ventral setae; ASlong, evenly curved, with small, subapical tooth.Glans with BK absent; ML apparently present,small; PSL long, ribbon-like, apically bifurcate;PSL2 present, transversely directed. S long, compressed;lacking BF and SA.Female (allotype): Total body length, 1.67.Scute length, 1.15; width, 1.31. Eye mound length,0.30; width, 0.27. Leg II length, 3.13. TrochanterIV with ventral tubercle.Ovipositor (Figs. 18-21): cuticle intricately folded;dorsal surface sparsely set with small spines;ventral with microtubercles; 1 pair apical teeth present;setal pattern typical.Variation.-Slight differences were detected inthe male genitalia. The tip of the stylus varies inshape and the subapical prongs of the PSL varyslightly in size and curvature.The number of ventral tubercles on Tr IV varies,even within a single population. Typically, maleshave two tubercles, the apical being twice the lengthand diameter of the basal, and females have a singlesmall tubercle. Some males were found with one(five out of 26 specimens examined) or three (5/26)tubercles, as well as some females with zero (2/11)or two (3/11). Males from Round Mountain haveone or two tubercles, all small. In the Oregonpopulations, males have only the single, apical tubercle(which is of moderate size) and the femaleslack them altogether.The individuals from Oregon are also paler andsomewhat smaller than those from California, having:SL = 1.00-1.10 (N = 5), compared to SL =1.13-1.31 for remaining epigean specimens measured(N = 18); SW = 1.08-1.15 (1.18-1.41 for theremaining specimens); LIIL = 2.92-3.10 (3.13-3.59for the remaining specimens). The single OregonFigs. 16-17.-Texella bifurcata (Briggs), female topotype: 16, anterior half of body, lateral view;17, trochanter IV, lateral view.177

Figs. 18-21.-Texella bijUrcata (Briggs), female topotype: 18, ovipositor, lateral view; 19, ovipositor tip, lateral view; 20,ovipositor, apical view; 21, ovipositor, enlarged view of dorsolateral surface.178

male examined closely has fewer setae on the VPprongs, 3 ventral and 6 or 7 lateral, but does not appearto differ from the Californian males in glanscharacters.The single female from a cave locality (ShastaLake Caverns) has distinctly longer appendages. ItsLII/SL = 3.27; for all remaining individuals examined(N = 22) the LII/SL ratio ranges from2.60-3.10 (mean = 2.79). This specimen is alsoslightly less pigmented but otherwise, including thetarsal count, resembles those from surface habitats.Natural History.-Collected from beneath rocksand logs and in duff in coniferous and riparian forests.One specimen is from a cave.Material Examined.-UNITED STATES: California:Shasta Co.: near west entrance CastleCrags State Park, 6 May 1967, under granite, pinewoodland (T. Briggs, K. Hom, and A. Jung, CAS),13 males (including type), 5 females; 5 mi SWRound Mountain, 7 May 1967, under limestone,pine woodland (T. Briggs and A. Jung, CAS), 6males, 3 females; Clear Creek Camp, ShastaNational Forest, 5 Aug. 1967, under rotten log,riparian woodland (T. Briggs, CAS), 1 male, 2females; Shasta Lake Caverns, 16 mi NNE Redding,10 Apr. 1979 (D. Rudolph, B. Martin, and S. Winterath,CAS), 1 female; N of Hazel Creek, 26 June1954 (R. Schuster and B. Adelson, AMNH), 2males, 6 females; 10 mi S Dunsmuir, 2 July 1954(R. Schuster and E. Gilbert, AMNH), 1 male;Squaw Creek, 122.40, 15 July 1937 (R. Chamberlin,AMNH), 1 male. Del Norte Co.: 0.3 mi SEeast entrance Jedediah Smith State Park, 25 June1966 (A. Jung and K. Hom, CAS), I male. Oregon:Jackson Co.: IS mi SW Ruch, Upper ApplegateRoad, T40S R3W sec 8, 1800', 13 Nov. 1971,berlese duff, debris, moss, bark (E. Benedict,WAS), 1 male, 1 female; 6 mi S Ruch, Upper ApplegateGrange, T39S R3W sec IS, 1600', 13 Nov.1971, berlese mixed duff, litter (E. Benedict,WAS), 1 male, 2 females.Distribution.-Known from northwestern Californiaand adjacent regions of Oregon (Map I).The kokoweeJGroupDiagnosis.-Males of this group differ fromthose of the other two groups by the followingcombination of characters: TrlV with well developedspur (Figs. 24, 25, 37); POP absent; BKabsent; BF absent; PSL2 medium sized to small,longitudinally directed (Figs. 26-29, 38-47). Females(only those of T. kokoweeJ clearly observed)have a straight ovipositor which lacks apical teeth(Figs. 30, 32, 33).Texe/La kokoweeJ, new speciesFigs. 22-33, 42, 43.Diagnosis.-Males of this species differ from allother Texella by the combined presence of a uniqueTrIV spur which is apically T-shaped (Figs. 24, 25)and a VPP which has a long, pointed AS (Figs.26-29).Type.-Male holotype from Kokoweef CrystalCave, San Bernardino County, California (30 Dec.1968; T. Briggs, B. Lum, and G. Leung), depositedin CAS.Etymology.-The specific name is a noun in appositiontaken from the type locality.Description.-Total body length, 1.67-1.82.Scute Length, 1.21-1.31. Leg II length, 3.97-4.72.Leg II/Scute length, 3.28-3.84. (N = 6).Color light brownish orange. Body coarsely rugose;tubercles present on eye mound, on the parsthoracica, tergite margins, and coxae. Carapacewith 4-8 pairs of AT arranged in two, poorly definedrows. Eye mound a broadly rounded cone,eyes well developed. Palpal megaspines: trochanter,1 ventral; femur, 1 mesoapical; patella, 2mesal; tibia, 3 mesal. Tarsal count: 3-5-4-5.Male (holotype): Total body length, 1.67. Scutelength, 1.26; width, 1.23. Eye mound length, 0.31;width, 0.31. Leg II length, 4.49. TrIV spur robust,curved, apically enlarged (T-shaped); length,0.54. POP absent.Penis (Figs. 26-29, 42, 43): VPP with 3 dorsal,10 lateral, 4 ventral setae; AS long, pointed, slightlycurved. Glans: BK absent; ML small but distinct;PSL long, ribbon-like, with subapical ectal tooth;PSL2 well developed, apically directed. S long,compressed, apically bent; BF absent; SA representedby laterobasal carina.Female (paratopotype): Total body length, 1.82.Scute length, 1.21; width, 1.33. Eye mound length,0.26; width, 0.27. Leg II length, 3.97. TrIV withtwo ventral tubercles.Ovipositor (Figs. 30, 32, 33): cuticle intricatelyfolded; surface sculpturing of minute serrations,sparse ventrally; apical teeth absent; setal pattern: Ipair ventral, 2 pairs dorsal, 4 pairs lateral.Note.-The glans of T. kokoweeJ shows dramaticchanges following treatment with KOH. Unlike thestrictly positional changes observed in the majorityof other species so treated, the glans structures inthis species are also altered in appearance. Specifically,the PSL are both inflated, especially in the179

Figs. 22-25.-Texella kokoweef, new species, male paratopotype: 22, anterior half of body, lateral view; 23, palpi and chelicerae,sublateral view; 24, trochanter IV spur, lateral view; 25, trochanter IV spur, apical view showing enlarged tip.180

Figs. 26-29.-Texella kokoweej, new species, male paratopotype: 26, penis, dorsal view; 27, penis, dorsolateral view; 28, penis,apical view; 29, penis, ventral view.181

Figs. 30-33.-Texella kokoweeJ, new species, female paratopotype: 30, ovipositor, lateral view; 31, trochanter IV, lateral view;32, ovipositor, subapical view; 33, ovipositor, enlarged view oflateral surface.182

apical region which loses the apical fold, andtwisted, with the lateral prongs becoming mesal(compare Figs. 26-29 to 42 and 43). Similarchanges, but of a lesser magnitude, occur to the PSLof T. shoshone.Natural History.-Specimens were collected beneathdecaying wood debris, along with several additionalinvertebrates. Given the isolation ofKokoweef Crystal Cave, it is likely that much of thefauna is endemic. At present, a species of spider,Usofila n. sp. (Telemidae), is known only from thiscave (Gertsch, personal communication).Other Material Examined (Paratypes).- UNITED STATES: California: San Bernardino Co.:Kokoweef Crystal Cave, 30 Dec. 1968 (T. Briggs,B. Lum, and G. Leung, CAS), 7 males, 1 female; 1Apr. 1972 (T. Briggs and D. Ubick, CDU,AMNH), 2 males, 1 female.Distribution.-Known only from the type locality(Map 1).Texella shoshone, new speciesFigs. 34-41,44.Diagnosis.-Males of this species may be distinguishedfrom all others in the group by the followingcombination of characters: VPP thin,parallel-sided; AS apparently absent; PSL2 slender;and GO with 2 pairs of apical tubercles (also foundin females) (Figs. 38-41, 44). This species is alsodistinct somatically in having the body surface studdedwith large tubercles; especially prominent on thetergite margins and anal operculum (Figs. 34-35).Type.-Male holotype from Shoshone Cave,Inyo County, California (29 Dec. 1971; T. Briggs),deposited in CAS.Etymology.-The specific name is a noun in appositiontaken from the type locality.Description.-Total body length, 1.40-1.49.Scute length, 0.90-1.05. Leg II length, 3.82-4.23.Leg II/Scute length, 3.86-4.24. (N = 3).Color yellowish orange. Body coarsely rugose;large tubercles on eye mound, pars thoracica, coxae,and especially tergite margins and anal operculum.Carapace with 2 pairs of large anterior tubercles.Eye mound rounded, dorsally flattened, eyes welldeveloped. Palpal megaspines: trochanter, 1 ventral;femur, 1 mesoapical; patella, 1 mesal; tibia, 2mesal. Tarsal count: 3-5-4-5.Male (holotype): Total body length, 1.49. Scutelength, 1.05; width, 1.08. Eye mound length, 0.23;width, 0.23. Leg II length, 4.05. TrIV spur relativelystraight; length, 0.25. POP absent. GO with2 pairs of apical tubercles.Penis (Figs. 38-41): VPP slender, parallel sided,apically pointed; with 2 dorsal, 6 lateral, and 4 ventralsetae; AS apparently absent. Glans: BK absent;ML absent or possibly represented by two laterallobes; PSL long, ribbon-like, apically folded; PSL2slender, apically directed. S long, compressed,relatively straight; SA represented by basolateralcarina; BF absent.Female (paratopotype): (The single availablespecimen was badly damaged during dissection.)Similar to males in size and general appearance. GOwith two pairs of prominent apical tubercles. Ovipositorwithdrawn, apical setae visible. Tr IV withone ventral tubercle.Note.-The PSL apparently inflate during glansexpansion (see note in T. kokoweef and compareFigs. 38-41 to 44).Natural History.- This species is known onlyfrom Shoshone Cave. This cave has a thermalspring, which maintains conditions at about 25°Cand 100% humidity. The habitat and cave conditionsare more thoroughly described by Briggs andHom (1972). Shoshone Cave is located in a saltbushdesert and has a rich fauna which, as a result of itsisolation, is high in endemicity. In addition to T.shoshone, the fauna includes an endemic shizomid[Hubbardia shoshonensis (Briggs and Hom, 1972)],milliped (Colactis briggsi Shear 1974), cricket(Ceuthophilus n. sp.), and carabid (Rhadine n. sp.).Other Material Examined (Paratypes).- UNITED STATES: California: Inyo Co.: ShoshoneCave, 29 Dec. 1971 (T. Briggs, CAS), 3 males, 1female.Distribution.-Known only from the type locality(Map 1).Texella deserticola, new speciesFigs. 45-47.Diagnosis.-This species may be distinguishedfrom the males of other species in the group by thecombined presence of a medium-sized AS, shortclaw-like PSL, and strongly reduced PSL2 (Figs.45-47).Type.-Male holotype from Whitewater Canyon,Riverside County, California (8 Apr. 1979; D.Ubick), deposited in CAS.Etymology.-The species name IS Latin for"dweller of the desert".Description.-Male (holotype): Total bodylength, 1.33. Scute length, 0.92; width, 0.85.Eyemound length, 0.21; width, 0.23. Leg II length,3.13. LII/SL = 3.40.183

Figs. 34-37.-Texella shoshone, new species, male paratopotype: 34, anterior half of body, lateral view; 35, middorsal section ofbody, lateral view showing prominent tubercles on tergite margins; 36, palpi and chelicerae, lateral view; 37, trochanter IV spur,lateral view.184

Figs. 38-41.-Texella shoshone, new species, male paratopotype: 38-39, penis, lateral view; 40, penis, ventrolateral view; 41,penis, dorsolateral view.185

Color yellowish orange. Body coarsely rugose;tubercles on eye mound, pars thoracica, tergitemargins, coxae, etc. Carapace with 2 pairs of AT.Eye mound low, rounded; eyes well developed.Palpal megaspines: trochanter, I ventral; femur, 1mesoapical; patella, 2 mesal; tibia, 2 mesal. Tarsalcount: 3-5-4-5. Tr IV spur relatively straight,length, 0.21. POP absent.Penis (Figs. 45-47): VPP apically pointed; with3 dorsal, 4? lateral, and 2 ventral setae; ASmedium-sized, stout. Glans: BK absent; MLapparently absent; PSL short, claw-like; PSL2vestigial, apically directed. S compressed; SAapparently represented by small, laterobasal carina;BF absent.Female: Unknown.Note.-The single specimen available for studywas unusually resistant in having its genitalia extractedand, unfortunately, was badly damaged inthe process.Natural History.-The holotype was collectedunder a rock (?granite) in a mesic southern chaparralcommunity along the canyon slope. A subsequentvisit to the locality, on 2 April 1985, failed to produceadditional specimens.Other Material Examined.-None.4344Figs. 42-47.-Texella species, male genitalia: 42-43, T. kokoweef, new species, paratopotype: 42, penis, suhventral view,expanded with KOH; 43, penis, lateral view. 44, T. shoshone, new species, paratopotype: Penis, lateral view, expanded with KOH.45-47, T. deserticola, new species, holotype: 45, penis, lateral view, expanded; 46, penis, lateral view, partially expanded; 47,penis, ventral view. Scale bar = 0.25 mm.186

Distribution.-Known only from the type locality(Map 1).The mulaiki GroupDiagnosis.-The following combination of malecharacters distinguishes the members of this group:POP present (secondarily lost in T. mulaiki,hardeni, bilobata, !eflai, and homi), glans peniswith BK (lost in T. brevistyla and the spiflopercainfragroup), and stylus with BF (small in T.brevistyla) (Figs. 7, 132, 133). No characters havebeen found which distinguish females.Range.-Known from southern New Mexico tocentral Texas (Maps 2-4).The brevistyla SubgroupDiagnosis.-The combination of S shorter thanPSL, and the absence of a SA (Figs. 52-55, 66-69,74-76) distinguishes males of these species from allother Texella.Range.-Known from western central Texas(Maps 3,4).Texella brevistyla, new speciesFigs. 48-59, 76.Diagnosis.-Males of this species are distinguishedfrom other Texella by the following combinationof characters: VPP apically convergent; PSLlong, attenuated, spiraled; S short, tube-like; BFpresent; SA absent; BK absent (Figs. 52-55, 76).Type.-Male holotype from Crom Cave, UvaldeCounty, Texas (Summer 1983; R. Waters), depositedin CAS.Etymology.- The speci fie name refers to theshort stylus characteristic of this species.Description.-Total body length, 1.56-2.18.Scute length, 1.18-1.44. Leg II length, 3.54-4.03.Leg II/Scute length, 2.58-3.17. (N = 11).Color orange to brownish orange. Body ofmedium rugosity; tubercles on eye mound, pars thoracica,tergite margins, and coxae. Carapace with 8pairs of AT arranged in two rows. Eye moundbroadly conical, eyes well developed, cervicalgroove prominent. Palpal megaspines: trochanter,1 ventral; femur, 1 mesoapical; patella, 2 mesal;tibia, 3 mesal. Tarsal count: 3-5-4-5.Male (holotype): Total body length, 1.85. Scutelength, 1.38; width, 1.44. Eye mound length, 0.41;width, 0.41. Leg II length, 3.56. TrIV spur robust,apically bent; length, 0.51. POP length, 0.26.Penis (Figs. 52-55, 76): VPP apically converging;with 2 dorsal, 10 lateral, and 4 ventral setae;AS short, curved. Glans: BK absent; ML present,of moderate sire; PSL elongate, tapering apically,spiraled. S short, tubular; BF present, reduced; SAabsent.Female (paratopotype): Total body length, 1.64.Scute length, 1.28; width, 1.33. Eye mound length,0.32; width, 0.31. Leg II length, 3.54. TrIV withtwo ventral tubercles.Ovipositor (Figs. 56, 58, 59): cuticle intricatelyfolded; dorsal surface sparsely set with microserrations,ventral smooth; 1 pair of small apical teethpresent; setal pattern: 1 pair dorsal, 4 pairs lateral, 1pair ventral.Variation.-The cervical groove is more prominentin specimens from Crom Cave. One male(SEM) has a short TrIV spur (L = 0.35) and astrongly reduced POP (L = 0.05) (Figs. 50, 51); inthe remaining males (6) these structures are uniformlylarge (about 0.50 and 0.30 respectively).The aberrant male is distinctly smaller than theothers (scute length = 1.18 compared to 1.31-1.44for the others). One male from BFS Cave (SEM)lacks an apical spine (Figs. 52-55) which, like thetip of one prong, may be broken. The ovipositor ofthe female from BFS Cave (SEM) lacks a pair ofapical teeth (Figs. 56, 58) which are present,although small, in all three females from CromCave.Note.-The PSL are spiral-shaped and enclosethe S in the unexpanded and partially expandedglans. With more complete expansion the PSL separateand become somewhat unspiraled (as in Fig.53).Natural History.-One female (Crom Cave) haswhat appears to be two large eggs (one visiblethrough the somewhat transparent sternites and thesecond revealed through dissection). The presumedeggs are quite large (0.36 x 0.46), much larger thanthe genital opening (0.18 x 0.23), and occupy muchof the abdominal cavity. We have seen specimens ofanother laniatorid harvestman, Hoplobunus species(Gonyleptoidea), from a nearby cave, Frio QueenCave.Other Material Examined (Paratypes).- UNITED STATES: Texas: Uvalde Co.: Crom Cave,Summer 1983 (R. Waters, TMM, CDU), 3 females;BFS Cave, 8 June 1989 (A. Grubbs, "A.c.", and R. Waters, TMM, CAS, CDU), 6 males, 1female.Distribution.-Known only from Crom and BFScaves, Uvalde County, Texas (Maps 2-4).187

Figs. 48-51.-Texella brevistyla, new species, male paratype from BFS Cave: 48, anterior half of body, lateral view; 49, palpi,lateral view; 50, subventral view of body showing genital operculum and reduced postopercular process (indicated by arrow); 51,trochanter IV spur, lateral view.188

Figs. 52-55.-Texella brevistyla, new species, male paratype from BFS Cave: 52-53, penis, lateral view; 54, penis, apical view;55, penis, dorsal view.189

Figs. 56-59.-Texella brevistyla, new species, female paratype from BFS Cave: 56, ovipositor, lateral view; 57, trochanter IV,lateral view; 58, ovipositor, apical view; 59, ovipositor, enlarged view of lateral surface.190

Texellajungi, new speciesFigs. 60-75.Diagnosis.-Males of this species are distinguishedfrom other Texella by the following combinationof characters: VPP apically divergent; PSLclaw-like; S shorter than PSL, apically tubular; SAabsent (Figs. 66-69, 74, 75).Type.-Male holotype from 7.1 mi E Campwood,Real County, Texas (23 Sep. 1971; A. lung)deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr Albert K. S. lung, the sole collector ofthis species.Description.-Total body length, 1.54-1.92.Scute length, 1.03-1.36. Leg II length, 2.47-3.13.Leg II/Scute length, 2.30-2.66. (N = 6).Color yellowish orange. Body finely rugose; tuberclessparse, most pronounced on dorsum fromeye mound to scute posterior and intercoxal regions,inconspicuous on tergite and sternite margins. Eyemound broadly conical, eyes well developed. Palpalmegaspines: trochanter, 2 ventral; femur, 1 mesoapical;patella, 2 mesal; tibia, 3 mesal. Tarsalcount: 3-5-4-5.Male (holotype): Total body length, 1.92. Scutelength, 1.36; width, 1.23. Eye mound length, 0.37;width, 0.33. Leg II length, 3.13. TrIV spurrobust, apically enlarged, bent; length, 0.51. POPlength, 0.26. Carapace with 5-6 pairs of AT, arrangedin two rows.Penis (Figs. 63, 64, 66-69, 74, 75): VPP apicallydiverging; with 2 dorsal, 13 lateral, 3 ventral setae;AS moderate sized, elbowed, apically pointed.Glans: BK conical; ML inconspicuous; PSLclaw-like. S short, tubular; BF well developed; SAabsent.Female (paratopotype): Total body length, 1.59.Scute length, 1.03; width, 1.03. Eye mound length,0.23; width, 0.26. Leg II length, 2.47. TrIV withone ventral tubercle. Carapace with 2-3 pairs ofAT.Ovipositor (Figs. 70-73): cuticle intricately folded;surface smooth, lacking microspinations; 1 pairapical teeth present; setal pattern: 1 pair ventral, 4pairs lateral, 2 pairs dorsal.Variation.-The two male paratypes are smallerthan the holotype (scute lengths 1.08 and 1.05) andhave shorter TrIV spurs (0.33 each) and POP (0.13and 0.08).Note.-This is the only species which has asexually dimorphic number of anterior tubercles(Figs. 60-62).Natural History.-This epigean species has beencollected in oak-juniper woodland.Other Material Examined (Paratypes).UNITED STATES: Texas: Real Co.: 7.1 mi ECampwood, 23 Sep. 1971, in oak-juniper woodland(A. lung, CAS), 2 males, 3 females.Distribution.-Known only from Real County,Texas (Maps 2-4).The mulaiki SubgroupDiagnosis.-Males of this SUbgroup are distinguishedfrom other Texella by the following combinationof characters: PSL subrectangular in shape(except in T. bilobara), somewhat folded, shorterFigs. 60-61.-Texellajl4ngi, new species, paratopotypes; 60, male, anterior half of body, lateral view;body, lateral view.61, female, anterior half of191

Figs. 62-65.-Texella jungi, new species, male paratopotype: 62, anterior part of scute, lateral view, showing anterior tubercles;63, apical spine of penis, dorsal view; 64, subventral view of body, showing postopercular process and genital operculum with apicaltubercles; 65, trochanter IV spur, lateral view.192

Figs. 66-69.-Texellajungi, new species, male paratopotype: 66, penis, ventral view; 67, penis, lateral view; 68, penis, apicalview; 69, penis, dorsolateral view ofglans.193

Figs.70-73.-Texellajungi, new species, female paratopotype: 70, ovipositor, lateral view; 71, genital operculum, ventral view,showing apical tubercles; 72, ovipositor, enlarged view of lateral surface; 73, ovipositor, apical view.194

than S; SA with laterobasal carina (except in T. welbourniand bilobata) and tooth (except in T.hardeni); apical portion of S tubular or compressed.Range.-Known from southern New Mexico tocentral Texas (Maps 2-4).The longistyla InfragroupDiagnosis.-Males of this infragroup are distinguishedfrom other Texella by the following combinationof characters: TC = 3-5-5-5; PTb with 2mesal megaspines; S compressed; SA prongs welldeveloped, parallel sided, blunt (Figs. 77-84).Range.-Known from southern New Mexico andadjacent Texas (Maps 2-4).Texella longistyla, new speciesFigs. 77-79.Diagnosis.-Males of this species are distinguishedfrom other Texella by the following combinationof characters: PTb with two mesal megaspines;S long, compressed; SA with lateral carinaand laterobasal prongs (Figs. 77-79).Type.-Male holotype from Musk Ox Cave,Carlsbad Caverns National Park, Eddy County,New Mexico (7 Aug. 1976; W. Weibourn), depositedin CAS.Etymology.-The specific name refers to thelong stylus characteristic of this species.Description.-(cavernicole)-Total body length,1.51-1.59. Scute length, 1.05-1.08. Leg II length,3.64-3.92. Leg II/Scute length, 3.47-3.63. (N =4) (epigean)-Total body length, 1.67. Scutelength, 1.05. Leg II length, 2.95. Leg II1Scutelength, 2.81. (N = 1).Color yellowish white (yellowish orange in epigeanspecimen). Body coarsely rugose; tuberclesmoderate on eye mound (dense in epigean specimen),sparse on pars thoracica, moderate on tergitemargins and coxae. Carapace with 2 or 3 pairs ofAT. Eye mound broadly conical, eyes well developed.Palpal megaspines: trochanter, 1 ventral;femur, 1 mesoapical; patella, 2 mesal; tibia, 2 mesal.Tarsal count: 3-5-5-5.Male (holotype): Total body length, 1.59. Scutelength, 1.10; width, 1.08. Eye mound length, 0.28;width, 0.29. Leg II length, 3.82. TrIV spurlength, 0.36. POP length, 0.10.Penis (Figs. 77-79): VPP with 2 dorsal, 11lateral, and 4 ventral setae; AS stout, pointed,geniculate. Glans: BK triangular, narrow; MLrectangular, wide; PSL quadrate, folded. S long,compressed; BF present; SA with basolateral prong(long, rounded apically) and mediolateral canna(evenly rounded).Female (paratype, Doc Brito Cave): Total bodylength, 1.51. Scute length, 1.08; width, 1.10. Eyemound length, 0.26; width, 0.31. Leg II length,3.92. TrIV with one ventral tubercle.Ovipositor: cuticle intricately folded, surface apparentlysmooth; 1 pair of apical teeth present; setalpattern: 2 pairs dorsal, 4 lateral, 1 ventral.Variation.-The paratype male from Musk OxCave, although subequal in size to the holotype, haslonger TriV spurs (0.49) and POP (0.18).Natural History.-The specimen from Texaswas collected beneath a large decaying prickly pearcactus, under conditions of high moisture, in anoak-juniper forest adjacent to a mesic riparian woodland.Other Material Examined (Paratypes).- UNITED STATES: Texas: Culberson Co.: GuadalupeMountains National Park, McKittrick Canyon, 7Sep. 1989 (D. Ubick, CDU), 1 male. New Mexico:Eddy Co.: Carlsbad Caverns National Park, MuskOx Cave, 5200', 27 Mar. 1976 (W. Welbourn,TMM), 1 male, 1 female; Doc Brito Cave, 3500',25 May 1975 (W. Welbourn, MLG), 1 female.Distribution.-Known only from CulbersonCounty, Texas, and Eddy County, New Mexico(Maps 2-4).Texella welbourni, new speciesFigs. 80-84.Diagnosis.-The single male representing thisspecies is distinguished from other Texella by thefollowing combination of characters: eyes absent(lacking both retina and cornea); TC = 3-5-5-5; SAwith a pair of long laterobasal prongs, lackingcarina; S long, compressed (Figs. 80-84).Type.-Male holotype from Jurnigan Cave(3500'), Eddy County, New Mexico (16 Feb. 1974;W. Welbourn), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr W. Calvin Welbourn, collector of thisand other species of Texella.Description.-Male (holotype): Total bodylength, 1.69. Scute length, 1.23; width, 1.10. Eyemound length, 0.31; width, 0.31. Leg II length,5.62. Leg II1Scute length, 4.57. TrIV spur length,0.08. POP length, 0.05.Color yellowish white. Body moderately rugose;tubercles sparse on eye mound, pars thoracica,tergite margins, and coxae. Carapace with 5 pairsof AT. Eye mound broadly conical, eyes absent(lacking retina and cornea). Palpal megaspines:195

trochanter, 1 ventral; femur, 1 mesoapical; patella,2 mesal; tibia, 2 mesal. Tarsal count: 3-5-5-5.Penis (Figs. 80-84): VPP apically rounded, with2 dorsal, 17 lateral, and 4 ventral setae; AS stout,curved, pointed. Glans: BK present; ML present;PSL subquadrate, intricately folded. S long, compressed;BF present; SA represented by a pair oflaterobasal prongs (long, parallel-sided, apicallyrounded), lacking lateral carinae.Female: Unknown.Other Material Examined.-None.Distribution.-Known only from Jurnigan Cave,Eddy County, New Mexico (Maps 2-4).The mulaiki InfragroupDiagnosis.-Members of this infragroup aredistinguished from other Texella by the following,7475 \~77Figs. 74-79.-Texella species, male genitalia, paratopotypes: 74-75, T. jungi, new species: 74, penis, lateral view, with fullyexpanded glans; 75, penis, ventral surface of glans. 76, T. brevisryla, new species: Penis, sublateral view. 77-79, T. longisryla,new species: 77, penis, lateral view; 78, penis, ventral surface ofglans; 79, penis, ventral view. Scale bar = 0.25 mm.196

~81//J/'/ ;'?/ ~~.........--~~I, IIIIIIIIIIII1IIII\\, ,, , , ,II,I\ , \ , , ,8384Figs. 80-84.-Texella we/boumi, new species, male holotype: 80-81, penis, lateral view, unexpanded; 82, penis, ventral view; 83,penis, lateral view, with partially expanded glans; 84, penis, dorsal view, with partially expanded glans. Scale bar = 0.50 mm forFig. 80 and 0.25 mm for the others.197

primarily troglomorphic, characters: eyes absent;0-1 pairs of AT; PTb with 2 mesal megaspines;LIIISL ~ 9.4; TC ~ 3-6-4-5 or 4-5-5-5. Males arefurther distinguished by the following genitaliccharacters: PSL subrectangular, folded; SA withlaterobasal carina bearing a short prong or tubercle;apical half of S a straight tube (Figs. 85-88, 93-96,102, 105-108).Species.-The three species show increasingtroglomorphism from T. hardelli to T. cokelldolpherito T. mulaiki. This cline is also reflected intwo genitalic characters associated with the SA. InT. hardeni, the SA has the largest carina and shortestteeth; these states are reversed in T. mulaiki andintermediate in T. cokelldolpheri. As pronged SAare found in virtually all species of the subgroup,and carinae absent in many, the loss of the prongs(Figs. 87, 88) seems to be derived.Range.-Known only from central Texas (Maps3,4).Texella hardeni, new speciesFigs. 85-88.Diagnosis.-This strongly troglomorphic speciesis most closely related to T. mulaiki and T. cokelldolpherifrom which it may be distinguished by itssomewhat lower TC (leg I with only three tarsomeres)and shorter legs (LII/SL = 9.9-11.6) andby its glans characters: PSL with dorsoapical spine,SA prong reduced to a short knob (Figs. 87, 88).Type.-Male holotype from Haby SalamanderCave, Bandera County, Texas (31 Oct. 1984; S.Harden and J. Ivy), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr S. J. Harden, one of the collectors ofthe holotype.Description.-Total body length, 1.23-1.67.Scute length, 0.95-1.18. Leg II length, 11.0-11.9.Leg II1Scute length, 9.9-11.6. (N = 3).Color pale orange, appendages lighter. Bodyfinely rugose; with several pointed tubercles anteriorlyon eye mound; with few tubercles on parsthoracica; with tergite margins smooth; with coxa Ibearing row of setate tubercles, coxae 2-4 smooth.Carapace with 1 pair of small AT. Eye mound anarrowed cone, somewhat cylindrical, retina andcornea absent. Palpal megaspines: trochanter, twoventral; femur, one (female) or two (male) mesoapical;patella, two mesal; tibia, two mesal. Tarsalcount: 3-6 to 8-4-5.Male (holotype): Total body length, 1.67. Scutelength, 1.18; width, 1.08. Eye mound length, 0.26;width,0.26. Leg II length, 11.72. TrIV with vestigialspur, length, 0.02. POP absent.Penis (Figs. 85-88): VPP with 2 dorsal, 16 lateral,and 6 ventral setae; lateral setae short, dorsaland ventral setae long; AS short, apically entire.Glans: BK conical; ML not evident; PSL subquadrate,folded, with dorsoapical prong. S with apicalhalf straight, lacking ornamentation; BF present; SAwith a broad lateral expansion and an apical knob.Female (paratopotype): Total body length, 1.46.Scute length, 1.13; width, 1.03. Eye mound length,0.23; width, 0.21. Leg II length, 11.92. TrIVlacking ventral tubercle.Ovipositor lacking apical teeth.Natural History.-AII populations occur sympatricallywith Hoplobunus species (Gonyleptoidea).Other Material Examined (Paratypes).- UNITED STATES: Texas: Bandera Co.: Haby SalamanderCave, 31 Oct. 1984 (S. Harden and J. Ivy,TMM), 1 female; 20 July 1986 (S. Harden, TMM),1 female; Station "COO Cave, 4 Sep. 1988 (P.Sprouse, TMM), 1 juvenile. KelT Co.: StowersCave, 20 Mar. 1965 (1. Reddell, TMM), 1 juvenile.Distribution.-Known definitely only from HabySalamander Cave, Bandera County, Texas (Maps2-4).Note.-The juvenile specimens from Station "C"Cave and Stowers Cave are assigned to this speciesonly tentatively.Texella cokendolpheri, new speciesFigs. 89-100.Diagnosis.-This strongly troglomorphic speciesis most closely related to T. mulaiki and T. hardellifrom which it may be distinguished by its glanscharacters: PSL lacking dorsoapical spine and SAwith lateral prong and apically serrated carina (Figs.93-96).Type.-Male holotype from Robber Baron Cave,San Antonio, Bexar County, Texas (3 Apr. 1982;A. Grubbs), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr James C. Cokendolpher, who first recognizedthis species as new.Description.-Total body length, 1.28-1.67.Scute length, 1.05-1.28. Leg II length, 13.1-13.6.Leg II1Scute length, 11.1-12.9. (N = 5).Color pale orange. Body finely rugose; withseveral pointed tubercles anteriorly on eye mound;with few tubercles on pars thoracica; with tergitemargins smooth; with coxa I bearing row of setatetubercles, coxae 2-4 smooth. Carapace with 1 pairof AT. Eye mound a narrowed cone, somewhat198

cylindrical, retina and cornea absent. Palpal megaspines:trochanter, one ventral; femur, one mesoapical;patella, one mesal; tibia, two mesal. Tarsalcount: 4-6 to 8-5-5.Male (holotype): Total body length, 1.28. Scutelength, 1.05; width, 0.92. Eye mound length, 0.20;width, 0.20. Leg II length, 13.4. TrIV spur vestigial,length, 0.02. POP length, 0.06.Penis (Figs. 93-96): VPP with medium-sized setaelaterally, long setae ventrally and dorsally; with2 dorsal, 13 lateral, and 4 ventral setae. Glans:BK small; ML small; PSL ventrally notched. S withapical half straight, tube-like; BF well developed;SA with lateral tooth and serrate carina.Female (paratopotype): Total body length, 1.33.Scute length, 1.18; width, 1.05. Eye mound length,0.18; width, 0.20. Leg II length, 13.1. TrIVlacking ventral tubercle.Ovipositor (Figs. 98, 100): cuticle intricatelyfolded, lacking microspines or tubercles; apical teethabsent.Other Material Examined (Paratypes).- UNITED STATES: Texas: Bexar Co.: Robber BaronCave, 10 Mar. 1982 (A. Grubbs, B. Steele, and R.Waters, TMM, CAS), 2 females; 6 Apr. 1983 (R.Waters, CAS), 1 male; Jan 1984 (R. Waters,TMM), 1 female; John Wagner Ranch Cave No.3,25 Jan. 1985 (S. Harden, TMM), 1 juvenile.Distribution.-Known from cave(s) in BexarCounty, Texas (Maps 2-4).Note.-The juvenile specimen from John WagnerRanch Cave No.3 is assigned to this species onlytentatively.Texella mulaiki Goodnight and GoodnightFigs. 3,4, 101-112.Texella mulaiki Goodnight and Goodnight, 1942: 10(female holotype from Hays County, Texas, inAMNH, examined). Goodnight and Goodnight,1967:6. Reddell, 1965: 177. Mitchell and Reddell,1971:46. Davis, 1979:34.Diagnosis.-This strongly troglomorphic speciesis most closely related to T. cokendolpheri and T.hardeni from which it may be distinguished by itsglans characters: PSL lacking dorsoapical spine andII I\ I....... _J185Figs. 85-88.-Tex-ella hardeni, new species, male holotype: 85-86, penis, lateral view; 87, penis, dorsal view; 88, glans, ventralsurface. Scale bar = 0.50 mm for Fig. 85 and 0.25 mm for the others.199

Figs. 89-92.-Texella cokendolpheri, new species, male paratopotype: 89, anterior half of body, lateral view; 90, eye mound,lateral view; 91, trochanter IV, lateral view, showing reduced spur; 92, genital region, lateral view, showing operculum and reducedpostopercular process.200

Figs. 93-96.-Texella cokendolpheri, new species, male paratopotype: 93, penis, ventrolateral view; 94, penis, dorsolateral view;95, penis, lateral view; 96, penis, dorsal view.201

Figs. 97-100.-Texella cokendolpheri, new species, female paratopotype: 97, palpi and chelicerae, lateral view; 98, ovipositor,apical view: 99, trochanter IV, lateral view: 100, ovipositor, sublateral view.202

SA with well developed prong and reduced carina(Figs. 105-108).Description.-Total body length, 1.49-2.21.Scute length, 1.31-1.49. Leg II length, 12.5-19.8.Leg II/Scute length, 12.5-15.3. (N = 20).Color of body yellowish orange, appendagesyellowish white. Body finely rugose; tuberclesabsent from eye mound, pars thoracica, tergitemargins, and coxae II-IV; coxa I with row of setatetubercles. Carapace without, or with 1 pair of verysmall, AT. Eye mound conical, reduced; retina andcornea absent. Palpal megaspines: trochanter, oneventral; femur, two mesoapical; patella, two mesal;tibia, two mesal. Tarsal count: 4-5 to 8-5 (or 6)-6(or 5).Male (Fern Cave): Total body length, 1.62.Scute length, 1.41; width, 1.31. Eye mound length,0.18; width, 0.23. Leg II length, 19.7. TrIV spurabsent. POP absent (Fig. 104).Penis (Figs. 102, 105-108): VPP with 2 dorsal,16 lateral, and 4 ventral setae; lateral setaemedium-sized, dorsal and ventral setae long; ASshort, curved, with trifurcate apex. Glans: BK along flap; 'ML broad with truncate apex; PSL arounded lobe lacking ventral notch and dorsal spine.S with apical half straight and thin; BF welldeveloped; SA with lateral prongs and reducedcarina, lacking serrations.Female (Fern Cave): Total body length, 1,69.Scute length, 1.44; width, 1.36. Eye mound length,0.21; width, 0.26. Leg II length, 18.5. TrIVlacking ventral tubercle.Ovipositor (Figs. 101, 109-112): cuticle intricatelyfolded; lacking microspines or tubercles;apical teeth absent.Variation.-The tarsal count is rather variable,especially in the tarsomere number of leg II wherealmost half of the specimens show variation fromthe typical number of -7-: one specimen has -5-,one -5/6- (-5- on the left leg, -6- on the right), one-617-, three -7/8-, and two -8-. In addition, twospecimens have counts of -5/6-, one on leg III theother on leg IV.The LII/SL varies from 12.5 to 15.3. The shortestleg lengths tend to occur in the southernpopulations (Boggus and Ezell's Caves); the longestin the northern population (Flint Ridge Cave).Natural History.-T. mulaiki occurs sympatricallywith T. renkesae and T. diplospina butoccupies relatively deeper portions of the respectivecaves. These harvestmen appear to be relativelyuncommon, at least in Ezell's Cave where a 15month faunal survey turned up only 7 specimens.On the other hand, we found a comparable numberof specimens during a few hours of collecting inFern Cave. That same study also suggests that theyare attracted to baits (6 specimens found at baitedrocks) and seem to prefer cheese (4 specimens)(Davis, 1979).Material Examined.-UNITED STATES:Texas: Hays Co.: [no specific locality], 15 Apr.1939 (S. Mulaik, AMNH), 1 female (type); SanMarcos, Ezell's Cave, 30 Apr. 1978 (J. Davis,WAS), 2 females; 2 July 1978 (1. Davis, WAS), 1female; Boggus Cave,. 16 Jan. 1988 (S. Harden,TMM), 2 females; McCarty Cave, 1988 [no specificdate] (A. Grubbs, TMM), 1 juvenile; McGlothlinSink, 26 May 1989 (A. Grubbs, J. Reddell, and M.Reyes, TMM), 1 juvenile; 3 Sep. 1989 (D. Ubick,S. Fend, and S. Renkes, CDU), 1 male, 3 juveniles;II mi W San Marcos, Fern Cave, 15 July 1988 (A.Grubbs, J. Evans, and L. Schneider, TMM), 1female; 26 May 1989 (A. Grubbs, J. Reddell, andM. Reyes, TMM), 1 female; 2 Sep. 1989 (D.Ubick, S. Fend, S. Renkes, and A. Grubbs, CDU,CAS), 3 males, 3 females; Ladder Cave, lowerlevel, 2 Sep. 1989 (D. Ubick, S. Fend, S. Renkes,J. Reddell, and M. Reyes, CDU), 1 female; 5 mi WKyle, Michaelis Cave, Jan. 1990 (A. Grubbs and L.Graves, TMM), 1 male. Travis Co.: Flint RidgeCave, 8-9 June 1984 (D. Pate, TMM), 1 juvenile;Apr. 1989 (M. Grimm, TMM), 1 male; Apr. 1989(J. Reddell and M. Reyes, TMM), 1 juvenile; 25Nov. 1989 (M. Grimm, AMNH), 1 female, 1juvenile; Slaughter Creek Cave, 24 Feb.!3 Mar.1990 (W. Russell, TMM), 1 male; Whirlpool Cave,8 Jun. 1980 (S. Robertson, TMM), 1 juvenile.Distribution.-Known from the caves of CentralTexas, Hays and Travis Counties (Maps 2-5).Notes.-The sex of the holotype was originallystated as male but, as was subsequently corrected(Goodnight and Goodnight, 1967:7), is in fact afemale. All other specimens previously assigned tothis species actually represent anew, and not veryclosely related, species (T. reyesi).The type locality, although not specificallyindicated on the locality label, is probably Ezell'sCave, a popularly visited cave at the time S. Mulaikcollected the holotype (J. Reddell, personalcommunication). This is supported by the fact thatthe three recently collected female specimens fromEzell's Cave closely resemble the holotype.The present interpretation of T. mulaiki must beconsidered preliminary until male specimens fromEzell's Cave become available and are shown to beconspecific with the known males. Even moretentative is the inclusion of the two populations atthe extreme ends of the distribution, from Whirlpool203

Figs. 101-104.-Texella mulaiki Goodnight and Goodnight, Fern Cave: 101, 103, female: 101, anterior half of body, lateralview; 103, palpi and chelicerae, lateral view. 102,104, male: 102, apical spine of penis, ventral view; 104, subventral view of bodyshowing absence of trochanter IV spur and postopercular process.204

Figs. 105-108.-Texelia mulaiki Goodnight and Goodnight, male from Fern Cave: lOS, penis, ventral view; 106, penis, lateralview; 107, penis, apical view; 108, penis, subdorsal view.205

Figs. 109-112.-Texella mulaiki Goodnight and Goodnight, female from Fern Cave: 109, ovipositor, lateral view; 110, ovipositor,apical view; III, ovipositor, enlarged view of lateral surface; 112, ovipositor, enlarged view of apical region.206

and McCarty Caves, which are known only fromjuveniles.Texella sp.Note.-Two females from Marguerite Caveresemble those from Haby Salamander Cave (T.harden;) and Robber Baron Cave (T. cokendolpheri)but differ in possessing a tubercle ventrally on TrIV.On the basis of this and, especially, their isolation,these specimens probably represent a new species.The positive identification of this population willawait the discovery of a male.Material Examined.-UNITED STATES: Texas:Med;na Co.: Marguerite Cave, 28 Apr. 1984(S. Harden, TMM), 1 female; 5 May 1984 (R.Waters, TMM), 1 female.The bi/obara InfragroupDiagnosis.-Males of the single known speciescomprising this infragroup may be distinguishedfrom other Texella by the following combination ofcharacters: epigean; TC = 3-5-4-5; BK large, bilobed;PSL intricately folded, apically hooked; SAwith lateromedial prong, lacking carina; apical halfof S tubular (Figs. 113-117).Range.-Known only from Kerr Co., Texas(Maps 3,4).Texella bi/obata, new speciesFigs. 113-117.Diagnosis.-As for infragroup.Type.-Male holotype from Raven Ranch, KerrCounty, Texas (Dec. 1939; S. and D. Mulaik),deposited in AMNH.Etymology.-The specific name is a reference tothe bilobed basal knob of the glans, unique to thisspecies.Description.-Total body length, 1.23-1.41.Scute length, 0.84-0.97. Leg II length, 2.41-2.44.Leg II/Scute length, 2.62-2.80. (N = 3).Color brownish orange. Body coarsely rugose;tubercles scattered on eye mound, pars thoracica,tergite margins, and coxae. Carapace with 2 pairsAT. Eye mound broadly conical, retina and corneapresent. Palpal megaspines: trochanter, one ventral;femur, one mesoapical; patella, two mesal;tibia, three mesal (mesoapical small). Tarsal count:3-5-4-5.Male (holotype): Total body length, 1.23. Scutelength, 0.92; width, 0.92. Eye mound length, 0.23;width, 0.23. Leg II length, 2.41. TrIV spurlength, 0.05. POP absent.Penis (Figs. 113-117): VPP with 2 dorsal, 3lateral, and 2? ventral setae; lateral setae short,dorsal and ventral setae medium to long; ASapparently broken, base distinct, large. Glans: BKbifurcate, large; ML broad; PSL complex, withapical hook. S with apical third straight, tube-like;BF present; SA with lateral prong, lacking carina.Female (paratopotype): Total body length, 1.41.Scute length, 0.97; width, 1.03. Eye mound length,0.21; width, 0.26. Leg II missing. TrIV withventral tubercle.Ovipositor damaged; apical teeth present.Note.-The available specimens are old and extremelybrittle; all have somewhat damaged genitalia.In both males the apical spine appears to bebroken and is represented by a clearly visible andquite large basal portion.Other Material Examined (Paratypes).- UNITED STATES: Texas: Kerr Co.: Raven Ranch,Dec. 1939 (S. and D. Mulaik, AMNH), 1 male, 1female.Distribution.-Known only from the typelocality (Maps 2-4).The reddell; SubgroupDiagnosis.-The males of this subgroup aredistinguished from all other Texella by the followingcombination of characters: PSL claw-like (except inT. fend; and T. homi); SA produced into a scroll orprong (except in T. hom;); apical half of S spatulate(except T. fendi and T. homi).Range.-Known only from central Texas (Maps3,4).The reddelli InfragroupDiagnosis.-The males of this infragroup aredistinguished from all other Texella by the followingcombination of characters: S apically spatulate; SAscrolled, clips onto the PSL (Fig. 7).Distribution.-Known only from the caves ofTravis and Williamson Counties, Texas (Maps 3-5).Texella reddelli Goodnight and GoodnightFigs. 118-123.Texella reddelli Goodnight and Goodnight, 1967:7(male holotype from Bee Creek Cave, Austin,Travis County, Texas, in AMNH, examined).207

Diagnosis.-This species is most easily distinguishednumber of setae on the VPP [10 lateral setae in T.\\Ifrom T. reyesi by its somatic characters. T. reddelli (Figs. 120, 121); 12-20 in T. reyesi (Figs.reddelli is only slightly troglomorphic and has well 7, 136-147)]. The two species are clear!y verydeveloped eyes, relatively shorter legs (LII/SL =3.8-5.2), and a higher number of AT (3-5 pairs;closely related and, using the standards of genitalicdistinctness applied to other Texella species, mayFigs. 118, 119). T. reyesi lacks retinae, has longer even be considered conspecific. However, givenlegs (LII/SL = 4.3-8.7), and has a lower number of that the two groups can be distinguished, they areAT (virtually all specimens with 2-0 pairs; Figs. treated here as species. A comparison of most of124, 125, 127).these characters is given in Table 3.The only genitalic differences so far detected are Description.-Total body length, 1.90-2.18.the relative lengths of the AS [shorter in T. reddelli Scute length, 1.21-1.66. Leg II length, 4.92-7.59.(Fig. 123) than in T. reyesi (Figs. 148-151)] and the Leg II/Scute length, 3.81-5.20. (N = 16).Figs. 113-117.-Taella bilobata, new species, male holotype: 113-114, penis, lateral view; 115, penis, ventral view; 116, penis,sublateral view; 117, penis, dorsal view. Scale bar = 0.50 mm for Fig. 113 and 0.25 mm for the others.208

Color orange. Body of medium rugosity, sparselytuberculate. Carapace with 3-5 pairs of AT. Eyemound broadly conical, eyes well developed. Palpalmegaspines: trochanter, 1 ventral; femur, 1mesoapical; patella, 2 mesal; tibia, 3 mesal. Tarsalcount: 3-5-5-6, rarely 3-5-5-5 or 3-6/5-5-5.Male (holotype): Total body length, 1.90. Scutelength, 1.44; width, 1.44. Eye mound length, 0.36;width, 0.38. Leg II length, 6.00. TrIV spur long,apically bent, length, 0.67. POP length, 0.44.Penis (Figs. 120, 121): VPP with 2 dorsal, tolateral, and 3 ventral setae; AS curved, apicallypointed. Glans: BK slender; ML present; PSLclaw-like. S spatulate; BF present; SA scrolled,clips onto PSL.Female (paratopotype): Total body length, 1.97.Scute length, 1.41; width, 1.44. Eye mound length,0.33; width, 0.33. Leg II length, 5.44. TrIV withtwo ventral tubercles.Ovipositor: cuticle intricately folded; 1 pair ofapical teeth present; setal pattern typical.Variation.-The TrIV spur of the seven knownmales varies in length from 0.51-0.77. Thespecimens from Bee Creek Cave and Cave Y have4-5 pairs of AT, those from Bandit Cave and JesterEstates Cave have 3 pairs. All known specimens ofT. reddelli have a TC of 3-5-5-6, except for the twofemales from Cave Y, one having a TC of 3-5-5-5and the other 3-6/5-5-5. The specimens from JesterEstates Cave have slightly longer legs (LIIISL =4.0-5.2) than the others (LIIISL = 3.8-4.3).These last two characters suggest a cline with theleast troglomorphic individuals (Cave Y) in thesouthernmost, and the most troglomorphic (JesterEstates Cave) in the northernmost, part of the range.Natural History.-"In Jester Estates Cave thisspecies was collected on the underside of smallrocks. One specimen was taken in dim twilight atthe bottom of the 8 ft. entrance drop. The remainingspecimens were in darkness." (J. Reddell, personalcommunication)Material Examined.-UNITED STATES: Texas:Travis Co.: Cave Y, 4 June 1990 (J. Reddelland M. Reyes, TMM), 1 female; 14 June 1990 (J.Reddell and M. Reyes, TMM), 1 female; BanditCave, 17 May 1965 (T. Barr, MLG), 1 female; BeeCreek Cave, 2 Oct. 1963 (1. Reddell and D.McKenzie, AMNH, CAS), 2 males (including type),2 females; 4 Oct. 1975 (A. Grubbs, M. Cossey, andT. Byrd, TMM), 1 male, 3 females; 7 June 1965 (J.Fish and J. Reddell, MLG), 1 female; Jester EstatesCave, 4 June 1990 (1. Reddell and M. Reyes,TMM), 1 female; 10 June 1990 (J. Reddell and M.Reyes, TMM, CDU), 4 males, 2 females.Distribution.-Known only from caves 10central Travis County, Texas (Maps 2-5).Figs. 118-119.-Texella redde/li Goodnight and Goodnight, male topotype: 118, anterior half of body, lateral view; 119, enlargedview of previous image showing anterior tubercles.209

Figs. 120-123.-Texella reddelli Goodnight and Goodnight, male topotype: 120, penis, sublateral view; 121, penis, ventral view;122, trochanter IV spur, lateral view; 123, apical spine of penis, dorsal view.210

Notes.-The type locality was erroneouslypublished (Goodnight and Goodnight, 1967) as"Pine Creek Cave. "Males will be necessary to verify the inclusion ofspecimens from Cave Y and Bandit Cave in thisspecies. However, the females resemble those fromBee Creek Cave and Jester Estates Cave in overallappearance and clearly lack the apomorphies of theother two species in the immediate vicinity, T.spinoperca and T. mulaiki.Texella reyesi, new speciesFigs. 7, 124-159.Texella mulaiki Goodnight and Goodnight, 1942.Goodnight and Goodnight, 1967:6 (in part,specimens from Cotterell, Beck Ranch, andMan-With-A-Spear Caves only).Texella reddelli Goodnight and Goodnight, 1967:7(in part, specimens from Tooth, Weldon, andBone Caves only): Mitchell and Reddell, 1971:46.Diagnosis.-See discussion under T. reddelli andTable 3.Type.-Male holotype from Bone Cave, WilliamsonCounty, Texas (4 June 1989; W. Elliott, J.Reddell, M. Reyes), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr. Marcelino Reyes, collector of this andmany other species of Texella.Description.-Total body length, 1.41-2.67.Scute length, 1.26-1.69. Leg II length, 6.10-11.79. Leg II1Scute length, 4.30-8.68. (N = 85).Color pale orange. Body finely rugose; fewsmall tubercles on eye mound, pars thoracica, andcoxae; tergite margins smooth. Carapace with 0-2pairs of anterior tubercles. Eye mound broadlyconical, retina absent, cornea well developed, reduced,or absent. Palpal megaspines: trochanter, 1ventral; femur, 1 mesoapical; patella, 2 mesal; tibia,3 mesal. Tarsal count from 3-5-5-5 to 4-7-6-8.Male (holotype): Total body length, 1.59. Scutelength, 1.28; width, 1.31. Eye mound length, 0.26;width, 0.28. Leg II length, 9.60. TrIV spurstraight; length, 0.56. POP length, 0.26.Penis (Figs. 145-147): VPP rounded apically;with 2 dorsal, 17 lateral, and 4 ventral setae; ASbent, apically pointed, length 0.05. Glans: BK narrowlyconical; ML long; PSL claw shaped. S long,curved, ventrally carinate, apically spatulate; BFwell developed; SA laterally scrolled, clips ontoPSL.(paratype, Tooth Cave): Total body length,2.13. Scute length, 1.46; width, 1.54. Eye moundlength, 0.34; width, 0.38. Leg II length, 6.67.TrIV spur bent; length, 0.59. POP length, 0.36.Penis (Figs. 7, 136, 138, 148): Similar toholotype except for the following: VPP with 12lateral setae; AS slightly shorter (0.04), apicallybifurcate.Female: (paratopotype) Total body length,1.92. Scute length, 1.33; width, 1.33. Eye moundlength, 0.26; width, 0.28. Leg II length, 9.64.TrIV with tiny ventral tubercle.Ovipositor: cuticle intricately folded; surfacesparsely covered with tiny microspines; 1 pair apicalteeth present; setal pattern: 2 pairs dorsal, 4 lateral,1 ventral.Juvenile: Color white to yellowish white;integument apparently unsclerotized. Tarsal countof middle instars 2-2-3-3 and 2-2-3-4; tarsomeres ofearlier instars not discernable. Six immature specimenshave short trochanteral spurs; one of these,presumably a penultimate, also has a short POP.Variation.-This species is extremely polymorphic,most notably in troglomorphic characters. Thecharacters examined are presented in Table 3. Acline is apparent, with the more troglomorphicindividuals occurring at the northern end of thedistribution.The size of sexually dimorphic structures appearsto vary with body size: the TrIV spur (Figs. 128,129) varies from 0.35-0.82; the POP (Figs. 132,133) varies from 0.09-0.51.A female from Tooth Cave (Figs. 152-155) isunique in having a hyperexpanded ovipositor (seediscussion under morphology). A specimen fromCotterell Cave, the second largest female specimen,is aberrant in having both a tiny POP (Figs. 134,135) and a short TrIV spur (Fig. 130). A similar(although smaller) TrIV spur is present on a femalefrom Beck Ranch Cave.Natural History.-"Most specimens of thisspecies have been taken on the underside of rockslightly buried in soil. In all instances they havebeen found only in the more remote parts of thecaves. None have been found in twilight, with theexception of the single juvenile from NewComanche Trail Cave. This specimen was found onthe talus slope below the entrance the day after aheavy rain had saturated the surface of the slope."(J. Reddell, personal communication)Other Material Examined (paratypes).- UNITED STATES: Texas: Travis Co.: 14 mi NW Austin,McDonald Cave, 18 May 1984 (D. Pate, J.Reddell, and M. Reyes, TMM), 1 female; 29 May211

Figs. 124-127.-Texella reyesi, new species, paratypes: 124, female (Cotterell Cave), anterior half of body, lateral view. 125-127,males: 125, anterior half of body, lateral view (Bone Cave); 126, palpi and chelicerae, sublateral view (Bone Cave); 127, eye mound,lateral view showing reduced cornea (Inner Space Caverns).212

Figs. 128-131.-Taella reyesi, new species, paratypes: 128-129, males: 128, trochanter IV, lateral view showing curved spur(Cotterell Cave); 129, trochanter IV, lateral view showing straight spur (Bone Cave). 130-131, females: 130, trochanter IV, lateralview showing aberrant development of short spur (Cotterell Cave); 131, trochanter IV, lateral view showing typical development ofsmall tubercle (Cotterell Cave).213

Figs. 132-135.-Taella reyesi, new species, paratypes: 132-133, males: 132, genital region, ventrolateral view showing welldeveloped postopercular process (Tooth Cave); 133, genital region, ventrolateral view showing reduced postopercular process (BoneCave). 134-135.,Ffmale (Cotterell Cave): 134, genital region, ventrolateral view showing aberrant development of small postopercularprocess; 135, enlarged view of postopercular process, lateral view (same specimen).214

Figs. 136-139.-Texella reyesi, new species, male paratypes: 136, 138, penis showing slightly expanded glans ([ooth Cave): 136,dorsolateral view; 138, lateral view. 137, 139, penis showing more fully expanded glans (Cotterell Cave): 137, sublateral view; 139,ventrolateral view.215

Figs. 140-143.-Taella reyesi, new species, male paratypes: 140, 142, penis showing even more fully expanded glans (BeckRanch Cave): 140, subventral view; 142, ventrolateral view. 141, 143, penis showing completely expanded glans (Inner SpaceCaverns): 141, ventrolateral view; 143, lateral view.216

Figs. 144-147.-Texella reyesi, new species, male paratypes: 144-145, penis, ventrolateral view: 144, partially expanded glans(Off Campus Cave); 145, more fully expanded glans (Bone Cave). 146, penis, dorsal view (Bone Cave); 147, penis, apical view(Bone Cave).217

Figs. 148-151.-Texella reyesi, new species, male paratypes, apical spines of penes: 148, Tooth Cave; 149, Beck Ranch Cave;ISO, Cotterell Cave; 151, Inner Space Cave.218

Figs. 152-155.-Texella reyesi, new species, female paratype (rooth Cave) showing hyperexpanded ovipositor: 152, lateral view;153, ventral view; 154, hyperexpanded segment, lateral view; 155, apical view.219

Figs. 156-159.-Taella reyesi, new species, female paratypes showing typically expanded ovipositors: 156, 158, lateral view:156, Inner Space Caverns; 158, Cotterell Cave. 157, apical view (Inner Space Caverns). 159, enlarged view of lateral surface (ToothCave).220

1989 (W. Elliott, J. Reddell, and M. Reyes, TMM),2 males, 1 female; 12 mi NW Austin, Root Cave,12 July 1984 (1. Reddell and M. Reyes, TMM), 3juveniles; 11 mi NW Austin, Tooth Cave, no date(no collector, WAS), 1 juvenile; 2 Feb. 1963 (J.Reddell and D. McKenzie, AMNH), 4 damagedspecimens; 16 May 1965 (T. Barr, R. Mitchell, andAndrews, MLG, CAS, CDU), 5 males, 5 females;14 May 1966 (J. Reddell, MLG, CAS), 5 damagedspecimens; 19 Aug. 1970 (R. Mitchell, MLG), Ifemale, 1 juvenile; 6 Apr. 1986 (J. Reddell and M.Reyes, TMM), 1 male; 21 Feb. 1988 (J. Reddell,TMM), 1 female; Gallifer Cave, 28 Aug. 1988 (J.Reddell and M. Reyes, TMM), 1 male; NewComanche Trail Cave, 26 Jan. 1989 (J. Reddell andM. Reyes, TMM), 2 juveniles; Cotterell Cave, 11Mar. 1964 (B. Russell, AMNH), 2 females; 18 May1989 (W. Elliott, J. Reddell, and M. Reyes, TMM,CAS, CDU), 6 males, 6 females, 1 juvenile;Mil1wood, Hole in the Road (cave), 27 Oct. 1985(W. El1iott, TMM), 1 juvenile; Cold Cave, 8 Apr.1990 (J. Reddell and M. Reyes, TMM), 1 female, 1juvenile; Beer Bottle Cave, 3 May 1990 (W.Russell, TMM), 1 male, 1 female; McNeil BatCave, 2 Mar. 1986 (J. Reddel1 and M. Reyes,TMM), 2 females; Weldon Cave, 11 June 1990 (J.Reddell and M. Reyes, TMM), I male, I female, 1juvenile; No Rent Cave, 6 June 1990 (J. Reddelland M. Reyes, TMM), 1 female, 1 juvenile; 11 June1990 (J. Reddell and M. Reyes, TMM), 1 juvenile;Fossil Garden Cave, 27 June 1990 (J. Reddell andM. Reyes, TMM), 1 male. Williamson Co.:Lakeline Cave, 21 Jan. 1990 (1. Reddell, TMM), Ifemale; 16 Feb. 1990, (J. Reddell and M. Reyes,CDU), 1 male, 1 female; Beck Sewer Cave, 23 Jan.1965 (R. Mitchell and J. Reddell, MLG), 1 male;Beck Ranch Cave, 23 June 1968 (J. Reddell, MLG),1 male, 3 damaged specimens; 9 Mar. 1988 (J.Reddel1 and M. Reyes, TMM, CAS), 3 males, 1female; 24 Mar. 1989 (J. Reddel1 and M. Reyes,TMM), 1 female; Brown's Cave, 23 Apr. 1989 (W.Elliott, J. Reddell, and M. Reyes, TMM), 2females; Bone Cave, 4 June 1989 (W. Elliott, J.Reddell, and M. Reyes, TMM, CAS), 1 male, 4females, 2 juveniles; 4 Aug. 1963 (J. Reddell,AMNH), 1 male, 2 females; Man-With-A-SpearCave, 24 Aug. 1963 [?] (J. Reddell and B. Russell,AMNH), 3 males, 4 females; 2 Sep. 1990 (D. Allenand W. Elliott, TMM), 1 female; Inner SpaceCaverns (=Laubach Cave), 9 July 1965 (1. Reddell,MLG), 1 male, 1 female; 22 Dec. 1968-5 Jan.1969, (W. El1iott, MLG, CAS, CDU), 4 males, 3females; 6 May 1979 (no collector, WAS), 1juvenile; Steam Cave, 19 May 1985 (J. Reddell andM. Reyes, TMM), 1 female; Off Campus Cave, 31Aug. 1985 (D. McKenzie, TMM), 1 juvenile; 8Apr. 1989 (W. Elliott, J. Reddell, and M. Reyes,TMM, CAS), 2 males, 2 females, 1 juvenile;Sore-ped Cave, 28 Apr. 1990 (B. Larsen, TMM), 1male; 28 Apr. 1990 (P. Sprouse, TMM), 2 females;28 July 1990 (B. Larsen, J. Reddell, and M. Reyes,TMM), 1 male, 2 females.Distribution.-Known from caves in Travis andWilliamson Counties, Texas (Maps 2-5).Note.-The populations from Hole in the Road,Cold Cave, No Rent Cave, Root Cave, NewComanche Trail Cave, and Brown's Cave are knownonly from femaleS or juveniles and are assigned tothis species somewhat tentatively. Beck's Tin CanCave of Goodnight and Goodnight (1967) is actuallyBeck Ranch Cave.The brevidenta InfragroupDiagnosis.-The males of this infragroup may bedistinguished from other Texella by the followingcombination of characters: AS with apical brush;VPP with 4 dorsal setae; SA with short, blunt,laterobasal prong; S apically spatulate.Range.-Known only from central Texas (Maps3,4).Texe/fa brevidenta, new speciesFig. 161.Diagnosis.-Texella brevidenta may be distinguishedfrom the other species in its infragroup bythe following combination of genitalic characters:penis large, basal knob robust, length shorter thanbasal diameter, and apical spine distally broad, withseveral points (Fig. 161).Type.-Male holotype from Honey Creek Cave,Comal County, Texas (15 May 1982; K. Menking,R. Waters, S. Harden), deposited in CAS.Etymology.-The specific name refers to theform of the AS, represented by a pair of short teeth.Description.-Male (holotype): Total bodylength, 1.95. Scute length, 1.36; width, 1.54. Eyemound length, 0.34; width, 0.33. Leg II length,4.44. Leg II/Scute length 3.26. TrIV spur length,0.51. POP length, 0.26.Color pale orange, abdomen dusky posteriorly,legs dusky. Body coarsely rugose; tubercles on eyemound, pars thoracica, tergite margins, and coxae.Carapace with 4 pairs AT. Eye mound a robust,rounded cone; eyes well developed. Palpal megaspines:trochanter, 2 ventral, small; femur, 1221

Table 3.-Partial character table for the reddelli infragroup. Abbreviations: # = number of cave as plotted on Map 5; m = male;f = female; • = specimens prepared for SEM; SL = scute length; LlI/SL = leg III scute length; AT = anterior tubercles; pres) =pair(s); TrSL = male trochanteral spur length (and number of female Trochanter IV tubercles); TrSS = male trochanteral spur shape;str = straight; POPL = postopercular process length; ret = retina; cor = cornea; ASL = apical spine (of penis) length; LS# =number of lateral setae on the ventral plate prong (of penis). Notes: Alternate values for tarsomere and tubercle numbers are indicatedwith the left side given first (eg., "617" = left leg with 6 tarsomeres, right with 7); a"?" = missing appendage. A" ±" = more or less;with regard to the cornea, a ± indicates that the cornea is strongly reduced, often barely discernable.# localitv sex SL UIISL tarsal count AT TrSL TrSS POPL ret cor ASL LS1 Cave Y f 1.21 4.07 3-5-5-5 5 prs 2 tubs + +1.21 4.26 3-6/5-5-5 5 prs 2 tubs + +2 Bandit f 1.36 3.81 3-5-5-6 3/2 prs 2/1 tubs + +3 Bee Creek m 1.56 4.09 3-5-5-6 4 prs 0.74 bent broken + + 10• 1.51 4.04 3-5-5-6 5 prs 0.62 bent 0.41 + + 0.034 101.44 4.17 3-5-5-6 4 prs 0.67 bent 0.44 + +f 1.54 3.92 3-5-5-6 4 prs 2 tubs + +1.41 3.86 3-5-5-6 4 prs 2 tubs + +1.38 3.92 3-5-5-6 4 prs 2 tubs + +4 Jester Estates m 1.66 4.37 3-5-5-6 3 prs 0.77 bent 0.51 + +1.49 4.23 3-5-5-6 3 prs 0.64 bent 0.44 + +1.46 5.20 3-5-5-6 3/4 prs 0.77 bent 0.46 + + 0.04 +81.31 4.11 3-5-5-6 3 prs 0.51 bent 0.23 + +f 1.54 4.03 3-5-5-6 2/3 prs 2 tubs + +1.33 4.77 3-5-5-6 3 prs 1 tub + +1.30 4.87 3-5-5-6 3 Drs 1 tub + +5 Tooth m 1.64 4.44 3-5-5-6 3 prs 0.69 +bent 0.42 - +1.62 4.60 3-6-5-6 1 pr 0.62 +bent 0.36 - +1.54 5.39 3-5-5-6 2 prs 0.62 +bent 0.41 -·+ 0.040 121.54 4.75 3-5-5-6 ?1 pr 0.64 +bent 0.41 - +1.46 4.57 3-6/5-5-6 2prs 0.59 +str 0.36 - +1.41 5.27 3-5-5-6/7 1 pr 0.51 +.tr 0.29 - +f 1.54 4.30 3-5-5-6 1 pr 1 tub - +1.51 4.36 3-5-5-6 1/2 prs 1 tub - +• 1.46 4.99 3-5-5-6 2 prs 1 tub - +1.44 4.92 3-5-5-6 2 prs I tub - +1.36 5.15 3-6-5-6 2 prs 1 tub - +1.36 4.48 3-6-5-7 2 Drs 1 tub - +6 Gallifer m 1.46 5.07 3-6/5-5-6/7 2 prs 0.51 ±bent 0.36 - +7 McDonald m 1.46 5.25 3/2-5-5-6 1/2prs 0.54 +str 0.33 - +1.41 5.20 3-5-5-6 I pr 0.51 +str 0.28 - +f 1.49 4.66 3-5-5-6 1 pr 1 tub - +1.41 5.21 3-5/6-5-5/6 I pr I tub - +8 Cotterell m 1.69 5.15 4-5-5-6 2 prs 0.82 bent 0.51 +1.59 5.97 4-6/7-5-6 2 Drs 0.74 bent 0.51 - +1.56 5.31 4-6/7-5-6 2 prs 0.72 bent 0.46 - +1.51 7.13 4-7/6-5-6 3 prs 0.62 bent 0.33 -·+ 0.040 131.44 6.60 4-6-5-6/7 2 prs 0.46 ±bent 0.31 - +1.38 6.81 4-7-5-6 2 prs 0.49 +bent 0.26 - +f 1.54 6.43 4-6/7-5-6/7 2 prs 1 tub - +·1.51 6.20 4-7-5-7 2 Drs 0.08 0.09 - +1.41 6.45 4-7-5-7 2 prs I tub - +1.41 6.43 4-7-5-6/7 2 prs I tub - +1.38 7.17 4-7/6-5-6 1 Dr 1 tub - +1.36 7.13 4-6-5-6/7 2 prs 0.06 - +1.33 7.37 4-6-5/6-7 2 prs 1 tub - +1.31 6.03 46/7 5 6 2 prs 1 tub - +9 Cold f 1.46 6.02 4-6-5-6 2 Drs 1 tub - +10 Beer Bottle m 1.69 6.04 4-6-5/?-7/6 2 prs 0.76 bent 0.47 - +f 1.31 6.36 4-?16-5-6 2 Drs 1 tub - +222

Table 3.-Continued.# locality sex SL ill/SL tarsal count AT TrSL TrSS POPL ret cor ASL LS11 McNeil Bat f 1.62 5.86 4-516-5-617 2 Drs I tub - +1.41 6.13 4-6-5-6 2 Drs I tub - +12 Weldon m 1.46 6.85 4-?17-5-617 2 DrB 0.59 +bent 0.37 - +f 1.38 6.76 4-6-5-7 1 pr o tub - +13 No Rent f 1.36 6.81 4-7-5-617 1 Dr otub - +14 Fossil Garden m 1.49 6.76 3-6-5-7/6 oDr 0.55 +bent 0.31 - +15 Beck Sewer m 1.36 6.82 314-5-?-6 2 Drs 0.49 benl 0.10 - +16 Beck Ranch m 1.51 5.86 4-6-5-6 2 Drs 0.62 benl 0.38 - +* 1.46 6.08 4/3-615-5-6 3-4 Drs 0.59 benl 0.31 - + 0.050 121.36 6.54 4/3-5-5-6 2prB 0.56 bent 0.36 - +f* 1.41 6.36 4-5-5-6 2 DrB 1 tub 12 - +1.36 6.43 4/2-5/4-5-6 2 DrB otub - +17 Lakeline m 1.41 7.51 3-6-5-5 1 Dr 0.46 str 0.21 - + 0.050 12f 1.36 7.11 3-5-5-6 1 pr 1 tub sm - +1.36 6.73 3-5-5-6 1 Dr otub - -18 Brown's f 1.28 7.66 4-6-5-6 1 pr 1 tub sm - -1.28 7.58 4-6-5-7 1 Dr otub - -10 Bone m 1.41 7.20 4-7/6-5/?-7 1 Dr 0.72 str 0.38 - -1.28 7.50 3-6-5/6-7/8 1 pr 0.56 sir 0.26 - +* 1.28 7.13 4/?-6/8-5/?-7 opr 0.39 str 0.11 - + 0.050 17f 1.44 7.48 4-817-6-617 1 Dr otub - -1.42 7.02 4-6-6-8 opr otub - -1.33 7.25 4-6-6/5-7 opr I tub sm - -* 1.31 7.21 4-6-5-7 1 pr otub - +1.31 7.18 4-6-5-7/6 oDr otub - -1.28 6.89 4-6-5-7 o pr 1 tub sm - -20 ManWSpear m 1.36 7.94 4-6-5-6 1 pr 0.36 str 0.10 - + 0.050 121.33 7.38 4-617-5-6 2prs 0.46 str 0.12 - +1.33 - 4-?-5-6 1 Dr 0.35 sir 0.15 - + 12f 1.38 7.02 4-6-5-7 2prs otub - -1.38 7.01 4-617-5-6 2 Drs otub - +1.33 7.02 4-6-5-7 1 pr otub - ±1.31 7.28 4-617-5-6 1 pr 1 tub - +1.26 8.10 4-617-5-5 2prs 1 tub - ±21 Inner SDace m 1.54 7.08 4-7-5-7/8 1 Dr 0.66 +8tr 0.36 - +1.33 8.68 4-7/8-5-7/8 1 pr 0.54 ±8tr 0.18 - -1.33 7.89 4-7-5-7/8 opr 0.64 ±Str 0.28 - ±1.33 7.74 4-6-5-7 1 pr 0.46 sir 0.21 - +* 1.31 7.28 4-617-5-7/8 2 prs 0.48 sir 0.09 - - 0.060 17f 1.54 7.29 415-6-?-7 I Dr - +1.54 6.34 4-6-5-7 o pr 1 tub - +1.46 7.12 4-7-5/6-? o pr o tub - -* 1.46 - 4-?-5-7 o pr o tub - +22 Steam f 1.44 6.81 4-5-6-7 1 pr otub - +23 Off Campus m* 1.49 7.91 4-517-6-7/8 1-2 pr 0.64 +benl 0.33 - - 0.060 201.38 7.47 4-7/6-6-8 1-2 Dr 0.56 +benl 0.28 - -f 1.36 7.72 4-6-5-7 1 pr 1 tub sm - -* 1.33 7.17 3-6-5-7 1 Dr 1 tub sm - -24 Sore-ped m 1.38 7.45 4-?/6-6/5-7/8 o pr 0.49 ±Str 0.15 - -1.36 7.15 3-6-5-716 opr 0.59 ±8tr 0.29 - -f 1.41 6.82 3-6-5-617 110 Dr 1 tub - -1.33 7.29 3-6-5-7/6 opr 1 tub - -223

24reyesi75,6_1----109-"7""---8312redde11io~seinoeercamu1aikidip1osoina-- -r•}--+'----renkes iMap 5.-Map of the Balcones Escarpment, central Texas, showing caves where Texella species have been collected. Degree oftroglomorphy indicated by symbols: open circle = troglophiles (eyes well developed); black dot = troglobites with partial to completeloss of cornea; mixed symbol = troglobites with well developed corneas. The numbers show the caves from which adults of thereddelli infragroup are known (see Table 3). (Although not indicated on the map, T. mulaiki also occurs sympatrically with T.diplospina and T. renkesi.)224

mesoapical; patella, 2 mesal; tibia, 3 mesal. Tarsalcount: 3-5-5-6.Penis (Fig. 161): VPP apically rounded, somewhatexpanded along mid-dorsal margin; with 4dorsal, 22? lateral, and 7 ventral setae; AS mediumsized, apically broad, brushy. Glans: BK robust,length shorter than basal diameter; ML broad; PSLclaw-like. S straight, apically spatulate, with ventralcarina; BF well developed; SA represented by apair of short lateral prongs.Female: Unknown.Natural History.-This species occurs sympatricallywith another laniatorid, Hoplobunus species(Gonyleptoidea). The collection data reads: "Mosquitonet placed in stream".Other Material Examined.-None.Distribution.-Known only from Honey CreekCave, Comal County, Texas (Maps 2-4).Texella grubbsi, new speciesFigs. 160, 162-165.Diagnosis.-Males of Texella grubbsi may bedistinguished from all other Texella by the large,robust basal knob distinctly longer than the basaldiameter (Fig. 160).Type.-Male holotype from Burnett Ranch Cave,7 mi W Wimberly, Hays County, Texas ("1982";A. Grubbs), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr. Andy G. Grubbs, collector of this andnumerous other species of Texella.Description.-Total body length, 1.62-1.82.Scute length, 1.21-1.31. Leg II length, 4.5'1-4.67.Leg II/Scute length, 3.56-3.73. (N = 4).Color brownish orange. Body coarsely rugose;tubercles on eye mound, pars thoracica, and coxae;tergite margins smooth. Carapace with 2 pairs AT.Eye mound a robust, rounded cone; eyes well developed.Palpal megaspines: trochanter, 2 ventral,small; femur, 2 mesoapical; patella, 2 mesal; tibia,3 mesal. Tarsal count: 3-5-5-5.Male (holotype): Total body length, 1.82. Scutelength, 1.31; width, 1.38. Eye mound length, 0.36;width, 0.36. Leg II length, 4.67. TrIV spur large,apically bent, length, 0.59. POP length, 0.26.Penis (Figs. 160, 162-165): VPP apically rounded;with 4 dorsal, 18 lateral, and 5 ventral setae;setae long on apical half of prong; AS of moderatelength, slightly curved, with apical brush. Glans:BK large, conical; ML broad; PSL claw-like. Slong, slightly bent, ventrally carinate, apicallyspatulate; BF present; SA represented by two, short,basolateral prongs.Female (paratopotype): Total body length, 1.62.Scute length, 1.23; width, 1.33. Eye mound length,0.31; width, 0.31. Leg II length, 4.51. TrIV withtwo ventral tubercles.Ovipositor: cuticle intricately folded; dorsal surfacelacking apparent microspines; 1 pair apicalteeth present; setal pattern: 1 pair dorsal, 4 lateral,1 ventral.Variation.-The TrIV spur varies in length from0.46-0.59, and the POP from 0.23-0.26, in the threeavailable males.Other Material Examined (Paratypes).- UNITED STATES: Texas: Hays Co.: Burnett RanchCave, 7 mi W Wimberly, "1982" (A. Grubbs,TMM, CAS), 2 males, 1 female.Distribution.-Known only from Burnett RanchCave, Hays County, Texas (Maps 2-4).The spinoperca InfragroupDiagnosis.-The females of this infragroup aredistinguished from all other Texella, except T. shoshoneand jungi, by the presence of 1-2 pairs ofspines or tubercles on the anterior margin of thegenital operculum (Figs. 171, 173, 182, 183, 194,197). The males are distinguished from all otherTexella by the following combination of characters:VPP with dorsal margin expanded; VPP setae long;AS apically polyfurcate to plumose; BK absent; SAapically produced into long prongs (except in T.homi); S apically spatulate (except T. fendi andhomi) (Figs. 168, 169-170, 174-181, 186-191,198-201).Range.-Known only from eastern central Texas(Maps 3,4).Texella diplospina, new speciesFigs. 166-171.Diagnosis.-Males of this species are distinguishedfrom others in the infragroup by the followingcombination of characters: GO lacking apicaltubercles; VPP with three dorsal setae; SA prongsoriginating medioventrally on stylus; S narrowlyspatulate (Figs. 168-170). Females differ fromthose of other Texella in having 2 pairs of apicalspines on the GO (Fig. 171).Type.-Male holotype from Ladder Cave, HaysCounty, Texas (2 Sep. 1989; D. Ubick, S. Fend, S.Renkes), deposited in CAS.Etymology.-The specific name refers to the 2pairs of apical spines on the female genital opercula.Description.-Total body length, 1.41-1.74.225

161~--)/// ////IJJIIJIIJIIJIIIIIIIIIIIII\\\\\"-\\,'- /....... __ // ////Figs. 160-161.-Taella species, male holotypes: 160, T. grnbbsi, new species: Penis, dorsolateral view. 161, T. brevidenta,new species: Penis, dorsolateral view. Scale bar = 0.25 mm.226

Figs. 162-165.-Texella gmbbsi, new species, male paratopotype: 162, penis, ventral view; 163, penis, sublateral view; 164,penis, apical view; 165, apical spine of penis, lateral view.227

Scute length, 1.13-1.23. Leg II length, 3.23-3.67.Leg II1Scute length, 2.63-3.09. (N = 9).Color brownish orange. Body rugosity medium;tubercles on eye mound, pars thoracica, tergitemargins, and coxae. Carapace with 4-6 pairs of AT;arranged in two, poorly deftned rows. Eye moundbroadly conical, eyes well developed. Palpal megaspines:trochanter, 1 ventral; femur, 1 mesoapical;patella, 2 mesal; tibia, 3 mesal. Tarsal count:3-5-5-5; two specimens have six tarsomeres on thefourth right tarsi.Male (holotype): Total body length, 1.54. Scutelength, 1.15; width, 1.15. Eye mound length, 0.28;width, 0.31. Leg II length, 3.46. TrIV spurlength, 0.36. POP length, 0.26.Penis (Figs. 168-170): VPP apically rounded,dorsally expanded; with 3 dorsal, 16 lateral, and 5ventral setae; AS evenly curved, apically polyfurcate.Glans: BK absent; ML not apparent; PSLclaw-like. S long, slightly sigmoid, apically spatulate,ventroapically with slight groove, ventrobasallywith longitudinal carina; BF present; SA representedby two prongs originating medioventrally on thestylus.Female (paratopotype): Total body length, 1.67.Scute length, 1.23; width, 1.21. Eye mound length,0.28; width, 0.28. Leg II length, 3.67. TrIV withventral tubercle. GO with two pairs of apicalspines.Ovipositor (Fig. 171): cuticle intricately folded;surface lacking microspination; apical teeth apparentlyabsent; setal pattern: 1 pair dorsal, 4 pairs lateral,1 pair ventral.Variation.-The second male, although similarin size to the holotype, has a shorter POP, 0.15 inlength.Natural History.-This species has been collectedunder rocks near the cave entrance (includingdark and twilight regions), whereas the sympatric T.mulaiki occupies the deeper zones.Other Material Examined (Paratypes).- UNITED STATES: Texas: Hays Co.: Ladder Cave, 26May 1989 (A. Grubbs, J. Reddell, and M. Reyes,TMM, CAS), 1 male, 2 females; 2 Sep. 1989 (1.Reddell and M. Reyes, TMM), 4 females, 1juvenile (D. Ubick, S. Fend, and S. Renkes, CDU),1 female, 1 juvenile.Distribution.-Known only from Ladder Cave,Hays County, Texas (Maps 2-5).Texella renkesae, new speciesFigs. 172-177.Te.xella sp. Davis, 1979:33.Diagnosis.-Males of this species are distinguishedfrom others in the infragroup by the followingcombination of characters: GO lacking apicaltubercles; VPP with two dorsal setae; SA prongsbasally fused; S broadly spatulate, with ventrobasalgroove (Figs. 172, 174-177).Type.-Male holotype from Ezell's Cave, HaysCounty, Texas (2 Sep. 1989; D. Ubick, S. Fend, S.Renkes), deposited in CAS.Etymology.-The specific name is a matronymin honor of Ms Saelon Renkes one of the collectorsof the holotype.Description.-Total body length, 1.54-1.92.Scute length, 1.15-1.31. Leg II length, 3.92-4.46.Leg II/Scute length, 3.21-3.41. (N = 5).Color brownish orange. Body of moderaterugosity; tubercles sparse on eye mound, pars thoracica,tergite margins, and coxae. Carapace with 4pairs of AT. Eye mound broadly conical, eyes welldeveloped. Palpal megaspines: trochanter, 1 ventral;femur, 1 mesoapical; patella, 2 mesal; tibia, 3mesal. Tarsal count: 3-5-5-5.Male (holotype): Total body length, 1.72. Scutelength, 1.15; width, 1.15. Eye mound length, 0.26;width, 0.28. Leg II length, 3.92. TrIV spurlength, 0.33. POP length, 0.12.Penis (Figs. 172, 174-177): VPP apically pointed,dorsal margin expanded; with 2 dorsal, 18 lateral,and 6 ventral setae; AS slightly curved,apically polyfurcate. Glans: BK absent; ML notevident; PSL claw-like. S long, slightly sinuate,apically spatulate, ventrobasally grooved; BF present;SA represented by a pair of long, pointed,basally fused, prongs.Female (paratopotype): Total body length, 1.79.Scute length, 1.31; width, 1.33. Eye mound length,0.27; width, 0.31. Leg II length, 4.21. TrIV withventral tubercle. GO with a pair of apical spines.Ovipositor (Fig. 173): cuticle intricately folded,surface lacking microspination; 1 pair apical teethpresent; setal pattern I pair dorsal, 4 pairs lateral, 1pair ventral.Variation.-The TrIV spur varies in length from0.33-0.54, the POP from 0.12-0.38, in the threeavailable males.Natural History.-This species has been collectedunder rocks near the cave entrance (includingdark and twilight regions), whereas the sympatric T.mulaiki occupies the deeper zones. One male wascollected beneath a rock baited with cheese (Davis,1979).Other Material Examined (Paratypes).- UNITED STATES: Texas: Hays Co.: Ezell's Cave, 15Aug. 1978 (J. Davis, WAS), 1 male; 2 Sep. 1989228

Figs. 166-169.-Texella diplospina, new species, male paratopotype: 166, eye mound, lateral view; 167, trochanter IV spur,lateral view; 168, penis, apical view; 169, penis, dorsolateral view.229

Figs. 170-173.-Te.xella species, paratopotypes: 170-171, T. dip/ospina, new species: 170, male, apical spine of penis, sublateralview; 171, female, ovipositor, subapical view. 172-173, T. renkesae, new species: 172, male, apical spine of penis, sublateral view;173, female, ovipositor, lateral view.230

Figs. 174-177.-Texella renkesae, male paratopotype: 174, penis, ventral view; 175, penis, lateral view; 176, penis, apical view;177, penis, dorsolateral view.231

(J. Reddell and M. Reyes, CAS, CDU), 1 male, 2females, 1 juvenile.Distribution.-Known only from Ezell's Cave,Hays County, Texas (Maps 2-5).Texella spinoperca, new speciesFigs. 2, 178-183.Diagnosis.-Males of this species are distinguishedfrom others in the infragroup by the followingcombination of characters: VPP with threedorsal setae; SA prongs long and narrow,originating basoventrally on stylus; S broadlyspatulate, lacking ventral groove or carina; GOsometimes with apical tubercles (Figs. 179-181).Type.-Male holotype from Airman's Cave,Austin, Travis County, Texas (3 Sep. 1989; D.Ubick, J. Reddell, and M. Reyes), deposited inCAS.Etymology.-The specific name is a reference tothe apical spines on the female genital opercula.Description.-Total body length, 1.36-2.08.Scute length, 1.13-1.36. Leg II length, 4.59-5.23.Leg II/Scute length, 3.38-4.30. (N = 13).Color pale orange. Body rugosity moderate;tubercles on eye mound, pars thoracica, tergitemargins, and coxae. Carapace with 3 or 4 pairs ofAT. Eye mound broadly conical, eyes well developed.Palpal megaspines: trochanter, 1 ventral;femur, 1 mesoapical; patella, 2 mesal; tibia, 3mesal. Tarsal count: 3-5-5-5; one female has 6tarsomeres on the fourth right tarsus and one malehas six tarsomeres on the fourth left tarsus.Male (holotype): Total body length, 1.74. Scutelength, 1.23; width, 1.28. Eye mound length, 0.28;width, 0.31. Leg II length, 4.87. TrIV spurlength, 0.49. POP length, 0.21.Penis (Figs. 2, 178-181): VPP apically rounded,dorsally expanded; with 3 dorsal, 14 lateral, and 8ventral setae; AS slightly curved, apically polyfurcate.Glans: BK absent; ML not evident; PSLclaw-like. S long, apically broadly spatulate, ventrallylacking carina or distinct groove; BF present;SA represented by two, slender prongs originatingbasoventrally on the S.Female (paratopotype): Total body length, 1.79.Scute length, 1.33; width, 1.41. Eye mound length,0.28; width, 0.31. Leg II length, 4.67. TrIV withone ventral tubercle. GO with a pair of apicalspines.Ovipositor (Figs. 182, 183): cuticle intricatelyfolded; surface smooth, lacking microspination; 1pair apical teeth present; setal pattern: 2 pairsdorsal, 4 pairs lateral, 1 pair ventral.Variation.-There is some variation in the sizeof TrIV spurs (0.46-0.54 long) and POP (0.21-0.33)in the seven males measured. Three males have apair of apical tubercles on the GO (Figs. 178, 181),the right one missing in one specimen, which aresmaller and blunter than those found in females.Natural History.-A large series of 10 individualswas recently collected under rocks in a small,narrow room (dark zone). Because of a scarcity ofrocks, most undersurfaces were examined few toseveral times (being carefully replaced in theiroriginal position after each search) over a period ofabout 1.5 hours. It is worth noting that only 2specimens were collected on the initial examination;the remainder came from rocks which had beenpreviously checked two or more times. As the rocksin this cave were situated on a loose, crumbly soil itappears that the harvestmen migrated up from thesoil stratum to the rock undersurface during theperiod of our collecting. This sudden appearance ofharvestmen is noteworthy, especially if it resultedfrom the disturbances to the environment (vibrations,movement of soil, etc.) caused by our collecting.Another possibility, and likewise interesting,is that the migration represented an endogenousrhythm of these harvestmen.Other Material Examined (Paratypes).-UNITED STATES: Texas: Travis Co.: Airman's Cave,3 Sep. 1989 (D. Ubick, J. Reddell, M. Reyes,TMM, CAS, CDU), 6 males, 3 females; 1 June1984 (1. Reddell and M. Reyes, TMM), 1 male; 27Sep. 1975 (A. Grubbs and L. Wilk, TMM), 2females.Distribution.-Known only from Airman'sCave, Travis County, Texas (Maps 2-5).Texella fendi, new speciesFigs. 184-197.Diagnosis.-Males of this species are distinguishedfrom others in the infragroup by the followingcombination of characters: TrIV spur reduced;POP absent; GO lacking apical spines; VPP lackingdorsal setae; SA prongs long and narrow,originating basoventrally on stylus; PSL attenuated;S tube-like, lacking ventral groove or carina (Figs.186-189). Females are unique in the infragroup inhaving a genital operculum with a pair of apicaltubercles, rather than spines (Figs. 194, 197).Type.-Male holotype from 9 mi N La Grange,232

Figs. 178-18\.-Teulla spinoperca, new species, male paratopotype: 178, anterior half of body, lateral view; 179, apical spineof penis, lateral view; 180, penis, subdorsal view; 181, penis, lateral view.233

Fayette County, Texas (I Sep. 1989; D. Ubick, S.Fend, S. Renkes), deposited in AMNH.Etymology.-The specific name is a patronym inhonor of Mr. Steven Fend, one of the collectors ofthe holotype.Description.-Total body length, 1.59-2.00.Scute length, 1.21-1.41. Leg 11 length, 3.28-3.69.Leg II1Scute length, 2.54-2.71. (N=9)Color brownish orange. Body moderately rugose;tubercles small and sparse on eye mound andpars thoracica, large and dense on tergite margins(both free and fused), sternite margins, and coxae.Carapace with 2-4 pairs of AT. Eye mound broadlyconical, eyes well developed. Palpal megaspines:trochanter, I ventral; femur, I mesoapical; patella,2 mesal; tibia, 3 mesal. Tarsal count: 3-5-5-5.Male (holotype): Total body length, 1.72. Scutelength, 1.38; width, 1.36. Eye mound length, 0.33;width, 0.36. Leg II length, 3.62. TrIV spurlength, 0.10. POP absent.Penis (Figs. 186-191): VPP apically pointed,dorsal margin expanded; with 0 dorsal, 17 lateral,and 9 ventral setae; AS long, straight, apicallypolyfurcate. Glans: BK absent; ML broad, rounded;PSL attenuated. S long, tube-like, lackingventral modifications; BF present; SA representedby a pair of long prongs originating basoventrallyon the S.Female (paratopotype): Total body length, 1.92.Scute length, 1.36; width, 1.38. Eye mound length,0.33; width, 0.33. Leg II length, 3.67. TrIV withone or two ventral tubercles. GO with 1 pair ofapical tubercles.Ovipositor (Figs. 194-197): cuticle intricatelyfolded; dorsal and ventral surfaces with tiny serrations;1 pair of apical teeth present; setal pattern: Ipair dorsal, 4 pairs lateral, I pair ventral.Natural History.-The type locality is a pineforest with oaks. At the time of the collection thesurface conditions were somewhat xeric. The specimenswere collected beneath imbedded limestoneboulders; one specimen was on the rock undersurface,the others in the loose sandy soil immediatelybelow the rocks. The harvestmen occurred roughly0.2-0.3 m below the surface of the ground, in cooland relatively mesic conditions.Other Material Examined (Paratypes).- UNITED STATES: Texas: Fayette Co.: 11 mi N LaGrange, 18 July 1966 (J. and W. Ivie, AMNH,CAS), 3 males, 2 females; 9 mi N La Grange, 1Sep. 1989 (D. Ubick, S. Fend, and S. Renkes,COD), I male, 2 females.Figs. 182-183.-Texella spinoperca, new species, female paratopotype: 182, ovipositor, lateral view; 183, ovipositor and genitaloperculum, apical view.234

Distribution.-Known only from FayetteCounty, Texas (Maps 2-4).Texella homi, new speciesFigs. 198-201.Diagnosis.-Males of this species are distinguishedfrom others in the infragroup by the followingcombination of characters: TrIV spur reduced;POP absent; GO lacking apical spines; VPP lackingdorsal setae; PSL attenuated; SA represented by alaterobasal carina; S short, tube-like, lacking ventralcarina (Figs. 198-201).Type.-Male holotype from 13.3 mi SEOakville, Live Oak County, Texas (14 Aug. 1966;T. Briggs and K. Hom), deposited in CAS.Etymology.-The specific name is a patronym inhonor of Mr. Kevin Hom, one of the collectors ofthis species.Description.-Total body length, 1.59-1. 80.Scute length, 1.13-1.23. Leg II length, 3.05-3.33.Leg II/Scute length, 2.48-2.90. (N = 4).Color brown. Body moderately rugose; tuberclessparse on eye mound, pars thoracica, tergite margins,and coxae. Carapace with 3 pairs of AT. Eyemound broadly conical, eyes well developed.Palpal megaspines: trochanter, 1 ventral; femur, 1mesoapical; patella, 2 mesal; tibia, 3 mesal. Tarsalcount: 3-5-4-5.Male (holotype): Total body length, 1.67. Scutelength, 1.15; width, 1.13. Eye mound length, 0.26;width, 0.28. Leg II length, 3.33. TrIV spurlength, 0.03. POP absent.Penis (Figs. 198-201): VPP apically pointed,dorsal margin expanded; with 0 dorsal, 19 lateral,and 7 ventral setae; AS long, robust, curved, apicalhalf densely plumose. Glans: BK absent; ML rectangular;PSL attenuated. S short, apically tubelike;BF present; SA represented by laterobasalcarina.Female (paratopotype): Total body length, 1.80.Scute length, 1.18; width, 1.18. Eye mound length,0.31; width, 0.31. Leg II length, 3.28. TrIV withventral tubercle. GO with 1 pair of apical spines.Ovipositor: cuticle intricately folded; dorsal surfacewith few scattered microtubercles, ventralsurface smooth; 1 pair apical teeth present; setalpattern: 1 pair dorsal, 4 pairs lateral, 1 pair ventral.Variation.-The second female has asymmetricalapical spination on the GO: one spine on the leftside, two on the right.Note.-The brown color of the specimens isprobably the result of prolonged storage in Oudemans'Fluid (see discussion in Ubick and Briggs,1989:96); it is expected that the natural color ofthese harvestmen is brownish orange.OtherMaterial Examined (Paratypes).-UNIT-EDSTATES: Texas: Live Oak Co.: 13.3 mi SEOakville, 14 Aug. 1966 (T. Briggs and K. Hom,CAS), 1 male, 2 females.Distribution.-Known only from the type locality(Maps 2-4).Figs. 184-185.-Texellafendi, new species, male paratype: 184. anterior half of body, lateral view;185, palpi. lateral view.235

Figs. 186-189.-Texellafendi, new species, male paratype: 186, penis, subdorsal view; 187, penis, sublateral view; 188, penis,ventral view; 189, penis, lateral view.236

Figs. 190-193.-Texellafendi, new species, paratypes: 190-192, male: 190, apical spine of penis, dorsal view; 191, apical spineof penis, lateral view of tip; 192, trochanter IV spur, lateral view. 193, female, trochanter IV spur, lateral view.237

Figs. 194-197.-Texellafendi, new species, female paratype: 194-195, ovipositor, lateral view;lateral surface; 197, ovipositor, apical view.196, ovipositor, enlarged view of238

CLASSIFICAnONTexellabifurcata groupbifurcata (Briggs)kokoweef groupkokoweefn. sp.shoshone n. sp.deserticola n. sp.mulaiki groupbrevistyla subgroupjungi n. sp.brevistyla n. sp.mulaiki subgrouplongistyla infragrouplongistyla n. sp.welboumi n. sp.mulaiki infragrouphardeni n. sp.cokendolpheri n. sp.mulaiki Goodnight and Goodnightbilobata infragroupbilobata n. sp.reddelli subgroupreddelli infragroupreddelli Goodnight and Goodnightreyesi n. sp.brevidenta infragroupbrevidenta n. sp.grubbsi n. sp.spinoperca infragroupdiplospina n. sp.renkesae n. sp.spinoperca n. sp.fendi n. sp.homi n. sp.198"- "- \\\\\\II\, >,III//I///\/201Figs. 198-201.-Texella homi, new species, male paratopotype: 198, penis, lateral view; 199, penis, ventral view; 200, penis,dorsal view; 201, glans, ventral surface. Scale bar = 0.50 mm.239

ACKNOWLEDGMENTSPrimary thanks go to James R. Reddell, withoutwhose important collection of Texas cave harvestmenthis project would never have beenendeavored, for his help in locating the cavelocalities and providing information on the habitatsof Texella, and for his hospitality and fieldassistance during a visit to Texas. Appreciation forassistance in field work is extended to Steven V.Fend and Saelon K. Renkes, who made possible atrip to Texas (for DU).Norman I. Platnick kindly loaned the AMNHholdings of Phalangodidae and made a special effortto make available unsorted material. Jonathan A.Coddington, James C. Cokendolpher, and WilliamA. Shear quickly responded to our request for specimens.W. Calvin Welbourn provided informationon the New Mexican cave localities. Willis J.Gertsch generously shared his unpublished informationon the distribution of Telemidae and othercavernicolous spiders.Many thanks go to Mary Ann Tenorio for invaluableassistance and expert advice on the SEM andfor help in artistic matters, especially in the preparationof the diagram and maps. Thanks also go toSaelon K. Renkes for assistance in developing theSEM prints. Paul H. Arnaud, Jr., is thanked forpermission to use the research facilities at the CAS.Appreciation is extended to James C. Cokendolpher,Warren C. Rauscher, James R. Reddell,and William A. Shear for their helpful commentsand criticisms on a preliminary draft of this manuscript.The Exline-Frizzell Fund for ArachnologicalResearch, from the California Academy of Sciences,defrayed the costs associated with use of thescanning electron microscope and the preparation ofplates.LITERATURE CITEDBriggs, T. S. 1968. Phalangids of the laniatorid genus Sitalcina(phalangodidae: Opiliones). Proc. California Acad. Sci.,36:1-32.Briggs, T. S. 1974. Phalangodidae from caves of the SierraNevada (California) with a redescription of the type genus(Opiliones: Phalangodidae). Occas. Pap. California Acad.Sci., 108:1-15.Briggs, T. S. and K. Hom. 1972. A cavernicolous whipscorpionfrom the northern Mojave Desert, California(Schizomida: Schizomidae). Occas. Pap. California Acad.Sci., 98:1-7.Briggs, T. S. and D. Ubick. 1981. Studies on cave harvestmenof the central Sierra Nevada with descriptions of new speciesof Banksula. Proc. California Acad. Sci., 42:315-322.Briggs, T. S. and D. Ubick. 1989. The harvestman familyPhalangodidae. 2. The new genus Microcina (Opiliones:Laniatores). J. Arachnol., 17:207-220.Brignoli, P. M. 1968. Note su Sironidae, Phalangodidae eTrogulidae italiani, cavernicoli ed endogei (Opiliones).Fragm. Entomol., 5:259-293.Davis, J. C. 1979. A faunal survey of Ezell's Cave, HaysCounty, Texas. Unpublished thesis. San Marcos: SouthwestTexas State University.Gertsch, W. J. 1974. The spider family Leptonetidae in NorthAmerica. J. Arachnol., I:145-203.Gertsch, W. J. 1984. The spider family Nesticidae (Araneae) inNorth America, Central America, and the West Indies. Bull.Texas Memorial Mus., 31: 1-91.Gertsch, W. J. 1992. Distribution patterns and speciation inNorth American cave spiders with a list of the troglobites andrevision of the cicurinas of the subgenus Cicurella. TexasMemorial Mus., Speleol. Monogr., 3:75-122Goodnight, C. J. and M. L. Goodnight. 1942. New Phalangodidae(Phalangida) from the United States. American Mus.Novitates, 1188:1-18.Goodnight, C. 1. and M. L. Goodnight. 1967. Opilionids fromTexas caves (Opiliones, Phalangodidae). American Mus.Novitates, 230I :1-8.Maddison, W. P. and D. R. Maddison. 1987. MacClade,version 2.1. A phylogenetics computer program distributedby the authors.Martens, J. 1972. Allsobskya athos, der erste Krallenweberknechtaus Greichenland. Senckenbergiana bioI.,53:431-440.Martens, J. 1978. Spinnentiere, Arachnida-Weberknechte,Opiliones. Die Tierwelt Deutschlands, 64:1-464. Jena:Fischer.Martens, 1., U. Hoheisel, and M. GolZe. 1981. VergleichendeAnatomie der Legerohren der Opiliones als Beitrag zurPhylogenie der Ordnung (Arachnida). Zool. Jb. Anat.,105:13-76.Mitchell, R. W. and 1. R. Reddell. 1971. The invertebratefauna of Texas caves, pp. 35-90. In: E. L. Lundelius andB. H. Slaughter, eds., Natural history of Texas caves.Dallas, Texas: Gulf Natural History Press. 174 pp.Murphree, C. S. 1988. Morphology of the dorsal integument often opilionid species (Arachnida, Opiliones). J. Arachnol.,16:237-252.Reddell, 1. R. 1965. A checklist of the cave fauna of Texas. I.The Invertebrata (exclusive of Insecta). Texas 1. Sci.,17(2):143-187.Shear, W. A. 1974. North American cave millipeds. II. Anunusual new species (Dorypetalidae) from southernCalifornia, and new records of Speodesmlls tllganbilLS(Trichopolydesmidae) from New Mexico. Occas. PapersCalifornia Acad. Sci.,112:1-9.Silberling, N. 1., D. L. Jones, M. C. Blake, Jr., and D. G.Howell. 1984. Lithoteclonic Terrane maps of the NorthAmerican Cordillera. U. S. Geol. Survey Open File Report84-523.Ubick, D. and T. S. Briggs. 1989. The harvestmen familyPhalangodidae. I. The new genus Calicina, with notes onSitulcina (Opiliones: Laniatores). Proc. California Acad.Sci., 46:95-136.This is publication No. N.S.-60 of the Texas Memorial Museum.240

Chandler, D.S. 1992. The Pselaphidae (Coleoptera) ofTexas caves. Texas Mem. Mus., Speleol. Monogr., 3:.241-253.THE PSELAPHIDAE (COLEOPTERA) OF TEXAS CAVESDonald S. ChandlerDepartment of EntomologyUniversity of New HampshireDurham, New Hampshire 03824ABSTRACTTwelve species of Pselaphidae have been found in Texascaves. Five species are true cavernicoles and are restricted tocaves in central Texas: Batrisodes grubbsi n. sp., B. reyesi n.sp., B. texanus n. sp., B. venyivi n. sp., and Texamauropsreddelli Barr & Steeves. Seven other species are probably foundonly near cave entrances, in litter washed into caves, or in debrisbrought in by rodents: B. uncicomis (Casey), B. clypeonotus(Brendel), B. globosus (LeConte), Cylindrarctus bicomisChandler, Tmesiphorus costalis LeConte, Hamotus electraePark, and Trimioarcus musamator n. sp. Batrisodes clypeonotusis associated with the ant, Camponotus americanus Mayr, basedon many collections from Oklahoma. Two new synonymies havebeen discovered concerning these species: B. schneiderensisPark is a junior synonym of B. uncicomis (Casey), and B.tridens Casey is a junior synonym of B. clypeonotus (Brendel).INTRODUCTIONPselaphids have been recorded from caves since1855 when Machaerites spelaeus Miller was describedfrom a cave at Struge, Yugoslavia. Sincethen the European pselaphid cave fauna has beenextensively documented by a number of authors,while the North American fauna has been treatedonly recently (Park, 1951, 1956, 1960). The firstpselaphid described from a Texas cave wasBatrisodes schneiderensis Park (1960), based on asingle female from Schneider Ranch Cave in KendallCounty. Four years later a second species wasadded, Texamaurops reddelli Barr and Steeves(1963), based on a single female from KretschmarrCave in Travis County. In the only other paperdealing with pselaphids from Texas caves, Barr(1974) figured the aedeagus of a specimen from InnerSpace Caverns believed to be the male of T. reddelli,but is here recognized as a new species, Batrisodestexanus. Texamaurops reddelli is now aregistered endangered species listed as the "KretschmarrCave mold beetle" (Chambers and Jahrsdoerfer,1988).Twelve species are listed from Texas caves in thispaper. Five species appear to be true cavernicolesbased on their reduced eyes and longer than typicallegs and antennae, and are only known from thekarst region of the Edwards Plateau in centralTexas: Batrisodes grubbsi n. sp., Batrisodes reyesin. sp., Batrisodes texanus n. sp., Batrisodes venyivin. sp., and Texamaurops reddelli. The other sevenspecies are free-living or myrmecophilous, includingthree species of Batrisodes: the myrmecophile clypeonotus(Brendel), the free-living troglophile uncicornis(Casey) (=schneiderensis Park), and globosus(LeConte). The free-living Hamotus electraePark was described from Veracruz, Mexico, and ishere recorded for the first time in the United Statesfrom Texas in caves and at ultra-violet light. Tmesiphoruscostalis LeConte has been collected with ants241

and in several leaf litters, and has been previouslyrecorded from a cave (Park, 1960: 1(0). A specimenof Cylindraretus bicomis Chandler was found in acave, but is presumed to be associated with leaf litters,based on the common association of the otherspecies in the genus. Trimioarcus musamator n. sp.has reduced eyes, but is most likely an inhabitant ofrodent nests or deep leaf litter since no other speciesin its tribe, the Euplectini, is known to be cavernicolous.In summary, five species are troglobites,one a myrmecophile, one a troglophile, and theother five species are probably only found in cavesnear their entrances or in litter piles washed in orbrought in by rodents.All measurements are in millimeters. The collectionswhere specimens are deposited are indicated bythe following codons: ANSP, Academy of NaturalSciences, Philadelphia; DSC, collection of author;FMNH, Field Museum of Natural History, Chicago;JAW, John A. Wagner collection, Evanston,Illinois; TMMC, Texas Memorial Museum Collection,Austin; USNM, National Museum of NaturalHistory, Washington, D.C. The types of all previouslydescribed species were examined, and theholotypes of species described here are to be placedin the Field Museum of Natural History, Chicago.Illustrations and descriptions were based on wholeand disarticulated specimens in temporary mountson slides, and checked against whole specimensmounted on points.SYSTEMATICS1. Batrisodes (Babnormodes) uncicomis (Casey)(Figs. 1-2)Batrisus uncicomis Casey, 1897:576. Type locality:New York City, New York. Male holotype(USNM).Batrisodes schneiderensis Park, 1960:75. Typelocality: Schneider Cave, Texas. Holotype female(FMNH). NEW SYNONYMY.Description.-Length 2.32-2.64. Males withhead transversely excavate anterior to antennalbases, frons lacking projections, largely covereddorsally by rounded angulate projection of vertex,dense setae projecting ventrally from anteriormargin of vertex between antennal bases, vertexcoarsely punctate especially on anterior half, lateralcarinae strong basally but fading before reachingantennal bases, median carina distinct but not strongand extending anteriorly to line between nudevertexal foveae, eyes with about 32 facets; firstantennomere (scape) not enlarged but conspicuouslypunctate on antero-ventral margin, antennomere IXprotruding laterally as a lamina, IX less than twiceas wide as long, X with large fovea occupying overhalf of ventral surface, XI with large ventraltubercle at base.Pronotum with large scattered punctures particularlyanteriorly, median longitudinal sulcus extendingto near apex; elytra with three basal foveae andsubhumeral fovea; sternite VI vaguely impressedlongitudinally; protibiae flattened near apex anddensely setate on medial margin, mesotibiaeexcavate on mesal margin in apical third, thick setaealong posterior edge of excavation, secondmesotarsomere sinuate ventrally near base.Females with coarse punctures on head and pronotum;eyes with about 17 facets, lacking modificationsof head, antennae, and legs; sternite VIrounded.Male from Schneider Ranch Cave: antennae 1.10,metafemora 0.84, metatibiae 0.80, metatarsi 0.37.Specimens examined.-43: TEXAS: BurnetCounty: Resurrection Well, VII- 2-1989, M. Grimm(1). Hays County: Anaqua Cave, VI-2I-I985, A.Grubbs (2); Ezell's Cave, 1-16-1979, J. C. Davis,cave area (2); Fern Cave, V-26-I989, A. Grubbs, J.Reddell, M. Reyes (5); Ladder Cave, IX-2-I989, D.Ubick, S. Fend, S. Renkes (1); McGlothlin Sink,V-26-I989, A. Grubbs, J. Reddell, M. Reyes (1).Carnal County: Klar's Cave, III-I2-I988, S. Spence,G. Veni (1). Kendall County: The Crack,V-28-I990, D. Pate (1); Schwarz Cave,IX-12-1987, J. Ivy, G. Veni (I); Schneider Cave,VIII-30-1959, T. C. Barr, Jr., H. M. Koepke (typeof schneiderensis); same locality, IX-8-1963, J.Reddell, D. McKenzie (4). Llano County:Enchanted Rock Cave, IV-I4-1985, J. Reddell, M.Reyes (1); Freshman Mountain Cave, IV-9-1989,W. Elliott, J. Reddell, M. Reyes (2). TravisCounty: Bandit Cave, V-I7-1965, T. C. Barr (I);Cave X, III-30-1974, W. Elliott, W. Russell, S. &R. Fieseler, C. Rogers (1); Ireland's Cave, 1-231989, J. Reddell, M. Reyes, E. Grimm, M. Grimm(10); laCrosse Cave, V-8-1990, J. Reddell, M.Reyes (2); Pickle Pit, V-2I-1990, J. Reddell, M.Reyes, L. Sherrod (2). Williamson County: BeckRanch Cave (=Beck's Cave), III-24-1989, J. Reddell,M. Reyes (1); same locality, V-16-I965, T. C.Barr (3). Specimens in collections of: DSC, FMNH,JAW, and TMMC.Comments.-Batrisodes uncicornis is a widespreadspecies whose range extends from Massachusettsand Florida to Texas. Habitat information hasbeen associated with only a few specimens, and242

Key to Species1. Abdominal segments II-IV with acute lateral margins 2Abdominal segments II-IV lacking lateral margins, abdomenround in cross section (Fig. 1) (Batrisinae, Batrisini) 52.(1) Last segment of maxillary palpi laterally angulate, previous twosegments with spine on outer face (Fig. 21) (Pselaphinae,Tmesiphorini10. Tmesiphorus costalisApical segment of maxillary palpi elongate, lacking lateralangulation or spines on previous segments (Figs. 20, 22) 33.(2) Last segment of maxillary palpi elongate, penultimate segmentangulate on mesal margin (Fig. 20) (Goniacerinae, Tychini9. Cylindrarctus bicornisLast segment of maxillary palpi enlarged, penultimate segmentmuch smaller (Figs. 22, 24) 44.(3) Antennal club of last three segments, penultimate antennomerelonger than wide, last segment of maxillary palpus withlongitudinal groove on inner margin (Fig. 22) (Pselaphinae, Tyrini) 11. Hamotus electraeAntennal club of last two segments, penultimate antennomeremuch wider than long, last segment of maxillary pal pus lackingany groove (Fig. 24) (Euplectinae, Euplectini)12. Trimioarcus musamator, n. sp.5.(1) Apex of metatibiae lacking elongate pencil of appressed setae(Fig. 18); all antennomeres more than twice as long as wide(Fig. 17)8. Texamaurops reddelliApex of metatibiae with elongate pencil of appressed setae (Fig. 11);with antennomere VIII distinctly less than twice as long aswide except texanus with VIII twice as long as wide (Figs. 3, 9)(Batrisodes) 66.(5) Eyes distinct, with 10 to 50 facets grouped together. 7Eyes apparently absent, possibly up to 10 disassociated palegranules in area where eyes should be 97.(6) Vertex of head and anterior portion of pronotum coarselypunctate; males with anterior margin of vertex angulate (Fig. 1);females with vertex coarsely punctate1. Batrisodes uncicornisBasal half of vertex and anterior portion of pronotum not coarselypunctate; males with anterior margin of vertex truncate tobroadly rounded (Figs. 3, 15); females with vertex at mostcoarsely punctate on antennal tubercles, area between vertexalfoveae smooth or sparsely granulate 88.(7) Males with anterior margin of vertex broadly rounded,penultimate antennomere with small fovea at base (Fig. 15);females with large eyes of more than 40 facetsMales with anterior margin of vertex broadly truncate,penultimate antennomere lacking basal fovea (Fig. 3); femaleswith small eyes of approximately 15 facets7. Batrisodes globosus2. Batrisodes clypeonotus243

9.(6) Antennomeres III-VIII twice as long as wide; sides of head ineye region gently curved (Fig. 9)5.Batrisodes texanus n. sp.At least antennomere VIII less than twice as long as wide; lateralmargins of head slightly angulate and bearing granules whereeyes should be (Figs. 7, 13) 1010.(9) Head lacking lateral and medial vertexal carinae (Fig. 13),vertex smoothHead with at least lateral carinae extending anteriorly from base(Figs. 5, 7), vertex in apical half distinctly punctate or rugulose11.(10) Apical half of vertex roughened (females) or coarsely punctate(males), with faint lateral vertexal carinae, lacking medianvertexal carina; males with anterior margin of vertex broadlytruncate, first antennomere not modified (Fig. 5)Apical half of vertex smooth (females) to transversely rugulose(males), with median and lateral vertexal carinae sharplydefined; males with anterior margin of vertex broadlybisinuate, first antennomere angulate ventrally (Fig. 7)6. Batrisodes venyivi n.sp.ll3. Batrisodes grubbsi n.sp.4. Batrisodes reyesi n. sp.these have been collected in sawdust, under pinebark, and under stones. This species has not beenpreviously noted as a troglophile, and it is interestingthat in surface habitats this species is alwayscollected in quite short series. The longest series Ihave seen is from Ireland's Cave.James R. Reddell (pers. comm.) has supplied thefollowing collection information. "It has been foundbelow rocks in moist sinkhole-type caves in associationwith troglobitic millipedes and other species,but also occurs in total darkness. Most specimenshave been found on the underside of rocks lightlyburied in silt. A specimen from Beck Ranch Cavewas found crawling on moist flowstone more than200 meters from the cave entrance. "Park (1960) described a single female fromSchneider Cave as a new species, schneiderensis.Several series from Texas caves contain both sexes,and it is obvious that the type of schneiderensis is atypical female of uncicornis. Batrisodes uncicornishas been placed close to riparius (Say) based on thevery similar male antennal characters (Park, 1947).The modifications of the legs are also similar, as isthe general structure of the head. The punctation ofthe head is coarser in uncicornis, and the apical angieof the vertex is only slightly depressed, while inriparius the punctation is not coarse on the pronotumand basal portion of the vertex, and the apicalportion of the vertex is strongly deflexed betweenthe antennal bases. Females of uncicornis may beconfused with those of globosus (LeConte), but maybe separated by the head being more completely andcoarsely punctate.2. Batrisodes (Excavodes) clypeonotus (Brendel)(Figs. 3-4)Batrisus clypeonotus Brendel, 1893:280, pI. IV.Type locality: Ponchatoula, Louisiana. Maleholotype (ANSP).Batrisodes tridens Casey, 1908:263. Type locality:St. Louis, Missouri. Male holotype (USNM).NEW SYNONYMY.Description.-Length 2.28-2.68. Males withhead transversely excavate just anterior to antennalbases, frons with acute medial and lateral tubercles,medial tubercle with short posterolaterally directedsetae, lateral tubercles glabrous, vertex sharply angledventrally between antennal bases, this area setwith short obvious setae arising from small distinctpunctures, vertexal apex with small median trilobedprojection bearing two small tufts of setae, basalportion of vertex smoother, sparsely setate withsetae arising from isolated granules, vertexal foveaenude, lateral vertexal carinae weak and extending toantennal bases, median carina weak and extending tomiddle of vertex, eyes with 12 (Texas) to 40(Missouri) facets grouped in crescentic pattern; firstantennomere (scape) enlarged, impressed at center ofanterior face, densely and minutely punctate in impression,antennomere X globose, slightly widerthan last antennomere and lacking a ventral fovea.Pronotum with median longitudinal sulcus extendingto near apex, disc sparsely punctate; elytrawith three basal foveae and subhumeral fovea;244

stemite VI slightly impressed at base with discrounded; second mesotarsomeres straight.Females with frons and anterior half of vertexminutely rugose, small eyes with 10-16 facets, lackingmodifications of head and antennae, stemite VIbroadly rounded.Male from Powell's Cave, Texas: antennae 1.08,metafemora 0.80, metatibiae 0.77, metatarsi 0.45.Specimen examined.-TEXAS: Menard County:Powell's Cave, IV-II-1989, E. Shoud, third crevice(TMMC).Comments.-This species has been poorlyknown since its description, and during preparationof this paper was found to be the senior synonym oftridens Casey. Park (1956:84) examined a specimenhe placed as this species from northern Alabama,and found a minute fovea of antennomere X. A slidepreparation of a specimen from Oklahoma did notreveal any trace of a fovea, and a fovea is not obviousin any of the specimens from several localitiesmounted on points. Park (1956) believed thatclypeonotus represented an undescribed subgenus,but it appears typical of Excavodes to me. Thisspecies is now known from Missouri to Alabamaand Texas. The largest series have been generatedby collecting specimens with their ant host, Camponotusamericanus Mayr. Whether this species isan obligate myrmecophile is not known, but thereduced number of eye facets indicates that clypeonotushas adapted to a subterranean life.3. Batrisodes (Excavodes) grubbsi n. sp.(Figs. 5-6)Description.-Length 2.32-2.48. Male head withtransverse excavation anterior to antennal bases,clypeus with erect blunt tubercle, tubercle denselysetate along postero-Iateral margin, vertex broadlytruncate in dorsal view, in anterior view vertex projectingventrally to medial point that is slightly angledanteriorly near apex, vertexal apex densely setateinto transverse excavation, vertex coarselypunctate, smoothly roughened in area betweenvertexal foveae, median carina faint, lateral carinaelow but distinct and extending to just anterior toeyes, eyes consisting of approximately 8 granulesFigs. 1-7.-Batrisodes spp.: I, B. uncicomis, habitus of male and anterior view, right antenna; 2, B. uncicomis, dorsal and rightlateral view, aedeagus; 3, B. clypeonotus, head and anterior view, right antenna of male; 4, B. clypeonotus, dorsal and right lateralview, aedeagus; 5, B. grubbsi, head and anterior view, right antenna of male; 6, B. grubbsi, dorsal and right lateral view, aedeagus; 7,B. reyesi, head and anterior view, right antenna of male. Scale line equals O. J. Apical three segments of male antennae presented inventral view.245

indicating facet remnants; antennomere X with largefovea covering ventral half.Pronotum with median longitudinal sulcus faintin basal half of disc, disc sparsely and minutelypunctate; elytra with 3 basal foveae and subhumeralfovea; stemite VI with lateral setate tubercles atapex, disc between tubercles flat; second mesotarsomeresstraight.Females lacking modifications of head, antennae,and sixth stemite; frons and anterior half of vertexcoarsely punctate, posterior half of vertex smoothlyroughened.Holotype male: antennae 1.40, metafemora 0.96,metatibiae 1.00, metatarsi 0.48.Specimens examined.-6: TEXAS: HOLOTYPE male, Hays County, Grapevine Cave,VII-I-1990, A. Grubbs, J. Reddell, M. Reyes(FMNH). PARATYPES, all same locality: 1 female,eutopotypical (TMMC); 2 males, 1 female,VI-IO-1990, A. Grubbs, L. Davis, J. Elliott (DSC,TMMC); 1 female, VII-9-1990, A. G. Grubbs(DSC). The specimens collected on June 10 areteneral.Etymology.-The species is named for theprincipal collector of the series, Andrew Grubbs.Comments.-The type series was collected underwashed-in leaf litter at the litter-clay floor interfacefrom the terminal room of Grapevine Cave in totaldarkness (James R. Reddell, pers. comm.). Themale modifications of the head are generally similarto those of clypeonotus, but differ in the form of thevertexal apex, lack of lateral clypeal spines, and inthe form of antennomeres I and X. The male antennae,median clypeal spine, and form of the aedeagusare also similar to those of venyivi, but may be readilyseparated by the truncate vertexal apex andcoarse punctation on the anterior half of the vertex.4. Batrisodes (Excavodes) reyesi, n. sp.(Figs. 7-8)Description.-Length 2.36-2.44. Males withhead transversely excavate just anterior to antennalbases, frontal hom prominent and with dense shortsetae directed posterolaterally, anterior margin ofclypeus broadly rounded, anterior half of vertexwith obscure transverse rugules and smooth onposterior half, anterior margin of vertex weaklybiemarginate over transverse excavation, short setaealong margin longest near lateral edges, stronglateral carinae of vertex extending anteriorly toantennal bases, median carina extending from baseto just anterior to line between nude vertexal foveae,eyes represented by a number of small scatteredfacet rudiments appearing as granules; first antennomere(scape) angulate on ventral margin, denselyand minutely punctate in lower half, antennomere Xwith large fovea occupying over half of ventralsurface.Pronotum smooth, with median longitudinal sulcusextending to near pronotal apex; elytra withthree basal foveae and subhumeral fovea; metasternumat middle with small deep circular impressionjust anterior to metacoxae; stemite VI medially impressedwith tuberculate raised areas laterally, lineof thickened setae extending from tubercles tomiddle; second mesotarsomeres straight.Females lack modifications of head, antennae,and stemite VI; vertex depressed between antennalbases and smoothly rugulose; metastemal impressionpresent.Male holotype: antennae 1.52, metafemora 1.04,metatibiae 1.05, metatarsi 0.44.Specimens examined.-13: HOLOTYPE male,Texas, Burnet County, Fenceline Sink, IV-17-1990,J. Reddell, M. Reyes (FMNH). PARATYPES: 1female, eutopotypical (TMMC); I male, same locality,V-27-1989, M. Reyes (DSC); 1 male, Snake PitSink, XI-20-1990, J. Reddell, M. Reyes, undersideof rock loosely buried in silt (TMMC). TravisCounty: 1 female, Armadillo Ranch Sink,IX-23-1990, J. Reddell, M. Reyes, C. Sexton, fromcrack in rotten wood (DSC); 3 females, YellowBerry Cave, XII-I 11990, J. Reddell, M. Reyes(DSC, TMMC); 2 males, 3 females, Moss Pit,III-5-1991, J. Reddell, M. Reyes, underside ofrocks deeply buried in silt (DSC, TMMC). The twospecimens collected in April are teneral.Etymology.-The name is derived from MarcelinoReyes, the principal collector of the type series.Comments.-Batrisodes reyesi was found on theunderside of rocks deeply buried in silt at thebottom of the entrance drop of Fenceline Sink,associated with troglobitic spiders, Cicurina (Cicurella)species, and millipedes, Speodesmus species;in Yellow Berry Cave beneath rocks at the bottom ofthe second drop in total darkness; and in Moss Piton the underside of rocks deeply buried in silt atcave bottom in association with blind isopods of thefamily Trichoniscidae (James R. Reddell, pers.comm.).The general form of the modified vertex and firstantennomeres are similar to those of clypeo/lotus(Brendel), a myrmecophile with reduced eyes. Themales of reyesi are easily distinguished by the relativelyflat vertex, the stronger lateral and median246

vertexal carinae, and the laterally tuberculate stemiteVI.5. Batrisodes (Excavodes) texanus n. sp.(Figs. 9-12)Description.-Length 2.60-2.88. Male head withvague transverse impression anterior to antennalbases, impression shallow and medianly angulate,vertex smooth and sparsely setate, antennal tuberclesprominent with a few coarse punctures dorsally,sides of head smoothly curved and flat with a fewgranules present where eyes should be, lateral carinaeextending sinuately from head base of outer angieof antennal tubercles, median carina extendinganteriorly to point between nude vertexal foveae; antennomeresall elongate, X nearly twice as wide asIX and narrowing in apical half, with large nude foveacovering one-third of surface in ventral view.Pronotum with median longitudinal sulcus shallowon disc; elytra with three basal foveae and subhumeralfovea; second mesotarsomeres straight.Females lacking transverse sulcus anterior to antennalbases, vertex merging smoothly with frons,antennomere X barely wider and longer than IX.Male holotype: antenna 1.63, metafemur, 1.32,mesofemur 1.32, metatarsi lost.Specimens examined.-4: TEXAS: WilliamsonCounty: HOLOTYPE male, Inner Space Caverns(=Laubach's Cave), V-23-1965, W. Russell(FMNH). PARATYPES (all females): same locality,VIII-1968, W. Elliott (DSC); Off CampusCave, IV-8- 1989, W. Elliott, J. Reddell, M. Reyes(DSC); Coffin Cave, 10 mi. NW Georgetown,XI-3-1963, J. Reddell (TMMC).Etymology.-The name is derived from itsknown occurrence restricted to Texas.Comments.-Batrisodes texanus was found inOff Campus Cave on the underside of a rock lightlyburied in silty clay in total darkness (James R. Reddell,pers. comm.). Since this species was includedas the Coffm Cave population of Texamaurops reddelliwhen reddelli was listed as an endangered species(Chambers and Jahrsdoerfer, 1988), texanusshould also be considered endangered under federallaw although no specific ruling has yet been published(Steve Chambers, pers. comm. to James R.Reddell).This species possesses elongate legs andantennae, and is superficially similar to T. reddelli.The lack of any ocular projection, and the presenceof the pencil of setae at the apex of the metatibiaereadily separate the two taxa. The aedeagus of thisspecies was figured by Barr (1974) as T. reddelli.Batrisodes texanus is placed in the subgenus Excavodesdue to the modified head anterior to theantennae and the straight mesotarsomeres of themale.6. Batrisodes (Excavodes) venyivi n. sp.(Figs. 13-14)Description.-Length 2.24. Males with headtransversely excavate just anterior to antennal bases,frontal hom prominent with short dense setae directedposterolaterally, anterior margin of clypeusangulate, anterior margin of vertex rounded withshallow medial emargination, margin anterior to antennalbases depressed, vertex smooth and lackinglateral and medial basal carinae; first antennomere(scape) not modified, antennomere X with fovea occupyingone-fourth of face in ventral view.Pronotum smooth, with median longitudinal sulcusbarely attaining middle of disk; elytra with threebasal foveae, lacking subhumeral fovea; stemite VIlight!Y impressed medially, smoothly granulate inimpression; second mesotarsomeres straight.Female unknown.Male holotype: antennae 1.05, metafemora 0.76,metatibiae 0.80, metatarsi 0.44.Specimen examined.-TEXAS: Bexar County:HOLOTYPE male, Helotes Hilltop Cave, IX-291984, J. Ivy, G. Veni (FMNH).Etymology.-The name is formed by combinationof the names of the collectors.Comments.-It is not close to any other speciesof Excavodes that I have seen. It is easily distinguishedby the lack of vertexal carinae, smooth vertex,and lack of a subhumeral fovea on the elytra.7. Batrisodes (Excavodes) globosus (LeConte)(Figs. 15, 16)Batrisus globosus LeConte, 1849: 100. Type localities:Pennsylvania and Georgia. Syntypes maleand female (MCZC).Description.-Length 2.36-2.48. Males withhead transversely excavated anterior to antennalbases, frons with acute medial tubercle bearing foursmall recurved setae at apex and longer laterally directedsinuate setae toward base, frontal tuberclenearly obscured in dorsal view by broadly roundedprojecting vertexal lobe, ventral margin of vertexallobe with small group of setae projecting ventrallylateral to frontal tubercle, vertex coarsely punctatein anterior third, generally smooth in posteriorportion with setae arising from small granules,247

lateral carinae strong and extending to antennalbases, medial carina distinct and extending fromcervix over moderately swollen vertex to point anteriorto nude vertexal foveae, eyes with approximately45 facets; antennomere X large and globose,wider than XI, X with small ventral fovea near base.Pronotum sparsely punctate, median longitudinalsulcus extending to near apex; elytra with threebasal foveae and subhumeral fovea; sternite VIslightly flattened medially; second mesotarsomeresstraight.Females with coarse punctures only on antennaltubercles and posteriorly to near vertexal foveae,eyes with approximately 42 facets; lacking modificationsof head and antennae, sternite VI barelyrounded at middle.Male from Simons Squirm-Around Cave: antennae1.12, metafemora 0.79, metatibiae 0.76, metatarsi0.40.Specimens examined.-5: TEXAS: BurnetCounty: 1 male, Simons Squirm-Around Cave,XI-20-1990, J. Reddell, M. Reyes, underside ofrock loosely buried in soil (TMMC); 1 female,Road Side Sink No.1, XI-20-1990, J. Reddell,underside of rock loosely buried in soil (TMMC); 1male, Persimon Sink, 1-17-1991, J. Reddell, M.Reyes (DSC); 1 male, Simons Pretty Pit, 1-17-1991,J. Reddell (DSC); 1 female, Snake Pit Sink,11-8-1991, G. Veni (TMMC).Comments.-Batrisodes globosus is foundthroughout eastern North America, with isolatedrecords from Colorado, Washington, and Alberta. ItFigs. 8-17.-8-16, Balrisodes spp.: 8, B. reyesi, dorsal and right lateral view, aedeagus; 9, B. lexanus, head and anterior view,right antenna of male; 10, B. lexanus, lateral view, right antenna of female; II, B. lexanus, posterior view, right metatibia of male; 12,B. lexanus, dorsal view, aedeagus; 13, B. venyivi, head and anterior view, right antenna of male; 14, B. venyivi, dorsal and right lateralview, aedeagus; 15, B. globosus, head and anterior view, right antenna of male; 16, B. globosus, dorsal and right lateral view,aedeagus. Apical three segments of male antennae of Balrisodes spp. presented in ventral'view. 17, Texamaurops reddelli, head andanterior view, right antenna of female. Scale line equals 0.I.248

is associated with rotten woods (Chandler, 1987,and many collection records), and may be foundwith ants in the northern half of its range (Park,1960). Park (1960) noted one population that wasfound in a cave in Alabama.Male globosus may be separated from uncicomisby the more broadly rounded anterior margin of thevertex, and by the form of the apical segments of theantennae. Females lack coarse punctures between thevertexal fovea on the head, and also on the pronotum,while in uncicomis coarse punctures will befound in both areas.8. Texamaurops reddelli Barr and Steeves(Figs. 17-19)Texamaurops reddelli Barr and Steeves, 1963: 118.Type locality: Kretschmarr Cave, Travis Co.,Texas. Holotype female (FMNH).Description.-Length 2.72-3.08. Male with vertexsmoothly merging into elongate frons, areasparsely setate with prominent antennal tubercles,base of vertex with median and lateral carinae distinctbut not prominent, lateral carinae reachingapex of antennal tubercles, median extending topoint anterior to nude vertexal foveae, all antennomereselongate, X with large nude fovea on ventralsurface near base, antennomere swollen in basalhalf, eye area a rounded angulation with approximately6 granules that appear to be vestigial eyefacets.Pronotum with median longitudinal sulcus faint,extending to near pronotal apex; elytra with 2 basalfoveae and subhumeral fovea; metasternum atmiddle with group of long sparse setae; metatibiaelacking apical pencil of setae, sternite VI laterallywith rounded tumuli bearing long setae that aredirected laterally.Females with lateral margins of antennomere Xstraight, lacking ventral fovea, sternite VI broadlyrounded.Male from Stovepipe Cave: antennae 2.12,metafemora 1.56, metatibiae 1.72, metatarsi 0.80.Specimens examined.-6: TEXAS: TravisCounty: Holotype female, Kretschmarr Cave, 15 mi.NW Austin, III-2-1963, J. R. Reddell, D. McKenzie(FMNH); 1 female, same locality, 1968, R. W.Mitchell (FMNH); 1 female, Tooth Cave, V-31964, J. R. Reddell (TMMC); 1 female, samelocality, V-14-1966, J. R. Reddell (FMNH); 1female, Amber Cave, IV-8-1984, J. Reddell, M.Reyes (DSC); 1 male, Stovepipe Cave, X-25-1990,L. Sherrod, under rock lightly buried in silty clay insmal1 side room in total darkness (DSC).Comments.-Texamaurops reddelli has onlybeen found in four closely situated caves on theunderside of rocks lightly buried in silt in totaldarkness (James R. Reddel1, pers. comm.), and hasbeen placed on the US. List of Endangered Species(Chambers and Jahrsdoerfer, 1988). It has elongatelegs and antennae with reduced eyes, and is clearly atroglobite. The holotype and the single male aredifferent from all the other specimens in possessingonly two basal foveae on each elytron. The otherfour specimens, including one from the typelocality, have three equal foveae at the elytral base.AI1 other features appear to be similar and these fourspecimens are placed as reddelli.Barr and Steeves (1963) initially believed thatthis genus might be best placed in the Amauropsini,a European tribe whose members are all troglobites.Barr later (1974) figured the aedeagus of a purportedmale, which clearly indicated the placement of Texamauropsin the Batrisini since the morphology ofthe aedeagus in members of the two tribes is quitedifferent (Jeannel, 1948). The male specimen inquestion has proven to be a new species of Batrisodesconvergent with Texamaurops in appearance,but discovery of a true male of Texamauropsconfirms the placement in the Batrisini since theaedeagus is similar in form to those of members ofBatrisodes. The absence of the metatibial pencil ofsetae is shared by members of BatriasymmodesPark, a group with a number of troglobitic speciesin the eastern United States, and some members ofthis group also have a fovea on antennomere X.However, the aedeagal form of this genus is consistentlydifferent from members of Batriasymmodes,and based on the form of the aedeagus andantennal characters Texamaurops is probably bestconsidered a lineage derived from Batrisodes thathas lost the metatibial pencil of setae.Texamaurops reddelli is superficial1y similar totexanus by the greatly elongate antennae and legs, aswell as body size. However, reddelli possesses adistinct rounded angulate knob where the eyesshould be, and the metatibiae lack an apical pencilof setae.9. Cylindrarctus bicomis Chandler(Fig. 20)Cylindrarctus bicomis Chandler, 1988: 135. Typelocality: Texas (probably Bosque County). Holotypemale (USNM).249

Description.-Length 2.20. Males with antennaltubercles on head close, forming a broad tubercleconstricted at base, vertexal foveae a distance of oneocular facet from eyes, eyes with about eight largefacets, maxillary palpi with last three segments elongate,together nearly as long as antennae, penultimatesegment angulate near base on mesal margin.Pronotum lacking any sulci, with small medianand lateral antebasal foveae. Elytra with two basalfoveae, lacking subhumeral fovea. Metastemumwith short longitudinal carina extending anteriorlyfrom near inner margins of each metacoxa; stemitesII-VI broadly flattened at middle; protrochanterswith prominent spine, metatrochanters with elongatebroad flange extending ventrally.Females are unknown.Male specimen from Gorman Cave, Texas: antenna1.19, metafemur 0.76, metatibia 0.79, metatarsus,0.47.Specimen examined.-Texas: San Saba County,Gorman Cave, 6 mi. SE Bend on Colorado River,III-15-1963, J. R. Reddell (DSC).Comments.-Cylindrarctus bicornis was takenfrom washed-in organic debris in total darkness severalhundred meters from the cave entrance (JamesR. Reddell, pers. carom.). This is the second knownspecimen of this species. The other specimen, theholotype, was produced by Belfrage from "Texas,"where he had collected most extensively at Cliftonand Norse in Bosque County. The other members ofthe genus are most commonly encountered in moistlitter near water, and bicornis will probably only befound in caves at their entrances since it lacks anyobvious cave adapted features. Cylindrarctus bicornisis readily separated from the other cave pselaphidsby the elongate last three segments of themaxillary palpi that together are nearly equal to theantennae in length.10. Tmesiphorus costalis LeConte(Fig. 21)Tmesiphorus costalis LeConte, 1849:77. Type locality:Pennsylvania. Holotype male (MCZC).Description.-Length 3.08-3.52. Males withhead and pronotum coarsely punctate, antennomereIX cylindrical and three times length of VIII, withlongitudinal groove on outer face, X slightly shorterthan IX and flattened ventrally with anterodistal angletuberculate, XI excavate in basal third on ventralsurface; maxillary palpi with apical segment triangular,previous two segments with lateral spine.Pronotal disc rounded; elytra with two basal foveae,lacking subhumeral fovea; abdomen with sternitesII-III lightly flattened at middle.Females with apical antennomeres narrower,lacking modifications; abdominal sternites broadlyrounded.Male specimen from Puberty Pit, Texas: antennae1.68, metafemora 1.04, metatibiae 1.20, metatarsi0.48.Specimens examined.-TEXAS: Burnet County:Simon Says Sink, XI-12-1990, J. Reddell, M.Reyes, berlese litter (2, TMMC); Snake Pit Sink,XI-20-1990, J. Reddell, M. Reyes, underside ofrock loosely buried in silt (2, DSC); SimonsSquirm-Around Cave, XI-20-1990, J. Reddell, M.Reyes, underside of rocks loosely buried in silt (4,DSC); Shin Oak Sink, 1-17-1991, J. Reddell, M.Reyes (4" DSC). San Saba County: Puberty Pit,IV-21-1990, D. Allen, W. Elliott, B. Fralia (1,TMMC). Travis County: Big Oak Cave, X-17-1990,J. Reddell, M. Reyes, on underside of rock indarkness (I, TMMC); Twin Dig Pit, XII-1l-1990,J. Reddell, berlese rodent nests at lowest point ofcave in total darkness (1, TMMC).Comments.-This species has been collected invarious leaf litters, under bark, with the antAphaenogaster fulva Roger (Park, 1933), and onespecimen has been found in a cave (Park, 1960).Tmesiphorus costalis and the other Nearctic speciesin the genus, carinatus (Say), are easily recognizedby the form of the maxillary palpi, and the coarselypunctate head and pronotum. Tmesiphorus costalis islarger than carinatus, the latter being easily distinguishedfrom costalis by a median longitudinalcarina on the first visible abdominal tergite.11. Hamotus (Hamotoides) electrae Park(Figs. 22-23)Hamotus electrae Park, 1942:327. Type locality:San Juan, Veracruz, MEXICO. Holotype male(FMNH).Description.-Length 3.20-3.40. Males with antennomeresIV-VIII transverse, IX twice as long asVIII, IX slightly shorter than X, both IX and Xslightly longer than wide; last segment of maxillarypalpus enlarged, twice as long as wide, with longitudinalsulcus on mesal margin.Pronotum with distinct transverse antebasal sulcus,width about half that of median antebasal fovea;elytra with two basal foveae, lacking subhumeral fovea;protrochanters carinate ventrally, a few specimenswith small denticle on carina, profemora with250

small ventral carina near base, protibiae with smalltooth on mesal margin at point about two-thirds oflength, mesotrochanters vaguely carinate on ventralmargin, metasternum deeply and transversely impressedin apical two-thirds, posterior margin raisedas flanges at mesal margins of metacoxae; abdominalsternites II-VI widely impressed.Females lack modifications of legs and metasternum;abdominal sternites broadly rounded.Male specimen from Porcupine Cave, Texas: antennae1.44, metafemora 1.00, metatibiae 1.08,metatarsi 0.54.Specimens examined.-3: TEXAS: KinneyCounty, Porcupine Cave, X-17-1987, G. Veni, J.Ivy (DSC and TMMC). Five other specimens werecollected at Bentsen-Rio Grande St. Pk., VI-I01975, by Robert Turnbow at ultraviolet light (DSCand RHT).Comments.-This species was previously knownonly from the holotype male collected in Veracruz,Mexico (Park, 1942). It is apparently a free-livingspecies, and will probably only be found when associatedwith caves at their entrances. Park (1942)placed this species in Group XII of Hamotus, wherethe other four North American species are placed.The males of electrae are readily distinguished fromall other Hamotus (Hamotoides) species by thetransverse impression of the metasternum, and themetasternal laminae projecting posteriorly near themesal margins of the metacoxae in males.Trimioiircus musamator n. sp.(Figs. 24-26)Description.-Length 1.16-1.24. Male head withvertex extended and flattened laterally, eyes not~~ Of;;18oClI/, r '22Figs. 18-26.-18-19, Texamaurops reMelli: 18, posterior view, right metatibia of female; 19, dorsal and right lateral view,aedeagus. 20, Cylindrarctus bicomis, dorsal view, male head. 21, Tmesiphorns costalis, dorsal view, male head. 22-23, Hamotuselectrae: 22, dorsal view, male head; 23, dorsal and right lateral view, aedeagus. 24-26, Trimioarcus musamator: 24, dorsal view,male head; 25 posterior view, male right middle leg; 26, dorsal and right lateral view, aedeagus. Scale line equals 0.1.251

visible in dorsal view, with median elongate tuberclepointed at apex and rounded at base, top of tuberclenarrowly rounded, area immediately anterior andlateral to tubercle shallowly impressed and impunctate,minute vertexal foveae lateral to middle oftubercle, anterior margin of vertex broadly rounded,eyes with 13 facets, genal process lateral to maxillarybase bearing three setae, antennal club of apicaltwo antennomeres.Pronotum with biarcuate transverse antebasal sulcus,median antebasal fovea indistinct, distinct lateralantebasal foveae nude, large procoxal foveaemeeting internally; elytra with 2 basal foveae; withsmall lateral mesosternal foveae, otherwise sternalfoveae similar to those in figure by Grigarick andSchuster (1980) for Trimioarcus incisiurus Park;profemora with 7-8 sensory pits on ventral margin,mesotrochanters with broadly rounded angulation onventral margin, mesotibiae gradually expanded topast middle on mesal margin, abruptly constricted toapex, apex with angulate spine; only abdominal tergiteI with basal longitudinal carinae; sternite VIbroadly impressed at middle, penial plate with setaeon apical half.Females with vertex smoothly rounded laterallyand eyes easily visible, eyes with 13 facets, vertexalfoveae nude; legs lacking modifications; disc ofsternite VI broadly rounded.Holotype male: antenna 0.33 long, metafemora0.29 long, metatibia 0.30 long, aedeagus 0.22 long.Specimens examined.-7: TEXAS: TravisCounty: HOLOTYPE male, Twin Dig Pit, XII-l11990, J. Reddell, berlese rodent nests (FMNH).PARATYPES: 3 females, eutopotypical (DSC,TMMC); 1 male, 2 females, Moss Pit, III-5-1991,J. Reddell, M. Reyes, berlese litter from entrancepit bottom (DSC, TMMC).Etymology.- The name was suggested by thecollection from rodent nests.Comments.-This species was found in rodentnests in total darkness at the lowest point in TwinDig Pit, and in litter from the bottom of the entrancepit of Moss Pit (J. R. Reddell, pers. comm.).Trimioarcus Park (1952) was created to hold thenew species, incisiurus Park, from Monterrey,Mexico. A second species, pajarito, was recentlydescribed from southern Arizona (Chandler, 1985).Trimioarcus musamator is closest to incisiurus bythe broadly expanded head vertex with a medianprotuberance, and the eyes not visible in dorsal viewin the males. The two species differ in the form ofthe vertexal tubercle; and in musamator the smalleyes are of equal size with about 13 facets in bothsexes, and the middle legs modified in the male.This species looks very much like a member ofTrimiomelba LeConte, and would be readily placedin this genus except for the presence of lateralmesocoxal foveae and procoxal foveae.The reduction in eye size is probably a responseof this species to its exploitation of rodent nests ordeep leaf litter, rather than a response to the lack oflight in caves. No other members of the tribe Euplectiniare known to be cavemicolous, while anumber of species in a variety of genera in this tribehave reduced eyes and are found in deep leaf litteror tree holes.ACKNOWLEDGMENTSI would like to thank J. R. Reddell (Texas MemorialMuseum, Austin) for suggesting and assistingwith this study, and T.C. Barr, Jr. (College of BiologicalSciences, University of Kentucky, Lexington)for his contribution of many of the specimenscovered in this paper. I would like to also thankthose who arranged the loan of specimens: D.Azuma (Academy of Natural Sciences, Philadelphia),A.F. Newton, Jr. (Division of Insects,Field Museum of Natural History, Chicago), R.H.Turnbow (Fort Rucker, Alabama), and J.A. Wagner(Field Museum of Natural History, Chicago). C.E.Carlton, Department of Entomology, University ofArkansas, and J.F. Burger and R.M. Reeves, bothDepartment of Entomology, University of NewHampshire, reviewed the manuscript.LITERATURE CITEDBarr, T.C., Jr. 1974. The eyeless beetles of the genus ArianopsBrendel (Coleoptera, Pselaphidae). Bull. American Mus.Nat. Hist., 154(1):51 pp.Barr, T.C., Jr., and H.R. Steeves. 1963. Texamaurops, a newgenus of pselaphids from caves in central Texas (Coleoptera:Pselaphidae). Coleopterists Bull., 17:117-120.Brendel, E. 1893. Notes and descriptions of Pselaphidae, withremarks on the Scydmaenidae. Trans. American Entomol.Soc., 20:277-284.Casey, T.L. 1897. Coleopterological notices. VII. Ann. NewYork Acad. Sci., 9:285-684.Casey, T.L. 1908. Remarks on some new Pselaphidae. CanadianEntomol., 40:257- 281.Chambers, S.M., and S. Jahrsdoerfer. 1988. Endangered andthreatened wildlife and plants; final rule to determine fiveTexas cave invertebrates to be endangered species. FederalRegister, 53(180) :36029-36033.Chandler, D.S. 1985. The Euplectini of Arizona (Coleoptera:Pselaphidae). Entomography, 3:107-126.Chandler, D.S. 1988. A revision of the Nearctic genus Cylindrarclus(Coleoptera: Pselaphidae). Trans. American Entomol.Soc., 114:129-146.252

Grigarick, A.A., and R.O. Schuster. 1980. Discrimination of thegenera of Euplectini of North and Central America(Coleoptera: Pselaphidae). Univ. California Publ. Entomol.,87:vi + 56 pp., 79 pis.Jeannel, R. 1948. Revision des Amaurops et genres voisins(pselaphidae). Rev. Fr. Ent., 15:1-19.LeConte, J.L. 1849. On the Pselaphidae of the United States.Boston J. Nat. Hist., 6:64-110.Park, O. 1933. The food and habits of Tmesiphoms coszalis Lec.(Co1eop.: Pselaphidae). Entomol. News, 44:149-151.Park, O. 1942. A study in Neotropical Pselaphidae.Northwestern Univ. Stud. BioI. Sci. Med., Number I.Evanston and Chicago: Northwestern University. x + 403pp.Park, O. 1947. Observations on Barrisodes (Coleoptera: Pse-laphidae), with particular reference to the American specieseast of the Rocky Mountains. Bull. Chicago Acad. Sci.,8(3):45-132, II pis.Park, O. 1951. Cavernicolous pselaphid beetles of Alabama andTennessee, with observations on the taxonomy of the family.Geol. Surv. Alabama, Mus. Papers, 31, 107 pp.Park, O. 1952. A revisionary study of Neotropical pselaphidbeetles. Part Two. Tribe Euplectini sensu lanore. ChicagoAcad. Sci., Spec. Publ. No. 9(1):1- 49.Park, O. 1956. New or little known species of pselaphid beetlesfrom southeastern United States. J. Tennessee Acad. Sci.,31 :54-100.Park, O. 1960. Cavernicolous pselaphid beetles of the UnitedStates. American MidI. Nat., 64:66-104.This is Publication No. N.S.-61 of the Texas Memorial Museum.253

INDEX OF NEW TAXACRUSTACEAAMPHIPODAArtesiidaeArtesia welbourni Holsinger 2HadziidaeMexiweckelia hardeni Holsinger 8Holsingerius smaragdinus Holsinger 12SebiidaeSeborgia (Relictoseborgia) hershleri Holsinger 16ISOPODACirolanidaeSpeocirolana hardeni Bowman 25ARACHNIDASCHIZOMIDAProtoschizomidaeProtoschizomus rowlandi Cokendolpher and Reddell 51Protoschizomus sprousei Cokendolpher and Reddell 55Protoschizomus purificacion Cokendolpher and Reddell 57Protoschizomus gertschi Cokendolpher and Reddell. 59Protoschizomus treaceyae Cokendolpher and Reddell 60Agastoschizomus stygius Cokendolpher and Reddell 67Agastoschizomus patei Cokendolpher and Reddell 69ARANEAEDictynidaeCiucurina (Cicurella) arkansa Gertsch 84Cicurina (Cicurella) harrietae Gertsch 88Cicurina (Cicurella) mirifica Gertsch 88Cicurina (Cicurella) secreta Gertsch 88Cicurina (Cicurella) marmorea Gertsch 90Cicurina (Cicurella) modesta Gertsch 90Cicurina (Cicurella) oklahoma Gertsch 92Cicurina (Cicurella) gatita Gertsch 92Cicurina (Cicurella) aenigma Gertsch 94Cicurina (Cicurella) dorothea Gertsch 94Cicurina (Cicurella) rosae Gertsch 94Cicurina (Cicurella) blanco Gertsch 95Cicurina (Cicurella) annadillo Gertsch 95Cicurina (Cicurella) sinronia Gertsch 95Cicurina (Cicurella) delrio Gertsch 96Cicurina (Cicurella) joya Gertsch 96Cicurina (Cicurella) browni Gertsch 98Cicurina (Cicurella) vibora Gertsch 98255

Cicurifla (Cicurella) holsiflgeri Gertsch 98Cicurina (Cicurella) menardia Gertsch 98Cicurina (Cicurella) ezelli Gertsch 99Cicurina (Cicurella) travisae Gertsch 101Cicurifla (Cicurella) wartofli Gertsch 101Cicurifla (Cicurella) elliotti Gertsch 101Cicurifla (Cicurella) coryelli Gertsch 103Cicurifla (Cicurella) uvalde Gertsch 103Cicurifla (Cicurella) watersi Gertsch 103Cicurifla (Cicurella) pablo Gertsch 105Cicurina (Cicurella) orellia Gertsch 105Cicurina (Cicurella) serena Gertsch 105Cicurina (Cicurella) selecta Gertsch 105Cicurina (Cicurella) reddelli Gertsch 105Cicurina (Cicurella) balldida Gertsch 107Cicurilla (Cicurella) cueva Gertsch 107Cicurifla (Cicurella) russelli Gertsch 107Cicurifla (Cicurella) reyesi Gertsch 107Cicurifla (Cicurella) ubicki Gertsch 109Cicurifla (Cicurella) baronia Gertsch 109Cicurina (Cicurella) madia Gertsch 109Cicurina (Cicurella) vespera GertschIIICicurina (Cicurella) vellii Gertsch 111Cicurina (Cicurella) recIusa GertschIIICicurifla (Cicurella) puenrecilla Gertsch 111Cicurifla (Cicurella) bandera GertschIIICicurilla (Cicurella) obscura Gertsch 113Cicurina (Cicurella) patei Gertsch 113Cicurina (Cicurella) sprousei Gertsch 113Cicurina (Cicurella) stowersi Gertsch 113Cicurifla (Cicurella) pastura Gertsch 114Cicurina (Cicurella) machete Gertsch 114Cicurina (Cicurella) sansaba Gertsch 114Cicurina (Cicurella) venefica Gertsch 114Cicurina (Cicurella) caverna Gertsch 115Cicurina (Cicurella) porteri Gertsch 115Cicurina (Cicurella) sheari Gertsch 115Cicurina (Cicurella) suttoni Gertsch 115Cicurina (Cicurella) mckenziei Gertsch 115Cicurilla (Cicurella) barri Gertsch 117Cicurilla (Cicurella) rainesi Gertsch 117Cicurifla (Cicurella) gruta Gertsch 117Cicurina (Cicurella) medina Gertsch 117Cicurilla (Cicurella) wiltoni Gertsch 119Cicurina (Cicurella) leona Gertsch 120TheridiidaeThymoites minero Roth 124PSEUDOSCORPIONIDAChtoniidaeTyra1l1lOchthonius texallus Muchmo:e 129256

NeobisiidaeTartarocreagris comanche Muchmore 133Tartarocreagris intermedia Muchmore 152SyarinidaeChitrella welbourni Muchmore 134Chitrella major Muchmore 135Chitrella elliotti Muchmore 137CheiridiidaeCheiridium reyesi Muchmore 138Apocheiridium reddelli Muchmore 139ChemetidaeNeoallochernes cubanus Muchmore 141Neoallochernes (?) incertus Muchmore 144Dinocheirus cavicolus Muchmore 149OPILIONIDAPhalangodidaeTexella kokoweeJUbick and Briggs ; 179Texella shoshone Ubick and Briggs 183Texella deserticola Ubick and Briggs................................................................................... 183Texella brevistyla Ubick and Briggs 187Texellajungi Ubick and Briggs 191Texella longistyla Ubick and Briggs 195Texella welbourni Ubick and Briggs 195Texella hardeni Ubick and Briggs 198Texella cokendolpheri Ubick and Briggs 198Texella bilobata Ubick and Briggs 207Texella reyesi Ubick and Briggs 211Texella brevidenta Ubick and Briggs 221Texella grubbsi Ubick and Briggs 225Texella diplospina Ubick and Briggs 225Texella renkesae Ubick and Briggs 228Texella spinoperca Ubick and Briggs 232TexellaJendi Ubick and Briggs 232Texella homi Ubick and Briggs 235INSECTACOLEOPTERAPselaphidaeBatrisodes (Excavodes) grubbsi Chandler 245Batrisodes (Excavodes) reyesi Chandler 246Batrisodes (Excavodes) texanus Chandler 247Batrisodes (Excavodes) venyivi Chandler 247Trimioarcus musamator Chandler 251257

searchable PDF - Association for Mexican Cave Studies

Create successful ePaper yourself

Delete template?

Save as template?